ETTR and ISO Invariance - [Should You Care?]
ฝัง
- เผยแพร่เมื่อ 18 พ.ค. 2020
- Should you care about ETTR and ISO Invariance. If ISO Invariance is something new to you. Maybe check this video out. I am explaining ISO Invariance and ETTR in the video.
👉 Subscribe to my Weekly Newsletter: bit.ly/PetersNewsletter
My Channel membership is now open!
👉 Join: / @forsgardpeter
Get a Hoodie!
👉 teespring.com/stores/peter-fo...
The One and Only Lens Hoodie!
👉 teespring.com/lens-hoodie
🎼 music from www.epidemicsound.com
Hovering by Jon Bjork
My Amazon Store: www.amazon.com/shop/peterfors...
Want to learn more about photography or using your Olympus camera. I have coaching sessions available in my Webstore:
► www.ajatuksiavalokuvauksesta....
Videos on my channel:
Photography tutorials:
📽 • Photography Tutorials
Best Settings for Olympus cameras:
📽 • Best settings for Olym...
OM-D E-M5-series:
📽 • Olympus E-M5 MKII
OM-D E-M1 MKII:
📽 • Olympus OM-D E-M1 MKII
OM-D E-M1 MKIII:
📽 • Olympus E-M1 MKIII
OM-D E-M10 MKII:
📽 • Olympus OM-D E-M10 MKII
E-M5 MKIII:
📽 • Olympus OM-D E-M5 MKIII
E-M1 X:
📽 • Olympus OM-D E-M1X
OM-D E-M10-series:
📽 • Olympus OM-D E-M10 MKI...
M.Zuiko lenses I have tested:
📽 • M.Zuiko lenses for m4/...
Video about the future of the photo industry:
📽 • Photo Industry - My Th...
Disclaimer: I am an Olympus Visionary. I make all my content with mirrorless Olympus micro four-thirds gear.
Links to my Olympus gear (disclaimer: using these links when doing purchases, you support this channel. You pay the same price and I get a few bucks to finance this channel.)
📷 www.ajatuksiavalokuvauksesta....
_____________________________________________
This video was made with the following gear:
My video gear:
Camera:
Olympus OM-D E-M1 MKII
Lens:
Olympus M.Zuiko Digital ED 17mm f1.2 Pro
Microphone:
Rode Wireless Go
Light:
Dörr flexible LED light
Memory card:
Sandisk Extreme Pro 64GB
Tripod:
Peak Design Travel Tripod
________________________________________
► Any professional inquires please email: peter@peterforsgard.com
Links to my accounts all around the web:
My portfolio: www.peterforsgard.com
My blog (eng): bit.ly/peterfblog
My Instagram: / jpeterf
My Olympus page: my.olympus-consumer.com/membe...
My Olympus gear: www.ajatuksiavalokuvauksesta....
My kit.co store: kit.co/PeterF
#Olympus #PeterForsgard
(disclaimer: Some of the links above are affiliate links. Using these links when doing purchases, you support this channel. You pay the same price and I get a few bucks to finance this channel.)
![Mastering Exposure - [5 Tips to Get the Right EXPOSURE]](http://i.ytimg.com/vi/33LPaj2rYhw/mqdefault.jpg)
![Mastering Exposure - [5 Tips to Get the Right EXPOSURE]](/img/tr.png)
![ISO - [5 Useful Tips]](http://i.ytimg.com/vi/X0WbpelfyaM/mqdefault.jpg)
Peter, it's good that you've raised this subject -- especially for those who work with jpeg files -- as it is important for digital photographers to understand that their camera's sensor is ALWAYS only operating at its "Native ISO". We can't really change the physics of the sensor's sensitivity to photons. All that happens when we dial up a certain ISO on the camera is that the image is post-processed within the camera to raise the signal amplitude to "effectively" raise the ISO of the camera-processed jpeg file. So, if you need to shoot at ISO 1600 to correctly expose the image within your preferred shutter speed and aperture (which you chose for the artistic effect) then the sensor sees an under-exposed image, which the camera's inbuilt electronics then post-processes to create an artificial ISO 1600 effect. It's that in-camera signal processing that creates the noise; quite simply, because the sensor wasn't hit by enough protons to create a perfect image.
