📛 Become a channel member: th-cam.com/users/thomaseislphotographyjoin ☕ Donate a coffee to support this channel: ko-fi.com/thomaseislphotography ❓📩 Direct support: thomaseisl.photography/shop/p/support-ticket
I have 5D classic and it is not that bad, it's slightly noisier than for example my Canon M50. I have shot a few weddings with it and it's a perfectly good camera. Nobody will ever notice the difference. Even the high ISO performance is quite good, i can push easily 2 stops images shot at 1600, so basically it can go up to 6400, and it has similar noise like the M50.
@@dtibor5903Yeah, old cameras just didn't let you turn it up as much, the maximum iso noise on my 5D is less than my 6D, which has better lowlight overall. They let you turn it up a lot, but it doesn't change the raw, just changes what your metering is, or how many stops above and below metered you get. Arri or such shoot normally at 3200 iso and they do that because with RAW video you only get so many stopd above and below, and faster gives you a more normal range compared to film, with more stops overexporure latitude than under.
You videos are so informative Thomas - I can't stop watching them and have learnt so much. Thank you. What I'm starting to realise is that a lot of 'it' is talk. When your friend in the pub says his camera is better as it has more megapixels, a bigger sensor and more dynamic range. They've probably spent a load more money yet don't actually know how to utilise the 'better bits' (if indeed they are even 'better'). I think the M43 sensor cameras are so overlooked because a lot of people rule them out based on having 'a tiny sensor' and 'only 20MP', but from my very unscientific testing, the OM-1.2 I have on my desk alongside my Fuji X-H2 is making a really good case for showing Fuji the door. The images are equally as good, the camera has more features I'll actually use, and the system as a whole is considerably lighter and more compact. I can't think of what I don't like currently... Thank you again and I look forward to your next videos.
Thank you very much, that is just great to read. Regarding your assessments - well, I completely agree. I'm planning to expand my series on some "photo misconceptions" in the future. Thank you very much for your support! Best, Thomas 📸
Since 2016, 90 % of my photography has been done with m4/3 bodies. I never missed on dynamic range. Learning to expose properly and shoot raw is key IMO. Exposure compensation is your best friend. Even the best metering system can use help from time to time. And let's not forget that the human eye doesn't have an unlimited DR either. In some extremely high contrast situations, having a blown highlights and/or blocked shadows looks more realistic.
Absolutely, I completely agree. I think that those super-muted ultra wide dynamic range images are very rarely a desirable. Many thanks for your great contributions on many videos!
I definitely agree with everything you said. Another thing to remember is that it appears manufacturers have stopped working to improve sensors. There's been no real improvement for the last 5 to 6 years (basically with the release of the D850 and Sony A7RIII). All improvements now seem to be concerned with AF, FPS, and ways to compete with dedicated video cameras. If image quality is your main concern there haven't been any improvements since that generation of sensors.
You have made a very valid point here! Things did not change much in terms of image quality, but the higher readout speeds - which are crucial for better AF, FPS, and EVF performance - were the major advancement. I have to admit that I am a huge fan of the OM-1's sensor, as I have seen IQ improvement, but the camera uses a fundamentally different sensor than the previous Olympus models. Be that as it may, the main improvements are in the AF, FPS and EVF department. As you've said - "old" cameras (also DSLRs!) are still relevant, and will be as they are just good enough.
Most of the research and development in sensors is for smartphones and then it drips into cameras. With a shrinking market no one wants to invest money in it.
The plateau of improvements in sensor tech is due to the limitations of current CMOS sensor manufacturing. We are seeing faster readouts, frame blending (mostly cellphones), dual gain circuit outputs etc… We won’t see leaps on DR until new sensor technology is developed. Panasonic has organic sensor in development but it’s a long way off and has issues.
Another fascinating video Thomas. This topic is definitely something that comes up regularly. In fact, I had just finished watching a vlog from another TH-camr who stated "of course M4/3 has lower dynamic range than full frame". It was so coincidental that yours was the next video I watched. In any case, I'm very happy with my OM-1 for the whole package of features it offers. Thanks for another great presentation Thomas!
Thank you very much for your kind words and appreciation! I myself am also very satisfied with the OM-1 and its performance, just shot another magazine editorial with it and the quality of the camera is just great, in every way. I'll soon follow up with new videos on the OM-1 and its functions! Thanks again, Stephen!
For me, photography is an "escape route". when everyday life takes over and adult life settles down like a blanket. In this bubble I want to be, just be. Enjoy the nature, listen to the stream flowing and the birds chirping. I don't want to be technical. Blown highlights. Crashed shadows. Soft corners. I do not care. But that said, I'm no professional. No weight on the shoulders to "deliver". By the way, did I say that I appreciate your technical videos? I do. Vielen Dank
Thank you for your excellent, eloquent statement - exactly these points are why I just like cameras that resonate with me. As soon as I get to post process my images, I'm crushing shadows and blowing highlights, because it just looks awesome, right! So who really cares if you have 25 stops of DR recorded, no client ever asked me about that or complained because they were not "there"
Another excellent and entertaining video, Thomas. Your depth of understanding and ability to explain it to others is without equal, and so appreciated.
Excellent presentation; data-based and practical! The only ones who might not like this are the manufacturers' marketing teams where it's primarily a numbers game: more is better...
Well, if you look this from manufacturers and marketing teams perspective, only value they can present is the theoretical maximum DR value, since there is no standardized method to measure "photographic DR" that could be compared between different brands and manufacturers. So they choose to publish value that they have measured in their own lab. We have to look for third party labs, such as photons photos old DxOlab, that measure and compare sensors.
I love your videos. the thorough investigative research that you do, video set up, sound and lighting quality and the excellent dialog delivery that goes into them. It's like watching inspector gadget teaching these facts that not very many people know. I've been binge watching many of your videos and learning a lot. I so much appreciate the effort you've put into them. I don't subscribe to many YT channels, but I have subscribed to you and will watch more and some of them over a few times. Great learning material. Thank you. The shutter simulation between scenes is a good touch.
Many thanks for your exceptionally kind comment, Richard. Feedback like yours means a lot to me and I very much appreciate that you took the time to share that with me. I'm very thankful to welcome you as a subscriber - I hope future content will be of equal interest to you! Best wishes, Thomas
Thank you Thomas, a very enlightening video! I am an amateur photographer. Years ago, I started shooting slide film, and just got used to the 'small' dynamic range of that film compared to negative film. As a new photographer, I longed for a film that had the same or close to the dynamic range of the human eye. Then I could 'catch everything'! As I gained experience, I learned that having limited dynamic range is not a bad thing, one can do a lot creatively with it, capturing sun beams through a windows in a dark room for example. When I switched to digital cameras, I felt I was on familiar ground, and just started using many of the same techniques as I used with slide film. I've never had a complaint or criticism of 'low' dynamic range in my photos. I just try to get the best exposure for the mood, scene, feeling, whatever of the image I see in my head, and I'm happy with it. It's actually fun to see what I can do with the range I have. What really needs to change is the dynamic range of the monitors we use to view our digital photos with; but then, that might be another creative avenue to work with. I'm Subscribed. Looking forward to exploring other videos of yours. thanks again.
Thank you very much for sharing your experiences - I've read everything carefully, I am totally with you. The "limit" or working with the limitations are actually ok! Thank you very much, also for subscribing! Hope to hear from you soon!
@@ThomasEisl.Photography Yeah for real, unless you're boosting shadows by 5 stops in all you images, you don't need that much. You can't put more dynamic range onto a 6 stops screen or print or it looks super flat and boring. Slide film was a mere 5-6 stops usually and some may have even been under 5, like Velvia. And yet everyone raves about the colour and contrast of something that the Nikon D1 beats for dynamic range.
Well said, sir! Back in the 70s and the first half of the 80s, I worked as a photojournalist. My biggest concerns were how far I might need to push my Ilford HP5 or Kodak Tri-X 400; dynamic range was more about how dynamic my images might be than about how many steps of grey I could stuff between black and white. Even when (on vacations) I put on my Stetson and emulated Ansel Adams, I realized that I needed visualization, image capture, and final image creation (in the darkroom) to achieve my goal. I am very well served today by a pair of Olympus OM-D EM5 Mk IIs - a newer unit might have slightly better resolution and a few better features, but the wonderful ease of use and lens selection keeps me in the M43 camp.The key advances of my particular camera bodies include the 40MP mode that uses pixel-shifting, and numerous choices for bracketing of focus, exposure, etc.
Well for my bird photography hobby where I prefer to shoot in very low light and/or strongly backlit situations 2 stops improvement (which is 4 times more light) is very helpful. When I switched from a Canon 7Dii to a 5Diii and 1DX the improvement was mind blowing. Also the high iso performance of these full frame sensors seems freakish to me after years of using only APS-c. Bird photography is an extreme form of photography where high speed action, low light and tiny erratic subjects mean that having high performance cameras and lenses are very helpful. I also avoid shooting wide open as it's likely to miss focus more on fast moving tiny birds so when I use my 400 f/2.8 I generally stop down to f/4 or preferably f/5.6 or even f/8 and keep the shutter speed to at least 1/400 second or preferably higher.
Thanks for saying this. I’m constantly trying to explain to my full frame friends how terrible the dynamic range is on my M50. I wish someone had explained this to me before I bought my camera.
@@eafortson you don't really appreciate the advantage of the bigger sensor until you try it out. I recommend trying a full frame dslr like the 5Diii which are very affordable and great cameras
"High ISO performance of these full frame sensors seems freakish" Ikr! I just bought my first full frame camera recently, it's crazy to me to shoot at ISO's like 1600 and 6400, no problem. And with a flash, I'm literally shooting at minimum flash power 😂😅😭
1:20 “It’s about time for a serious reality check about photography silicon” *sweats profusely in excitement* *nerdy pathways activated, serotonin dispensed*
To be quite frank this video left me a bit confused, Thomas. At first I learnt that dynamic range is a bit like image resolution: As long as you've got enough for what you want to do everything is ok. And that dynamic range above 8 EVs isn't relevant and even a bit less isn't a problem all by itself. That sounds a lot like "the max 9.7 EVs of the sensor in my E-M1X is more than enough" just like "the 20 MP of the sensor of my E-M1X is more than enough for taking the kind of pictures I'm taking, with very humble cropping and not deliberately higher than ISO 6400 sensitivity setting" (which in a nutshell really is my point of view). And it also fits my experience for rating picture quality on my E-M1X on different ISO settings: 200/excellent/9.7EVs 1600/very good/6.9EVs 3200/good/6.1 6400/still-satisfying/5.2EVs >6400/try-to-avoid-if-possible/
What a wonderfully thorough and informative video! I used to shoot weddings on Canon 1Ds MkII - about 8.5 stops similar to the 5D. Several times I could rescue highlights and shadows to make perfect exposures from RAW even at that level. I think the dynamic range argument says more about the ability of the photographer to expose correctly than it does about what camera format you need.
Agreed - having more DR is great, and I also have to rescue photos from time to time. But it is exactly as you've said - it is us photographers screwing up, not the limitation of a given camera system. Many thanks for the comment!
Spot on! Regardless of the slight image quality increase, the large-sensor camera buys me nothing if I’ve left it home because it’s no fun to carry. I’m seriously considering a switch from full-frame to the OM1, and your insights are slowly helping me to make that decision. The best camera in the world is the one you have with you when you need it.
Completely agree. Thank you very much. I'd say you would be happy. I've still got everything from digital medium format to 43, and I'm using the latter practically all the time. Best, Thomas
At 09:45 you talk about a 32 to 1 lighting ratio. You then say you need 9 stops of dynamic range. Looks more like 5 to me: 2^5 = 32 (1,2,4,8,16,32) Or do I not fully understand it?
Yes, you are correct - the reason why I stated 9 stop is: you need at least 4 stops for recording every nuance of a flat, non-reflective subject. The contrast ratio as you have stated adds the need for additional 5 stops - this means we now need at least 9. This is where the 9 comes from. Also see this video were I state this more clearly: th-cam.com/video/uYOr6t8llgc/w-d-xo.html
@@ThomasEisl.Photography Correct. I did not catch that you were talking about light... I watch the video you recommended. Very nice and clear explanation. I posted a related comment under that video, about how the histogram is related to the Picture Style when the camera is set to RAW only.
Hello Thomas, 32:1 is 5 stops and not 9 stops. 2^5=32. And are you sure your most dynamic range on set was only 32:1? With my old Canon APS-C 500D camera: You couldn't even look at ISO 1600 at 15.1 Megapixel. With my new Sony A1 at 51 Megapixel photographing with ISO 3200, 6400 or 12800 or more is even better. Of course ISO 100 would be better, but you don't always have so much light. So I don't agree that a new camera doesn't bring much: There is a lot of difference in image quality, noise and dynamic range with newer sensors and cameras. Also keep in mind, that readout speed with newer stacked sensors (e.g. Sony A1 has 4ms) and is around at the time a mechanical shutter has. So there is in typical situations no rolling shutter effect at all with electronic shutter. I guess in 2003 we had 50ms+ on sensor readout speeds. From a practical standpoint I couldn't should a lot of picture with the old Canon body handheld.
Hello Gerhard, thanks for contributing, I'll gladly address your much appreciated inputs! Regarding 32:1 being 5 stops A non-reflective subject requires about 4 stops of dynamic range. Now, increasing the lighting in one stop increments on one side of the subject until we arrive at +5 stops (32:1 as you have stated correctly): In order to reproduce the full tonal scale on both, unevenly lit sides of the subject, we now need 4+5 stops of usable dynamic range. Regarding Old Cameras / New Sensors Especially when shooting high ISO, low dynamic range can result in bad results like with your 500D. Let's say at ISO 1600 only 4 stops of high fidelity DR remain, then you have to expose perfectly to capture the full tonal scale of the main (4 stop DR) subject. Every small error will lead to a significant, noticeable loss in quality. That is why I said in the video that these older models need a flash and are sometimes tedious to work. Not an issue with your A1, obviously! Does the A1 have more dynamic range than some older DSLRs? No, but does that matter: No! Because, as you have stated, the main advantage newer sensors for mirrorless cameras have is not significantly better DR, but faster readout speeds, something that was not so relevant for DSLRs (not talking about something like the Olympus E-10P, of course, which is limited by the readout speed although an SLT. I'd like to conclude that we are exactly on the same page here! Fun fact: I've been typing this response twice, as the browser crashed. This also happened when I was replying to your previous comment on my previous video (th-cam.com/video/uYOr6t8llgc/w-d-xo.html&si=EnSIkaIECMiOmarE) , seems to be a thing haha Best wishes!
Nice overview that puts things into perspective. It's a most convenient truth for those of us on a tight budget who want a great performing digital with excellent traditional photographic functionally and don't need to shoot a gazillion frames per second, a focus point for every other pixel on the sensor, or any heavy in-camera processing / AI hocus-pocus nonsense. For hobbies like astrophotography, there is not much between a 10+ y.o. DSLR and a modern mirrorless costing 10-30x more. Cameras like the Canon 6D and Nikon D600 costing less than a low-end smartphone are already (pardon the pun) stellar performers and I see no need to go to huge expense buying a new or recent camera. Even for more general use, for the average hobbyist or even serious enthusiast, there are many great second hand options out there for a fraction of new camera prices that do the business where it counts. I appreciate though that pro photographers will need the latest features & functionality to keep up with the competition, but more importantly to keep up appearances in front of clients.
Thank you for sharing that! I have to admit that I am fully comfortable with showing up with a MFT or old DSLR to a shoot, no one ever asked any questions. I think it is mostly in our heads.
Thank you for a good video! Informative in many ways. However author of this video never described what stop means in cameras. As far as I know one stop in exposure or DR (dynamic range) means doubling of light, so if DR is increased from 10 to 11, DR has not increased by 10%, but by 100%! Correct me if I am wrong. These DR values are compared in this video as if they are linear values, which is not the case. If you compare presented DR’s of om-1 and Nikon d800, it shows slightly over 2 stop advantage to Nikon. Full frame sensor is four times larger in surface area than MFT sensor, so it will gather four times more light by frame, which means two stops. So these findings are exactly what we should expect! Nothing surprising here. However is this significant in normal photography such as portrait or landscape, is completely different thing. This was addressed in this video very well! As a nature and bird photographer myself (who shoots with om-1 and A1), I think there is difference in DR and noise when shooting moving subjects in less than optimal conditions. It’s matter of preference whether you consider it to be significant enough. Happy shooting!
Hey Heikki! Regarding doubling the light: you are correct! However, it is way less significant than it sounds. If you have not watched already, I recommend watching this video th-cam.com/video/uYOr6t8llgc/w-d-xo.html I think it illustrates why 8 stops is basically enough and why one stop more or less is hardly ever noticeable in practice. Thank you for sharing your very valid remarks, sorry for the late reply!
@@ThomasEisl.Photography I agree with you regarding 8 stops being more than enough. But this applies only if shooting in native ISO or slightly faster. It's funny since I have always been reluctant to push ISO values above 800-1600 in OM-1. When looking at charts by photonstophotos DR drops below 8 stops at those ISO values. On FF cameras I usually push ISO to 5000 range, but avoid pushing further. Stopping motion in less than ideal conditions is when bigger glass and sensor helps (even 1 stop can make a difference). Thats why I never gave up on my FF camera (I was very close selling). OM-1 is brilliant camera, could have better video capabilities though. I have watched many of your videos, don't agree on everything, but still very good and informative content! Keep up the good work!
