Thank you so much for this extremely informative and well put together video. And your comic humor in the intro was absolutely hysterical I laughed out loud for quite a while. Thank you thank you very much
Greetings from Tasmania 42 South-good to see you have dodged the minute particle. Some experienced photographers I know still believe the crop sensor myth. Hearing it from another one who knows may help to set them on the right path. Keep up the great work.😊🇦🇺🦘🔭
I hope so - but judging from some of the feedback, we may not be there yet. I have a followup demonstration coming out in a couple of days. Maybe that'll do it!
I watched this one after your Part 2 follow on, it ll makes sense and Like Chris below, I love your sense of humour with the videos, especially Audry, Winston and Maggot, great family portrait!
Great stuff, Allan. Count me in for the detailed courses. And Allan, you are weird to a T: Wonderful, Enlightening, Informative, Rare, Dynamic Teacher.
I'm afraid not. DoF is a function of lens aperture, focal length, object distance, and sensor resolution. But not sensor size. A smaller sensor gives the impression that you are using a longer focal length lens (compared to a 50mm on a FF sensor, the same lens on APS-C "looks like" you are using a significantly longer lens), but that is only because the smaller sensor can only see a small part of the available field of view. Remember that lens performance is constant, regardless of the camera you put it on. Putting it another way, if you compare the actual photographs from a crop frame sensor and a full frame sensor with the same lens, you would see that the depth of field is actually identical, it is the field of view that changes.
@@AllanWallsPhotography DoF is a function of magnification (not focal length - try photographing a ruler at angle with two macro lenses of different focal length set to the same mag perhaps 1:1 and see for yourself) and entrance pupil size. As you've alread shown, mag does not rely on sensor size!
@@AllanWallsPhotography Functionally though, you do get more dof for the same size image using the whole frame of the camera. If you have a 20mp m4/3 camera and a 20mp ff camera and want the same final reproduction ratio with the same dof, your apeture on the ff camera will have to be 2x what it is on the m4/3 camera because one camera will be using a 30mm vs the others 60mm. You could also crop half the image away on the ff but then functionally resolution drops. This is always somewhat of a silly argument though, you are correct in that the lens functions identically on either camera but the thing people talk about this is in practical applications, its much easier to think you are getting more magnification despite the fact that the camera is just using less of the lens. Mathematically though, the reproduction ratio is the same on both cameras if you use the same lens because the sensors are different sizes (ie a 18mm subject will fill a frame on mft and be 1:1 as it fill the whole frame, and it will only fill half the frame on ff because the frame is 36mm wide, and thus it will be 1:1 magnification). This is where I think most people go wrong, as this is strictly a bad assumption to make because ff is the default standard for nearly everything when we talk about it (its easier to say and understand for instance that something is a 50mm equiv or 2:1 magnification equivalent on a ff sensor than it is to say 1:1 magnification on this camera system). I think this is why crop sensors show equivalences, because we are so used to ff as the default, for instance if I want a 50mm lens then I will have to use a 25mm on mft because the camera uses half of the lens. This is why people say you get 2x the magnification on a mft camera despite the formula saying it is exactly the same. The thing is that we look at our images on prints and monitors as exactly the same thing despite an image coming from a different camera. If I see a super macro of an insect on instagram I could not tell what camera sensor size it came from unless they told me. It could be a 30mm lens at f8 on a mft camera or a 60mm lens at f16 on a ff camera. Functionally the increased magnification is there, but mathematically it isnt, just because we are printing and viewing things as images rather than imprints on physical film. I think magnification is generally understood to be perceived magnification as well, given the equivalence. Dof is also completely based off of the sensor size because people are taking images with field of view in mind not the physical length of the lens. I am not taking an image with a 30mm lens because I want 30mm lens bokeh, I'm using it because I want the field of view it achieves on the crop sensor camera I am using. You are right that technically it isnt based off of it because of the formulas, but given that people choose the lens based off of its fov (ie I want to use a 35mm lens for street photos so I will shoot on a 17mm on mft or a 50mm on medium format and not the idea that the focal length gives us different depth of field) I would say it is inherently linked to it. I really liked how you explained how everything works and I am glad this information is easily digestible through your video, but I just wanted to point out that it really is just easier to say certain things despite them being not 100% correct. The truth is that yes the lens is what is doing the magnification, and nothing is really changing depending on sensor size, but that functionally in usability and practicality there is more "zoom" and "magnification" on smaller sensors because we view the images on a 5x7 print or a 14 inch computer monitor no matter what size the sensor was that captured the image. It is exactly the same as cropping an image in post, but given that most people dont have ff sensors with the same pixel density as smaller sensors, I think it might be sort of silly to say that its exactly the same (lower res for instance for the same size crop). When counting diffraction as well the diffraction starts at different points on the different sensors, perhaps at f8 on a mft camera but with the same equivalent lens at f16 on a ff, but these roughly equal each other out and you will probably be able to get the same exact image with the same sharpness without diffraction on either system. This whole debate is silly and basically the trade-offs end up being functionally the same. Higher iso on a ff to account for the higher fstop and less light to the less noisy photoreceptors because they are larger and less dense, and lower iso on the crop camera because the fstop can be lower because the equivalent focal length is .75 or .5 the lens size which has more dof at the same fov- yet because the pixels are denser the noise is usually about the same as the image you would shoot on the full frame camera with the higher iso rating. The only thing that changes is the weight and size of the lens for the same exact picture, so I think its quite pointless outside of super extreme examples with tripods and super slow shutters in order to remove the light hurdle. Most of this talk is completely around functional usage of a camera, and not the physical science behind it. Optics perform exactly the same no matter what, but given that you are only using half the lens or almost 2x the lens means that functionally the dof and magnification changes based off of the same field of view equivalent lens at the same f stop (25, 35, 50, or 75/80 all look different on their own but will end up the same fov on each respective system (obviously different dof if you compare the same fstop as well because that is 100% based off of focal length and focus distance)). Obviously the optics dont perform any different technically on different cameras, but they do end up looking completely different in terms of fov if used on the different sized sensors... Its not that you GET more depth of field from the sensor being smaller, it is just that you get more depth of field from using the shorter lens that is equivalent to the larger lens you have to use on the ff camera for the exact same shot. great video, sorry for the super long comment. I didnt really proof read this I just slammed it out but I just wanted to point some things out.