[ Edit: I should have begun by defining "Invariant". It means 'unchanging'. So, Invariant ISO refers to the fact that your sensor will ALWAYS, or unchangingly, operate at its Native ISO (typically 200), whatever pseudo ISO you set on the ISO dial. As stated above, the ISO dial only determins how the camera's internal post-processing will treat the image when producing a jpeg file. It has absolutely no effect on the RAW files, if you choose to save in RAW, other than adding some meta data for your third party software to use. ]
What does this mean in the real world? Well, you have a choice. You can use your camera's ISO dial to have the camera boost the camera sensor's ISO 200 under-exposed image up to your preferred image brightness, or you can shoot at the camera sensor's Native ISO 200, then do the ISO correction in post, using something like Lightroom or Capture One.
Which is best? Typically, if the camera manufacturer is any good, then the camera's onboard software will do a fractionally better job than some more generalist third-party software.
As for those of us shooting in RAW. The captured raw image file will always be in the Native ISO 200. This may not be immediately obvious, as the file contains meta data which tells Lightroom or Capture One what ISO setting you had on the camera. Therefore, those post processing softwares will automatically apply the appropriate ISO correction to your image, and it's then up to you whatever other adjustments you want to make.
What does this mean for ETTR? Well, for the RAW photographer, ETTR is meaningless, as we can't recover what the sensor is technologically incapable of capturing. BUT , for jpeg photographers, ETTR is important, as you can never completely recover lost data from a jpeg file, and that detail on the right of the histogram can be permanently lost in the jpeg. So, for jpeg shooting, pay attention to Exposing to the Right or ETTR as it's known.
Oh, and by the way, the E-M5 MkII is a so-called ISO Invariant camera.
Rick
Ettr works just fine for raw as long as you know how far you can push it without clipping.
It's all about exposing the sensor with as much light as possible.
But it will never work good with iso change as you really do not let more light in with iso change.
@@photosvein Very true, but I was trying to keep this (complex) subject as simple as possible for the enthusiasts. My comments are generalizations, while the expert in post-processing can always do better.
Good observation.
Rick
In this particular test E-M5 MKII seemed not to be. I could be wrong.
Thanks for this explanation.
@@ForsgardPeter Various "scientific" tests suggest that all of the Olympus OM-D cameras are Agnostic. Whereas, in some manufacturer's cameras it's clear cut. Anyway, it was a fun and totally worthwhile evaluation.
Rick
Derek Forss, an Olympus mentor always under exposes and corrects in post. He produces spectacular images.
I do not agree that under expose will give you the results, but then if some one is happy then it is ok. I know his images and his style and nothing wrong with that. The correct exposure is the one that you are happy with.
Just my opinion, I think that the fear of noise was a ploy to sell full frame cameras. Most of my photographic life I dealt with film grain, which I usually liked because I liked the texture that it added. I often add grain to my digital files for a better look, especially my B&W photos. Now about ISO or I.S.O. [Should you care?] I don’t think so; at this point they are very interchangeable. As an American that can only speak English I would never criticize someone that is bilingual, I have my own problems with English. Love your blogs, love my Olympus camera, keep up the good work. Stay safe, Lee in Dallas, TX
Thanks.
Peter, thanks for a great discussion. I have only just stumbled on this video as I work through your library, and as I investigate ISO invariance with my EM5 III. I take night sky landscape photographs and this is an area that is a real test for M43 (well according to the full frame world). ISO invariance, if it exists, is useful for this type of photography by allowing for lower ISO to get higher dynamic range, and avoid the reputed noise problems of M43 at high ISO. I conducted a test using the approach by night sky photographer Alyn Wallace and found my EM5 III ISO invariant from ISO 800 up. I then got bluntly told in another M43 forum that no M43 camera can be ISO invariant, due to design, but they did not offer any further explanation. All very interesting. On my next night sky outing I am going to try both an ISO invariant approach (ISO 800 and crank the exposure in post), and a traditional approach of higher ISO and noise reduction in post to see which is better.