We need to distinguish two kinds of Dynamic Range (DR). (1) The operating range of a camera (cf. temperature operating range) that is generally referenced as DR, and (2) a camera's contrast envelope that is the DR available to a single shot. These two can be far apart. When you say what you need in 1 shot, this is the contrast envelope version of DR. Not sure where you got the 4 stops from for "an evenly lit subject", but in the film days in our discipline of densitometry (and sensitometry) we used a 10-log base for i*t and most film would give a contrast envelope in the range of 10-log=4, some went to 4.5 or 5. "Stops" are 2-log and compared to 10-log "4" is peanuts - abysmal. In studio photography under controlled light, we can keep contrast relatively low, but elsewhere? Note that LV 10 already is darkish in available light photography - and light level dictates contrast.
I agree 100% with your thoughtful assessment. In 2019 I made a TH-cam video that said that any decent prosumer/professional camera from the last 10 years could accomplish most professional photography needs, and those cameras still hold up today. My first DSLR was purchased in 2003 and was the Canon Digital Rebel (Canon 300D). In some normal situations, its dynamic range seemed lacking. However, that was not the case with my 2009 Nikon D90, which still takes great photos. I now use a Canon 5D IV and a Canon 5D III for professional work, which serves my needs well. Newer cameras have advanced features like better tracking, possibly quicker focusing, and better video capabilities. However, such things are optional in the work that I do. Thanks again for this very good video.
Hey Mike! Thank you very much for the comment, your kind words and the sharing of your experiences! It is great to read that we are on the same page here - I'll also check out your video ASAP. The real issue is that the 5D-series cameras you are using are actually so good that they can get every (!) job done. Bad times for manufacturers trying to sell you new stuff. I mean, even the focus systems on these cameras are top notch and absolutely future proof.
It’s all about using the right tool. I use for indoor and some landscape a Sony A1. For wildlife and outdoors I shoot a OM-1. There is no way I a hauling around or shooting effectively with FF and lens. The lens is massive and heavy vs the OM-1 and my 300 f4 MFT.
I shoot a lot of HDR landscapes, and I have something perhaps curious to add. Sometimes I do not want all of the dynamic range and enjoy the fact that I can easily create black shadows in sunsets and sunrises. In fact, I have often been disappointed with HDR because it gets rid of mystery. I like that I have a choice.
Absolutely, you are completely right! In the following video, I illustrate exactly what you have observed - having less is not necessarily bad, as this is the way human perception works. th-cam.com/video/uYOr6t8llgc/w-d-xo.html&si=EnSIkaIECMiOmarE Thanks for your contribution!
??? I am confused here, why don't you use HDR just when you need it. Beside, if you keep all the shots you can definitely go back to the one you need. Tools are as good as we know how to use them, don't you think?
@@BrunoChalifour that is exactly what I do. My point is that if you make a sensor that inherently incorporates a huge dynamic range, equivalent to today's HDR images, sometimes, even often, that is not desireable. My point is that I don't care about buying a camera with the extra dynamic range, especially if it is a hindrance. Although, to be fair, manufacturers can probably create a limited dynamic range mode.
@@craignichols21 Well unfortunately we are not there yet and the dynamic range of today's sensors is still limited compared to our eyes/brain's potential adaptation to light. HDR has a look that unless mastered does appear artificial. A lot of people have a tendency to overuse it I the same way some did with sharpening when it came out. Using unfortunately does not mean mastering.
@@craignichols21 I guess it wouldnt be hard to add an SDR mode. Its all digital anyway. But I dont see a scenario where I personally would want less info in my raw files.
I just got my OM-1 Mark II and I can say, it’s an amazing camera, it’s only 20MP but you can’t even see the difference when you compare the quality with full frame cameras.
Very good overview and myth busting. I would add some slight caveats around the need for higher shutter speeds in landscape photography: 1. Mountains and buildings do not move. Clouds, trees and water do move. Freezing their movement into an exposure may require a single exposure with a high enough shutter speed, not allowing bracketing. 2. Higher shutter speeds may also be needed when there isn't necessarily subject movement, but the camera itself may move (even when on a tripod). Vibrations from traffic, people walking by your tripod, of a very heavy camera/lens setup on a tripod fully extended may require a fast enough shutter speed after all for sharp images. Everything you said still applies and is very sound advice.
Many thanks - I completely agree with the two caveats you mentioned! It was a bit of a generalisation on my part to simplify the argument. Many thanks again, much appreciated!
Professional indeed ;) Further to your experiences with scene dynamic range, I'm coming into photography from a background in graphic design, and I feel pretty clued up on brief specification - be it for magazines, bus stop advertisements, websites or html5 banner ads etc. And with that, the best printers money can buy, will in ideal conditions reproduce
Hehe 😉 Thanks for sharing your professional experiences as well - it is really astounding to read very similar statements in the comments repeatedly. Proves the point - DR in photography is only an issue if you are looking at spec sheets only...
Hello Tomas I had the om1 yesterday and done a shot of a venue, and I tried the high res mod. All of the photos came out washed, the color was terrible. What did I do wrong? I also took my Nikon z8 and I did get some great phot for my client.
Only started watching the video, interesting what you will say, without knowing it - I came from an APS-C mirorless camera to the mirorless Full Frame world and that was my best decision made so far! How much it speed up my photography game (shooting and processing pictures - retouch) I cannot explain.
@@laiebi_3639 Stay. There are probably scenarios where Aps-C be better, wildlife, sport, but I shoot portraiture and with FF my workflow got much faster and the quality of pictures got better. I was using a Nikon Z50 and now I am using a Lumix S5. Also the videos and grading is day and night.
@@iammz81 Cool, looking forward to shoot cars and wildlife and some occasional family gatherings so its a bit of both worlds. For traveling sth small and light would be cool too but must be worth the quality, otherwise I'd never really use it
Great to hear! Personally, I am not sure why you are seeing such a difference, but as long as it works for you that is just awesome! Maybe you really prefer the large sensor look or you are using better lenses than you did with your apsc cameras. In any case, also consider medium format if you really like the big sensor look and feel!
The A7 III has a photographic dynamic range of 11.6 stops. The 15 stops are the engineering dynamic range, which is not the actual "usable" DR of the camera. th-cam.com/video/uYOr6t8llgc/w-d-xo.html&si=EnSIkaIECMiOmarE
Dynamic range is more useful for general use cameras than studio or landscape photography, where filters, lighting and multiple exposures can be employed. It's also important to remember that most stops of dynamic range are for shadow recovery, not highlights. I had a Canon 5D until 2 years ago, and while the colours were good and 12mp resolution was not a problem, it was difficult to recover detail in any backlit scene.
The 5D Mark II has awful shadow banding and crosshatching, especially in underexposed images. Attempting to push the shadows makes this very apparent. Today's sensors have clean shadows which can be pushed two stops without major issue.
@@raksh9 The 5D had highlight banding, which was its most annoying feature. By comparison my 24mp Lumix S5 has extraordinary dynamic range - perhaps due to its modest megapixel count - and is the only camera I have owned where ISO can be effectively ignored. However, the Canon 5D is an 18 year old digital camera, which is light years technologically speaking.
@@xcx8646 you can't generalize like that. If you want to preserve blue skies you have to underexpose, so the midtones will be very dark and if your camera lacking dynamic range you will end up with noisy images trying to push the shadows. Dynamic range is quite important in nature photography unless you like to jump through hoops with multiple exposures.
When moving from a D750 to a Z7, I found, at first, I kept bracketing. Then, after some time, it became clear to me that I didn’t need bracketing because, post-processing tools and better dynamic range together got the job done with 1 image. Less work, better outcome. And then … there’s one of the best reasons to go to a digital camera … size and weight of the camera and lenses!
Absolutely - better DR can be useful indeed, no doubt about it! But as you've stated, the most important aspect is that you bring the camera with you, and size and weight do matter a lot when it comes to this!
As you juggle with numbers and say that 32:1 needs 9 stops DR, what do you mean with 32:1? If you use aperture stops then 32:1 is 10 stops, if this is about light levels in Lumen/lux or so, then 32:1 is 5 stops.
@jpdj2715 9 stops is referring to 4 stops for the subject plus 5 stops for the added contrast. Please refer to the linked page for the full calculation. Best, Thomas
My experience about dynamic range shooting pro interiors during many years already : Canon Canon 5d , 5Dm2 5dm3 : just enough , but no so impressive , in today standards. Canon R5 : not enough for the price . Pentax 645z : Super Amazing ..! Fuji gfx 50 R, 50 S and 50sII : Amazing ..! Beyond this , the quality of color rendition , tonalities of medium format is something appreciable . Schöne Grüsse .
Hi Thomas, I've just recently discovered your channel. Really great to see someone approach camera systems from a technical, balanced, and realistic point of view. I wanted to share a technique I've recently started to really lean on for HDR situations with my GH5. The couple of times I've tried shooting HDR brackets, I was never really satisfied with the end result, since I would get halos and an "over processed look". That's when I realized that Bill Claff's (owner of photons to photos) photographic dynamic range concept was based on the idea of scaling all images to a common 8x10 print output size, and that I could gain dynamic range scaling my images down further. So now, I use a longer focal length and make a panorama that's larger than my intended output, and then scale the whole stitched image down to the intended output size. Since so often those HDR shots are landscape, for me this kills two birds with one stone: gives me both more photographic dynamic range and more fine detail. Plus, I find that I get fewer stitching artifacts than with HDR stitching, and I don't ever get the "HDR" look. Theoretically, software noise reduction should work better if you give it more detail to "bite" onto, but I haven't tested this specifically. I'm not entirely sure how the PDR will scale with the number of stops, my guess is each doubling of total pixels should be another stop of PDR.
Many thanks for the exceptionally kind comment. I'm sorry for the late reply, I somehow missed it! Regarding your tip, I think the photonstophotos noise chart is what you mean? DR should not change when you downscale the photo. In any case, I'm working on a video on HDR photography. How to set it up in practice and how to get natural results in post processing. Just like you, I'm not a fan of the overly processed HDR look. Again, I very much appreciate the feedback and welcome to the channel! Hope to read from you in the future!
@@ThomasEisl.Photography I don't think I worded my initial comment very well. Shrinking the output size should increase the Photographic Dynamic range, because it's a ratio of maximum highlight (not impacted by downscaling) to minimum "noise-free" shadows (noise suppressed by downscaling, so the minimum useful signal is lowered). I don't think engineering dynamic range would change though, just photographic dynamic range. If I am understanding correctly, the PDR chart on Photons to Photos is based on the ratio between the highlight clipping point and the shadow point of having a 20:1 signal to noise ratio, after being downscaled to a resolution appropriate for an 8x10 print (and if I'm not, that means I get to learn something new today!) I think that means starting with a panorama instead of a single exposure means either a bigger output for the same PDR, or more downscaling and more PDR for the same output size. Also, the GH5 doesn't have any high-resolution pixel-shift mode; if it did I would be using it and taking far fewer panoramas. But I'm considering a switch to the OM-1, and trying to learn more about it and it's built-in computational photography features is how I came across your channel.
Is a one stop improvement in DR a doubling of dynamic range as a one stop increase in aperture is a doubling of the exposure? Or am I reading it wrong?
is a one stop improvement noticeable, at correct exposure its half a stop at light and dark ends? now a one stop improvement at each end is worth the effort, you can see that in sunset shots for example
Well, you can detect one stop more light. And one stop more luminance means that there is something in the frame that is twice as bright as something else. However, human perception does not work linearly so this is usually not a big deal. Maybe this video helps to clear things up: th-cam.com/video/uYOr6t8llgc/w-d-xo.html&si=EnSIkaIECMiOmarE
Thank you for a very nice video. Some other factors to consider, which are also noted by many others might include: The number of cameras, sampled to derive the dynamic range and other data and how “tight” are the values (standard deviation). DR, like iso sensitivity and color rendition, is best appreciated as part of a system. Perhaps to consider comparisons using various 50mm full frame equivalent lenses with various f stops in budget, mid and high price categories. Also using various common lens filters at budget, mid-range, and professional price ranges. One must also consider the roles of the viewer on camera and the computer monitor used to view the images. Data from a few common laptop and desktop display panel types could be presented. As a final step in system assessment, one could consider printed photos, including those available at store kiosks, along with medium and higher cost inks, toners, and papers. If one compares budget vs medium vs higher cost items, how great are the visually notable differences? (What is visibly gained with higher end equipment?) It might be useful to provide a few images so that the viewer can see where DR seems acceptable and where one should note a lack of DR, perhaps in dark or bright areas. Maybe even noting visually how higher DR using multiple photos in fast succession or photoshop can improve things - either a bit, or completely. Then the person viewing the video can decide what is most reasonable for their budget and abilities. Again, many thanks !
Absolutely fantastic suggestions and thanks for the feedback - you are completely right and your comment highlights why this topic is far more complex than just "camera x has more DR than camera y". Oversimplification leads to false conclusions, and people assuming that one thing is far better than the other although it is not. I definitely put this on my list, thanks!
Regardless of age, the general trend is ~1 stop between digital MF/FF, FF/APSC, and ~2 stops between FF/MFT as long as they're in the same generation (and not a Canon made sensor). We also have to keep in mind that SDR publishing i.e. JPEG exports can only hold ~6+ stops of dynamic range max. Having more DR helps with post processing. 1 or 2 stops is nothing to laugh at. It's what all of us are paying through the nose for f/1.4 or f/1.2 lenses. 1 stop is the difference between having useless shots on a f/2.8 zoom and a usable shot with a f/2 or faster prime. I think it makes sense to buy better gear for that extra 1 or 2 stops, but only if it makes business sense and pays itself off. For hobby use, it's not very worth it.
Thanks for your contribution! A JPEG can represent any amount of dynamic range, but as you have stated, it can only display a contrast ratio of 1:255 (th-cam.com/video/aWGIjXutyKU/w-d-xo.html). The usefulness of having more than 8 stops is therefore highly debatable, as you've stated. Regarding the f/2.8 f/2 - I think we have to keep in mind that we do not loose DR linearly when the ISO goes up. Therefore, even cameras with less DR can perform well at high ISO. I can recommend the charts on photonstophotos, very insightful I'd say. The one stop can be useful if you mess up the exposure, or you want to capture one stop more.
@@ThomasEisl.Photography Yes, I always reference photons to photos. Great site. Indeed. The DR loss is not linear. In fact, after factoring in equivalence for DoF, for most average shooters in non-extreme conditions, DR does not pose a huge issue as even the new sensors perform about the same. e.g. 6400 ISO on A7iv vs 3200 ISO on a crop sensor.
I think an important consideration is, when does the image become unusable for your application. FF, APS-C, … they all fall apart… is there a difference?
actually dynamic range is just one factor. If I'm shooting in the m43 format, ISO 200, f2.8, I would need to stop down on FullFrame at least to f5.6 and increase ISO to at least 800 to keep the same shutter speed. And there - the dynamic range is the same. Also for static landscape or architectural scenery - I'm happy to use Olympus' HHHR mode or LND64 to reach really high DR. So yes - the FF camera (or larger sensor cameras in general) has better DR but one needs to know how to use it and when :)
Hey and thanks for asking: So you definitely ran out of DR in the highlights with the X-T100. The problem is that manufacturers can "put" the gray point and therefore the exposure meter reference point wherever they want. It seems that Fujifilm decided to put it more to the highlight side of things than Nikon, who might have put it more in the center of the camera's DR. It still could be the case that the X-T100 has more DR at the end of the day, but it is just more on the other side. Hope this helped!
I think you got a few issues about DMs wrong, the R5 in photos to photons is 11.85 and the GFX100 is 12.3. And all DMs are base ISO so when you push the ISO you loose DM. If you start from 12 you have 4 stops to spare till you reach 8, if you start from 8 you don't have much to spare. And you need DM for post processing. I also don't know where you got the idea that you only need 4 stops of DM for an evenly lit subject, so pls elaborate.
Thanks for contributing! I did not mention the two cameras you mentioned at all in this video. DR and ISO are not related linearly - therefore you do not loose one stop per iso step. For all other questions, please watch my video on Dynamic Range in photography 📸
@@ThomasEisl.Photography Tnx i watched the said the video. The evenly illuminated, 2:1, etc story you tell has to do with the variation of illumination that you created in your scene/subject subject (e.g. one cannot 'see' the texture of the black suit/fabric). Nothing to do with the sensor. I also never claimed that ISO v DM is linear, each sensor has unique characteristics. Eg the one in R5 has two ‘bases’ at 100 and 400 ISO. And some manufacturers claim to have ISO invariant, etc. which is bogus but we have to accept that some sensors are more ISO-invariant than others. I also appreciate you did not mention the camera models I reported, which is why I reported them as they are contemporary ones you excluded and ended up as a result drawing the wrong conclusion. You also have to raise ISO for a variety of reasons and not just those you mention, e.g. you have slow glass, you want a large depth of field, you are in low light and do not have a tripod, etc. I shot film, slides, crop, and FF for 30 years and DM of 8 doesn’t cut it for me and for a lot of other people. If it's good for what you do and how you use your cameras, etc. fine by me but that's not the full or the only story.
Thanks for this comparison video. I have a reasonable understanding of sensor size, lenses, f-stops and light. I know that a Sony A7SIII is going to get a "better" image than my mid level Lumix G85, but to what purpose? What are you shooting, and how is it to be used? I would say that most any camera produced by any of the major manufacturers in the last 8-10 years will produce a good/usable image & video. I use MFT because of the form factor & weight. The camera fits my hand perfectly. My results are good enough for my needs. Camera companies are in the business to sell cameras (just as are car and mobile phone companies.) There is always going to be a bigger, better, more shiny new and improved model of something. Cheers. Peace.
Absolutely, Randal. Your comment is spot on - cameras have been good enough for many many years now, and let's not get started about film, which was also good enough and still is. Thank you for your educated contribution, much appreciated!