@@SuperZarkosis thank you Jude. I appreciate you taking the time to make all of these points and in most instances I think we agree. I do think that today's equipment has advanced to the point that most of these formerly critical choices just become a matter of taste. But there is one idea of yours that I do take exception to. And that is that accuracy and rigor are indispensable when discussing the technical aspects of something like photography. You don't need to know this stuff to be a competent photographer, or even a good photographer, but to really understand why you're a good photographer, or not, you do. The fact that some misunderstandings have been propagated through so many generations of photographers that they have become accepted as fact, is the primary reason I take the time and trouble to correct them. If young, learning photographers are encouraged to believe that magnification and enlargement are the same thing, how can they possibly understand more difficult concepts like depth of field. I have actually received so much feedback about this video, much of it strongly disagreeing with my position, I have already made a follow up video, that will be released in a couple of days, that attempts to clarify this all important point. I would be interested to hear how you respond to it. Thank you again for your thoughtful and considerate comments. There is always room for conversation.
@@migranthawker2952 That is not correct - to calculate depth of field you need to know the focal length of the lens, the object distance, the f#, and the circle of confusion (DoF=2 x object distance squared x f# x CoC / all divided by the square of the focal length of the lens). This may be simplified by removing f# and focal length, but the calculation is slightly more reliable if you don't. But the point I was making was simply that if there is no magnification increase, the DoF, derived from the former, also cannot increase.
hi Uncle Alan, could you let me know that is the "Mitutoyo M Plan Apo 5X Objective" lens mount in RMS to M42? or any other size? I am using a 24-240mm zoom lens, the screw ring is 77mm. Thanks.
I would recommend using a series of step down rings, if you already have them - most sets go all the way down to M26. But you can also order a single adapter (M26 to M77) from eBay. I believe WeMacro also has an inexpensive M26/M77 adapter.
Hi Alan, again another great video, Been trying to get hold of mitutoyo Objectives for a long time, but can't get them anywhere in nz, so emailed Jeff, at 3.40pm over here, he replied in 5 mins over there 8.40pm, said yes he ships to nz, so bought 2 of them, the 5x and the 10x he, looking forward to getting them. Cheers keep up the great work. Dave. Ps I don't think your a weird old man, slightly warped sense of humor, but I find it very funny
I bought the 10x hr which cost more than my d850, so I'm really hoping it's worth it Again great videos built my macro cage on your design, loved the lighting series, although I have to say the led lighting that i use now, I do love it. Bought the splashart because of you and that's really cool, love using 2 colored milk great results, learnt so much from you as far as lighting and composition. Ps totally agree with your video today, full sensor gives you the freedom to crop and still get high resolution
@@andrewchalmers3902 hi Andrew, doing to nz was $206 nz flying FedEx. Nothing arrived yet but have received a couple of emails saying it's on its way. Will keep you posted
Good stuff this discussing the crop factor in relation to the quality of the image you’re going to get an magnification is A great topic ….worthy of a mini course.
Oh absolutely! I have dozens of different diffusion strategies - this is one that works best with very small reflective subjects. I always recommend you use the method you are most comfortable with. But now you know about it, you might want to give it a try sometime. Can never have too many lighting options!
Hi Alan, I tried the ping pong ball trick, but no matter how I move the lighting I see a white cross X in the background, I’m using a 4x objective on a bellows, the objective has to enter the ball to get focus. Do you have any advice, on how to overcome this. I have been using constant light, I will give the flash a go and see if it helps
Hi Peter, I really should have mentioned this, but this method, if you are going to do it using a hole in the back of the PP ball, works best with long-working distance objectives (30+ mm). When you are working at 10mm, you would be better off leaving the back wall intact, as a gradient backdrop.