Anyway, thanks for your excellent and informative channel.
Thanks. Glad you enjoy it.
Hi Peter,
Good stuff as expected. My photography is simple. On camera, I use compensation, aperture, shutter speed. I am a day shooter and use the lowest ISO. In post, I change highlights, shadows, whites, blacks, and contrast. That’s it. I just look for what looks good to me, using composing rules more or less. You have seen my stuff and I think I do pretty good 😊
🤓 Thanks Peter, I need to try this out😃
You should!
Should you care ? Do I care ? No. ISO, while useful in certain respects, is the bane of digital photography if one thinks about it too much. After watching Robin Wong's video last evening and now yours, I'm perplexed by the need for the conversation. As you've both illustrated, while coming at the discussion from opposing perspectives, the ultimate conclusion seems to be that none of the ETTR hailstorm really matters. Regardless of the in-camera protocol results, post production techniques can return images with little 'to the eye' difference. The world would be simpler if a camera came with one fixed native ISO, and we were forced to properly expose images based solely on shutter speed and aperture. Just like in the pre-digital film days.
Totally ISO less camera would be great.
So always use ETTR, if it happens to coincide with shooting at ISO 200 most of the time (really strong highlights, such as a full moon by day and by night), so be it. Yet shooting at ISO 200 no matter what is kind of silly. Thanks for the insight.
Love your props! Olympus Square! That was great. I want a Ferrari camera.
Thanks Rob. Make a search in eBay and see if there are any on the market.
Peter, I'm glad that you did the test, and I can add another camera to your list: The Pen F Digital is NOT ISO invariant! I found that (in low-light situations) it is crucial to not underexpose and better raise the ISO setting, or otherwise banding in the shadows creeps in (a clear sign that they are changing the analog gain with ISO, at least to a certain extent and for high ISO, maybe in larger steps). Sadly, most camera manufacturers (Olympus included) keep all this technical detail from its customers. Is it important or useful? Sometimes it is.
Your set up made me laugh. Thank you!
:D
Oh, now I wanna test my M10 Mark IV!
I have tested this on PEN-F with RAW. Time and f-stop the same, only ISO changed. On the underexposed photo the noise is a bit higher but the difference is only visible if zoomed to 200%. Nothing really to care about.
I use ETTR with but I always change exposure time. F-stop remains constant.
I had this discussion with Robin Wong. Like you, I think there are benefits in increasing the exposure in post and take the photo with the base ISO. You have more control over it (masking). Of course, there are limits on how far you can go.
With some Olympus cameras it might be sometimes wise to use the base ISO200 and raise the exposure in post.
Well, I'm going to have to test for myself now. On my old Nikon, under-exposing & raising exposure in post was a bad idea. Your results with Olympus seem at odds with Robins. Ultimately, I think I'll prefer to shoot at correct exposure regardless. Anything else will feel wrong at time of capture.
Correct exposure is the way to go. ETTR does not mean overexposure. Overexposure is making the image look bad and cannot be corrected. ETTR so that the highlights are not clipped is a good way.
OK as a scientist its all kinda sounds superfluous, I think I will just stick with what I do now, fundamentals, works fine for me and I am very critical.
Thanks for the test.
Probably my lack of good processing skills but I find lifting an under exposed image turns out a little bit flatter, lack of contrast possibly, don't know.
I have tried ETTR but only using exposure compensation to push the histogram to the right without clipping and I think it works. Never had to change the ISO for the ETTR. Differently lit image types seem to give me different results. Say close up image compared to landscape. but then my skills are not that good.
I think Robins examples were more towards the extreme side compared to these examples, but that is good as it shows results from a different perspective.
Could be that the images look a bit flatter. Then just add some contrast to make it look better.
Changing the ISO not technically ETTR, but it seems to work with the same logic.
Robins example was interesting and not sure if he actually used ETTR in his images. The exposure with ISO3200 was 0.0 and very seldom when using ETTR that is the case. The histogram was not shown. The ISO6400 was just +1 and on that one too no histogram. ETTR is not overexposing, it is making the exposure correct.