Good day, Thomas. It's a good video with good information. I do have one quibble with you on the number of stops per image needed for creating an HDR image. You state at the 12-minute mark the need to take images at one-stop increments to make an HDR image with the older cameras. This is not true. Even with the older cameras, only three images (-2, 0, +2) were sufficient (not five). And this was if you were shooting JPG or raw. As you know, an 8-bit image, such as a JPG, has about 7-8 stops (depending on the signal-to-noise ratio, but that's getting away from the point). Around 2012, I was getting annoyed with folks showing their ghastly, over-saturated HDR images and bragging about how they shot every third stop or half stop. So, I did an experiment. I set up my camera on a tripod, and I shot from -2 - +2 at every half-stop. This gave me nine images that I saved as JPG images. I then processed them in Photomatix (the best HDR Tone-Mapping software at the time) and processed them all with no adjustments. I did -2, -1.5, -1, .-5, 0, .5, 1, 1.5, 2. Then I did -2, -1, 0, 1, 2. Finally, I processed -2, 0, 2. The result was that there was absolutely no difference between any of the resultant photos. But, as you stated, the newer cameras do have more dynamic range, and because of that (and Chan from Adobe (their raw image master)) states that all you really need is the -2 and +2 shots, as that provides more than enough overlap. I have confirmed this, but I still mostly take all three shots because I usually hand-hold my camera. If the +2 shot is blurry due to camera movement, I can always fall back on the 0 shot. Nonetheless, I am always glad for things to be tested before being convinced of something, and I thank you for your observations.
Alles richtig - alles perfekt und didaktisch geschickt und immer in Bezug auf die Relevanz für die Praxis erklärt! Entspricht auch meiner Erfahrung. Wenn der Kontrastumfang einer Landschaftsszene, - klassischerweise: Sonnenuntergang, Himmel gleissend hell, Vordergrund schon im dunklen Schatten - sehr hoch ist meistert das auch kein noch so grosser Sensor, man muss zu HDR oder zu einem Verlaufsfilter greifen wenn man noch Zeichnung in den Lichtern und auch in den Schatten haben will. Aber die Hersteller wissen, dass höhere Zahlen - seien es mehr Pixel, mehr Stufen Dynamikumfang oder eine höher Kadenz - ein wirkungsvolles Verkaufsargument sind, auf das eben viele abfahren, weil sie zu technikverliebt und zu wenig Praktiker sind. Insofern kannst du dich als Aufklärer sehen, als Mann der aus dem wirklichen fotografischen Leben kommt, der die Kirche wieder in's Dorf stellt und sagt was Sache ist. Und je mehr Videos ich von dir anschaue desto weniger Gründe sehe ich die für Vollformat sprechen ... Vielleicht teste ich nächstens die OM-1 bei meinem Händler in Bern, reizen tät es mich schon. Inzwischen habe ich nämlich herausgefunden, dass man die Bildqualität noch erheblich steigern kann wenn man die RAW- Dateien mit DxO Pure RAW 2 entwickelt, statt mit Adobe RAW - unglaublich was man da noch herausholt.
Vielen Dank auch für diesen fachkundigen, umfassenden Kommentar. Das weiß ich sehr zu schätzen, er kommt aus der Praxis eines Fotografen, das merkt man sofort. Probieren würde ich die OM-1 auf jeden Fall - vor allem wenn man die Features richtig nutzt kann man wirklich extrem viel aus der Kamera rausholen. Ich finde, dass man diesen "Kompromiss" durchaus eingehen kann, vor allem wenn man eine kleine, leichte Ausrüstung schätzt und damit vielleicht sogar noch mehr fotografiert. Danke auch besonders für die Tipps hinsichtlich Bildverarbeitungssoftware!
@@ThomasEisl.Photography Ja ich war wirklich verblüfft über DxO Pure RAW-2 - ein Versuch lohnt sich! Die Software produziert rauscharme Dateien ohne dabei die Auflösung zu reduzieren, jedes noch so feine Detail wird aus den Daten geradezu herausgemeisselt, man könnte in's Schwärmen kommen 🤩 Habe heute mit der Lumix GX80 und dem NullAchtFünfzig Kit- Objektiv 12-32mm/3.5-5.6 - wahrlich nicht das Prunkstück im mFT- Objektivpark - einige Aufnahmen von ganz verschiedenen Motiven geschossen und das Ergebnis hat mich fast umgehauen!
The DR difference is real. It's there and this can't be denied. Do you need it? Do you want to pretend it's not there? Do you want to tell yourself it doesn't matter to you? That's all personal preference and may be completely valid depending on one's personal situation. Personally, I've gone back to FF. I'd much rather have the DR and not need it, than need it and not have it. But like you say, that's a whole different subject.
You didn't mention the effect of image format. Should I assume that you automatically meant [.raw] format? And that other formats (especially [.jpg]) would not convey the same dynamic range?
While I'm definitely referring to RAW here, keep in mind that the RAW image format does not necessarily increase the dynamic range (but often only improves gradation smoothness) more on that here: th-cam.com/video/uYOr6t8llgc/w-d-xo.html
Thanks for this summary, The one place where more dynamic range is essential and how I learned it was a thing was with time laps photography. When your camera is set, and you want to recover day-night sunsets without significant incremental adjustments in aperture or shutter speed, more dynamic range is critical.
@@stevenlang7709 newer cameras allow aperture ramping while using the inbuilt intervalometer. After that, the way I know is post-processing in LRtimelaps. I've seen a few impressive Davinci Resolve tutorials but never tried it.
Does anyone know the D850...it's incredible dynamic range .. I don't know the numbers but I have shot photos in long exposures we're of underexposed or made a mistake in calculation and the photo has been almost black and I have been able to recover and image I would be curious to know where the Nikon d850 sits I know it has more dynamic range than the Z7 series cameras as I have tested those and they aren't as good
The Nikon D850 has 11.63 stops of PDR. With that, it outperforms even the Z9. DSLRs have great sensors and the recent mirrorless hype has often overshadowed how good these cameras are in terms of raw image quality.
@@ThomasEisl.Photography hi, I have resisted in getting a Z9 I’m mainly shooting landscapes but I just could not see the advantage in a Z9 over a D850 when you consider the dynamic range and the megapixels. They are very similar. Yes I understand I make it slightly sharper images with the Z9, but you can always add sharpening in editing anyway I just couldn’t come to the conscious decision to pay an extra $5000 for a minimal improvement in image quality, I am a professional landscape photographer in Australia- not full-time gig. I now have a Fuji film, XT5 and a Nikon D850 I have the best of both worlds. Do not need a Z9
Because I'm a journalist, dynamic range is critical to me. I have an Olympus E-M1 Mk I, and a Sony A7 IV, and I often have to pull down highlights on faces and heads, while bringing up shadow detail. So the wider the dynamic range the better. I can say that dynamic range measurements are often much different in practice, since shadow noise isn't always reflected in those measurements, and my Sony blows my Olympus out of the water in that response, and my Ricoh GR III is better then my Olympus. Usable dynamic range is different than just dynamic range.
@@ThomasEisl.Photography I think that dynamic range in practice isn't unlike the difference between measured ISO and E.I. Whereas Sensitometry might measure a film speed at one rating, if you do your own work with the film, often there is a mismatch. I worked on a couple of projects for Kodak, one of which resulted in the decision Kodak made to label T-Max films with an Exposure Index derived from use vs. only sensitometry. With new AI masking tools, dynamic range becomes eminently usable, for all types of situations. And ISO absolutely figures into use, since "grain" isn't linear, as it tends be worse in shadows. So much so, that on older cameras, it's almost useless to attempt it. I'd love to see a methodology that reveals useful dynamic range. Astronomers can demonstrate sensor sensitivity in low light and extended exposure. So, I'm wondering if there is a way to utilize that approach?
Hi Thomas, I am from Tyrol, I now live in Reunion island, plenty to take pictures of here. I have been an amateur photographer for over 30 years. Currently I am doing a lot research to buy a new camera but I just don't feel like I am near a decision. I have been using a Canon 6d mk i for the last year and I don't see that even a €4000 camera from Sony does give noticeable improvement in terms of dynamic range. I want to buy a hybrid camera so at the moment I am looking at the Sony a7iv but I don't like the rolling shutter.
To be honest - you should definitely not upgrade because you get better image quality. Because the differences are practically non-existent for all practical purposes. Better get a 5DIV or a 5DS you can use all the lenses and they are absolutely outstanding cameras. Don't go to mirrorless expecting better image quality. You might even have to deal with some negative issues you never heard of - like heat-induced noise and heat-induced camera shutdowns in super hot summers. Thank you for the nice comment and best wishes to Reunion! Hope you don't miss the wonderful mountains too much!
Interesting points. You gave me a flashback to the zone system, but that is besides the point. A really good assessment. Thank you Thomas. Shoutout from Denmark. /Per
Many thanks! I think you are making a very valid reference here when mentioning the zone system. Correct exposure is as important for digital as it was for analog. Best wishes from Vienna!
HDR is no panacea for less dynamic range. Take it from a guy that's shot M43 for 12+ years. The HDR feature in LR or PS will make your shot look like mud in most cases. Bracketing over 5 stops have been my solution but is a PITA. Dealing with all the files can be a headache. In fast changing landscape circumstances bracketing is a hindrance. Manual exposure blending is the way to go rather than auto HDR selecting 2-3 of best shots but if you have a couple extra stops hard core bracketing would be avoidable and the blending chore!
Yes, there are many things a lot more pleasing than HDR haha - the problem is that you really end up in HDR territory really quickly, regardless of which camera you are using. BTW, I'm working on an interesting video about HDR for professional photography with the OM-1 - I'd love to have your feedback when it comes out. Cheers!
Well, the Pentax K-1 has one of the best sensors / imaging pipelines even by today's standards. It is not easily outperformed by anything. Such a great camera, I'm considering one as well to be honest. Thanks for sharing, Anna!
When digital sensor came big challenge wasdynamic range. Good topic and answered in video.. If we understand deeper problem as u said Can solve by composition and composite or graduated nd etc gives Solutions.great job!
I think often the problem of dynamic range and small sensors is more tied to low light performance. If the sky is so bright that you have to underexpose the foreground elements to preserve the highlights, sensor size makes a big difference to how much usable shadow detail you can recover, because it comes down to the low light performance in capturing the lower light parts of the scene. Yes you can use CPL and NDs in many situations, but not always, and you can’t bracket if you have moving elements in the scene.
Yes, but one or two stops (which is the actual difference) does not help with HDR scenes. Regardless of format, you have to resort to tricks. Maybe that is of interest to you as well: th-cam.com/video/jJQgJMCQJc8/w-d-xo.htmlsi=wfVvx_TMfzZkZ5bm
Regarding high-iso situations, one may find oneself a wildlife photographer in low-light conditions needing to raise their iso, which is a situation where dynamic range makes a difference.
The great thing is that practically all cameras will be able to provide 5 stops of DR at even high ISO numbers. And those 5 stops are more than enough for wildlife. Thanks for contributing!
@@ThomasEisl.Photography I understand that. However, in my experience, 5 likely won't be enough. I have done portraits with people in the shadows and a sunny background. The images were exposure balanced and at iso 100. With those conditions, my background was blown out. DXOMark claims 11.9 stops of dynamic range to my camera. In higher-iso environments, this might be more prevalent.
Yes! Just a food for thought - measure the DR in those situations you have just described with a light meter - it is pretty likely that even 15 stops of technical DR won't be enough to avoid blown out stuff in these situations. A bit of flash, however, will! All the best, Thomas
Thanks for asking: An evenly lit subject, no reflections - requires about 4 stops of DR for the full tonal scale to be reproduced. If you increase the lighting on one side by 5 stops (32:1) you now need 9 stops to reproduce its full tonal scale on both sides. The full version is here: th-cam.com/video/uYOr6t8llgc/w-d-xo.html&si=EnSIkaIECMiOmarE
Yes - it is a bit misleading and you are the second person to ask, rightly so! It is because in another video (th-cam.com/video/uYOr6t8llgc/w-d-xo.html&si=EnSIkaIECMiOmarE) I staged a sample portrait session using lighting ratios (but not the contrast ratios) to illustrate the DR you "actually need". In any case, we are totally on the same page here, thanks for getting in touch and engaging. I really appreciate that!
@@ThomasEisl.Photography I'm watching that video now and it's excellent. I wasn't aware of the base 4 stops needed to begin with (for the evenly lit scene), even though you did mention it in the above video too. Thanks for your responses and for these videos. I've been shooting for 35 years BTW, so there's always something new to learn regardless of how long we shoot.
I was geeking out about a new camera and my grandson, who also is a photographer, accused me of being a spec sheet masturbater. That stopped me cold in my tracks but it did get me thinking. I've actually only once had a client that questioned what equipment I was using. I've never had a client that could tell the difference between my digital Canon images, images from my Olympus EM1X or ones that were from my Hasselblad 503CX. In the end my experience is that good composition, good exposure and good editing of the images is way more important than which camera or sensor or how much dynamic range I had.
Thanks for sharing that Sophie - and I completely agree with what you said. As long as the full tonal scale of the main subject (usually 4 stops) is reproduced, the difference in DR does not matter at all.
Absolutely - regarding print, completely correct. The thing is that tonal compression should be considered as well, so you can "kinda" compress more stops into the limited print: th-cam.com/video/uYOr6t8llgc/w-d-xo.html&si=EnSIkaIECMiOmarE In any case, even with 8 stops of DR everything is usually perfectly fine!
Thank you very much for your comment! I recently did a very insightful research into ISO invariance - to find out, that almost no sensor is completely ISO invariant, although many ppl believe so, including many reviewers. The results are very hard to spot in practice, but the charts on photonstophotos clearly show that.
Sure, all modern cameras are very good in DR, even 1inch sensors. But I realized that APS-C is the minimum sensor for me, because I tried MFT cameras, but there was something wrong with rendering of such nature objects like water, clouds and greenery (if you pixel peep a bit). Looks like a sort of tonality lack. Even the newer 20mpx MFT sensor, I tried RAWs from Dpreview, but again, didn't like it. So m43 is very good for travel, street and architecture photography, but not for landscape, nature or portraiture in my opinion. But I have to agree about the older sensors, even the 16mpx Nikon D7000 sensor from 2010 is good enough in most scenarios today.
Color reproduction and so on is really very subjective. There are so many aspects to consider: Which program was used, was the file actually exposed correctly, and much more. Thank you very much for sharing your views, and I agree: The D7000 has an awesome sensor.
I've come to the same conclusion. And I am now hunting for information why I see it that way. As this video and other data suggests, the dynamic range is enough. I wish that the small vs big sensor debate would focus more on the difference in tonality, micro contrast, lens technology etc. Instead we are getting caught up in how much of an underexposed photo I can recover, or how the bokeh with a smaller sensor.
@@mumrik Probably it's important to us to see the photos which are closer to reality and looks cleaner. I also find that Canon APS-C photos look better than M4/3, but anyway they look a bit "rough" (I don't how to explain it correctly), Canon sensors just a bit noisier even at base ISO. So Nikon, Sony and Fuji files looks cleaner to me and just right.
Within the Canon line, I can affirm that the sensors on the mirrorless R series are vastly superior to the one in my old 60D but I cannot speak as to the difference on the higher-end Canon DSLR cameras. There is however, something to be said about older cameras. In fact, there are certain times that I still use my 2008 model Fuji S1500 with it's 10 megapixel sensor to shoot landscape images. The primary reason for doing this is to be able to set it up with the "FujiChrome" settings to produce some seriously intense color depth. I could recreate the same thing in Photoshop from my Canon mirrorless RAW files but sometimes it's nice to just "set it and forget it," as George Forman has stated in his well-known ads for his grill. Also, there are times when I do not want to lug around a full size camera on a long hike. The relatively small S1500 can easily be stuffed into a coat pocket or one of the exterior water bottle sleeves on my backpack. Best of all, should I happen to lose or damage the old camera, no big deal. I can easily get another for under fifty bucks!
Thank you very much for sharing that - and I completely agree: old cameras are still awesome. It seems that manufactures put a lot of thought in the picture profiles and jpg processing back then. Also, the reviews when these cameras came out often focused on this aspect. I also really like to use them nowadays, as the results still hold up, if you know what you are doing. Thanks again
Dynamic range does come into play when it comes to shadow detail recovery under low light conditions. For the average photographer, any camera these days is prob good enough. But for the demanding pro's there's a host of factors to consider when it comes to technical specs/design of the sensor and performance. Plus there's a number of technical factors that go into sensor design and performance as it relates into practical shooting, raw data capture and final image quality output. Sensor size does matter to a degree, but more importantly pixel pitch and photosite size, not to mention the A/D converters and computational algorithms that process the incoming analog signal to digital signal and how that signal is being processed and recorded as raw data.... just something to keep in mind and consider making additional videos on these discussion topics. Keep up the production, you're doing great!
Hey Stefan! Thanks for the comment and sharing your thoughts! Many factors like the one you mentioned AD / processing and so on are often overlooked, you are right - that is why the Pentax K-1II gives you better DR than the D800 with practically the same sensor, right! Thanks for engaging, the suggestions and the kind words!
@@weizenobstmusli8232 don't have to be a pro, but it's dependent on what you're looking to achieve in your final image... how one shoots/captures an image in-camera is not how the final image is processed... the goal is to capture as much of highquality RAW data as possible in camera to expose in-post for the final image rendition... as a general rule one should always ETTR without blowing the highlights, unless the final composition calls for otherwise... Frankly, I'd never recommend a micro 4/3ds sensor camera to anyone... but to each their own... (right tool for the right job principle always applies) Camera gear is just a tool... regardless of the brand or censor size... but one should always do their own research based on their needs application... and the technical benefits/limitations of the hardware/firmware specs... also, let's not forget the ecosystem as a whole for future expansion potential... which is something most people getting into photography do not consider.
I have been shooting with Sony A6700 for around a year since its release and looking to upgrade to FF A7cii because of low light performance and dynamic range as I got more into landscape and occasionally astro when opportunity arises. With your video now I am not sure anymore if I should upgrade to FF + more expensive lenses!
