Age old question revisited

Does APSC really have deeper DOF for macro?

Been researching this and this link:

http://www.bobatkins.com/photography/technical/format_vs_macro_dof.html

have some really good info.

But there are a lot of arguments and confusion.

Base one BobAtkins's page, I am trying to figure out if I consider pixel on target or a fixed output number of pixel (say 4680 × 3166) for my frame, does APSC fair better?

Basically, I am ignoring magnification definition of 1X equal image size on sensor == real object size. But just want to get a 4680 × 3166 framing of the subject.

So if we consider the MP the sensor have:

(using macro DOF calculator from http://www.cambridgeincolour.com/tutorials/macro-lenses.htm)

FF DOF at 1X is 1.02 mm

So to have the same 4680 × 3166 frame as the FF ref above, my 24MP APSC (1.5x) should have the following magnification (ie magnification that can give the same framing above).

FF MP : APSC Magnification : DOF

22 : 0.4074074074 : 2.86 mm

35 : 0.6481481481 : 1.32 mm

42 : 0.7777777778 : 0.99 mm

50 : 0.9259259259 : 0.76 mm

Note: According to BobAtkins's page, Mag of 0.625x on APSC give the same framing as the 1X FF.

So, am I correct to conclude that if I want the same MP on subject, unless my FF is 45MP and above, my 24 MP APSC will be slightly better in terms of DOF?

If I include pixel pitch (pixel size) and its impact on noise into consideration, how will this chance my choice of sensor?

And if I include consideration for diffraction softening, will the 24MP APSC be less sharp (ie have more diffraction softening) compare to 35MP FF?

-kk
I believe the short, and correct, answer to your question "Does APSC really have deeper DOF for macro?" is what John K has said. "No".

Perhaps I could tackle this from a different angle.
Nick, you post is very helpful. Thanks and sorry for the late reply (see my prev post for reason). I took your advise and try to test things out with the cameras I have.
I use cameras with different sensor sizes. I don't have a full frame camera but I do have an APS-C camera, a micro four thirds camera and a bridge camera with a very small sensor.

As I understand it (happy to be corrected of course)...

I can get a very similar maximum DOF with each of them, but I have to use a different aperture on each to do so. If I capture the same scene (same angle of view) I will get pretty much the same DOF using f/28 on my APS-C camera, f/22 on my micro four thirds camera and f/8 on my bridge camera. This accords both with my own experience and with what the Depth of Field Equivalents calculator says on this page from Cambridge in Colour.
That make a lot of sense.
The loss of sharpness/detail from diffraction is the same in these three cases. For example see http://www.josephjamesphotography.com/equivalence/#diffraction where it says that "All systems suffer the same diffraction softening at the same DOF". Note though that it goes on to say "but do not necessarily resolve the same detail at the same DOF, as diffraction softening is merely one of many forms of blur (e.g. lens aberrations, motion blur, large pixels, etc.)"

Also note what it says next: "As the DOF deepens, all systems asymptotically lose detail, and by f/32 on FF (f/22 on APS-C, f/16 on mFT -- 4/3), the differences in resolution between systems is trivial, regardless of the lens, sensor size, or pixel count." That is why I can get as good resolution with my bridge camera as with my other cameras because I use minimum aperture to get maximum DOF, and the apertures I use are one stop less than those stated here.
I have also see very good macro shots for some small sensor camera. FZ series are very good. Which one are you using?

This means I can use the most convenient camera I have for macro (weight, size etc). Which is good, but also sad that the expensive gears don't pay off. :)
You are quite probably not using minimum aperture; most people don't. However you need to be aware that (unless you are using a Nikon camera and lens) it is the effective aperture (the actual aperture) that you are using that matters rather than the nominal aperture (what the camera says you are using). And the effective aperture depends on the magnification. The usual formula (an approximation I believe) used to calculate the effective aperture is:

Effective aperture = Nominal aperture x (1 + magnification)

Therefore, if you have set the camera lens to f/11 and are working at a magnification of 1:1 the effective aperture will be f / (11 x (1 + 1)) = f/22. So if you are using APS-C you will be in (and if you are using FF you will be near) the range of "trivial" difference in resolution "regardless of lens, sensor size, or pixel count".