The issue with high iso in low light is also false colors and w.r.t ETTR , the later one is actually more suitable for landscape and higher dynamic range.Anyway both ain't gone compensate if one sit to far off with exposure.On some camera's such as Nikon D4 is more easy to push and pull exposure in post then on micro loser camera's.
Good video with one of the most important theoric and practical issues in digital photography. Mr Robin Wong has talked about it few days ago. I thing you both should record a face-to-face video about this topic. At what time, it's up to you :-)
We could do that. Lets see what we can do about it.
An update, the M10 Mark IV is ISO invariant!
I compared the same photo with different levels of ISO from 200 to 6400, at 200 there's a slight magenta tone when you increase exposure through Lightroom, but it disappears after ISO 400
The noise level doesn't really increase until you go beyond 6400, at extended ISO's though you see how the dynamic range decreases after 3200!
Greta news. Thanks for sharing.
Hi Peter, you Day it is hard to compose in low light situation with iso 200. But have you try live boost ON2 mode?
Yes, LiveBoost is a good way to see better through the EVF in low light. But reviewing the the images from the LCD is hard. Also with Live boost it harder to make the correct exposure.
Peter, from the outset, you mentioned that the Invariant ISO technique is really something for the Landscape photographers, and perhaps less useful in photography in general. You then omitted to say why it is useful in Landscape photography. As someone who does a lot of Landscape photography, and who uses Invariant ISO exposure myself, I think it'd be useful if you explained how and why Invariant ISO is useful in this genre. This is your tutorial, so it's for you to explain how/why Invariant ISO and ETTR is so useful to Landscape photographers.
Thanks,
Rick
ISO certainly affects my RAW files
When I started photography, 40 years ago, (was it that long ago?)...we were always told ‘in difficult light expose for the highlights and let the shadows look after themselves’....
Oops, sorry...it was actually 50 years ago!...😎
Expose to the right is better in high contrast daylight scenes, where you have dark shadows and bright highlights. In low light you are not going to have that problem unless you have a lot of back-light such as taking a photo of a sunset or sunrise. But most landscape photographers do not want noise in their images so iso 200 is about max anyway. And since the dynamic range of a sunset or sunrise is so great, most landscape photographers are either going to use graduated neutral density filters, or multiple exposure bracket with post process blending or composite. So there is no real reason to use ETTR in low light or to underexpose for that matter.
Using ETTR is a good way to avoid underexposure. But as I said in the video, there are many other things in photography that matters more.
Great video Peter! You're talking here about the ETTR in post but how about raising exposure just before taking a shot? I bumped into a problem trying to take a pic in low light at 150mm/f5.6 with shutter priority (1/400s or faster). Increasing ISO helps as expected but increasing exposure up to +5 on my em5-3 doesn't shift the histogram to the right in the captured shot. The scene becomes brighter only in the viewfinder before pressing Shutter but the shot remains dark. Can you explain this? Thank you much.
I cannot say anything for sure what is going on. Need to test it. I try to came back with this.
@@ForsgardPeter I am using your best settings for raw + ettr. In daylight with aperture priority at high shutters that works like a charm, also at night but with lower shutters of course.
Any particular reason why EM5 II not included other than perhaps you don’t have one in your possession?
It is included. E-M5 and E-M1X is not. I do not have those two cameras.
Thank you doing these tests. I found the results interesting. I thought the E-M1 Mark II and III (and E-M5 Mark III) used the same sensor. If that is the case, is the ISO Invariance due to firmware? Or do they, in fact use different sensors? Thank you.
I think this is a great question and look forward to seeing Peter’s answer too!
I have the same question as Jack Kurtz also. Also, when I look at the EM-1 ii images at 200 iso versus 1600 iso the TH-cam compression makes it impossible to see a difference .
The base sensor might be the same, but the analog signal processing (amplifiers and analog-to-digital converters) most likely is not.
Christian answered, it has to do with how the camera is amplifying analog vs. digital. That seems to be different with the newer cameras.