Thanks for sharing - great to hear that I was able to offer a different perspective. I'm sure that your A6700 is definitely (more than) good enough in terms of dynamic range and image quality! Best, Thomas 📸
@@ThomasEisl.Photography I am surprised too honestly that it holds up well but I guess it's because of the lack of experience that led me to believe that FF is going to be much much better as per many people claim it to be. As you mentioned HDR scenario can be overcome easily with multiple exposure blend and I have done that many many times too in Gimp. If FF is not going solve this issue then I guess I would not waste my money on the camera and, more important, the more expensive lenses. Let's not forget the fact that focus bracketing is more likely to happen in FF for landscape shot I usually do. I rarely have to do that on this A6700 but I was told that I would do it a tad more frequent with FF due to shallower depth of field.
@@ThomasEisl.Photography By the way, found your channel for the first ime and subscribed immediately after I finished this video. Thanks for the insightful explanation on DR which has always been a myth to me!
Well, it's not just about DR, is it? Newer and bigger sensors also come with improvements in other areas: higher resolution, better image processing/colour science, better AF, better noise control, better EVF, better screens, etc. Obviously, bracketing can replicate greater DR, but it adds more stages to work flow, doesn't it? You can mimic the DR of the Phase One with a lot of bracketing with the D2H, but that's not going to get you the high resolution the Phase One offers, is it? Don't get me wrong, I still use an old Pentax K5 and I am happy with the image quality.
Absolutely, the main point of this video was to illustrate that 1) 8 stops are usually enough and that 2) when you need more (landscape, etc.) you usually need a lot more than any camera has to offer. Of course, the D2H vs Phase One was tongue in cheek. At the end of the day, more is always better, but it is important to find out how much is good enough. That is probably exactly the reason why you are still working with the (btw fantastic) K-5.
the simple answer is that a digital file has "bit depth" in its data elements that define what I call "gradation resolution". But this informs nothing about "dynamic range". You could use the 8 bits per colour channel in JPEG to print your image within 4 EV dynamic range that looking at paper is likely to give you. When you depict the bits in a much larger dynamic range (contrast envelope), then you'll most likely see "banding" from your JPEGs. Imagine a rectangle hypothetical reference card - one square pure black, the other square pure white. Note that pure black reflects no light and by definition has no colour and so there is only 1 type of black. Note that pure white reflects all light and by definition has no colour and so there is only 1 type of white. (I.e. "blacks and "Whites" in Lightroom are a misnomer - they're all shades of grey.) Imagine that reference card in a completely dark/black room. You don't see it at all - it's all black and contrast or dynamic range is zero. Now slowly increase the light level and the black remains black while the white becomes lighter and with that contrast increases. Dynamic range in images is a bit like that card. The contrast you get from it depends on how you work with it. The "possibility" from your question hence is not in the JPEG file but in your brain and hands. The problem with JPEG is its low bit depth and if you depict your images on a medium that has better potential you may see banding. This happens when a subtle continuous gradation (change of tone) in the subject was not captured in enough discrete levels (because that's what the bits give us), or you processed you imaged with insufficient bit depth. In such a case, a continuous subtle gradation becomes a set of bands perpendicular to the gradation change's direction, where each band has no gradation change in that direction. If you raw process with Adobe Camera Raw (ACR) in Lightroom Classic (LrC) then take the processed image from LrC into Photoshop (Ps) and convert it into 32 bits per channel. The effect is very well visible on a good 8 bits per channel monitor/display. Internally, ACR and LrC use ProPhoto colour space, by the way. Imagine you now have 32 bits depth per colour channel and gradation steps are so refined now that you cannot see them as banding. But nothing got changed to "dynamic range" at this "post" side of photography. Note that raw processing has become more sophisticated with better AI and so the way gradation is converted from higher bit depth to lower JPEG is done smarter - the risk of banding is less today than, say, a decade ago.
Excellent video, as always. I especially like the thoughtful presentation. It would be helpful to hear/see an explanation of exactly how/when the 2 stop advantage of FF is helpful to have (post processing of landscape shots, maybe?). The conclusion here seems to be that neither the OM1 nor the FF can completely capture the entire dynamic range of some landscape scenes, therefore they are equally suited (or not) to the task -- and I doubt that is actually the case.
Many thanks! I agree, that is the conclusion indeed. To be honest, I don't think that the two stops difference are very relevant in practice. Maybe my video on dynamic range will clarify why: th-cam.com/video/uYOr6t8llgc/w-d-xo.html
@@ThomasEisl.Photography Thank you for the reply and yes, that is another very informative video that points out that there are many factors that play a role in the quality of the final print. IIRC, though, in one of your other videos you stated that the OM1 has essentially 6 stops of quality DR, which would be less than the "8 Stops are Enough." I'll dig through my history and rewatch that one as well.
Many thanks - maybe to clarify: You need 8 stops of total DR (realistically), with 4 stops of high fidelity DR at most. The OM has plenty, as practically all other cameras. What I define as High fidelity DR is the DR which can be moved to middle gray with full color information and little to no noise. It sounds a bit contradictory at first, maybe this helped to clarify things a bit. Many thanks for the excellent conversation!
@@ThomasEisl.Photography Wow, great clarification! Thanks again for the wonderful, informative videos and also for taking the time to read and respond to comments -- that sets you apart from many others on YT!
In my opinion dynamic range is the most important thing in a camera. Depending on the source the human eye has a dynamic range of 21-24 stops and this is the goal. Especially in Video, comparing footage of a the new arri 35 in log compared to recording in rec709 it is simply amazing. In photography in high dynamic range scenes or simply as a buffer for human failure is so important. Imagine a soccer game were the players are correctly exposed but the sky is white because of a bad dynamic range ...
I do not believe the human eye sees that many stops of dynamic range at all. We can only focus on specific things at a time. Whatever we focus on, we expose for (so to speak). Try having your blinds open a small amount, such that it's dark inside and very bright outside. Adjust your eyes to the interior and the window is blown out, just like it would be in a photo. Look at the scene outside and everything inside is pure black, just like in a photo. I think we have maybe 3-4 stops more dynamic range, but not up to 9 stops more (24 stops).
Dynamic range is indeed a very important aspect of every camera, I totally agree! The big question is: how much is enough. While we all agree, more is always better, we can usually get the job done with less as well. In videography, having more DR to work with is more important than in photography. You cannot shoot HDR brackets in video, right! Thanks for contributing!
@cooloox - I am also inclined to state that we humans do not perceive 24 stops simultaneously!
ปีที่แล้ว +2
@@ThomasEisl.Photography yeah thatis true, sources differ a lot on that but it is actually the other way around you have less dynamic rang at the spot you are looking at. Because the spot where you have sharper vision. This is because of the fovea centralis which has more cone cells ant there fore more color perception. The rest of the eye has more rod cells and is more sensitive to light and has a better dynamic range. other than that idk i think around 21 stops are realistic the typical scene would be an interview setting inside front of an bright window. idk but cool video man i enjoyed watching it!
ปีที่แล้ว
@@ThomasEisl.Photography no problem, thanks for writing back! I love that topic! Imo I woul always have at least 3 stops more than i need. 1 to have safe space from the noisefloor, one for highlight and one for human failure and personally that works for me. i am pretty happy with cameras with 12 stops and up. And there are of course extreme scenes where i am happy about every stop i get out of the camera
Dynamic range is just one component involved in creating an image. The sensor read speed, the DoF, the number of megapixels, all these elements together will allow either to get a good frame or not to get it.
I like you're approach Thomas, but there is one huge thing you haven't covered: rasing shadows in post. I am a wedding photographer and I often shoot (the dance, mostly) at F1.8 and ISO 12,800. The image quality like that is good enough, but I often want to raise the shadows in post and then I have a serious problem. I used old cameras to do this and I've used new cameras and the difference is very, very noticable. Oh, and generally only use available light.
Great point! Plus there's a number of other technical factors that go into sensor design and performance as it relates into practical shooting, raw data capture and final image quality output.
Thank you very much Matt! I agree - the problem is that you are actually running out of "high fidelity DR" - and therefore, you are brightening the shadows but only get noisy results. Technically, you are still working "within the DR" of your camera. I think this video would be interesting for you in this context: th-cam.com/video/uYOr6t8llgc/w-d-xo.html&si=EnSIkaIECMiOmarE
Well presented. Though I think the number one consideration is this - if I really want extra details in my landscape images then buy a camera with more pixels. Otherwise e.g. Phase One and Hasselblad wouldn't produce expensive camera backs with millions and millions of pixels, even though they don't give significantly higher DR. So, it's the extra details that cost your socks off.
Thank you! You are right that a larger recording format will resolve more detail, however, the relationship between size and improved image quality is not linear, see for example: th-cam.com/video/FVHTLFD_7o8/w-d-xo.htmlsi=y5DTRmNGmAxJ1hOj It is quite surprising how little the extra detail matters in practice, as the output media is usually limiting the reproducible details. Thanks again for the very valid comment!
@@ThomasEisl.Photography Agree, the extra detailing is primarily used for large printing or other large scale presentation. Hardly noticeable on even good screens - unless you own the format and can "feel" the details.
Thank you very much for this feedback! Coincidentally, I've used the Z7II last year for a few jobs, so my experience is not extensive. However - if you are looking for a high resolution mirrorless camera for repro work, portraiture or landscape, the Z7II is for sure a great camera. Very good build quality, nice handling. Solid video. I ended up with my DSLRs nonetheless, as I just really like OVFs & DSLR focusing systems, plus the Z7II setup was not really lighter than my D800 setup and performed about the same in all (for me relevant) aspects. Hope this helped!
By all this talk about the dynamic range of today's and yesterday's photo sensors, one must not forget that the dynamic range of perfectly normal consumer diapositive film (slides) was about TWENTY stops and not meager 10 or even less. Capturing landscapes never needed any sort of bracketing. On a good day with a good lens the resolution was about 16 MP. The limiting factor was the graining of the film material. The main performance drawback was the limited ISO. It was hard to get substantially above 200 ISO. For color 400 ISO was really stretching things.
Thank you for your comment! I would rate the DR of film lower, but I definitely agree that it is a lot better than people assume! I think this would be of interest, it covers what you have stated: th-cam.com/video/uYOr6t8llgc/w-d-xo.html&si=EnSIkaIECMiOmarE
@@ThomasEisl.Photography Very interesting. Thanks a lot. The figures I remember are a little different. As color TV was introduced there was the problem of playing cinema films and color stills across analog color TV. The contrast was much too high. At the time a TV CRT had - very much depending on the amount of ambient light - a contrast ratio of about 1:30 - i.e. a little less than 2**5 = 10 stops. A color diapositive was given as accommodating a max. of 1:1000 = approx. 1:1024 = 2**10 = 20 stops. From then onwards filming and photographing for usage in color TV programs had to be "special low contrast" .
10:18 "It wouldn't be too bad. We would just have to accept that some parts are clipping white or black". It's like saying "It's not too bad, you just can't see some things in your photograph, so what."
Well, not really - I know this is what one would think initially, but if you examine how final images look and how humans perceive DR, then you will find out that it does not matter at all - see my video on photographic DR
Interesting video, thank you. Digital photography was starting to make inroads with print media photographers just as Kodak introduced Ektachrome 100 Professional. Using Kodak's new (at the time) T-grain technology, E100 was the ultimate in color transparency film, the best ever, said to have a dynamic range of about 5 stops. It was great film, but missed the boat due to the emerging demand for digital, which became more cost effective in pre-press than transparencies. Anyway, for digital to achieve a dynamic range of 9 stops is a huge improvement over the best film-16X better, with each additional stop double its predecessor. With film, the over-exposure/under-development method to increase tonal range can go too far, producing muddy images with too little contrast. If sensor dynamic range were to get too extreme, couldn't digital images also suffer undesirably low contrast?
You are absolutely correct about the low contrast issue - I invite you to watch my video on Dynamic Range in Photography th-cam.com/video/uYOr6t8llgc/w-d-xo.html&si=EnSIkaIECMiOmarE Why do prints which have to represent 8 stops or more DR look muddy? Because they actually can only produce a contrast ratio of around 6 stops. The compression of dynamic range leads to this muddy appearance. Another reason why 8 stops fo dynamic range are enough for photography. Thank you very much for your very valid contribution, much appreciated!
@@ThomasEisl.Photographywithout HDR video or picture formats on screens that can produce true HDR levels, even screens are only 6-7 stops of usable dynamic range. Only 2000+nits in a dark room OLED panel will get you that 10 stops of range with a proper contrast.
I has other system ( SONY ) And I keep my APS-C still because I can youse crop mode on long lenses, and is still good for backup, First and Second is Sony A7IV and A7RV and third always is Sony 6400, still 24MP camera and APS-C so with lens 70-200 on FF I have 105-400 on APS-C :) Maybe is not 2,8 but F5.6 but still good and with OSS I can catch nice captures. Good Videao anyway and proper explanation. All the best Thomas!
Great setup! Sony has a lot to offer and those are great cameras and lenses 👍 Thank you very much for sharing that, keeping an APSC is totally reasonable. Thanks for the kind words!
@@ThomasEisl.Photography Not a problem, I did extra step now and bought teleconverter 2.0 SEL from Sony, and with 70-200 I get on APS-C 210-600mm F5.6 which is beast for affordable money, thank you for answer! have a nice day! Leo3City
One factor you did not mention, or perhaps even consider, in this analysis was the effect of pixel size on dynamic range. Larger pixels capture more light. Hence a larger sensor with more densely packed pixels could have either lower, or perhaps no better, dynamic range than a smaller sensor with less densely packed pixels. No?
Whenever somebody asks me what camera to buy I say "take any top end product from the last 12 years". I think the needs in pixelcount, framerate and autofocus speed matter more than sensor tech. And megapixels only for people who want to print large.
I still love and often favour for size my Olympus 450 with a 14mm prime rather than lugging my indestructible Pentax around. I just watch the weather report first 😊
📛 Become a channel member:
th-cam.com/users/thomaseislphotographyjoin
☕ Donate a coffee to support this channel:
ko-fi.com/thomaseislphotography
❓📩 Direct support:
thomaseisl.photography/shop/p/support-ticket
I have 5D classic and it is not that bad, it's slightly noisier than for example my Canon M50. I have shot a few weddings with it and it's a perfectly good camera. Nobody will ever notice the difference. Even the high ISO performance is quite good, i can push easily 2 stops images shot at 1600, so basically it can go up to 6400, and it has similar noise like the M50.
@@dtibor5903Yeah, old cameras just didn't let you turn it up as much, the maximum iso noise on my 5D is less than my 6D, which has better lowlight overall. They let you turn it up a lot, but it doesn't change the raw, just changes what your metering is, or how many stops above and below metered you get.
Arri or such shoot normally at 3200 iso and they do that because with RAW video you only get so many stopd above and below, and faster gives you a more normal range compared to film, with more stops overexporure latitude than under.
I thought my PC lagged when you said "36 megapixel". Anyways, nice video!
I’m thinking of getting ad800 or a d4 & trading in my Fuji equipment to go full frame.
Howie G what’s your opinion thank you
You videos are so informative Thomas - I can't stop watching them and have learnt so much. Thank you.
What I'm starting to realise is that a lot of 'it' is talk. When your friend in the pub says his camera is better as it has more megapixels, a bigger sensor and more dynamic range. They've probably spent a load more money yet don't actually know how to utilise the 'better bits' (if indeed they are even 'better').
I think the M43 sensor cameras are so overlooked because a lot of people rule them out based on having 'a tiny sensor' and 'only 20MP', but from my very unscientific testing, the OM-1.2 I have on my desk alongside my Fuji X-H2 is making a really good case for showing Fuji the door. The images are equally as good, the camera has more features I'll actually use, and the system as a whole is considerably lighter and more compact. I can't think of what I don't like currently...
Thank you again and I look forward to your next videos.
Thank you very much, that is just great to read. Regarding your assessments - well, I completely agree. I'm planning to expand my series on some "photo misconceptions" in the future.
Thank you very much for your support! Best, Thomas 📸
Since 2016, 90 % of my photography has been done with m4/3 bodies. I never missed on dynamic range. Learning to expose properly and shoot raw is key IMO. Exposure compensation is your best friend. Even the best metering system can use help from time to time.
And let's not forget that the human eye doesn't have an unlimited DR either. In some extremely high contrast situations, having a blown highlights and/or blocked shadows looks more realistic.
Absolutely, I completely agree. I think that those super-muted ultra wide dynamic range images are very rarely a desirable.
Many thanks for your great contributions on many videos!
Spot on mate. Sometimes it’s nice for shadows to behave like shadows
I definitely agree with everything you said. Another thing to remember is that it appears manufacturers have stopped working to improve sensors. There's been no real improvement for the last 5 to 6 years (basically with the release of the D850 and Sony A7RIII). All improvements now seem to be concerned with AF, FPS, and ways to compete with dedicated video cameras. If image quality is your main concern there haven't been any improvements since that generation of sensors.
You have made a very valid point here! Things did not change much in terms of image quality, but the higher readout speeds - which are crucial for better AF, FPS, and EVF performance - were the major advancement.
I have to admit that I am a huge fan of the OM-1's sensor, as I have seen IQ improvement, but the camera uses a fundamentally different sensor than the previous Olympus models. Be that as it may, the main improvements are in the AF, FPS and EVF department.
As you've said - "old" cameras (also DSLRs!) are still relevant, and will be as they are just good enough.
image quality probably reached its peak. Not much to improve
Most of the research and development in sensors is for smartphones and then it drips into cameras. With a shrinking market no one wants to invest money in it.
@@sexysilversurfer same AI painted oversaturated crp on mobile , atrifical clean sky with same color every year despite claims, no thanks.