Note 1: The qualification about Nikon cameras and lenses is that they may tell you the effective aperture rather than the nominal aperture.

Note 2: The formula applies to macro lenses, extension tubes, teleconverters, reversed lenses and bellows, but not to close-up lenses.
Good info. I am using canon so need to take this into account. I do use f11 for FF on some of my test shots. But even at F22 (effective F32, I think), things still looks ok. Will test out more.

So for APS-C, I will try out F16 (effective F22).
I found this hard to believe and tried hard to prove it wrong, trying to capture the same scene with my Canon 70D and first (and briefly) a Canon 100L IS macro and later with a Sigma 105 Macro DG HSM. At f/22 with the either of these lenses on my 70D I did not obtain significantly better resolution than I could obtain with my bridge camera using a close-up lens at f/8. Similarly using a close-up lens at f/22 on a telephoto zoom with my micro four thirds camera or up to f/32 with a close-up lens on a telephoto zoom on my 70D.

To the extent that you will be using effective apertures which take you into the asymptotically resolution-degraded area it seems that it won't matter too much which camera you use from the point of view of achievable resolution.
I too find it hard to believe. But you are not the first person who share this. I have read other good macro photographer said the same thing.
It might be a good idea to do some practical experiments rather than getting too concerned about the theoretical aspects. Various equipment and techniques work fine for some people and not so well or not at all for other people. Only by practical testing will you be able to establish what works for you - and by what works I mean the whole sequence from finding subjects through image capture, selection, post processing and display. For example, I use several techniques which many other people know are not effective/practical/useful/possible, but they work fine for me, for my purposes. There are lots of ways of handling close-up and macro work. Try some of them and see how it goes.
Definately. A few test shots in the garden (spring is finally here!!) will be easy enough. :)
Well, that's how it seems to me anyway. FWIW. :)
Thanks again Nick.
 
As I said earlier in my reply, What if you stop down 1 stop with the FF camera?
 
In other words (to paraphrase), to get 'the less' magnification on a crop-sensor, the subject needs to be further away with the same lens/focal length. And with the subject furter away one 'automatically' gets more DOF.

Right, right?
No, the magnification that matters for DOF is the difference in size between life size and final display size. You use less optical magnification on the APS-C sensor, but the image from the sensor is then magnified more to get to the same display size, so it cancels out and DOF is the same. As for IQ, the larger sensor has the advantage at base ISO because you can capture more light before you saturate the sensor. Macro is just like every other type of photography in that regard, the largest sensor delivers the best results as long as you have enough optical magnification to use the whole sensor. If you have to crop the FF to match the FOV of the smaller sensor, the playing field is leveled and the sensor with the highest pixel density wins.
I think it is easy to believe that FF (or bigger sensor) is always better (for IQ).

The equivalent say so. But what Nick and others in other sources that I have read do seems to point at the fact that at max DOF, things are not better for larger sensor. What do you think?

BTW, macro DOF calculator (such as the one at CambridgeInColor) do take the larger "magnification" on the APS-C output to get the same display size into account when they compute the DOF isn't it? If I use that to estimate my DOF, will I still have to consider that?
 
I am very confused. As you guys can see from my reply, I am trying to make sense of things. :)
 
Thanks you for your patient and your effort in trying to help me understand.

I don't get the point about IQ is separated from magnification and DOF. So let me show my thinking process and you can point out what I start to get it wrong.

I think magnification (sensor area on subject) is directly link to IQ. Because we could just take a photo and crop like crazy and get a fly/bee in the photo frame but end up with just using a tiny area of the sensor. So the IQ will be terrible.

So macro shooter aim to get the best magnification i.e trying to use their sensor area as much as they could for a given subject. If you said IQ and magnification is not linked, then I am at a lost.

That is how I start to think about pixel on subject. Thinking that if we maintain a similar pixel on subject, the IQ will be maintained (or it limit how much cropping I can do). Since we have different sensor size and different MP count, I just try to focus on what I get as a final result, a photo with a particular pixel count on my subject.
You're correct in terms of cropping. However, let's take two cameras, one full frame and the other APS-C. Each camera is 20mp. You take a picture of a fly. The fly is 36mm in length. When you take a picture of the fly with a full frame camera at 1:1 magnification, the entire fly will fit. So, your picture will be of the ENTIRE fly at 20mp.