Thinking it over about the difference between magnification and enlargement-a microscope is a good example, in that to achieve increased magnification one has to change the objective (lens) itself, for example, going from low to high power with many microscopes generally is 4x, 10x, 20x, 40x, 100x. with a fixed tube length and say a 10x eyepiece. Using a higher power eyepiece only enlarges the image-there is no increase in resolution. Many modern and more expensive microscopes can start with a 1x and 2x objectives and go all the way to 100x oil immersion, such as those I used in my work in pathology, mind you, those microscopes were around $30,000, plus all the camera attachments and accessories could bump the cost up nearer $45.000. We had a departmental budget of quarter million dollars for microscope upgrades each year to keep pace with modern developments.
Rather than use a blade "as hot as the sun", you might try this idea: Take an X-acto or what ever replaceable blade craft knife is available in your area. Place the edge of the blade against the face of a file and hit the back of the blade with a small hammer. This will dent the sharp edge of the blade with a series of ridges from the file. In effect, this creates a very fine saw blade. I have seen this used for other crafts although never sawing into a ping-pong ball. Hope this helps and is less dangerous.
When I view my Nikon NEF files in XMP viewer, they look uncorrupted. But when I convert the files to TIFF and then view them, I see a few corrupted files. So, what software/technique do you use to weed out corrupted files prior to stacking?
My preferred workflow is to go through Adobe Lightroom. On import into LR my files are converted to .dng which can undergo any preliminary editing before export to ZS using the Zerene Plugin. This has worked flawlessly for many years. The files return to LR as .tif, which can be further edited or retouched in Photoshop.
if i'm getting 2mm of ruler on the M50 crop sensor what times magnification would that be? also does the crop sensor have better pixel density than a ff camera of te same mp
The M50 has a sensor size of 22.3 mm x 14.9 mm, therefore if you are measuring 2 mm across, you are getting 11 times magnification. You are correct to say that the pixel density of the M50 would be greater than that of a full frame 24 megapixel sensor. However, pixel density it's not a terribly useful characteristic to consider in isolation. Packing a larger number of photosites onto a smaller sensor, does not necessarily improve the performance of a sensor, and may have some negative consequences (smaller photosites limit low light performance and dynamic range, for example). But pixel density, as a yardstick for system performance, is not very useful compared to the diffraction characteristics of your lens, say.
What about it? Pixel density, by itself, is a relatively meaningless concept, in terms of image quality. Pixel density does not take into account the fact that the photosites have to be correspondingly smaller or that the sensor architecture must be modified. Further, any correspondence between pixel density and image quality must take into account the smaller APS-C image size. Even if the increased pixel density on a smaller sensor offers a theoretical improvement in resolution, it would be irrelevant compared to the far greater impact of diffraction on image quality. Choosing a crop frame camera because of the increased pixel density of its sensor, in the expectation of getting sharper macro images, will disappoint.
I think it be worth exploring what the crop factor does to the effective of aperture calculations ….always dodging the diffraction problem. Wide-open on a full frame SEEMs the highest quality . Condensing the available light down to 4/3 would give concentrated light and still 24 meg pixel? Which leads to the discussion isn’t the highest quality on almost any lens one f-stop down from the max aperature? So many rabbit holes to follow!
Many rabbit holes indeed - but to take your first point... The effective aperture is calculated by multiplying (the system magnification ratio + 1) by the nominal f/# of the lens. As the magnification effect of the crop frame sensor is illusory, the diffraction characteristics of the lens (reflected by the effective aperture) will not be affected by a change in sensor size.
I’d argue that for a given pixel count you are less likely to bump into diffraction limits on larger sensors. At 1x it’s maybe less important but at 5x or more, I often annoyingly bump into diffraction limits on a 24Mp full frame sensor at f/4. Simply put, smaller pixels = worse diffraction issues leading to softer images
Yes - absolutely. There are a lot variables to consider and the choices are not always intuitive, but trading pixel size for a higher pixel density has never made sense.
@@AllanWallsPhotography it does make sense to have a higher pixel density in macro if you have 1) a very high quality lens 2) sufficient light intensity or a non- living subject that does not move 3) you have continuous LED illumination (for example three cheap150 watt non-flickering photo LED bulbs) and a pixel shifting sensor to reduce noise and increase sharpness 4) use basal ISO [and 5) you want a smaller camera and lens.] Under these circumstances it becomes very relevant to compare the linear pixel density ( number of pixels per millimetre).
Good morning Allen, thanks for another nice piece. I see you avoided the dreaded Pixel Density Challenge, which can be a horrible rabbit hole. I am interested in printing, not so much in posting. So at the very least, we can say prints can be considered pixel pigs (pixel density oriented)! With this in mind, one of the most important things to me is pixel density, and so I pose the question: Does a D7200 24mp APC-C give me a bigger print (cropped to the same area) as my D850 48mp FF? And from my calcs, (or dare I say it, a Canon 90D) the answer is no, the APS-C at 24mp (and the Canon 90D by a lot more) beats the 48mp FF camera by a mile. To be the same, my D850 would need to be at least 58mp. This is why I am hoping for a Z8 with a 60+mp sensor... Calcs: A given 24MP APS-C sensor is 23.4mm x 15.6mm and produces images 6000 x 4000 pixels A generic 35mm (FF) sensor is 36.0mm x 24.0mm 6000 / 23.4 = 256.4 - or 256.4 pixels per mm of APS-C sensor horizontally 256.4 * 36 = 9230.8 - calculated horizontal pixels for a matching FF sensor 4000 / 15.6 = 256.4- pixels per mm of APS-C sensor vertically 256.4 * 24 = 6153.8 - this is the amount of vertical pixels in our FF sensor Calculating a total in a similar FF sensor: 9230.8 * 6153.8 = 56805 - in our matching FF sensor, or ~57 megapixels As usual, you're a pretty smart cookie, and maybe you'll prove me wrong, and teach me something along the way. So please, be snarky, and show me (in this narrow interpretation) where I errored. Thank you...