The plateau of improvements in sensor tech is due to the limitations of current CMOS sensor manufacturing. We are seeing faster readouts, frame blending (mostly cellphones), dual gain circuit outputs etc…
We won’t see leaps on DR until new sensor technology is developed. Panasonic has organic sensor in development but it’s a long way off and has issues.
Another fascinating video Thomas. This topic is definitely something that comes up regularly. In fact, I had just finished watching a vlog from another TH-camr who stated "of course M4/3 has lower dynamic range than full frame". It was so coincidental that yours was the next video I watched. In any case, I'm very happy with my OM-1 for the whole package of features it offers. Thanks for another great presentation Thomas!
Thank you very much for your kind words and appreciation!
I myself am also very satisfied with the OM-1 and its performance, just shot another magazine editorial with it and the quality of the camera is just great, in every way. I'll soon follow up with new videos on the OM-1 and its functions! Thanks again, Stephen!
For me, photography is an "escape route". when everyday life takes over and adult life settles down like a blanket. In this bubble I want to be, just be. Enjoy the nature, listen to the stream flowing and the birds chirping. I don't want to be technical.
Blown highlights. Crashed shadows. Soft corners. I do not care.
But that said, I'm no professional. No weight on the shoulders to "deliver". By the way, did I say that I appreciate your technical videos?
I do. Vielen Dank
Thank you for your excellent, eloquent statement - exactly these points are why I just like cameras that resonate with me.
As soon as I get to post process my images, I'm crushing shadows and blowing highlights, because it just looks awesome, right! So who really cares if you have 25 stops of DR recorded, no client ever asked me about that or complained because they were not "there"
Another excellent and entertaining video, Thomas. Your depth of understanding and ability to explain it to others is without equal, and so appreciated.
Thank you very much for your kind words. Cannot thank you enough, really!
Excellent presentation; data-based and practical! The only ones who might not like this are the manufacturers' marketing teams where it's primarily a numbers game: more is better...
Yes, I think you are absolutely right haha.
Thank you very much for your kind words!
Well, if you look this from manufacturers and marketing teams perspective, only value they can present is the theoretical maximum DR value, since there is no standardized method to measure "photographic DR" that could be compared between different brands and manufacturers. So they choose to publish value that they have measured in their own lab. We have to look for third party labs, such as photons photos old DxOlab, that measure and compare sensors.
I love your videos. the thorough investigative research that you do, video set up, sound and lighting quality and the excellent dialog delivery that goes into them. It's like watching inspector gadget teaching these facts that not very many people know. I've been binge watching many of your videos and learning a lot. I so much appreciate the effort you've put into them. I don't subscribe to many YT channels, but I have subscribed to you and will watch more and some of them over a few times. Great learning material. Thank you. The shutter simulation between scenes is a good touch.
Many thanks for your exceptionally kind comment, Richard.
Feedback like yours means a lot to me and I very much appreciate that you took the time to share that with me.
I'm very thankful to welcome you as a subscriber - I hope future content will be of equal interest to you!
Best wishes, Thomas
Thank you Thomas, a very enlightening video! I am an amateur photographer. Years ago, I started shooting slide film, and just got used to the 'small' dynamic range of that film compared to negative film. As a new photographer, I longed for a film that had the same or close to the dynamic range of the human eye. Then I could 'catch everything'! As I gained experience, I learned that having limited dynamic range is not a bad thing, one can do a lot creatively with it, capturing sun beams through a windows in a dark room for example. When I switched to digital cameras, I felt I was on familiar ground, and just started using many of the same techniques as I used with slide film. I've never had a complaint or criticism of 'low' dynamic range in my photos. I just try to get the best exposure for the mood, scene, feeling, whatever of the image I see in my head, and I'm happy with it. It's actually fun to see what I can do with the range I have. What really needs to change is the dynamic range of the monitors we use to view our digital photos with; but then, that might be another creative avenue to work with. I'm Subscribed. Looking forward to exploring other videos of yours. thanks again.
Thank you very much for sharing your experiences - I've read everything carefully, I am totally with you. The "limit" or working with the limitations are actually ok!
Thank you very much, also for subscribing! Hope to hear from you soon!
@@ThomasEisl.Photography Yeah for real, unless you're boosting shadows by 5 stops in all you images, you don't need that much. You can't put more dynamic range onto a 6 stops screen or print or it looks super flat and boring. Slide film was a mere 5-6 stops usually and some may have even been under 5, like Velvia. And yet everyone raves about the colour and contrast of something that the Nikon D1 beats for dynamic range.
Well said, sir! Back in the 70s and the first half of the 80s, I worked as a photojournalist. My biggest concerns were how far I might need to push my Ilford HP5 or Kodak Tri-X 400; dynamic range was more about how dynamic my images might be than about how many steps of grey I could stuff between black and white. Even when (on vacations) I put on my Stetson and emulated Ansel Adams, I realized that I needed visualization, image capture, and final image creation (in the darkroom) to achieve my goal. I am very well served today by a pair of Olympus OM-D EM5 Mk IIs - a newer unit might have slightly better resolution and a few better features, but the wonderful ease of use and lens selection keeps me in the M43 camp.The key advances of my particular camera bodies include the 40MP mode that uses pixel-shifting, and numerous choices for bracketing of focus, exposure, etc.
Well for my bird photography hobby where I prefer to shoot in very low light and/or strongly backlit situations 2 stops improvement (which is 4 times more light) is very helpful.
When I switched from a Canon 7Dii to a 5Diii and 1DX the improvement was mind blowing.
Also the high iso performance of these full frame sensors seems freakish to me after years of using only APS-c.
Bird photography is an extreme form of photography where high speed action, low light and tiny erratic subjects mean that having high performance cameras and lenses are very helpful.
I also avoid shooting wide open as it's likely to miss focus more on fast moving tiny birds so when I use my 400 f/2.8 I generally stop down to f/4 or preferably f/5.6 or even f/8 and keep the shutter speed to at least 1/400 second or preferably higher.
That is great to hear - the performance of these cameras is truly impressive, and they are great value for money!
Thanks for saying this. I’m constantly trying to explain to my full frame friends how terrible the dynamic range is on my M50. I wish someone had explained this to me before I bought my camera.
@@eafortson you don't really appreciate the advantage of the bigger sensor until you try it out. I recommend trying a full frame dslr like the 5Diii which are very affordable and great cameras
"High ISO performance of these full frame sensors seems freakish" Ikr! I just bought my first full frame camera recently, it's crazy to me to shoot at ISO's like 1600 and 6400, no problem. And with a flash, I'm literally shooting at minimum flash power 😂😅😭
@@blanked3 I've shot at 40,000 iso and still got a reasonable image so long as it's correctly exposed. I don't restrict my iso range at all
1:20
“It’s about time for a serious reality check about photography silicon”
*sweats profusely in excitement*
*nerdy pathways activated, serotonin dispensed*
Haha 😆
That comment made my day
To be quite frank this video left me a bit confused, Thomas. At first I learnt that dynamic range is a bit like image resolution: As long as you've got enough for what you want to do everything is ok. And that dynamic range above 8 EVs isn't relevant and even a bit less isn't a problem all by itself. That sounds a lot like "the max 9.7 EVs of the sensor in my E-M1X is more than enough" just like "the 20 MP of the sensor of my E-M1X is more than enough for taking the kind of pictures I'm taking, with very humble cropping and not deliberately higher than ISO 6400 sensitivity setting" (which in a nutshell really is my point of view). And it also fits my experience for rating picture quality on my E-M1X on different ISO settings: 200/excellent/9.7EVs 1600/very good/6.9EVs 3200/good/6.1 6400/still-satisfying/5.2EVs >6400/try-to-avoid-if-possible/
What a wonderfully thorough and informative video!
I used to shoot weddings on Canon 1Ds MkII - about 8.5 stops similar to the 5D. Several times I could rescue highlights and shadows to make perfect exposures from RAW even at that level. I think the dynamic range argument says more about the ability of the photographer to expose correctly than it does about what camera format you need.
Agreed - having more DR is great, and I also have to rescue photos from time to time. But it is exactly as you've said - it is us photographers screwing up, not the limitation of a given camera system. Many thanks for the comment!
Spot on! Regardless of the slight image quality increase, the large-sensor camera buys me nothing if I’ve left it home because it’s no fun to carry. I’m seriously considering a switch from full-frame to the OM1, and your insights are slowly helping me to make that decision. The best camera in the world is the one you have with you when you need it.
Completely agree. Thank you very much. I'd say you would be happy. I've still got everything from digital medium format to 43, and I'm using the latter practically all the time. Best, Thomas
At 09:45 you talk about a 32 to 1 lighting ratio.
You then say you need 9 stops of dynamic range.
Looks more like 5 to me: 2^5 = 32 (1,2,4,8,16,32)
Or do I not fully understand it?
Yes, you are correct - the reason why I stated 9 stop is:
you need at least 4 stops for recording every nuance of a flat, non-reflective subject. The contrast ratio as you have stated adds the need for additional 5 stops - this means we now need at least 9. This is where the 9 comes from.
Also see this video were I state this more clearly: th-cam.com/video/uYOr6t8llgc/w-d-xo.html
@@ThomasEisl.Photography Correct. I did not catch that you were talking about light...
I watch the video you recommended.
Very nice and clear explanation.
I posted a related comment under that video, about how the histogram is related to the Picture Style when the camera is set to RAW only.
Hello Thomas, 32:1 is 5 stops and not 9 stops. 2^5=32. And are you sure your most dynamic range on set was only 32:1?
With my old Canon APS-C 500D camera: You couldn't even look at ISO 1600 at 15.1 Megapixel. With my new Sony A1 at 51 Megapixel photographing with ISO 3200, 6400 or 12800 or more is even better. Of course ISO 100 would be better, but you don't always have so much light. So I don't agree that a new camera doesn't bring much: There is a lot of difference in image quality, noise and dynamic range with newer sensors and cameras. Also keep in mind, that readout speed with newer stacked sensors (e.g. Sony A1 has 4ms) and is around at the time a mechanical shutter has. So there is in typical situations no rolling shutter effect at all with electronic shutter. I guess in 2003 we had 50ms+ on sensor readout speeds.
From a practical standpoint I couldn't should a lot of picture with the old Canon body handheld.
Hello Gerhard, thanks for contributing, I'll gladly address your much appreciated inputs!
Regarding 32:1 being 5 stops
A non-reflective subject requires about 4 stops of dynamic range. Now, increasing the lighting in one stop increments on one side of the subject until we arrive at +5 stops (32:1 as you have stated correctly): In order to reproduce the full tonal scale on both, unevenly lit sides of the subject, we now need 4+5 stops of usable dynamic range.
Regarding Old Cameras / New Sensors
Especially when shooting high ISO, low dynamic range can result in bad results like with your 500D. Let's say at ISO 1600 only 4 stops of high fidelity DR remain, then you have to expose perfectly to capture the full tonal scale of the main (4 stop DR) subject. Every small error will lead to a significant, noticeable loss in quality. That is why I said in the video that these older models need a flash and are sometimes tedious to work. Not an issue with your A1, obviously! Does the A1 have more dynamic range than some older DSLRs? No, but does that matter: No! Because, as you have stated, the main advantage newer sensors for mirrorless cameras have is not significantly better DR, but faster readout speeds, something that was not so relevant for DSLRs (not talking about something like the Olympus E-10P, of course, which is limited by the readout speed although an SLT.
I'd like to conclude that we are exactly on the same page here!
Fun fact: I've been typing this response twice, as the browser crashed. This also happened when I was replying to your previous comment on my previous video (th-cam.com/video/uYOr6t8llgc/w-d-xo.html&si=EnSIkaIECMiOmarE) , seems to be a thing haha
Best wishes!
Thats assuming that ISO is the same on each camera - it isn’t - ISO differs between cameras and brands
Nice overview that puts things into perspective. It's a most convenient truth for those of us on a tight budget who want a great performing digital with excellent traditional photographic functionally and don't need to shoot a gazillion frames per second, a focus point for every other pixel on the sensor, or any heavy in-camera processing / AI hocus-pocus nonsense. For hobbies like astrophotography, there is not much between a 10+ y.o. DSLR and a modern mirrorless costing 10-30x more. Cameras like the Canon 6D and Nikon D600 costing less than a low-end smartphone are already (pardon the pun) stellar performers and I see no need to go to huge expense buying a new or recent camera. Even for more general use, for the average hobbyist or even serious enthusiast, there are many great second hand options out there for a fraction of new camera prices that do the business where it counts. I appreciate though that pro photographers will need the latest features & functionality to keep up with the competition, but more importantly to keep up appearances in front of clients.
Thank you for sharing that!
I have to admit that I am fully comfortable with showing up with a MFT or old DSLR to a shoot, no one ever asked any questions. I think it is mostly in our heads.
Thank you for a good video! Informative in many ways. However author of this video never described what stop means in cameras. As far as I know one stop in exposure or DR (dynamic range) means doubling of light, so if DR is increased from 10 to 11, DR has not increased by 10%, but by 100%! Correct me if I am wrong. These DR values are compared in this video as if they are linear values, which is not the case. If you compare presented DR’s of om-1 and Nikon d800, it shows slightly over 2 stop advantage to Nikon. Full frame sensor is four times larger in surface area than MFT sensor, so it will gather four times more light by frame, which means two stops. So these findings are exactly what we should expect! Nothing surprising here.
However is this significant in normal photography such as portrait or landscape, is completely different thing. This was addressed in this video very well! As a nature and bird photographer myself (who shoots with om-1 and A1), I think there is difference in DR and noise when shooting moving subjects in less than optimal conditions. It’s matter of preference whether you consider it to be significant enough.
Happy shooting!
Hey Heikki!
Regarding doubling the light: you are correct! However, it is way less significant than it sounds. If you have not watched already, I recommend watching this video th-cam.com/video/uYOr6t8llgc/w-d-xo.html
I think it illustrates why 8 stops is basically enough and why one stop more or less is hardly ever noticeable in practice.
Thank you for sharing your very valid remarks, sorry for the late reply!
@@ThomasEisl.Photography I agree with you regarding 8 stops being more than enough. But this applies only if shooting in native ISO or slightly faster. It's funny since I have always been reluctant to push ISO values above 800-1600 in OM-1. When looking at charts by photonstophotos DR drops below 8 stops at those ISO values. On FF cameras I usually push ISO to 5000 range, but avoid pushing further. Stopping motion in less than ideal conditions is when bigger glass and sensor helps (even 1 stop can make a difference). Thats why I never gave up on my FF camera (I was very close selling). OM-1 is brilliant camera, could have better video capabilities though.
I have watched many of your videos, don't agree on everything, but still very good and informative content! Keep up the good work!
We need to distinguish two kinds of Dynamic Range (DR). (1) The operating range of a camera (cf. temperature operating range) that is generally referenced as DR, and (2) a camera's contrast envelope that is the DR available to a single shot. These two can be far apart.
When you say what you need in 1 shot, this is the contrast envelope version of DR.
Not sure where you got the 4 stops from for "an evenly lit subject", but in the film days in our discipline of densitometry (and sensitometry) we used a 10-log base for i*t and most film would give a contrast envelope in the range of 10-log=4, some went to 4.5 or 5. "Stops" are 2-log and compared to 10-log "4" is peanuts - abysmal.
In studio photography under controlled light, we can keep contrast relatively low, but elsewhere? Note that LV 10 already is darkish in available light photography - and light level dictates contrast.
Interesting contribution - thank you very much.
I agree 100% with your thoughtful assessment. In 2019 I made a TH-cam video that said that any decent prosumer/professional camera from the last 10 years could accomplish most professional photography needs, and those cameras still hold up today. My first DSLR was purchased in 2003 and was the Canon Digital Rebel (Canon 300D). In some normal situations, its dynamic range seemed lacking. However, that was not the case with my 2009 Nikon D90, which still takes great photos. I now use a Canon 5D IV and a Canon 5D III for professional work, which serves my needs well. Newer cameras have advanced features like better tracking, possibly quicker focusing, and better video capabilities. However, such things are optional in the work that I do. Thanks again for this very good video.
Hey Mike!
Thank you very much for the comment, your kind words and the sharing of your experiences! It is great to read that we are on the same page here - I'll also check out your video ASAP.
The real issue is that the 5D-series cameras you are using are actually so good that they can get every (!) job done. Bad times for manufacturers trying to sell you new stuff. I mean, even the focus systems on these cameras are top notch and absolutely future proof.
It’s all about using the right tool. I use for indoor and some landscape a Sony A1. For wildlife and outdoors I shoot a OM-1. There is no way I a hauling around or shooting effectively with FF and lens. The lens is massive and heavy vs the OM-1 and my 300 f4 MFT.
I shoot a lot of HDR landscapes, and I have something perhaps curious to add. Sometimes I do not want all of the dynamic range and enjoy the fact that I can easily create black shadows in sunsets and sunrises. In fact, I have often been disappointed with HDR because it gets rid of mystery. I like that I have a choice.
Absolutely, you are completely right! In the following video, I illustrate exactly what you have observed - having less is not necessarily bad, as this is the way human perception works.
th-cam.com/video/uYOr6t8llgc/w-d-xo.html&si=EnSIkaIECMiOmarE
Thanks for your contribution!
??? I am confused here, why don't you use HDR just when you need it. Beside, if you keep all the shots you can definitely go back to the one you need. Tools are as good as we know how to use them, don't you think?
@@BrunoChalifour that is exactly what I do. My point is that if you make a sensor that inherently incorporates a huge dynamic range, equivalent to today's HDR images, sometimes, even often, that is not desireable. My point is that I don't care about buying a camera with the extra dynamic range, especially if it is a hindrance. Although, to be fair, manufacturers can probably create a limited dynamic range mode.