Now, let's take the same fly and shoot it with an APS-C camera so that it will fit the entire frame. When you take a picture of the entire fly on an APS-C camera, your magnification will be 1:2 (half lifesize), but you STILL have the entire fly in the frame at 20mp.

So, your image quality is the same with both cameras, BUT your DOF will be GREATER with the APS-C camera because you're shooting at a LESSER magnification in order to get the fly into the frame. Why are you shooting at a lesser magnification to get the fly into the frame? Because the APS-C sensor is SMALLER, therefore, you need LESS magnification to get the SAME SIZE object to fill a SMALLER frame.

I hope that makes sense.
Sorry for the late reply. Was sick and been trying to think about this and trying some experiments to check out what I read here.

I think that make sense. I have tested it and it seems to be correct.
Great! We're making progress

But what if I stop down my FF lens by one stop to match the APS-C? Would that not make the DOF the same for the two system?
Yes

FF one one stop of noise advantage, with the same amount of light, I can stop down 1 stop and raise iso by one stop. Isn't it?
Theoretically yes, but it depends on the camera's implementation. If you're using cameras that are of the same generation, have the same resolution and the same implementation of the A/D conversion. However, that's a lot of "ifs".

It's not entirely accurate to say that FF frame cameras have a 1 stop noise advantage. It depends on the factors outlined above. For example, if a full frame camera has considerably greater resolution than a crop frame, it may not necessarily have a noise advantage.

So, if you want to increase the depth of field on a FF camera, you'd be most likely better off by knocking down your shutter by 1 stop.
 
Since APS-C sensors were introduced everyone has been trying to make them sound as if they are somehow magical.

An APS-C sensor is just a physical crop of the image circle -and fundamentally no different than cropping in post. Since cropping in post does not increase magnification and APS-C sensor does not increase magnification. Some peeps might jump into this tread with some really twisted logic, but it's all BS...
The problem is not magnification. It is resolution. You can take a blurry picture of a bee and blow it up to 2 meters wide, and have enormous but useless magnification.

What you want is to see the finest details on the subject, not the biggest picture. This is where the sizes of pixels come in. You want whichever setup gives the highest number of pixels on the subject.

Depth of field is a different problem again. It is increased by stopping down (with problems from diffraction) and by using lower on-sensor magnification. So again, smaller pixels are good, because they give you some leeway to use a lower on-sensor magnification, without losing too much detail.

The number of pixels on the sensor doesn't matter much. Their size does. And in most cases, FF sensors have larger pixels than APS-C sensors. (And the cameras cost more.)
After my initial attempts, I had been putting number of pixels on subject aside. Trying to focus on just IQ you get from just considering the sensor size.

I am getting the hang of things, but adding in "number of pixels on subject" still confuses me. Because, when I start thinking about "number of pixels on subject", I am normally doing crop. Ie more pixels allow more cropping freedom.

If, as Nick said, all sensor size are equal at max DOF. Then I will choose a sensor with max pixel pitch (ie max "number of pixels on subject"). So comparing a 22MP FF and 24MP APS-C, I think the second will capture more details, If I want to crop.

This is important because, with my current skill level, I find that framing is quite challenging, so able to crop after capture will make things easier for me.
 
That make sense. Slowly things are falling into places for me (in my brain).

Thanks for taking the effort to help me out.
 
That make sense. Slowly things are falling into places for me (in my brain).

Thanks for taking the effort to help me out.
No problem. I'm glad to be able to help. Once you understand the relationships among sensor size, magnification and image filling the frame, we can talk about resolution, pixel size and so on.
 
Ok, when I am ready, will go into pixel and cropping etc.

Will take some time to test out what I have learn from this thread, capture some samples and maybe they can be use as examples for the pixel and cropping discussion.

-kk
 
Ok, when I am ready, will go into pixel and cropping etc.

Will take some time to test out what I have learn from this thread, capture some samples and maybe they can be use as examples for the pixel and cropping discussion.

-kk
Sounds like a plan! Keep us posted.
 
I think it is easy to believe that FF (or bigger sensor) is always better (for IQ).