When I initially made this video, it contained two parts that were subsequently removed. A demonstration of the magnification issue, which I didn't think was necessary, and a long discussion of sensor size, pixel density, and how they relate to magnification and enlargement. I think I should've left both parts in. There is nothing wrong with your logic, as far as it goes. But it's a mistake to put all your image quality/resolution eggs in one basket. I would avoid inferring pixel density from the number of pixels along one edge of the sensor - you can get away with it when the pixels are square, but it is safer to divide the pixel count by the total sensor size in mm² - which yields a pixels density of about 62,000 pixels per mm² for the D7500 and 55,550 pixels per mm² for the D850. If our only concern was pixel density then there would be no question, the best camera for the job would be the D 7500. But pixel density, taken in isolation is of very little value. The pixels on the D850 are a whopping 7% larger than the D7500. This is why the camera performs so well in low light and has such a high dynamic range. The pixel pitch is also 3% higher on the D850 sensor, which even further improves the signal to noise ratio. You would think that the spatial resolution of the D7500 would be better, and that may be true under certain circumstances, but pixel density alone does not translate to image quality. It is one, relatively small part of the quality equation, and must be considered in the context of the resolution of the lens and the signal to noise ratio. I would take the D850 over the D7500 if image quality was the only consideration, but I have both, because it seldom is. Pixel count and pixel density are important, but not that important.
@@AllanWallsPhotography Thanks Allan, all reasonable stuff. Since my comment centered around printing, I guess similar setups of the same subject out put at 300dpi and actually printed would really help. Maybe that's a rabbit hole worthy of your investigative talents.
Congratulations, I've been telling people this for years. Being an Optician, I understand the physics of optical systems, but there are some who simply won't believe this basics of optical physics, that the lens controls the magnification.
One point not really clarified, though, is if both the ff sensor and crop sensor have the same number of MP (which is quite often the case) then to crop the ff image down to the same field of view as the crop sensor image, you will end up with an image with less MP, therefore not potentially as sharp due to lower resolution. I think this is why many macro shooters reach for the crop camera first. Sorry if this was mentioned at the end of the video...I got a bit peckish towards the end and made a grilled cheese.
As it happens, the last 10 minutes of the video dealt with this question in some depth, but I cut all of that in the interest of time, and because I was feeling a bit peckish too. You bring up some good points. Firstly, it is not that helpful to consider pixel density as an independent arbiter of image quality. For one thing, the performance characteristics of the larger photosites on a FF sensor (of the same MP count) have a much greater impact on image quality than the reduced megapixel density might imply. Secondly, the resolution differences are far beyond our ability to discern, even at fairly large print sizes. And lastly, it is the diffraction characteristics of the lens we chose and how we configure it, that actually influences image quality. The impact of diffraction on image quality is far greater, and becomes apparent much sooner than any theoretical impact on resolution related to the reduced pixel density of cropping a FF image. But possibly more to the point, both the full frame and crop frame images are going to be enlarged for display or printing. The smaller, crop frame image, is going to require 1.5 times as much enlargement to create a final image of the same size as that from a full frame sensor. This will result in a decrease in the image pixel density that will offset any decrease in pixel density from cropping the Image made on the largest sensor. It all comes back to my central point, we have to understand that the crop frame camera image is a smaller image than that of the FF sensor. Any conclusions drawn from the false assumption that the initial FF and APS-C images are the same size (as we would usually view them), simply perpetuates the primary misunderstanding of magnification as a function of sensor size.
amazingly helpful. Thank you for taking the time to talk about sensor size!
Great stuff as ever Uncle Alan. Love your dry sense of humour... regards, Chris, Staffordshire, UK.
Allan, Thank you very, very much for all this treasure of knowledge and the tips you're sharing!
My pleasure!
Thank you so much for this extremely informative and well put together video. And your comic humor in the intro was absolutely hysterical I laughed out loud for quite a while. Thank you thank you very much
Why is the background in portraits then blurred with a full frame camera ? And with a apsc much sharper ?
Greetings from Tasmania 42 South-good to see you have dodged the minute particle. Some experienced photographers I know still believe the crop sensor myth. Hearing it from another one who knows may help to set them on the right path. Keep up the great work.😊🇦🇺🦘🔭
I hope so - but judging from some of the feedback, we may not be there yet. I have a followup demonstration coming out in a couple of days. Maybe that'll do it!
Why is the background in portraits then blurred with a full frame camera ? And with a apsc much sharper ?
Thanks for the tip on the Mitutoyo…I decided to grab one. 👍
You are going to love it! A truly awesome lens.
I watched this one after your Part 2 follow on, it ll makes sense and Like Chris below, I love your sense of humour with the videos, especially Audry, Winston and Maggot, great family portrait!