I assume that “The right” in “Exposing to the right” stand for brighter part of light meter.
Is that correct?
Thanks
Yes.
Sounds to me you are saying the opposite as what Robin said in his last video.
Anyway, ettr is all about exposing the sensor with as much light as possible and by changing the iso you don't let in more light so changing iso does not really work with ettr.
What does work is the same as astro photograpers do, stack images and/or using dark frames.
:-)
Actually my results are different from what Robin had. It was pure coincident that we had made these videos right now. This video is kind of a second part to my ISO-video from last Friday. Parts of it was made before Robin published his video.
This has confused me! I have a simple question: if two images are taken with the same ISO but one is ETTR and exposure reduced in post, which image will be better? I’d like a comparison of two images at the same ISO, one a correctly exposed image and one ETTR but adjusted later...
The one with was exposed using ETTR is better. It has more info. If you have the same ISO the image will have different shutter and/or aperture.
Thanks Peter for questioning yourself about ETTR...I assume that this video is based on argue with Robin Wong because you both made a video on this topic...with different point of view.
Anyway, I personally conclude that it is usable for manual mode only and that I need to test this in my real life condition shooting to check which method is more suitable for me when I will decide to use M instead of A mode 😁.
Actually this video was partly made before Robins video. Did not know that Robin was making a video about ETTR. This video was kind of a second part to my ISO-video. Testing yourself is the right way to go.
Peter, I prefer I.S.O. because it is an acronym for International Standard Orgasm. Never Iso. Or, just use ASA. However, I am inconsistent because I say J-Peg not J.P.E.G. Us humans are weird animals. Cheers, Antoine
I have a question off topic.
I have an e-m10 iii and when I record video, it doest let me control the iso and shutter. As if the camera is in auto. I set up my settings the way I want, press record, and it does what it wants. Is that normal? And is there a solution? Someone please help me.
Set the Mode Dial on the top to video. Press the OK button and there will be a menu on the side of the EVF/LCD from where you can set the shooting mode to M. Then you can adjust the shutter and ISO.
@@ForsgardPeter thank you so much for the help. Really appreciate it.
Dear Peter, I am writing you from Barcelona not only beacuse I follow regularly your very interesting videos about Olympus, but also because I like the brand very much.Today I read a notice saying tha Olympus is pulling out the camera-business of Southcorea because a heavy drop of sales in that country. They will alspo close the onlineshop in that market. Do you know some more about that? Thank you.
I just read about it and need to look into it some more. Sad news anyways.
@@ForsgardPeter Thanks Peter and good morning.
I would put it this way: Use base (or lower) ISO in low-light situations ONLY on an ISO invariant camera! One more reason to think about buying one of the two newest Olympus models (more leeway for shadow recovery).
A reminder: The ideas discussed here assume that it is not possible to use a larger lens (e.g., 0.95) and to increase the exposure time. This was explained at the beginning of the video. In the realm of movie making, the makers use artificial lighting or shoot during daytime and then process the material to make it look like night time.
Any findings with the EM1X?
Unfortunately I do not have the E-M1X right now, so I was not able to include it in the test.
My opinion: If under-exposure and boosting in post gives better results than increasing the iso value, then the camera manufacturer has done something wrong. However, I find shooting in low light conditions hard when I can’t see my subjects in the evf or on the display and for that reason I will stick to correct exposure anyway.
Try S.OVF or Live Boost to make the EVF brighter. It won't be WYSIWYG but you can see you composition better.
Peter would you agree with my interpretation of your video, or provide any corrections?:
(Even better if you find any part of this enlightening)...
In the video, which was released around the same time as was Robin Wong's video, Peter took two photos both at 1/250th a second at ƒ/4.5 using an Olympus EM-1 Mark III, which is a very similarly ISO-invariant camera to that which Robin used (the Olympus EM-5 Mark III, which Peter later ALSO used in the video). The first photo was taken using ETTR at ISO 1600 and looked to be overexposed by one stop, while the second photo was taken at ISO 200 and looked to be underexposed by two stops. The exposures of both photos were corrected post-capture in Adobe Lightroom to look more like they were taken at ISO 800 - thus with the ISO 1600 photo being taken down a stop and the ISO 200 photo being taken up two stops. The result was that both photos looked to have no differences between them in noise level.