The equivalent say so. But what Nick and others in other sources that I have read do seems to point at the fact that at max DOF, things are not better for larger sensor. What do you think?
The larger sensor always has the potential for better IQ. If you shoot macro with flash as many of us do, you can use base ISO and set your flash power to capture as much light as possible without over saturating the sensor (ETTR ). In this situation the larger sensor can capture more light and deliver better image quality. The large sensor loses it's advantage if you have to crop to match the field of view of the smaller sensor, or if you are shooting in natural light above base ISO where the larger sensor will use higher ISO to match the DOF and shutter speed of the smaller sensor.
BTW, macro DOF calculator (such as the one at CambridgeInColor) do take the larger "magnification" on the APS-C output to get the same display size into account when they compute the DOF isn't it? If I use that to estimate my DOF, will I still have to consider that?
I think the DOF calculator takes everything into account, but DOF calculators aren't really very useful for macro. The reality is that you are never going to have enough DOF, so you just shoot at the smallest aperture possible before diffraction becomes unacceptable to you. This is a matter of personal taste rather than something that can be calculated.
 
I have also see very good macro shots for some small sensor camera. FZ series are very good. Which one are you using?
I use an FZ200.I use it for mid-sized invertebrates - wasps, bees, flies, spiders etc. Obviously there are smaller species of these, but I'm talking about the ones of a size you can easily see when you are walking around. I mainly use a Raynox 150 close-up lens on the front of the FZ200 for this sort of subject, although I use a Raynox 250 for subjects at the smaller end of the mid-sized spectrum. Here are a couple of typical shots captured a few days ago with the Raynox 150 on the FZ200. They were captured at f/8, which is equivalent to f/45 on your full frame camera.

c1b57e63d51a44789a2a522d2c4076c8.jpg

67b959aef6f240f0a6a34c60de943ba8.jpg

The FZ200 is notoriously noisy (even compared to other small sensor cameras), and because of the way I capture my images (often underexposed to preserve highlights), I raise shadows a lot and this can bring out noise even at ISO 100 on the FZ200. This makes two things especially important: lighting and post processing.

I generally use flash with the FZ200 and this allows me to stick to ISO 100. These albums at Flickr illustrate my current flash setup, although this is subject to change at any time as I frequently experiment with alternatives.

https://www.flickr.com/photos/gardenersassistant/sets/72157665186197321/

https://www.flickr.com/photos/gardenersassistant/sets/72157666072580655/

I have put a lot of time and effort into trying to find a combination of post processing products and techniques that works well for me. Like my diffusion arrangements I experiment a lot with alternatives. As an example the first of the above images was processed with DXO Optics Pro 10 followed by Lightroom 6, the second with DXO followed by Silkypix 7 Pro. (These are observations btw, not recommendations of particular software products. Like cameras, diffusion arrangements, aperture usage, use of autofocus, use of a tripod etc, post processing preferences are a personal matter and what works best for one person may be useless for someone else.)
This means I can use the most convenient camera I have for macro (weight, size etc). Which is good, but also sad that the expensive gears don't pay off. :)
It may be a case of using the best equipment for particular tasks.

The approach I use only works if you want large DoF with single captures. If you are doing stacking for example you will get better results with a large sensor camera used with a macro lens at its "sweet spot" aperture. Similarly if you prefer narrower DoF shots you will do better with a larger sensor camera.

I routinely have two cameras with me and sometimes three. (I usually go around with one bag of equipment and sometimes two, depending on the circumstances.)

For flowers I find I can sometimes get reasonable images from an FZ200. I'm ok with this sort of thing for example.

FZ200, f/8, FF equivalent f/45

1f3452d62e57457b893bc8d5df2893b2.jpg

FZ200 f/4, FF equivalent f/22

1d2893b2594747588aa844f4946a5305.jpg

However, for flowers (for which I almost never use flash btw whatever camera I am using, except perhaps very occasionally fill flash ) I typically use my Canon 70D because I think I can get subtler and more pleasing colours and textures from the raw files it produces (I shoot raw exclusively btw).