Sold! Thanks for the pointer!
Great stuff, Allan. Count me in for the detailed courses. And Allan, you are weird to a T: Wonderful, Enlightening, Informative, Rare, Dynamic Teacher.
Great material Allan! Thanks for the shoutout. I don't see the setups as strange at all.....I have many like the fly eating the spider ;)
Your best ever vid - very entertaining!
absolutely love your dry sense of humour! Checked the tip, but no x10 in stock sadly :(
Isn't the extra depth of field you get w a crop sensor handy when shooting really tiny subjects? meaning you wouldnt have to stop down so much..?
I'm afraid not. DoF is a function of lens aperture, focal length, object distance, and sensor resolution. But not sensor size. A smaller sensor gives the impression that you are using a longer focal length lens (compared to a 50mm on a FF sensor, the same lens on APS-C "looks like" you are using a significantly longer lens), but that is only because the smaller sensor can only see a small part of the available field of view. Remember that lens performance is constant, regardless of the camera you put it on. Putting it another way, if you compare the actual photographs from a crop frame sensor and a full frame sensor with the same lens, you would see that the depth of field is actually identical, it is the field of view that changes.
@@AllanWallsPhotography DoF is a function of magnification (not focal length - try photographing a ruler at angle with two macro lenses of different focal length set to the same mag perhaps 1:1 and see for yourself) and entrance pupil size. As you've alread shown, mag does not rely on sensor size!
@@AllanWallsPhotography Functionally though, you do get more dof for the same size image using the whole frame of the camera. If you have a 20mp m4/3 camera and a 20mp ff camera and want the same final reproduction ratio with the same dof, your apeture on the ff camera will have to be 2x what it is on the m4/3 camera because one camera will be using a 30mm vs the others 60mm. You could also crop half the image away on the ff but then functionally resolution drops. This is always somewhat of a silly argument though, you are correct in that the lens functions identically on either camera but the thing people talk about this is in practical applications, its much easier to think you are getting more magnification despite the fact that the camera is just using less of the lens. Mathematically though, the reproduction ratio is the same on both cameras if you use the same lens because the sensors are different sizes (ie a 18mm subject will fill a frame on mft and be 1:1 as it fill the whole frame, and it will only fill half the frame on ff because the frame is 36mm wide, and thus it will be 1:1 magnification). This is where I think most people go wrong, as this is strictly a bad assumption to make because ff is the default standard for nearly everything when we talk about it (its easier to say and understand for instance that something is a 50mm equiv or 2:1 magnification equivalent on a ff sensor than it is to say 1:1 magnification on this camera system). I think this is why crop sensors show equivalences, because we are so used to ff as the default, for instance if I want a 50mm lens then I will have to use a 25mm on mft because the camera uses half of the lens. This is why people say you get 2x the magnification on a mft camera despite the formula saying it is exactly the same.
The thing is that we look at our images on prints and monitors as exactly the same thing despite an image coming from a different camera. If I see a super macro of an insect on instagram I could not tell what camera sensor size it came from unless they told me. It could be a 30mm lens at f8 on a mft camera or a 60mm lens at f16 on a ff camera. Functionally the increased magnification is there, but mathematically it isnt, just because we are printing and viewing things as images rather than imprints on physical film.
I think magnification is generally understood to be perceived magnification as well, given the equivalence. Dof is also completely based off of the sensor size because people are taking images with field of view in mind not the physical length of the lens. I am not taking an image with a 30mm lens because I want 30mm lens bokeh, I'm using it because I want the field of view it achieves on the crop sensor camera I am using. You are right that technically it isnt based off of it because of the formulas, but given that people choose the lens based off of its fov (ie I want to use a 35mm lens for street photos so I will shoot on a 17mm on mft or a 50mm on medium format and not the idea that the focal length gives us different depth of field) I would say it is inherently linked to it.
I really liked how you explained how everything works and I am glad this information is easily digestible through your video, but I just wanted to point out that it really is just easier to say certain things despite them being not 100% correct. The truth is that yes the lens is what is doing the magnification, and nothing is really changing depending on sensor size, but that functionally in usability and practicality there is more "zoom" and "magnification" on smaller sensors because we view the images on a 5x7 print or a 14 inch computer monitor no matter what size the sensor was that captured the image. It is exactly the same as cropping an image in post, but given that most people dont have ff sensors with the same pixel density as smaller sensors, I think it might be sort of silly to say that its exactly the same (lower res for instance for the same size crop). When counting diffraction as well the diffraction starts at different points on the different sensors, perhaps at f8 on a mft camera but with the same equivalent lens at f16 on a ff, but these roughly equal each other out and you will probably be able to get the same exact image with the same sharpness without diffraction on either system. This whole debate is silly and basically the trade-offs end up being functionally the same. Higher iso on a ff to account for the higher fstop and less light to the less noisy photoreceptors because they are larger and less dense, and lower iso on the crop camera because the fstop can be lower because the equivalent focal length is .75 or .5 the lens size which has more dof at the same fov- yet because the pixels are denser the noise is usually about the same as the image you would shoot on the full frame camera with the higher iso rating. The only thing that changes is the weight and size of the lens for the same exact picture, so I think its quite pointless outside of super extreme examples with tripods and super slow shutters in order to remove the light hurdle.