Peter made a comment that the ISO 200 image might be "better" since more dynamic range would have been available to it (than for the ISO 1600 photo), but I personally did not see ANY difference between both photos or any obvious use of more dynamic range for the ISO 200 image. At the very least, the ISO 1600 photo, as represented (if perhaps imperfectly) by the TH-cam video anyway, did not look worse to me - not less sharp or more painterly, for instance, as was the problem in Robin Wong's video due it seemed to the effects of in-camera noise reduction (if not at all due to any clipping of one or more of the three colour channels, although that can't really be confirmed given that no histograms were provided in that video).
Peter then took two more photos at different exposures. One was a normally exposed reference photo at 1/250th of a second, ƒ/4.0 and ISO 1250. The other photo was an ETTR photo at the same shutter speed and f-stop but using an additional stop of ISO to overexpose the photo at ISO 2500, after which the photo was brought down a stop to correct the exposure. Again I could see no significant difference between the two photos including any degradation with the ISO 2500 photo, while Peter felt that the ISO 2500 was actually a bit BETTER in terms of noise. Peter concluded then that using an ISO greater than the base ISO for intentionally overexposed ETTR work has its merits for low light photography (while ETTR in general can be used for any situation really).
I would say that that approach did not appear to do any harm at least under Peter's testing (even if not in Robin's testing). However, given that the improvement in quality that Peter perceived was very small, and given the ISO invariance of the camera, Peter also concluded that using a lower ISO than one meant for ETTR would be fine as well.
Peter noted that the ISO 2500 photo was quite a bit noisier/grainier than the original ISO 200 photo, but this made sense I think as the exposure settings where similar for both photos EXCEPT for the ISO which was much greater for the ISO 2500 photo and therefore naturally introduced quite a bit more noise, while the ISO invariance of the camera allowed the ISO 200 photo to be brought up 2.15 stops to roughly match the same exposure WITHOUT adding extra noise. Also in the video, Peter illustrated how NON-ISO-invariant cameras produced better results at higher ISO's than at lower ones.
In conclusion to what both videos demonstrated, I would say that for shooting a low light scene, between the available choices of using a low ISO that underexposes the scene, or using a higher ISO that normally exposes the scene and is comfortable to use for composing your scene, or using an even higher ISO that creates overexposure but is short of clipping highlights (in true ETTR fashion),…
…the best or at least "safest" choice (for a low light scene), especially when otherwise in doubt, would be the ISO for normal exposure.
That said, really it could be YOUR choice based on you testing YOUR camera, if you wanted to take things that far. You also have to be mindful as to whether your low light scene has a large dynamic range such that there are many highlight details that you want to record. If it does, then using a lower ISO makes more sense than using a higher one, and using an ISO-invariant camera that can well handle that lower ISO makes sense as well. But if a high dynamic range is not a concern and/or your camera is not ISO-invariant, a higher ISO makes more sense.
In any situation where the base ISO CAN be used for ETTR (with full sensor saturation short of unwanted clipping), that is the best way to go! Nothing beats real light for creating the best signal! Reducing noise through intentional overexposure using real light and correcting the exposure post-capture works well on certain scenes (like a somewhat contrasty landscape) but not great with very LOW light scenes nor well with very BRIGHT scenes with detailed highlights that need preserving.
Thanks!
Quite accurate interpretation.
@@ForsgardPeter Much appreciation for your response and your video! Didn't want to be mistaken on anything.
Confusederer, changing iso to affect exposure in landscape shooting isnt something I do, setting f first, lowest iso second then let speed run in A??? Maybe indoor event shooting I can see letting iso auto. Ok off to watch mr wong ....
Ok Wongs scenario was inside, dark, wide open using iso to give acceptable speed, read his edit comments .... that doesnt disagree with forsgards comments that I see.
You are right the difference in opinion is actually not that big. The approach was a bit different.
I learned how to properly expose images by using Kodachrome. Very little room for error, so you learn quickly.