@@craignichols21 Well unfortunately we are not there yet and the dynamic range of today's sensors is still limited compared to our eyes/brain's potential adaptation to light. HDR has a look that unless mastered does appear artificial. A lot of people have a tendency to overuse it I the same way some did with sharpening when it came out. Using unfortunately does not mean mastering.
@@craignichols21 I guess it wouldnt be hard to add an SDR mode. Its all digital anyway. But I dont see a scenario where I personally would want less info in my raw files.
I just got my OM-1 Mark II and I can say, it’s an amazing camera, it’s only 20MP but you can’t even see the difference when you compare the quality with full frame cameras.
Very good overview and myth busting.
I would add some slight caveats around the need for higher shutter speeds in landscape photography:
1. Mountains and buildings do not move. Clouds, trees and water do move. Freezing their movement into an exposure may require a single exposure with a high enough shutter speed, not allowing bracketing.
2. Higher shutter speeds may also be needed when there isn't necessarily subject movement, but the camera itself may move (even when on a tripod). Vibrations from traffic, people walking by your tripod, of a very heavy camera/lens setup on a tripod fully extended may require a fast enough shutter speed after all for sharp images.
Everything you said still applies and is very sound advice.
Many thanks - I completely agree with the two caveats you mentioned!
It was a bit of a generalisation on my part to simplify the argument.
Many thanks again, much appreciated!
Professional indeed ;)
Further to your experiences with scene dynamic range, I'm coming into photography from a background in graphic design, and I feel pretty clued up on brief specification - be it for magazines, bus stop advertisements, websites or html5 banner ads etc. And with that, the best printers money can buy, will in ideal conditions reproduce
Hehe 😉
Thanks for sharing your professional experiences as well - it is really astounding to read very similar statements in the comments repeatedly. Proves the point - DR in photography is only an issue if you are looking at spec sheets only...
Subscribing mostly because of your refined wardrobe. The camera part is excellent, too.
I knew that wearing a tie would finally pay off! 😂 Thank you!
Hello Tomas I had the om1 yesterday and done a shot of a venue, and I tried the high res mod. All of the photos came out washed, the color was terrible. What did I do wrong? I also took my Nikon z8 and I did get some great phot for my client.
Uh, I need more details to help you - send me an email.
There are far more issues than dynamic range to consider, great video.
Completely agree! Thanks for the comment, David!
Years ago, I got the Fujifilm Finepix S5 Pro -- which has a unique 6/12MP sensor, and about 12 stops dynamic range. Thoughts?
I've been trying to get one of those to test - had an S3. Great cameras IMHO. And yes, more than enough dynamic range hehe
Also, I love that this camera does things differently. Nowadays, digital cameras are often times so boring. Most of them are more or less the same 🥱
Only started watching the video, interesting what you will say, without knowing it - I came from an APS-C mirorless camera to the mirorless Full Frame world and that was my best decision made so far! How much it speed up my photography game (shooting and processing pictures - retouch) I cannot explain.
What is the difference between pictures? still not sure if staying ff or going aps-c
@@laiebi_3639 Stay. There are probably scenarios where Aps-C be better, wildlife, sport, but I shoot portraiture and with FF my workflow got much faster and the quality of pictures got better. I was using a Nikon Z50 and now I am using a Lumix S5. Also the videos and grading is day and night.
@@iammz81 Cool, looking forward to shoot cars and wildlife and some occasional family gatherings so its a bit of both worlds. For traveling sth small and light would be cool too but must be worth the quality, otherwise I'd never really use it
Great to hear!
Personally, I am not sure why you are seeing such a difference, but as long as it works for you that is just awesome!
Maybe you really prefer the large sensor look or you are using better lenses than you did with your apsc cameras. In any case, also consider medium format if you really like the big sensor look and feel!
@@ThomasEisl.Photography It's probably not ONLY the sensor but the camera itself that made my workflow easier, but all in all happier with the result.
What about sony a7iii it has 15 stops?
The A7 III has a photographic dynamic range of 11.6 stops.
The 15 stops are the engineering dynamic range, which is not the actual "usable" DR of the camera.
th-cam.com/video/uYOr6t8llgc/w-d-xo.html&si=EnSIkaIECMiOmarE
Dynamic range is more useful for general use cameras than studio or landscape photography, where filters, lighting and multiple exposures can be employed. It's also important to remember that most stops of dynamic range are for shadow recovery, not highlights. I had a Canon 5D until 2 years ago, and while the colours were good and 12mp resolution was not a problem, it was difficult to recover detail in any backlit scene.
Thank you very much for adding that - completely agree!
The 5D Mark II has awful shadow banding and crosshatching, especially in underexposed images. Attempting to push the shadows makes this very apparent. Today's sensors have clean shadows which can be pushed two stops without major issue.
@@raksh9 The 5D had highlight banding, which was its most annoying feature. By comparison my 24mp Lumix S5 has extraordinary dynamic range - perhaps due to its modest megapixel count - and is the only camera I have owned where ISO can be effectively ignored. However, the Canon 5D is an 18 year old digital camera, which is light years technologically speaking.
@@xcx8646 you can't generalize like that. If you want to preserve blue skies you have to underexpose, so the midtones will be very dark and if your camera lacking dynamic range you will end up with noisy images trying to push the shadows. Dynamic range is quite important in nature photography unless you like to jump through hoops with multiple exposures.
When moving from a D750 to a Z7, I found, at first, I kept bracketing. Then, after some time, it became clear to me that I didn’t need bracketing because, post-processing tools and better dynamic range together got the job done with 1 image. Less work, better outcome. And then … there’s one of the best reasons to go to a digital camera … size and weight of the camera and lenses!
Absolutely - better DR can be useful indeed, no doubt about it! But as you've stated, the most important aspect is that you bring the camera with you, and size and weight do matter a lot when it comes to this!
I don't know about you but i never needed to bracket with a D750 either, it has awesome dynamic range itself.
@3:17 you said the Canon EOS 5d is 16Megapixels, it is a 12Megapixel camera.
Already added a correction - it was a slip of tongue. Thanks for pointing that out!
As you juggle with numbers and say that 32:1 needs 9 stops DR, what do you mean with 32:1? If you use aperture stops then 32:1 is 10 stops, if this is about light levels in Lumen/lux or so, then 32:1 is 5 stops.
I'm not sure which passage you are referring to, please provide a time code.
@@ThomasEisl.Photography the section starting 9:18 about 9:40 the most challenging ...
@jpdj2715 9 stops is referring to 4 stops for the subject plus 5 stops for the added contrast. Please refer to the linked page for the full calculation. Best, Thomas
My experience about dynamic range shooting pro interiors during many years already :
Canon Canon 5d , 5Dm2 5dm3 : just enough , but no so impressive , in today standards.
Canon R5 : not enough for the price .
Pentax 645z : Super Amazing ..!
Fuji gfx 50 R, 50 S and 50sII : Amazing ..!
Beyond this , the quality of color rendition , tonalities of medium format is something appreciable .
Schöne Grüsse .
Thanks for sharing your perspective!
How much dynamic range is there in a jpeg?
And
How much dynamic range is there in a monitor or phone?
All these questions and more are covered in this video:
th-cam.com/video/uYOr6t8llgc/w-d-xo.html
Hi Thomas, I've just recently discovered your channel. Really great to see someone approach camera systems from a technical, balanced, and realistic point of view.
I wanted to share a technique I've recently started to really lean on for HDR situations with my GH5. The couple of times I've tried shooting HDR brackets, I was never really satisfied with the end result, since I would get halos and an "over processed look". That's when I realized that Bill Claff's (owner of photons to photos) photographic dynamic range concept was based on the idea of scaling all images to a common 8x10 print output size, and that I could gain dynamic range scaling my images down further. So now, I use a longer focal length and make a panorama that's larger than my intended output, and then scale the whole stitched image down to the intended output size. Since so often those HDR shots are landscape, for me this kills two birds with one stone: gives me both more photographic dynamic range and more fine detail. Plus, I find that I get fewer stitching artifacts than with HDR stitching, and I don't ever get the "HDR" look. Theoretically, software noise reduction should work better if you give it more detail to "bite" onto, but I haven't tested this specifically.
I'm not entirely sure how the PDR will scale with the number of stops, my guess is each doubling of total pixels should be another stop of PDR.
Many thanks for the exceptionally kind comment. I'm sorry for the late reply, I somehow missed it!
Regarding your tip, I think the photonstophotos noise chart is what you mean? DR should not change when you downscale the photo.
In any case, I'm working on a video on HDR photography. How to set it up in practice and how to get natural results in post processing. Just like you, I'm not a fan of the overly processed HDR look.
Again, I very much appreciate the feedback and welcome to the channel! Hope to read from you in the future!
@@ThomasEisl.Photography I don't think I worded my initial comment very well. Shrinking the output size should increase the Photographic Dynamic range, because it's a ratio of maximum highlight (not impacted by downscaling) to minimum "noise-free" shadows (noise suppressed by downscaling, so the minimum useful signal is lowered). I don't think engineering dynamic range would change though, just photographic dynamic range. If I am understanding correctly, the PDR chart on Photons to Photos is based on the ratio between the highlight clipping point and the shadow point of having a 20:1 signal to noise ratio, after being downscaled to a resolution appropriate for an 8x10 print (and if I'm not, that means I get to learn something new today!)
I think that means starting with a panorama instead of a single exposure means either a bigger output for the same PDR, or more downscaling and more PDR for the same output size.
Also, the GH5 doesn't have any high-resolution pixel-shift mode; if it did I would be using it and taking far fewer panoramas. But I'm considering a switch to the OM-1, and trying to learn more about it and it's built-in computational photography features is how I came across your channel.
Is a one stop improvement in DR a doubling of dynamic range as a one stop increase in aperture is a doubling of the exposure? Or am I reading it wrong?
is a one stop improvement noticeable, at correct exposure its half a stop at light and dark ends? now a one stop improvement at each end is worth the effort, you can see that in sunset shots for example
Well, you can detect one stop more light. And one stop more luminance means that there is something in the frame that is twice as bright as something else. However, human perception does not work linearly so this is usually not a big deal. Maybe this video helps to clear things up: th-cam.com/video/uYOr6t8llgc/w-d-xo.html&si=EnSIkaIECMiOmarE
@@ThomasEisl.Photography 👍
Thank you for a very nice video.
Some other factors to consider, which are also noted by many others might include:
The number of cameras, sampled to derive the dynamic range and other data and how “tight” are the values (standard deviation).
DR, like iso sensitivity and color rendition, is best appreciated as part of a system. Perhaps to consider comparisons using various 50mm full frame equivalent lenses with various f stops in budget, mid and high price categories.
Also using various common lens filters at budget, mid-range, and professional price ranges.
One must also consider the roles of the viewer on camera and the computer monitor used to view the images. Data from a few common laptop and desktop display panel types could be presented.
As a final step in system assessment, one could consider printed photos, including those available at store kiosks, along with medium and higher cost inks, toners, and papers.
If one compares budget vs medium vs higher cost items, how great are the visually notable differences? (What is visibly gained with higher end equipment?)
It might be useful to provide a few images so that the viewer can see where DR seems acceptable and where one should note a lack of DR, perhaps in dark or bright areas. Maybe even noting visually how higher DR using multiple photos in fast succession or photoshop can improve things - either a bit, or completely. Then the person viewing the video can decide what is most reasonable for their budget and abilities.
Again, many thanks !
Absolutely fantastic suggestions and thanks for the feedback - you are completely right and your comment highlights why this topic is far more complex than just "camera x has more DR than camera y".
Oversimplification leads to false conclusions, and people assuming that one thing is far better than the other although it is not.
I definitely put this on my list, thanks!
Regardless of age, the general trend is ~1 stop between digital MF/FF, FF/APSC, and ~2 stops between FF/MFT as long as they're in the same generation (and not a Canon made sensor).
We also have to keep in mind that SDR publishing i.e. JPEG exports can only hold ~6+ stops of dynamic range max. Having more DR helps with post processing. 1 or 2 stops is nothing to laugh at. It's what all of us are paying through the nose for f/1.4 or f/1.2 lenses. 1 stop is the difference between having useless shots on a f/2.8 zoom and a usable shot with a f/2 or faster prime. I think it makes sense to buy better gear for that extra 1 or 2 stops, but only if it makes business sense and pays itself off. For hobby use, it's not very worth it.
Thanks for your contribution!
A JPEG can represent any amount of dynamic range, but as you have stated, it can only display a contrast ratio of 1:255 (th-cam.com/video/aWGIjXutyKU/w-d-xo.html). The usefulness of having more than 8 stops is therefore highly debatable, as you've stated.
Regarding the f/2.8 f/2 - I think we have to keep in mind that we do not loose DR linearly when the ISO goes up. Therefore, even cameras with less DR can perform well at high ISO. I can recommend the charts on photonstophotos, very insightful I'd say. The one stop can be useful if you mess up the exposure, or you want to capture one stop more.
@@ThomasEisl.Photography Yes, I always reference photons to photos. Great site. Indeed. The DR loss is not linear. In fact, after factoring in equivalence for DoF, for most average shooters in non-extreme conditions, DR does not pose a huge issue as even the new sensors perform about the same. e.g. 6400 ISO on A7iv vs 3200 ISO on a crop sensor.
I think an important consideration is, when does the image become unusable for your application. FF, APS-C, … they all fall apart… is there a difference?
Absolutely, there will be a detailed examination of this in the future.
actually dynamic range is just one factor. If I'm shooting in the m43 format, ISO 200, f2.8, I would need to stop down on FullFrame at least to f5.6 and increase ISO to at least 800 to keep the same shutter speed. And there - the dynamic range is the same. Also for static landscape or architectural scenery - I'm happy to use Olympus' HHHR mode or LND64 to reach really high DR. So yes - the FF camera (or larger sensor cameras in general) has better DR but one needs to know how to use it and when :)
Could not agree more - the whole equivalence thing works both ways!
I don't know about that, i can use my sony on iso12800 and have better shadow recovery than mft.
@@cristibaluta you missed the point, but obviously, you will get better "shadow recovery" from a larger format camera on the same ISO settings
Why isnt the highlights blown out in pictures taken with the Nikon D5000, but it is in the Fuji X-T100? Are you saying DR is about something else?
Hey and thanks for asking:
So you definitely ran out of DR in the highlights with the X-T100. The problem is that manufacturers can "put" the gray point and therefore the exposure meter reference point wherever they want. It seems that Fujifilm decided to put it more to the highlight side of things than Nikon, who might have put it more in the center of the camera's DR.
It still could be the case that the X-T100 has more DR at the end of the day, but it is just more on the other side. Hope this helped!
I think you got a few issues about DMs wrong, the R5 in photos to photons is 11.85 and the GFX100 is 12.3. And all DMs are base ISO so when you push the ISO you loose DM. If you start from 12 you have 4 stops to spare till you reach 8, if you start from 8 you don't have much to spare. And you need DM for post processing. I also don't know where you got the idea that you only need 4 stops of DM for an evenly lit subject, so pls elaborate.
Thanks for contributing!
I did not mention the two cameras you mentioned at all in this video.
DR and ISO are not related linearly - therefore you do not loose one stop per iso step.
For all other questions, please watch my video on Dynamic Range in photography 📸
@@ThomasEisl.Photography Tnx i watched the said the video. The evenly illuminated, 2:1, etc story you tell has to do with the variation of illumination that you created in your scene/subject subject (e.g. one cannot 'see' the texture of the black suit/fabric). Nothing to do with the sensor. I also never claimed that ISO v DM is linear, each sensor has unique characteristics. Eg the one in R5 has two ‘bases’ at 100 and 400 ISO. And some manufacturers claim to have ISO invariant, etc. which is bogus but we have to accept that some sensors are more ISO-invariant than others. I also appreciate you did not mention the camera models I reported, which is why I reported them as they are contemporary ones you excluded and ended up as a result drawing the wrong conclusion. You also have to raise ISO for a variety of reasons and not just those you mention, e.g. you have slow glass, you want a large depth of field, you are in low light and do not have a tripod, etc. I shot film, slides, crop, and FF for 30 years and DM of 8 doesn’t cut it for me and for a lot of other people. If it's good for what you do and how you use your cameras, etc. fine by me but that's not the full or the only story.
Thanks for this comparison video. I have a reasonable understanding of sensor size, lenses, f-stops and light. I know that a Sony A7SIII is going to get a "better" image than my mid level Lumix G85, but to what purpose? What are you shooting, and how is it to be used?
I would say that most any camera produced by any of the major manufacturers in the last 8-10 years will produce a good/usable image & video. I use MFT because of the form factor & weight. The camera fits my hand perfectly. My results are good enough for my needs. Camera companies are in the business to sell cameras (just as are car and mobile phone companies.) There is always going to be a bigger, better, more shiny new and improved model of something.
Cheers. Peace.
Absolutely, Randal. Your comment is spot on - cameras have been good enough for many many years now, and let's not get started about film, which was also good enough and still is.
Thank you for your educated contribution, much appreciated!
Thanks for the video. But, I've always looked at sensor size in terms of high Iso noise and depth of field...
You will love my new video then!
Great Video Thomas, you explained the topic clearly. Well done :)
Many thanks!
Good day, Thomas. It's a good video with good information. I do have one quibble with you on the number of stops per image needed for creating an HDR image. You state at the 12-minute mark the need to take images at one-stop increments to make an HDR image with the older cameras. This is not true. Even with the older cameras, only three images (-2, 0, +2) were sufficient (not five). And this was if you were shooting JPG or raw. As you know, an 8-bit image, such as a JPG, has about 7-8 stops (depending on the signal-to-noise ratio, but that's getting away from the point). Around 2012, I was getting annoyed with folks showing their ghastly, over-saturated HDR images and bragging about how they shot every third stop or half stop. So, I did an experiment. I set up my camera on a tripod, and I shot from -2 - +2 at every half-stop. This gave me nine images that I saved as JPG images. I then processed them in Photomatix (the best HDR Tone-Mapping software at the time) and processed them all with no adjustments. I did -2, -1.5, -1, .-5, 0, .5, 1, 1.5, 2. Then I did -2, -1, 0, 1, 2. Finally, I processed -2, 0, 2. The result was that there was absolutely no difference between any of the resultant photos.