Most of this talk is completely around functional usage of a camera, and not the physical science behind it. Optics perform exactly the same no matter what, but given that you are only using half the lens or almost 2x the lens means that functionally the dof and magnification changes based off of the same field of view equivalent lens at the same f stop (25, 35, 50, or 75/80 all look different on their own but will end up the same fov on each respective system (obviously different dof if you compare the same fstop as well because that is 100% based off of focal length and focus distance)). Obviously the optics dont perform any different technically on different cameras, but they do end up looking completely different in terms of fov if used on the different sized sensors... Its not that you GET more depth of field from the sensor being smaller, it is just that you get more depth of field from using the shorter lens that is equivalent to the larger lens you have to use on the ff camera for the exact same shot.
great video, sorry for the super long comment. I didnt really proof read this I just slammed it out but I just wanted to point some things out.
@@SuperZarkosis thank you Jude. I appreciate you taking the time to make all of these points and in most instances I think we agree. I do think that today's equipment has advanced to the point that most of these formerly critical choices just become a matter of taste. But there is one idea of yours that I do take exception to. And that is that accuracy and rigor are indispensable when discussing the technical aspects of something like photography. You don't need to know this stuff to be a competent photographer, or even a good photographer, but to really understand why you're a good photographer, or not, you do. The fact that some misunderstandings have been propagated through so many generations of photographers that they have become accepted as fact, is the primary reason I take the time and trouble to correct them. If young, learning photographers are encouraged to believe that magnification and enlargement are the same thing, how can they possibly understand more difficult concepts like depth of field. I have actually received so much feedback about this video, much of it strongly disagreeing with my position, I have already made a follow up video, that will be released in a couple of days, that attempts to clarify this all important point. I would be interested to hear how you respond to it. Thank you again for your thoughtful and considerate comments. There is always room for conversation.
@@migranthawker2952 That is not correct - to calculate depth of field you need to know the focal length of the lens, the object distance, the f#, and the circle of confusion (DoF=2 x object distance squared x f# x CoC / all divided by the square of the focal length of the lens). This may be simplified by removing f# and focal length, but the calculation is slightly more reliable if you don't. But the point I was making was simply that if there is no magnification increase, the DoF, derived from the former, also cannot increase.
hi Uncle Alan, could you let me know that is the "Mitutoyo M Plan Apo 5X Objective" lens mount in RMS to M42? or any other size? I am using a 24-240mm zoom lens, the screw ring is 77mm. Thanks.
I would recommend using a series of step down rings, if you already have them - most sets go all the way down to M26. But you can also order a single adapter (M26 to M77) from eBay. I believe WeMacro also has an inexpensive M26/M77 adapter.
@@AllanWallsPhotography Let me try, Thanks Alan
Hi Alan, again another great video,
Been trying to get hold of mitutoyo
Objectives for a long time, but can't get them anywhere in nz, so emailed Jeff, at 3.40pm over here, he replied in 5 mins over there 8.40pm, said yes he ships to nz, so bought 2 of them, the 5x and the 10x he, looking forward to getting them.
Cheers keep up the great work.
Dave.
Ps I don't think your a weird old man, slightly warped sense of humor, but I find it very funny
Brilliant! That is so cool! I was joking when I said you guys had better leave one for me - now it looks like I need get mine right now!
I bought the 10x hr which cost more than my d850, so I'm really hoping it's worth it
Again great videos built my macro cage on your design, loved the lighting series, although I have to say the led lighting that i use now, I do love it.
Bought the splashart because of you and that's really cool, love using 2 colored milk great results, learnt so much from you as far as lighting and composition.
Ps totally agree with your video today, full sensor gives you the freedom to crop and still get high resolution
@@Puffer001 Thanks Dave! If I ever suggest you pick up a new Bentley, you might want to check with the wife first - just saying.
Hi Dave, can you tell me how much shipping was?? His website estimates $172 shipping to Australia???
@@andrewchalmers3902 hi Andrew, doing to nz was $206 nz flying FedEx.
Nothing arrived yet but have received a couple of emails saying it's on its way.
Will keep you posted
Good stuff this discussing the crop factor in relation to the quality of the image you’re going to get an magnification is A great topic ….worthy of a mini course.
Could one use a small light box instead of the ping pong thingamajig?
Oh absolutely! I have dozens of different diffusion strategies - this is one that works best with very small reflective subjects. I always recommend you use the method you are most comfortable with. But now you know about it, you might want to give it a try sometime. Can never have too many lighting options!
I would.buy into a mini- course that mapped out the pro-cons of Full vs APS vs 4/3 ….in Macro snd Extreme Macro
Hi Alan, I tried the ping pong ball trick, but no matter how I move the lighting I see a white cross X in the background, I’m using a 4x objective on a bellows, the objective has to enter the ball to get focus. Do you have any advice, on how to overcome this. I have been using constant light, I will give the flash a go and see if it helps
Hi Peter, I really should have mentioned this, but this method, if you are going to do it using a hole in the back of the PP ball, works best with long-working distance objectives (30+ mm). When you are working at 10mm, you would be better off leaving the back wall intact, as a gradient backdrop.