Slide film has very little latitude for proper exposer compared to print film, but slides have amazing images due to transmitted light through the image. With digital and HDR and multi exposer the image can have a greater range than shoot with film and RAW files.
In relation to physics only amount of the light matters and noise depends of it solely. And what I think is that when denoise AI will power cameras then ISO will become obsolete and this will greatly simplify photographing process - the photographers will finally forget ISO.
That would be great.
The results of your experiment as well as Wong's one are absolutely expected. We have only one sensor in a camera. In the old good days of films we could choose ISO100 or ISO200 film. But now we cannot chage a sensor. Changing ISO to a value higher than 200 is the same as using ranges or levels in post. No more, no less. And quality of the result depends solely on quality of algorithms in a camera and desktop software. May be Photoshop and others are using float-point arithmetics while cameras integer only? I don't know.
It's the same situation as with software zoom. We have only one lense. And it doesn't matter where we change the image size in pixels. Here as well we have differences depending on algorithms: quadratic, cubic extrapolation and so on.
But! If we decrease ISO to 100 or less than we actually loose an amount of information gathered by the sensor. It's for JPEG of course, RAW is always 200.
Would it not be nice if they made a camera like some high end video cameras that have dual native iso? :-)
"Changing ISO to a value higher than 200 is the same as using ranges or levels in post. No more, no less."
Technically speaking it is not, at least for ISO variant cameras. There is always electronic noise unrelated to ISO or exposure. Such noise originated after ISO amplification will be more pronounced as you lower ISO decreasing signal level. In the case of low light photography it can be pretty visible, and this is not software problem but hardware, because ISO in cameras amplifies with analog devices.
petapixel.com/2017/03/22/find-best-iso-astrophotography-dynamic-range-noise/
Hello Peter, the rest of your video notwithstanding, I'm curious as to why on 19 May you're pronouncing ISO, as if it was a word, 'eye-so', while on 15 May you regarded it as an abbreviation by saying I-S-O. In the old days film speed was said as a number following the abbreviation A-S-A, not pronounced as a word pronounced 'ay-sa. In general, whichever is easier to say would be the preferred way, such as NASA is easier to say as a word, 'nassa", then to say the letters N-A-S-A. But CIA is easier to say as an abbreviation, rather than saying 'see-ah'. Whichever sounds better would seem to be the better way, but 'eye-so' and I-S-O would both be equally good. So why change?
I was told that I-S_o is a wrong way. Thats why I changed it.
I love your videos. They're always very informative and entertaining. I hope I didn't sound rude questioning your pronunciation of ISO. I was just curious. Your command of English is quite excellent. BTW, I loved your comparison of the 12-50 mm lens (my favorite) with the, I think it was, pro-grade 12-40 mm. That lens is much more expensive and only (I believe) very slightly better optically.
I.S.O. versus 'Iso'; thought I.S.O. was International Standards Organisation. Hence I.S.O., cf ASA - American Standards Association in the old money.
I appreciate the subject, but found the presentation confusing. Too much maybe yes maybe no’s. With 1mk2, do I or do I not Ettr?
Yes I realised that the presentation was abit confusing. Yes, in my opinion you should use ETTR. ETTR is always the best way to make the exposure no matter if your camera is ISO invariant or not.
Will be there ever be a consensus on this topic ? The more video’s I see, the more confusing it gets.
Most likely there won't be. I was planning on publishing the video earlier yesterday, but I wanted to do my test again and be sure.
It's not you, it's the subject!
Technical discussion on photography is good -- picture content is the essential. Without the eye, the rest is academic. First there was the Focus Tracking War, followed by the Bokeh Wars, and then came the Noise War. It appears that Olympus has an edge over Panasonic in MFT Noise War, with better ISO handling. I think most people just like little MFT cameras, and don't care about the wars going on. I can go up the scale to FF or Medium Format which will be the easy win to the war. And that is fine. Especially for the street photography, and as a travel camera, the MFT cameras are just friendly little guys that get the job done. Fujifilm and Sony may be bigger guns to take to the Wars, IMHO. I will just try not to clip the highlights when shooting in JPG and enjoy the day shooting away. -Loren lorenschwiderski.smugmug.com/
Ok, now it gets confusing! @___@
Real life test is what it is really about!