But, as you stated, the newer cameras do have more dynamic range, and because of that (and Chan from Adobe (their raw image master)) states that all you really need is the -2 and +2 shots, as that provides more than enough overlap. I have confirmed this, but I still mostly take all three shots because I usually hand-hold my camera. If the +2 shot is blurry due to camera movement, I can always fall back on the 0 shot.
Nonetheless, I am always glad for things to be tested before being convinced of something, and I thank you for your observations.
Alles richtig - alles perfekt und didaktisch geschickt und immer in Bezug auf die Relevanz für die Praxis erklärt! Entspricht auch meiner Erfahrung. Wenn der Kontrastumfang einer Landschaftsszene, - klassischerweise: Sonnenuntergang, Himmel gleissend hell, Vordergrund schon im dunklen Schatten - sehr hoch ist meistert das auch kein noch so grosser Sensor, man muss zu HDR oder zu einem Verlaufsfilter greifen wenn man noch Zeichnung in den Lichtern und auch in den Schatten haben will.
Aber die Hersteller wissen, dass höhere Zahlen - seien es mehr Pixel, mehr Stufen Dynamikumfang oder eine höher Kadenz - ein wirkungsvolles Verkaufsargument sind, auf das eben viele abfahren, weil sie zu technikverliebt und zu wenig Praktiker sind. Insofern kannst du dich als Aufklärer sehen, als Mann der aus dem wirklichen fotografischen Leben kommt, der die Kirche wieder in's Dorf stellt und sagt was Sache ist.
Und je mehr Videos ich von dir anschaue desto weniger Gründe sehe ich die für Vollformat sprechen ... Vielleicht teste ich nächstens die OM-1 bei meinem Händler in Bern, reizen tät es mich schon. Inzwischen habe ich nämlich herausgefunden, dass man die Bildqualität noch erheblich steigern kann wenn man die RAW- Dateien mit DxO Pure RAW 2 entwickelt, statt mit Adobe RAW - unglaublich was man da noch herausholt.
Vielen Dank auch für diesen fachkundigen, umfassenden Kommentar. Das weiß ich sehr zu schätzen, er kommt aus der Praxis eines Fotografen, das merkt man sofort.
Probieren würde ich die OM-1 auf jeden Fall - vor allem wenn man die Features richtig nutzt kann man wirklich extrem viel aus der Kamera rausholen. Ich finde, dass man diesen "Kompromiss" durchaus eingehen kann, vor allem wenn man eine kleine, leichte Ausrüstung schätzt und damit vielleicht sogar noch mehr fotografiert.
Danke auch besonders für die Tipps hinsichtlich Bildverarbeitungssoftware!
@@ThomasEisl.Photography Ja ich war wirklich verblüfft über DxO Pure RAW-2 - ein Versuch lohnt sich! Die Software produziert rauscharme Dateien ohne dabei die Auflösung zu reduzieren, jedes noch so feine Detail wird aus den Daten geradezu herausgemeisselt, man könnte in's Schwärmen kommen 🤩
Habe heute mit der Lumix GX80 und dem NullAchtFünfzig Kit- Objektiv 12-32mm/3.5-5.6 - wahrlich nicht das Prunkstück im mFT- Objektivpark - einige Aufnahmen von ganz verschiedenen Motiven geschossen und das Ergebnis hat mich fast umgehauen!
The DR difference is real. It's there and this can't be denied. Do you need it? Do you want to pretend it's not there? Do you want to tell yourself it doesn't matter to you? That's all personal preference and may be completely valid depending on one's personal situation. Personally, I've gone back to FF. I'd much rather have the DR and not need it, than need it and not have it. But like you say, that's a whole different subject.
Thanks for the comment - completely agree!
You didn't mention the effect of image format. Should I assume that you automatically meant [.raw] format? And that other formats (especially [.jpg]) would not convey the same dynamic range?
While I'm definitely referring to RAW here, keep in mind that the RAW image format does not necessarily increase the dynamic range (but often only improves gradation smoothness) more on that here: th-cam.com/video/uYOr6t8llgc/w-d-xo.html
Thanks for this summary, The one place where more dynamic range is essential and how I learned it was a thing was with time laps photography. When your camera is set, and you want to recover day-night sunsets without significant incremental adjustments in aperture or shutter speed, more dynamic range is critical.
How would you get around that if the camera has limited dynamic range?
Very interesting! Although a very specialized case, I think it is relevant to keep that in mind nonetheless.
@@stevenlang7709 newer cameras allow aperture ramping while using the inbuilt intervalometer. After that, the way I know is post-processing in LRtimelaps. I've seen a few impressive Davinci Resolve tutorials but never tried it.
What about advances in auto white balance and auto focus? I have a dad era Nikon D3400, and I often do not like the off cast it produces in snaps.
There were some improvements, but for professional applications I would always set WB manually. So for me it is not really a deal breaker/must have.
Does anyone know the D850...it's incredible dynamic range .. I don't know the numbers but I have shot photos in long exposures we're of underexposed or made a mistake in calculation and the photo has been almost black and I have been able to recover and image I would be curious to know where the Nikon d850 sits I know it has more dynamic range than the Z7 series cameras as I have tested those and they aren't as good
The Nikon D850 has 11.63 stops of PDR. With that, it outperforms even the Z9. DSLRs have great sensors and the recent mirrorless hype has often overshadowed how good these cameras are in terms of raw image quality.
@@ThomasEisl.Photography hi, I have resisted in getting a Z9
I’m mainly shooting landscapes but I just could not see the advantage in a Z9 over a D850 when you consider the dynamic range and the megapixels. They are very similar. Yes I understand I make it slightly sharper images with the Z9, but you can always add sharpening in editing anyway
I just couldn’t come to the conscious decision to pay an extra $5000 for a minimal improvement in image quality, I am a professional landscape photographer in Australia- not full-time gig. I now have a Fuji film, XT5 and a Nikon D850 I have the best of both worlds. Do not need a Z9
Because I'm a journalist, dynamic range is critical to me. I have an Olympus E-M1 Mk I, and a Sony A7 IV, and I often have to pull down highlights on faces and heads, while bringing up shadow detail. So the wider the dynamic range the better. I can say that dynamic range measurements are often much different in practice, since shadow noise isn't always reflected in those measurements, and my Sony blows my Olympus out of the water in that response, and my Ricoh GR III is better then my Olympus. Usable dynamic range is different than just dynamic range.
Thanks for sharing your experiences!
@@ThomasEisl.Photography I think that dynamic range in practice isn't unlike the difference between measured ISO and E.I. Whereas Sensitometry might measure a film speed at one rating, if you do your own work with the film, often there is a mismatch. I worked on a couple of projects for Kodak, one of which resulted in the decision Kodak made to label T-Max films with an Exposure Index derived from use vs. only sensitometry. With new AI masking tools, dynamic range becomes eminently usable, for all types of situations. And ISO absolutely figures into use, since "grain" isn't linear, as it tends be worse in shadows. So much so, that on older cameras, it's almost useless to attempt it.
I'd love to see a methodology that reveals useful dynamic range. Astronomers can demonstrate sensor sensitivity in low light and extended exposure. So, I'm wondering if there is a way to utilize that approach?
is the scale linear? also, the more stops, the merrier? 😅
Well, it depends how you look at it.
And yes, the more the merrier - however, there is an upper limit to what makes sense in practice, I'd say.
Most impressive video. I enjoyed and took in every frame. Thank you. So Clever is your presentation.
Tom, this is really a very nice comment. I am honored and you made my day. Thank you so much!
Hi Thomas, I am from Tyrol, I now live in Reunion island, plenty to take pictures of here. I have been an amateur photographer for over 30 years. Currently I am doing a lot research to buy a new camera but I just don't feel like I am near a decision. I have been using a Canon 6d mk i for the last year and I don't see that even a €4000 camera from Sony does give noticeable improvement in terms of dynamic range. I want to buy a hybrid camera so at the moment I am looking at the Sony a7iv but I don't like the rolling shutter.
To be honest - you should definitely not upgrade because you get better image quality. Because the differences are practically non-existent for all practical purposes. Better get a 5DIV or a 5DS you can use all the lenses and they are absolutely outstanding cameras. Don't go to mirrorless expecting better image quality. You might even have to deal with some negative issues you never heard of - like heat-induced noise and heat-induced camera shutdowns in super hot summers.
Thank you for the nice comment and best wishes to Reunion! Hope you don't miss the wonderful mountains too much!
@@ThomasEisl.Photography Thank you for the quick reply. Do you have a video which explains the difference between Color Depth and Dynamic Range?
Interesting points. You gave me a flashback to the zone system, but that is besides the point. A really good assessment. Thank you Thomas. Shoutout from Denmark. /Per
Many thanks! I think you are making a very valid reference here when mentioning the zone system. Correct exposure is as important for digital as it was for analog.
Best wishes from Vienna!
How would compare monochrome versus colour?
There will be an in-depth video about that - stay tuned! A monochrome sensor definitely has certain undeniable benefits - but also drawbacks.
HDR is no panacea for less dynamic range. Take it from a guy that's shot M43 for 12+ years. The HDR feature in LR or PS will make your shot look like mud in most cases. Bracketing over 5 stops have been my solution but is a PITA. Dealing with all the files can be a headache. In fast changing landscape circumstances bracketing is a hindrance. Manual exposure blending is the way to go rather than auto HDR selecting 2-3 of best shots but if you have a couple extra stops hard core bracketing would be avoidable and the blending chore!
Yes, there are many things a lot more pleasing than HDR haha - the problem is that you really end up in HDR territory really quickly, regardless of which camera you are using.
BTW, I'm working on an interesting video about HDR for professional photography with the OM-1 - I'd love to have your feedback when it comes out. Cheers!
Still keep my pentax k1. Does everything i need (macro photography and portraits) no need to upgrade.
Well, the Pentax K-1 has one of the best sensors / imaging pipelines even by today's standards. It is not easily outperformed by anything. Such a great camera, I'm considering one as well to be honest. Thanks for sharing, Anna!
@ThomasEisl.Photography the price is right for k1 now. I also love some old film lenses i can use with it
Yes, that is true! That is the K mount bonus, right!
@@ThomasEisl.Photography You won’t be sorry, Thomas. I still even have my k-5. Never selling my k’s. Brilliant vid, fam.
Thanks!
When digital sensor came big challenge wasdynamic range.
Good topic and answered in video..
If we understand deeper problem as u said
Can solve by composition and composite or graduated nd etc gives Solutions.great job!
Agreed!
Thank you for the kind words!
I think often the problem of dynamic range and small sensors is more tied to low light performance. If the sky is so bright that you have to underexpose the foreground elements to preserve the highlights, sensor size makes a big difference to how much usable shadow detail you can recover, because it comes down to the low light performance in capturing the lower light parts of the scene. Yes you can use CPL and NDs in many situations, but not always, and you can’t bracket if you have moving elements in the scene.
Yes, but one or two stops (which is the actual difference) does not help with HDR scenes. Regardless of format, you have to resort to tricks.
Maybe that is of interest to you as well: th-cam.com/video/jJQgJMCQJc8/w-d-xo.htmlsi=wfVvx_TMfzZkZ5bm
Regarding high-iso situations, one may find oneself a wildlife photographer in low-light conditions needing to raise their iso, which is a situation where dynamic range makes a difference.
The great thing is that practically all cameras will be able to provide 5 stops of DR at even high ISO numbers. And those 5 stops are more than enough for wildlife.
Thanks for contributing!
@@ThomasEisl.Photography I understand that. However, in my experience, 5 likely won't be enough. I have done portraits with people in the shadows and a sunny background. The images were exposure balanced and at iso 100. With those conditions, my background was blown out. DXOMark claims 11.9 stops of dynamic range to my camera. In higher-iso environments, this might be more prevalent.
Yes!
Just a food for thought - measure the DR in those situations you have just described with a light meter - it is pretty likely that even 15 stops of technical DR won't be enough to avoid blown out stuff in these situations.
A bit of flash, however, will!
All the best, Thomas
How are you getting 32:1 brightness ratio as 9 stops? Isn't that 5 stops (2^5=32)?
Thanks for asking:
An evenly lit subject, no reflections - requires about 4 stops of DR for the full tonal scale to be reproduced. If you increase the lighting on one side by 5 stops (32:1) you now need 9 stops to reproduce its full tonal scale on both sides. The full version is here: th-cam.com/video/uYOr6t8llgc/w-d-xo.html&si=EnSIkaIECMiOmarE
Yes - it is a bit misleading and you are the second person to ask, rightly so! It is because in another video (th-cam.com/video/uYOr6t8llgc/w-d-xo.html&si=EnSIkaIECMiOmarE) I staged a sample portrait session using lighting ratios (but not the contrast ratios) to illustrate the DR you "actually need". In any case, we are totally on the same page here, thanks for getting in touch and engaging. I really appreciate that!
@@ThomasEisl.Photography I'm watching that video now and it's excellent. I wasn't aware of the base 4 stops needed to begin with (for the evenly lit scene), even though you did mention it in the above video too. Thanks for your responses and for these videos. I've been shooting for 35 years BTW, so there's always something new to learn regardless of how long we shoot.
Very clear. How does film compare? Do films vary more in dynamic range?
Hello Ken!
This is a very relevant question - I talk about that in detail in this video: th-cam.com/video/uYOr6t8llgc/w-d-xo.html
I was geeking out about a new camera and my grandson, who also is a photographer, accused me of being a spec sheet masturbater. That stopped me cold in my tracks but it did get me thinking. I've actually only once had a client that questioned what equipment I was using. I've never had a client that could tell the difference between my digital Canon images, images from my Olympus EM1X or ones that were from my Hasselblad 503CX. In the end my experience is that good composition, good exposure and good editing of the images is way more important than which camera or sensor or how much dynamic range I had.
Thanks for sharing that Sophie - and I completely agree with what you said. As long as the full tonal scale of the main subject (usually 4 stops) is reproduced, the difference in DR does not matter at all.
What's the dynamic range of a print ? 6 stops maybe. The little Ricoh GR 3 has a dynamic range of 11.5.
Absolutely - regarding print, completely correct. The thing is that tonal compression should be considered as well, so you can "kinda" compress more stops into the limited print:
th-cam.com/video/uYOr6t8llgc/w-d-xo.html&si=EnSIkaIECMiOmarE
In any case, even with 8 stops of DR everything is usually perfectly fine!
Source of those dynamic range values please.
photonstophotos - mentioned right at the beginning, you might missed that
@@ThomasEisl.Photography thx!
Completely right about dynamic range. I do think real progress has been made with ISO invariant full frame sensors introduced 5 or 6 years ago.
Thank you very much for your comment!
I recently did a very insightful research into ISO invariance - to find out, that almost no sensor is completely ISO invariant, although many ppl believe so, including many reviewers. The results are very hard to spot in practice, but the charts on photonstophotos clearly show that.
Sure, all modern cameras are very good in DR, even 1inch sensors.
But I realized that APS-C is the minimum sensor for me, because I tried MFT cameras, but there was something wrong with rendering of such nature objects like water, clouds and greenery (if you pixel peep a bit). Looks like a sort of tonality lack. Even the newer 20mpx MFT sensor, I tried RAWs from Dpreview, but again, didn't like it. So m43 is very good for travel, street and architecture photography, but not for landscape, nature or portraiture in my opinion.
But I have to agree about the older sensors, even the 16mpx Nikon D7000 sensor from 2010 is good enough in most scenarios today.
Color reproduction and so on is really very subjective. There are so many aspects to consider: Which program was used, was the file actually exposed correctly, and much more.
Thank you very much for sharing your views, and I agree: The D7000 has an awesome sensor.
I've come to the same conclusion. And I am now hunting for information why I see it that way. As this video and other data suggests, the dynamic range is enough. I wish that the small vs big sensor debate would focus more on the difference in tonality, micro contrast, lens technology etc. Instead we are getting caught up in how much of an underexposed photo I can recover, or how the bokeh with a smaller sensor.
@@mumrik Probably it's important to us to see the photos which are closer to reality and looks cleaner. I also find that Canon APS-C photos look better than M4/3, but anyway they look a bit "rough" (I don't how to explain it correctly), Canon sensors just a bit noisier even at base ISO. So Nikon, Sony and Fuji files looks cleaner to me and just right.
Within the Canon line, I can affirm that the sensors on the mirrorless R series are vastly superior to the one in my old 60D but I cannot speak as to the difference on the higher-end Canon DSLR cameras. There is however, something to be said about older cameras. In fact, there are certain times that I still use my 2008 model Fuji S1500 with it's 10 megapixel sensor to shoot landscape images. The primary reason for doing this is to be able to set it up with the "FujiChrome" settings to produce some seriously intense color depth. I could recreate the same thing in Photoshop from my Canon mirrorless RAW files but sometimes it's nice to just "set it and forget it," as George Forman has stated in his well-known ads for his grill. Also, there are times when I do not want to lug around a full size camera on a long hike. The relatively small S1500 can easily be stuffed into a coat pocket or one of the exterior water bottle sleeves on my backpack. Best of all, should I happen to lose or damage the old camera, no big deal. I can easily get another for under fifty bucks!
Thank you very much for sharing that - and I completely agree: old cameras are still awesome. It seems that manufactures put a lot of thought in the picture profiles and jpg processing back then. Also, the reviews when these cameras came out often focused on this aspect. I also really like to use them nowadays, as the results still hold up, if you know what you are doing.