Thinking it over about the difference between magnification and enlargement-a microscope is a good example, in that to achieve increased magnification one has to change the objective (lens) itself, for example, going from low to high power with many microscopes generally is 4x, 10x, 20x, 40x, 100x. with a fixed tube length and say a 10x eyepiece. Using a higher power eyepiece only enlarges the image-there is no increase in resolution. Many modern and more expensive microscopes can start with a 1x and 2x objectives and go all the way to 100x oil immersion, such as those I used in my work in pathology, mind you, those microscopes were around $30,000, plus all the camera attachments and accessories could bump the cost up nearer $45.000. We had a departmental budget of quarter million dollars for microscope upgrades each year to keep pace with modern developments.
Precisely! That is the key!
Rather than use a blade "as hot as the sun", you might try this idea: Take an X-acto or what ever replaceable blade craft knife is available in your area. Place the edge of the blade against the face of a file and hit the back of the blade with a small hammer. This will dent the sharp edge of the blade with a series of ridges from the file. In effect, this creates a very fine saw blade. I have seen this used for other crafts although never sawing into a ping-pong ball. Hope this helps and is less dangerous.
That is an insanely brilliant idea, Francis! As soon as my lacerations and burns heal, I'll give it a try!
When I view my Nikon NEF files in XMP viewer, they look uncorrupted. But when I convert the files to TIFF and then view them, I see a few corrupted files.
So, what software/technique do you use to weed out corrupted files prior to stacking?
My preferred workflow is to go through Adobe Lightroom. On import into LR my files are converted to .dng which can undergo any preliminary editing before export to ZS using the Zerene Plugin. This has worked flawlessly for many years. The files return to LR as .tif, which can be further edited or retouched in Photoshop.
TYVM. I'll need to follow suit but only after beefing up my computer - faster cpu, SSD in Raid 0 etc.
if i'm getting 2mm of ruler on the M50 crop sensor what times magnification would that be?
also does the crop sensor have better pixel density than a ff camera of te same mp
The M50 has a sensor size of 22.3 mm x 14.9 mm, therefore if you are measuring 2 mm across, you are getting 11 times magnification. You are correct to say that the pixel density of the M50 would be greater than that of a full frame 24 megapixel sensor. However, pixel density it's not a terribly useful characteristic to consider in isolation. Packing a larger number of photosites onto a smaller sensor, does not necessarily improve the performance of a sensor, and may have some negative consequences (smaller photosites limit low light performance and dynamic range, for example). But pixel density, as a yardstick for system performance, is not very useful compared to the diffraction characteristics of your lens, say.
Pixel density !?
What about it? Pixel density, by itself, is a relatively meaningless concept, in terms of image quality. Pixel density does not take into account the fact that the photosites have to be correspondingly smaller or that the sensor architecture must be modified. Further, any correspondence between pixel density and image quality must take into account the smaller APS-C image size. Even if the increased pixel density on a smaller sensor offers a theoretical improvement in resolution, it would be irrelevant compared to the far greater impact of diffraction on image quality. Choosing a crop frame camera because of the increased pixel density of its sensor, in the expectation of getting sharper macro images, will disappoint.
Winston and Audrey, LOL! I love it.
I think it be worth exploring what the crop factor does to the effective of aperture calculations ….always dodging the diffraction problem. Wide-open on a full frame SEEMs the highest quality . Condensing the available light down to 4/3 would give concentrated light and still 24 meg pixel? Which leads to the discussion isn’t the highest quality on almost any lens one f-stop down from the max aperature? So many rabbit holes to follow!
Many rabbit holes indeed - but to take your first point... The effective aperture is calculated by multiplying (the system magnification ratio + 1) by the nominal f/# of the lens. As the magnification effect of the crop frame sensor is illusory, the diffraction characteristics of the lens (reflected by the effective aperture) will not be affected by a change in sensor size.
I’d argue that for a given pixel count you are less likely to bump into diffraction limits on larger sensors. At 1x it’s maybe less important but at 5x or more, I often annoyingly bump into diffraction limits on a 24Mp full frame sensor at f/4. Simply put, smaller pixels = worse diffraction issues leading to softer images
Yes - absolutely. There are a lot variables to consider and the choices are not always intuitive, but trading pixel size for a higher pixel density has never made sense.
@@AllanWallsPhotography bigger pixels is also why I opted for the 24Mp sensor on the A7iii vs the 42.4Mp on the A7Riii.
@@AllanWallsPhotography oh and also, I loved the humor in your acronyms lol
@@AllanWallsPhotography it does make sense to have a higher pixel density in macro if you have 1) a very high quality lens 2) sufficient light intensity or a non- living subject that does not move 3) you have continuous LED illumination (for example three cheap150 watt non-flickering photo LED bulbs) and a pixel shifting sensor to reduce noise and increase sharpness 4) use basal ISO [and 5) you want a smaller camera and lens.]
Under these circumstances it becomes very relevant to compare the linear pixel density ( number of pixels per millimetre).
Good morning Allen, thanks for another nice piece.
I see you avoided the dreaded Pixel Density Challenge, which can be a horrible rabbit hole.