I use the 70D with an f/4-5.6 55-250 STM zoom lens rather than a prime f/2.8 macro lens (I have a Sigma 105 macro but I don't use it, preferring the zoom functionality). Even with this zoom lens I get a greater range of options for (narrower) DoF than with the FZ200. (The maximum aperture I can use with the FZ200 is f/2.8, which is FF equivalent f/16)

For the smallest subjects I deal with I use a different camera again, a micro four thirds Panasonic G5 with a Raynox 202 or 505 on a 45-175 lens. I don't do much with very small subjects, but when I do I find the G5 setup works best for me for subjects like this. (These were captured with the G5 setup, but they are not really tiny, like mites for example, but I have virtually no images of subjects that small.)

5a55a4df87a24d9ab3a3dab6068cab66.jpg

e3a52737823242978fed068d1075be5e.jpg

Keep the full frame. Use it for what it is best at. Horses for courses. :)

--
Nick
http://www.flickr.com/photos/gardenersassistant/collections/
GardenersAssistant Photography Videos - https://www.youtube.com/channel/UCmBgEwRDfiQMYTPORSzDxvw
https://www.talkphotography.co.uk/t...-dslr-primes-a-journey-of-exploration.531050/
 
Last edited:
I have also see very good macro shots for some small sensor camera. FZ series are very good. Which one are you using?
I use an FZ200.I use it for mid-sized invertebrates - wasps, bees, flies, spiders etc. Obviously there are smaller species of these, but I'm talking about the ones of a size you can easily see when you are walking around. I mainly use a Raynox 150 close-up lens on the front of the FZ200 for this sort of subject, although I use a Raynox 250 for subjects at the smaller end of the mid-sized spectrum. Here are a couple of typical shots captured a few days ago with the Raynox 150 on the FZ200. They were captured at f/8, which is equivalent to f/45 on your full frame camera.

c1b57e63d51a44789a2a522d2c4076c8.jpg
The FZ200 is notoriously noisy (even compared to other small sensor cameras), and because of the way I capture my images (often underexposed to preserve highlights), I raise shadows a lot and this can bring out noise even at ISO 100 on the FZ200. This makes two things especially important: lighting and post processing.

For flowers I find I can sometimes get reasonable images from an FZ200. I'm ok with this sort of thing for example.
The FZ-200 is certainly giving you excellent results with insects. Very nice indeed.

The flowers, however, are not very good, in my opinion. Horses for courses.
 
I have also see very good macro shots for some small sensor camera. FZ series are very good. Which one are you using?
I use an FZ200.I use it for mid-sized invertebrates - wasps, bees, flies, spiders etc. Obviously there are smaller species of these, but I'm talking about the ones of a size you can easily see when you are walking around. I mainly use a Raynox 150 close-up lens on the front of the FZ200 for this sort of subject, although I use a Raynox 250 for subjects at the smaller end of the mid-sized spectrum. Here are a couple of typical shots captured a few days ago with the Raynox 150 on the FZ200. They were captured at f/8, which is equivalent to f/45 on your full frame camera.

c1b57e63d51a44789a2a522d2c4076c8.jpg

The FZ200 is notoriously noisy (even compared to other small sensor cameras), and because of the way I capture my images (often underexposed to preserve highlights), I raise shadows a lot and this can bring out noise even at ISO 100 on the FZ200. This makes two things especially important: lighting and post processing.

For flowers I find I can sometimes get reasonable images from an FZ200. I'm ok with this sort of thing for example.
The FZ-200 is certainly giving you excellent results with insects. Very nice indeed.

The flowers, however, are not very good, in my opinion. Horses for courses.
Indeed so. That's why I generally use the 70D for flowers. Sometimes though I'd rather have an FZ200 flower image than none at all when those are the only available options.



--
Nick
GardenersAssistant Photography Videos - https://www.youtube.com/channel/UCmBgEwRDfiQMYTPORSzDxvw
 
Since APS-C sensors were introduced everyone has been trying to make them sound as if they are somehow magical.

An APS-C sensor is just a physical crop of the image circle -and fundamentally no different than cropping in post. Since cropping in post does not increase magnification and APS-C sensor does not increase magnification. Some peeps might jump into this tread with some really twisted logic, but it's all BS...
The problem is not magnification. It is resolution. You can take a blurry picture of a bee and blow it up to 2 meters wide, and have enormous but useless magnification.

What you want is to see the finest details on the subject, not the biggest picture. This is where the sizes of pixels come in. You want whichever setup gives the highest number of pixels on the subject.