Thanks again
Dynamic range does come into play when it comes to shadow detail recovery under low light conditions.
For the average photographer, any camera these days is prob good enough. But for the demanding pro's there's a host of factors to consider when it comes to technical specs/design of the sensor and performance.
Plus there's a number of technical factors that go into sensor design and performance as it relates into practical shooting, raw data capture and final image quality output. Sensor size does matter to a degree, but more importantly pixel pitch and photosite size, not to mention the A/D converters and computational algorithms that process the incoming analog signal to digital signal and how that signal is being processed and recorded as raw data.... just something to keep in mind and consider making additional videos on these discussion topics.
Keep up the production, you're doing great!
May I ask which pro raises shadows?
Hey Stefan!
Thanks for the comment and sharing your thoughts!
Many factors like the one you mentioned AD / processing and so on are often overlooked, you are right - that is why the Pentax K-1II gives you better DR than the D800 with practically the same sensor, right!
Thanks for engaging, the suggestions and the kind words!
@@weizenobstmusli8232 don't have to be a pro, but it's dependent on what you're looking to achieve in your final image... how one shoots/captures an image in-camera is not how the final image is processed... the goal is to capture as much of highquality RAW data as possible in camera to expose in-post for the final image rendition...
as a general rule one should always ETTR without blowing the highlights, unless the final composition calls for otherwise...
Frankly, I'd never recommend a micro 4/3ds sensor camera to anyone... but to each their own... (right tool for the right job principle always applies)
Camera gear is just a tool... regardless of the brand or censor size... but one should always do their own research based on their needs application... and the technical benefits/limitations of the hardware/firmware specs... also, let's not forget the ecosystem as a whole for future expansion potential... which is something most people getting into photography do not consider.
I’ve been rolling my eyes at the Dynamic Range warriors for years.
Yep 👍 rightly so I'd say
I have been shooting with Sony A6700 for around a year since its release and looking to upgrade to FF A7cii because of low light performance and dynamic range as I got more into landscape and occasionally astro when opportunity arises. With your video now I am not sure anymore if I should upgrade to FF + more expensive lenses!
Thanks for sharing - great to hear that I was able to offer a different perspective. I'm sure that your A6700 is definitely (more than) good enough in terms of dynamic range and image quality! Best, Thomas 📸
@@ThomasEisl.Photography I am surprised too honestly that it holds up well but I guess it's because of the lack of experience that led me to believe that FF is going to be much much better as per many people claim it to be. As you mentioned HDR scenario can be overcome easily with multiple exposure blend and I have done that many many times too in Gimp. If FF is not going solve this issue then I guess I would not waste my money on the camera and, more important, the more expensive lenses. Let's not forget the fact that focus bracketing is more likely to happen in FF for landscape shot I usually do. I rarely have to do that on this A6700 but I was told that I would do it a tad more frequent with FF due to shallower depth of field.
@@ThomasEisl.Photography By the way, found your channel for the first ime and subscribed immediately after I finished this video. Thanks for the insightful explanation on DR which has always been a myth to me!
@rsat9526 Better get an excellent lens for your existing camera, if necessary. That is always worth the investment!
@rsat9526 Thank you very much, that is great to read! Welcome to the channel, then!
Well, it's not just about DR, is it? Newer and bigger sensors also come with improvements in other areas: higher resolution, better image processing/colour science, better AF, better noise control, better EVF, better screens, etc. Obviously, bracketing can replicate greater DR, but it adds more stages to work flow, doesn't it? You can mimic the DR of the Phase One with a lot of bracketing with the D2H, but that's not going to get you the high resolution the Phase One offers, is it? Don't get me wrong, I still use an old Pentax K5 and I am happy with the image quality.
Absolutely, the main point of this video was to illustrate that 1) 8 stops are usually enough and that 2) when you need more (landscape, etc.) you usually need a lot more than any camera has to offer.
Of course, the D2H vs Phase One was tongue in cheek. At the end of the day, more is always better, but it is important to find out how much is good enough. That is probably exactly the reason why you are still working with the (btw fantastic) K-5.
Great video thank you for the enlightenment, I am looking forward to some more videos
More to come - thanks for the feedback!
What is the maximum dynamic range possibility of a jpeg file ?
This question has no easy answer, but in this video I talk exactly about that: th-cam.com/video/uYOr6t8llgc/w-d-xo.html&si=EnSIkaIECMiOmarE
the simple answer is that a digital file has "bit depth" in its data elements that define what I call "gradation resolution". But this informs nothing about "dynamic range".
You could use the 8 bits per colour channel in JPEG to print your image within 4 EV dynamic range that looking at paper is likely to give you. When you depict the bits in a much larger dynamic range (contrast envelope), then you'll most likely see "banding" from your JPEGs.
Imagine a rectangle hypothetical reference card - one square pure black, the other square pure white. Note that pure black reflects no light and by definition has no colour and so there is only 1 type of black. Note that pure white reflects all light and by definition has no colour and so there is only 1 type of white. (I.e. "blacks and "Whites" in Lightroom are a misnomer - they're all shades of grey.)
Imagine that reference card in a completely dark/black room. You don't see it at all - it's all black and contrast or dynamic range is zero. Now slowly increase the light level and the black remains black while the white becomes lighter and with that contrast increases. Dynamic range in images is a bit like that card. The contrast you get from it depends on how you work with it.
The "possibility" from your question hence is not in the JPEG file but in your brain and hands.
The problem with JPEG is its low bit depth and if you depict your images on a medium that has better potential you may see banding. This happens when a subtle continuous gradation (change of tone) in the subject was not captured in enough discrete levels (because that's what the bits give us), or you processed you imaged with insufficient bit depth. In such a case, a continuous subtle gradation becomes a set of bands perpendicular to the gradation change's direction, where each band has no gradation change in that direction.
If you raw process with Adobe Camera Raw (ACR) in Lightroom Classic (LrC) then take the processed image from LrC into Photoshop (Ps) and convert it into 32 bits per channel. The effect is very well visible on a good 8 bits per channel monitor/display. Internally, ACR and LrC use ProPhoto colour space, by the way.
Imagine you now have 32 bits depth per colour channel and gradation steps are so refined now that you cannot see them as banding.
But nothing got changed to "dynamic range" at this "post" side of photography.
Note that raw processing has become more sophisticated with better AI and so the way gradation is converted from higher bit depth to lower JPEG is done smarter - the risk of banding is less today than, say, a decade ago.
Does hi-res mode increase DR?
Yes, by reducing image noise the usable photographic DR is expanded. Note that Live ND expands the DR even more.
Excellent video, as always. I especially like the thoughtful presentation. It would be helpful to hear/see an explanation of exactly how/when the 2 stop advantage of FF is helpful to have (post processing of landscape shots, maybe?). The conclusion here seems to be that neither the OM1 nor the FF can completely capture the entire dynamic range of some landscape scenes, therefore they are equally suited (or not) to the task -- and I doubt that is actually the case.
Many thanks!
I agree, that is the conclusion indeed. To be honest, I don't think that the two stops difference are very relevant in practice. Maybe my video on dynamic range will clarify why: th-cam.com/video/uYOr6t8llgc/w-d-xo.html
@@ThomasEisl.Photography Thank you for the reply and yes, that is another very informative video that points out that there are many factors that play a role in the quality of the final print. IIRC, though, in one of your other videos you stated that the OM1 has essentially 6 stops of quality DR, which would be less than the "8 Stops are Enough." I'll dig through my history and rewatch that one as well.
Many thanks - maybe to clarify:
You need 8 stops of total DR (realistically), with 4 stops of high fidelity DR at most.
The OM has plenty, as practically all other cameras.
What I define as High fidelity DR is the DR which can be moved to middle gray with full color information and little to no noise.
It sounds a bit contradictory at first, maybe this helped to clarify things a bit.
Many thanks for the excellent conversation!
@@ThomasEisl.Photography Wow, great clarification! Thanks again for the wonderful, informative videos and also for taking the time to read and respond to comments -- that sets you apart from many others on YT!
@tommilton6775 many thanks!
In my opinion dynamic range is the most important thing in a camera. Depending on the source the human eye has a dynamic range of 21-24 stops and this is the goal. Especially in Video, comparing footage of a the new arri 35 in log compared to recording in rec709 it is simply amazing. In photography in high dynamic range scenes or simply as a buffer for human failure is so important. Imagine a soccer game were the players are correctly exposed but the sky is white because of a bad dynamic range ...
I do not believe the human eye sees that many stops of dynamic range at all. We can only focus on specific things at a time. Whatever we focus on, we expose for (so to speak). Try having your blinds open a small amount, such that it's dark inside and very bright outside. Adjust your eyes to the interior and the window is blown out, just like it would be in a photo. Look at the scene outside and everything inside is pure black, just like in a photo. I think we have maybe 3-4 stops more dynamic range, but not up to 9 stops more (24 stops).
Dynamic range is indeed a very important aspect of every camera, I totally agree!
The big question is: how much is enough. While we all agree, more is always better, we can usually get the job done with less as well.
In videography, having more DR to work with is more important than in photography. You cannot shoot HDR brackets in video, right! Thanks for contributing!
@cooloox - I am also inclined to state that we humans do not perceive 24 stops simultaneously!
@@ThomasEisl.Photography yeah thatis true, sources differ a lot on that but it is actually the other way around you have less dynamic rang at the spot you are looking at. Because the spot where you have sharper vision. This is because of the fovea centralis which has more cone cells ant there fore more color perception. The rest of the eye has more rod cells and is more sensitive to light and has a better dynamic range. other than that idk i think around 21 stops are realistic the typical scene would be an interview setting inside front of an bright window. idk
but cool video man i enjoyed watching it!
@@ThomasEisl.Photography no problem, thanks for writing back! I love that topic! Imo I woul always have at least 3 stops more than i need. 1 to have safe space from the noisefloor, one for highlight and one for human failure and personally that works for me. i am pretty happy with cameras with 12 stops and up. And there are of course extreme scenes where i am happy about every stop i get out of the camera
Dynamic range is just one component involved in creating an image. The sensor read speed, the DoF, the number of megapixels, all these elements together will allow either to get a good frame or not to get it.
Absolutely!
Exactly what I experience when I buy the EOS M some days ago! Pictures are amazing. EOS M is king of 2023!!❤
Yes! Great to hear and thanks for sharing!
I like you're approach Thomas, but there is one huge thing you haven't covered: rasing shadows in post. I am a wedding photographer and I often shoot (the dance, mostly) at F1.8 and ISO 12,800. The image quality like that is good enough, but I often want to raise the shadows in post and then I have a serious problem. I used old cameras to do this and I've used new cameras and the difference is very, very noticable. Oh, and generally only use available light.
Great point! Plus there's a number of other technical factors that go into sensor design and performance as it relates into practical shooting, raw data capture and final image quality output.
Thank you very much Matt!
I agree - the problem is that you are actually running out of "high fidelity DR" - and therefore, you are brightening the shadows but only get noisy results. Technically, you are still working "within the DR" of your camera. I think this video would be interesting for you in this context:
th-cam.com/video/uYOr6t8llgc/w-d-xo.html&si=EnSIkaIECMiOmarE
@@stefanski8287 Yes Stefan! That is why it is so important to do a bit of testing with the camera you are using, to determine the limits.
Well presented. Though I think the number one consideration is this - if I really want extra details in my landscape images then buy a camera with more pixels. Otherwise e.g. Phase One and Hasselblad wouldn't produce expensive camera backs with millions and millions of pixels, even though they don't give significantly higher DR. So, it's the extra details that cost your socks off.
Thank you!
You are right that a larger recording format will resolve more detail, however, the relationship between size and improved image quality is not linear, see for example: th-cam.com/video/FVHTLFD_7o8/w-d-xo.htmlsi=y5DTRmNGmAxJ1hOj
It is quite surprising how little the extra detail matters in practice, as the output media is usually limiting the reproducible details.
Thanks again for the very valid comment!
@@ThomasEisl.Photography Agree, the extra detailing is primarily used for large printing or other large scale presentation. Hardly noticeable on even good screens - unless you own the format and can "feel" the details.
@Eigil_Skovgaard yes, and one thing that is really cool if the detail is there is zooming in on a high quality monitor. It is fun haha
Где твои примеры фотографий Томас? Было бы интересно увидеть примеры в видео
There is a other video about low light with small sensors ans dynamic range which illustrates the topic with photos and graphics. Best, Thomas
amazing presentation.....congratulations ....AND A BIIIG like from mewhat is your opinion about nikon z7markII? THX
Thank you very much for this feedback!
Coincidentally, I've used the Z7II last year for a few jobs, so my experience is not extensive. However - if you are looking for a high resolution mirrorless camera for repro work, portraiture or landscape, the Z7II is for sure a great camera. Very good build quality, nice handling. Solid video.
I ended up with my DSLRs nonetheless, as I just really like OVFs & DSLR focusing systems, plus the Z7II setup was not really lighter than my D800 setup and performed about the same in all (for me relevant) aspects.
Hope this helped!
By all this talk about the dynamic range of today's and yesterday's photo sensors, one must not forget that the dynamic range of perfectly normal consumer diapositive film (slides) was about TWENTY stops and not meager 10 or even less. Capturing landscapes never needed any sort of bracketing.
On a good day with a good lens the resolution was about 16 MP. The limiting factor was the graining of the film material.
The main performance drawback was the limited ISO. It was hard to get substantially above 200 ISO. For color 400 ISO was really stretching things.
Thank you for your comment!
I would rate the DR of film lower, but I definitely agree that it is a lot better than people assume! I think this would be of interest, it covers what you have stated:
th-cam.com/video/uYOr6t8llgc/w-d-xo.html&si=EnSIkaIECMiOmarE
@@ThomasEisl.Photography Very interesting. Thanks a lot.
The figures I remember are a little different. As color TV was introduced there was the problem of playing cinema films and color stills across analog color TV. The contrast was much too high. At the time a TV CRT had - very much depending on the amount of ambient light - a contrast ratio of about 1:30 - i.e. a little less than 2**5 = 10 stops. A color diapositive was given as accommodating a max. of 1:1000 = approx. 1:1024 = 2**10 = 20 stops.
From then onwards filming and photographing for usage in color TV programs had to be "special low contrast" .
10:18 "It wouldn't be too bad. We would just have to accept that some parts are clipping white or black".
It's like saying "It's not too bad, you just can't see some things in your photograph, so what."
Well, not really - I know this is what one would think initially, but if you examine how final images look and how humans perceive DR, then you will find out that it does not matter at all - see my video on photographic DR
Watch this; th-cam.com/video/uYOr6t8llgc/w-d-xo.html&si=EnSIkaIECMiOmarE
Interesting video, thank you. Digital photography was starting to make inroads with print media photographers just as Kodak introduced Ektachrome 100 Professional. Using Kodak's new (at the time) T-grain technology, E100 was the ultimate in color transparency film, the best ever, said to have a dynamic range of about 5 stops. It was great film, but missed the boat due to the emerging demand for digital, which became more cost effective in pre-press than transparencies. Anyway, for digital to achieve a dynamic range of 9 stops is a huge improvement over the best film-16X better, with each additional stop double its predecessor.
With film, the over-exposure/under-development method to increase tonal range can go too far, producing muddy images with too little contrast. If sensor dynamic range were to get too extreme, couldn't digital images also suffer undesirably low contrast?
You are absolutely correct about the low contrast issue - I invite you to watch my video on Dynamic Range in Photography th-cam.com/video/uYOr6t8llgc/w-d-xo.html&si=EnSIkaIECMiOmarE
Why do prints which have to represent 8 stops or more DR look muddy? Because they actually can only produce a contrast ratio of around 6 stops. The compression of dynamic range leads to this muddy appearance. Another reason why 8 stops fo dynamic range are enough for photography.
Thank you very much for your very valid contribution, much appreciated!
@@ThomasEisl.Photographywithout HDR video or picture formats on screens that can produce true HDR levels, even screens are only 6-7 stops of usable dynamic range. Only 2000+nits in a dark room OLED panel will get you that 10 stops of range with a proper contrast.
I has other system ( SONY ) And I keep my APS-C still because I can youse crop mode on long lenses, and is still good for backup, First and Second is Sony A7IV and A7RV and third always is Sony 6400, still 24MP camera and APS-C so with lens 70-200 on FF I have 105-400 on APS-C :) Maybe is not 2,8 but F5.6 but still good and with OSS I can catch nice captures. Good Videao anyway and proper explanation. All the best Thomas!
Great setup! Sony has a lot to offer and those are great cameras and lenses 👍
Thank you very much for sharing that, keeping an APSC is totally reasonable.
Thanks for the kind words!
@@ThomasEisl.Photography Not a problem, I did extra step now and bought teleconverter 2.0 SEL from Sony, and with 70-200 I get on APS-C 210-600mm F5.6 which is beast for affordable money, thank you for answer! have a nice day! Leo3City
One factor you did not mention, or perhaps even consider, in this analysis was the effect of pixel size on dynamic range. Larger pixels capture more light. Hence a larger sensor with more densely packed pixels could have either lower, or perhaps no better, dynamic range than a smaller sensor with less densely packed pixels. No?
The factors you named as well as other factors are the reason for different DR capabilities, that is correct
Great comparison Thomas. You’ve been killing it lately!
Thank you so much! I'm honored!
Very well said, thanks for sharing
Thanks!
Whenever somebody asks me what camera to buy I say "take any top end product from the last 12 years". I think the needs in pixelcount, framerate and autofocus speed matter more than sensor tech. And megapixels only for people who want to print large.
Absolutely! And even for printing large, not a whole lot of megapixels are actually needed.
I still love and often favour for size my Olympus 450 with a 14mm prime rather than lugging my indestructible Pentax around. I just watch the weather report first 😊