Sorry Peter, but the argument to use higher ISO to compose better is a camouflage! For example using an ND filter I compose and set the exposure and focus without the filter and then having reset the filter I adjust the exposure - its a simple calulation.
Higher ISO ALWAYS means UNDEREXPOSURE and thus noise!
Unfortunately the Olympus cameras are not very good in low light even with ISO 200 underexposure may cause ugly noise. I always expose to the most illuminated area neglecting underexposure in the other areas. Take it or leave it,
I consider to buy in addition to my Olympus cameras either a Nikon Z6 or even a Fuji X-T4 ( even only because the sensor dimensions are not that big - I like Fuji). My experience with Fuji X-T3, X-H1 and Nikon Z6 and 7 are superior in low light.
Raising shadows always brings in more noise no matter camera.
The best ways to remove noise is with image stacking (if possible, works horrible with moving subjects) and/or dark frames.
Think astrophotography.
They also uses something named bias frames but I don't know what that is.
Must be the only one to dislike iso invariant and ettr I just try to get the correct exposure with the camera, blacks stays blacks, white, white, used to slide films...
Peter, very confusing video please allow me some feedback. First you do "ETTR" by using a higher iso and keeping the speed and aperture the same. I do no think most people consider that ETTR, as ETTR means to allow more light to hit the sensor, not to push the boosting of the signal in camera. Secondly, you reach the opposite conclusion from Robin Wong (look at his last video) for the same situation... What is going on, did you two have an argument about using ISO to do ETTR and each one of you did a video to prove their point? And lastly I am not sure about your ISO invariant categorization of the Olympus cameras, if you look at the EM5 mark III photos the one on the left seems to have a strong green cast. Are you sure it is ISO invariant? Actually I thought that most Olympus cameras were ISO invariant based on the sensor measurements I have seen. Thanks for your effort
No we did not have any argument. I was not even aware that Robin was making a video about ETTR. Part of this video was made before Robins video was out. This video is kind of a second part to my ISO-video that was published last Friday.
I agree that ETTR is considered to be only when adjusting light with shutter or aperture. Using ISO is not the usual way, but it seems to give better results than just using lower ISO. ETTR always means that exposure needs to be lower in post. That method gives me better results.
Some say that ETTR is overexposing. It is not. It making the image as bright as possible, without clipping the highlights, to get the best possible IQ. Overexposing will clip highlights. If we look at the light meters readings in camera, then according to that we are in many times "overexposing". Sometimes ETTR can mean that your light meter in camera shows "underexposing". It all depends on the scene. Light meters is cameras have no use in my opinion. Only histogram and the help of blinks/flag colors matter.
@@ForsgardPeter Thank you Peter, I appreciate your reply. Keep up your good work!
I wouldn't call this ETTR. You're using analog or digital gain to boost the signal and not actually increasing the light level on set.
Technically yes, but the idea is there. Expose as bright images as possible.
The only time a terrestrial photographer should worry about a sensor's ' invarianace ' is if and when the lens is turned out into the universe for long exposure deep field imaging. Apparently a couple of Fujis. a couple of Nikons and a Sony camera according to the published tests have proven to be iso invariant or the rumour has it. Personally I couldn't care less . Thanks.
I normally really enjoy and appreciate Peter's talks, but sorry this one lost me, couldn't watch any more at about 10 minutes,
Sorry to hear that.
@@ForsgardPeter No problems Peter Thanks for taking the time to reply, I am not a novice, I have been a long time OLY 4/3 user and am familiar with the concept you are describing, I just felt that on this occasion you made sound a lot more complex than it is, but I still look forward to your work. Interestingly I have subscribed and ticked the bell but I never get a notification of new work. John
why do you talk breaking sentences "in parts"? are you alright? or are you just slow? It's very annoying. Otherwise, the video is ok.
All good. I do not remember why. This video is 4 years old.