Crop sensors better than full frame for macro photography?

[faustino]

Many people claim that cropped sensors are actually better than full frame and larger sensors for macro applications.
Here is an example:
http://www.43rumors.com/micro-four-thirds-and-macro-photography-by-eugene-kitsios/

Their argument is that for macro photography DoF is the limiting factor. They say that to get the same DoF, the larger the sensor, the smaller the aperture needs to be, which would offsett the light gathering capacity of larger sensors.

As an example, if we compare full frame vs. 4/3 sensors:
- full frame is roughly four times larger than 4/3 (measuring the surface), thus has virtually a two stop advantage in terms of light gathering capacity;
- anyhow, if with four thirds sensor we need to close the aperture to f/8, then on a full frame sensor we would need to close the aperture at f16 to get the same DoF, thus loosing two stop of light (offsetting completely the larger sensor advantage).
I am not sure if the second point is correct. The comparison shall be done considering different lenses on the two systems:
- if we use a 50mm macro on the 4/3 sensor, for correct comparison (to get the same angle of view), a 100mm lens should be used on the full frame (I know there would be slight differences in the final image due to the different aspect ratio).

I am skeptical about the above argument, anyhow I am unable to tell why it would be wrong.
What do you think, is really a crop sensor better than full frame for macro photography? Or is the full frame better? In the latter case, can you explain why?

Thanks!
Fausto

[neuroanatomist]

The argument is wrong.  The reason you have deeper DoF with a smaller sensor is that to frame the subject identically, you're further away from the subject or using a shorter focal length.  If you're using the same focal length, you're trading optical magnification for that extra DoF (at macro distances and at the same magnification, the crop sensor actually has very slightly shallower DoF than FF, but the area framed is smaller).  Also, the extent that you need to stop down the FF camera to match the DoF, you can boost the ISO to compensate and have the same image noise, so nothing lost.

I discussed the trade-offs a short time ago in another thread:

Fundamentally, in most use cases, the only meaningful benefit of an APS-C camera is the lower cost….

Macro is potentially another one of those 'exception' use cases.  I say potentially, becuase it really depends on what you mean by macro.  Formally, 'macro' generally means 1:1 magnification (or higher), and anything less than that (0.1x - 1x mag) is considered 'close-up' photography.  If you're at 1:1 with a macro lens, you're at the MFD (or closer if you go higher with tubes) - that means there's no DoF 'advantage' (i.e. more of it) with APS-C.  The APS-C sensor (e.g. 18-20 MP) is going to give you more pixels on target at a given magnification compared to a FF sensor (e.g. 20-22 MP), whereas the FF sensor will give you a wider AoV at that magnification. 

Below is an example of that (shot with the MP-E 65mm on a 1D X and EOS M, both at the same distance from the coin for equivalent optical magnification, and since both use 18 MP sensors, no down- or up-scaling is required for the comparison).  At 1x mag with FF you can frame nearly the whole quarter while APS-C crops much of the coin away but gives you higher resolution at the pixel level.



Personally, I usually choose FF for macro (and have done so since having both the 7D and 5DII).

[alexturton]

Take the argument to the extreme and consider something like and iphone 5s. It has a tiny sensor so everything is always in focus. But it still has an f2 lens in terms of light transition. So sometimes I can get a better macro shot in low light on my 5s than I could on my 5d mk iii. Because to get an equivalent depth of field I'd have to stop down to f18 which would push up the ISO to extreme.

[faustino]

Thank you for your super quick replies.
I have only one additional question, probably redundant, to be sure my understanding is correct:

Shooting the same subject at the same distance with the following setups, Will I get the same image in terms of quality, perspective, and DoF (more or less)?
- Crop sensor setup: EOS 7d + EF-S 60mm f2.8 Macro USM @f/8 aperture
- Full frame sensor setup: EOS 1dx + EF 100mm f/2.8L Macro IS USM @f/11 aperture

If yes, then there would be no technical advantage of one setup over the other, except weight in favor of the crop setup, and flexibility in favor of the full frame setup (e.g. 1dx + 100mm used @f2.8 for portrait, combination that would not give the same result with the crop setup).

[Albi86]

Many people claim that cropped sensors are actually better than full frame and larger sensors for macro applications.
Here is an example:
http://www.43rumors.com/micro-four-thirds-and-macro-photography-by-eugene-kitsios/

Their argument is that for macro photography DoF is the limiting factor. They say that to get the same DoF, the larger the sensor, the smaller the aperture needs to be, which would offsett the light gathering capacity of larger sensors.

As an example, if we compare full frame vs. 4/3 sensors:
- full frame is roughly four times larger than 4/3 (measuring the surface), thus has virtually a two stop advantage in terms of light gathering capacity;
- anyhow, if with four thirds sensor we need to close the aperture to f/8, then on a full frame sensor we would need to close the aperture at f16 to get the same DoF, thus loosing two stop of light (offsetting completely the larger sensor advantage).
I am not sure if the second point is correct. The comparison shall be done considering different lenses on the two systems:
- if we use a 50mm macro on the 4/3 sensor, for correct comparison (to get the same angle of view), a 100mm lens should be used on the full frame (I know there would be slight differences in the final image due to the different aspect ratio).