I am interested in printing, not so much in posting. So at the very least, we can say prints can be considered pixel pigs (pixel density oriented)!
With this in mind, one of the most important things to me is pixel density, and so I pose the question: Does a D7200 24mp APC-C give me a bigger print (cropped to the same area) as my D850 48mp FF? And from my calcs, (or dare I say it, a Canon 90D) the answer is no, the APS-C at 24mp (and the Canon 90D by a lot more) beats the 48mp FF camera by a mile. To be the same, my D850 would need to be at least 58mp. This is why I am hoping for a Z8 with a 60+mp sensor...
Calcs:
A given 24MP APS-C sensor is 23.4mm x 15.6mm and produces images 6000 x 4000 pixels
A generic 35mm (FF) sensor is 36.0mm x 24.0mm
6000 / 23.4 = 256.4 - or 256.4 pixels per mm of APS-C sensor horizontally
256.4 * 36 = 9230.8 - calculated horizontal pixels for a matching FF sensor
4000 / 15.6 = 256.4- pixels per mm of APS-C sensor vertically
256.4 * 24 = 6153.8 - this is the amount of vertical pixels in our FF sensor
Calculating a total in a similar FF sensor: 9230.8 * 6153.8 = 56805 - in our matching FF sensor, or ~57 megapixels
As usual, you're a pretty smart cookie, and maybe you'll prove me wrong, and teach me something along the way. So please, be snarky, and show me (in this narrow interpretation) where I errored.
Thank you...
When I initially made this video, it contained two parts that were subsequently removed. A demonstration of the magnification issue, which I didn't think was necessary, and a long discussion of sensor size, pixel density, and how they relate to magnification and enlargement. I think I should've left both parts in. There is nothing wrong with your logic, as far as it goes. But it's a mistake to put all your image quality/resolution eggs in one basket. I would avoid inferring pixel density from the number of pixels along one edge of the sensor - you can get away with it when the pixels are square, but it is safer to divide the pixel count by the total sensor size in mm² - which yields a pixels density of about 62,000 pixels per mm² for the D7500 and 55,550 pixels per mm² for the D850. If our only concern was pixel density then there would be no question, the best camera for the job would be the D 7500. But pixel density, taken in isolation is of very little value. The pixels on the D850 are a whopping 7% larger than the D7500. This is why the camera performs so well in low light and has such a high dynamic range. The pixel pitch is also 3% higher on the D850 sensor, which even further improves the signal to noise ratio. You would think that the spatial resolution of the D7500 would be better, and that may be true under certain circumstances, but pixel density alone does not translate to image quality. It is one, relatively small part of the quality equation, and must be considered in the context of the resolution of the lens and the signal to noise ratio. I would take the D850 over the D7500 if image quality was the only consideration, but I have both, because it seldom is. Pixel count and pixel density are important, but not that important.
@@AllanWallsPhotography Thanks Allan, all reasonable stuff. Since my comment centered around printing, I guess similar setups of the same subject out put at 300dpi and actually printed would really help. Maybe that's a rabbit hole worthy of your investigative talents.
Congratulations, I've been telling people this for years. Being an Optician, I understand the physics of optical systems, but there are some who simply won't believe this basics of optical physics, that the lens controls the magnification.
Why is the background in portraits then blurred with a full frame camera ?
One point not really clarified, though, is if both the ff sensor and crop sensor have the same number of MP (which is quite often the case) then to crop the ff image down to the same field of view as the crop sensor image, you will end up with an image with less MP, therefore not potentially as sharp due to lower resolution. I think this is why many macro shooters reach for the crop camera first. Sorry if this was mentioned at the end of the video...I got a bit peckish towards the end and made a grilled cheese.
As it happens, the last 10 minutes of the video dealt with this question in some depth, but I cut all of that in the interest of time, and because I was feeling a bit peckish too. You bring up some good points. Firstly, it is not that helpful to consider pixel density as an independent arbiter of image quality. For one thing, the performance characteristics of the larger photosites on a FF sensor (of the same MP count) have a much greater impact on image quality than the reduced megapixel density might imply. Secondly, the resolution differences are far beyond our ability to discern, even at fairly large print sizes. And lastly, it is the diffraction characteristics of the lens we chose and how we configure it, that actually influences image quality. The impact of diffraction on image quality is far greater, and becomes apparent much sooner than any theoretical impact on resolution related to the reduced pixel density of cropping a FF image. But possibly more to the point, both the full frame and crop frame images are going to be enlarged for display or printing. The smaller, crop frame image, is going to require 1.5 times as much enlargement to create a final image of the same size as that from a full frame sensor. This will result in a decrease in the image pixel density that will offset any decrease in pixel density from cropping the Image made on the largest sensor. It all comes back to my central point, we have to understand that the crop frame camera image is a smaller image than that of the FF sensor. Any conclusions drawn from the false assumption that the initial FF and APS-C images are the same size (as we would usually view them), simply perpetuates the primary misunderstanding of magnification as a function of sensor size.
Every time I go out to take macros I reach for my crop sensor camera because it's the only one I´ve got.
Brilliant, Peter! Hard to argue with that reason!
@@AllanWallsPhotography Sorry Alan, but I couldn't resist that one. I suspect we share a similar sense of humour.