Depth of field is a different problem again. It is increased by stopping down (with problems from diffraction) and by using lower on-sensor magnification. So again, smaller pixels are good, because they give you some leeway to use a lower on-sensor magnification, without losing too much detail.

The number of pixels on the sensor doesn't matter much. Their size does. And in most cases, FF sensors have larger pixels than APS-C sensors. (And the cameras cost more.)
After my initial attempts, I had been putting number of pixels on subject aside. Trying to focus on just IQ you get from just considering the sensor size.

I am getting the hang of things, but adding in "number of pixels on subject" still confuses me. Because, when I start thinking about "number of pixels on subject", I am normally doing crop. Ie more pixels allow more cropping freedom.

If, as Nick said, all sensor size are equal at max DOF.
No. I said/suggested/quoted that in terms of resolution there is no significant difference past a certain (equivalent) aperture irrespective of sensor size (or lens or number of pixels). I believe/suspect/imagine (I have no evidence, it is not something I have delved into) that a camera with a larger sensor may have advantages even when operating at the same (effective) resolution as a camera with a smaller sensor. (It's similar to the argument that given two cameras with the same number of pixels, the one with the larger sensor - and larger pixels - will produce better results other things being equal, because its pixels are bigger/deeper/more discriminating.)
Then I will choose a sensor with max pixel pitch (ie max "number of pixels on subject"). So comparing a 22MP FF and 24MP APS-C, I think the second will capture more details, If I want to crop.

This is important because, with my current skill level, I find that framing is quite challenging, so able to crop after capture will make things easier for me.
 
we live in the 2016 not the 1950s the most pixels on the subject wins. lens and sensor has nothing to do with viewing the most eye detail on your monitor.

cheers don
 
we live in the 2016 not the 1950s the most pixels on the subject wins. lens and sensor has nothing to do with viewing the most eye detail on your monitor.
Don, thank you so much for this illuminating comment. It means that my 18 megapixel TZ60 (currently £199 at Amazon UK) gives me better detail than I could get with a 12 megapixel full frame Sony Alpha 7S II with 70-200 F2.8 G SSM 2 lens (currently £4,811 at Amazon UK). That is excellent news.
cheers don
 
we live in the 2016 not the 1950s the most pixels on the subject wins. lens and sensor has nothing to do with viewing the most eye detail on your monitor.
Don, thank you so much for this illuminating comment. It means that my 18 megapixel TZ60 (currently £199 at Amazon UK) gives me better detail than I could get with a 12 megapixel full frame Sony Alpha 7S II with 70-200 F2.8 G SSM 2 lens (currently £4,811 at Amazon UK). That is excellent news.
cheers don

--
Olympus xz1, e-pL5 , EM5 my toys.
http://www.dpreview.com/galleries/9412035244
past toys. k100d, k10d,k7,fz5,fz150,500uz,canon G9.
--
Nick
http://www.flickr.com/photos/gardenersassistant/collections/
GardenersAssistant Photography Videos - https://www.youtube.com/channel/UCmBgEwRDfiQMYTPORSzDxvw
https://www.talkphotography.co.uk/t...-dslr-primes-a-journey-of-exploration.531050/
im sure your 70 200 can do better :)





--
Olympus xz1, e-pL5 , EM5 my toys.
past toys. k100d, k10d,k7,fz5,fz150,500uz,canon G9.
 
wow that combo would capture 500 pixels on the fly, amazing detail don't you think ?

cheers don
 
we live in the 2016 not the 1950s the most pixels on the subject wins. lens and sensor has nothing to do with viewing the most eye detail on your monitor.
Don, thank you so much for this illuminating comment. It means that my 18 megapixel TZ60 (currently £199 at Amazon UK) gives me better detail than I could get with a 12 megapixel full frame Sony Alpha 7S II with 70-200 F2.8 G SSM 2 lens (currently £4,811 at Amazon UK). That is excellent news.
im sure your 70 200 can do better :)

Of course you know that diffraction puts a limit on resolution that more pixels can't improve. In the case of your fly shot with the FZ150 at f/8 that limit is about 6MP. This affects all systems equally at the same DOF, but a larger sensor could have helped with all that noise.
 
Last edited:

Keyboard shortcuts

Back
Top