I am skeptical about the above argument, anyhow I am unable to tell why it would be wrong.
What do you think, is really a crop sensor better than full frame for macro photography? Or is the full frame better? In the latter case, can you explain why?

Thanks!
Fausto

For tripod-based macro work, where you can work at ISO 100 and maybe set the lighting how you like, the advantages of a bigger sensor don't really show all that much. The higher pixel density can actually be the decisive advantage, since macro lenses are typically very sharp.

[faustino]

Take the argument to the extreme and consider something like and iphone 5s. It has a tiny sensor so everything is always in focus. But it still has an f2 lens in terms of light transition. So sometimes I can get a better macro shot in low light on my 5s than I could on my 5d mk iii. Because to get an equivalent depth of field I'd have to stop down to f18 which would push up the ISO to extreme.

Well... I have seen people taking better pictures with their iPhone than I with my 5dmk3 + a bunch of L series lenses. Though, I have to candidly admit, the problem was not in the reflex; rather, it was located few centimeters behind the viewfinder.
 

[9VIII]

Once again how much of an advantage you get depends on how much cropping you're doing. If you're shooting a bug that only takes up a fraction of the frame then crop sensors will have the advantage, and if you're like me and you prefer looking at things at 100% all the time anyway, then crop should also be the way to go.

The only time I would care about using a lower aperture vs. a smaller sensor is if moving to such a small aperture that it starts to affect image quality.
In my last bunch of macro shots I could see the difference in detail between f16 and f22, but my framing wasn't such that the loss of fine detail actually ruined the image.
The bigger problem was actually long exposure sensor noise, which can be corrected in post, but it's still at a cost to IQ.
After that I ordered a proper soft box and multi-bulb CFL adapters to stick in lamps (should produce around 8000 lumens), getting my shutter speed down should help more than anything.

[neuroanatomist]

Take the argument to the extreme and consider something like and iphone 5s. It has a tiny sensor so everything is always in focus. But it still has an f2 lens in terms of light transition. So sometimes I can get a better macro shot in low light on my 5s than I could on my 5d mk iii. Because to get an equivalent depth of field I'd have to stop down to f18 which would push up the ISO to extreme.

Fine.  Now, because image noise is determined by the total light gathered, which is proportional to the area of the sensor, you have to figure in the effect of sensor size on ISO noise.  What that means is a shot on the iPhone 5s at base ISO (ISO 32) has the same level of noise as a shot on your 5DIII at ISO 1600.  Want to see what the noise at max ISO on the iPhone 5s (ISO 3200) would look like on the 5DIII?  Set ISO 102400 (H2) and push the exposure 2/3-stop in post.  Ok, the iPhone doesn't look that bad, but that's only because of massive NR that obliterates the detail in the iPhone shot.

So, no - there's no DoF advantage to a smaller sensor.

[faustino]

What about the previous comparison I was suggesting?

Which setup would give a better image?
- Crop sensor setup: EOS 7d + EF-S 60mm f2.8 Macro USM @f/8 aperture shooting from 10cm
- Full frame sensor setup: EOS 1dx + EF 100mm f/2.8L Macro IS USM @f/11 aperture shooting from 10 cm

Or would it be a tie? My understanding is that it would more or less be a tie.

Thanks,
Fausto

[neuroanatomist]

You'd get a sharper image from the FF sensor.

[mackguyver]

You'd get a sharper image from the FF sensor.

Yep - check this out:
http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=335&Camera=736&Sample=0&FLI=0&API=4&LensComp=674&CameraComp=453&SampleComp=0&FLIComp=0&APIComp=5

Or how about 100L @f/8 on crop vs. 180L @f/11 on FF:
http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=674&Camera=736&Sample=0&FLI=0&API=4&LensComp=109&CameraComp=453&SampleComp=0&FLIComp=0&APIComp=4

[faustino]

You'd get a sharper image from the FF sensor.

Yep - check this out:
http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=335&Camera=736&Sample=0&FLI=0&API=4&LensComp=674&CameraComp=453&SampleComp=0&FLIComp=0&APIComp=5

Or how about 100L @f/8 on crop vs. 180L @f/11 on FF:
http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=674&Camera=736&Sample=0&FLI=0&API=4&LensComp=109&CameraComp=453&SampleComp=0&FLIComp=0&APIComp=4

Very informative! Thanks!

[digital paradise]

I had far better success with my FF over my crop bodies for IQ. I'm talking extreme macros. MP-E lens   

[CarlTN]

Many people claim that cropped sensors are actually better than full frame and larger sensors for macro applications.
Here is an example:
http://www.43rumors.com/micro-four-thirds-and-macro-photography-by-eugene-kitsios/

Their argument is that for macro photography DoF is the limiting factor. They say that to get the same DoF, the larger the sensor, the smaller the aperture needs to be, which would offsett the light gathering capacity of larger sensors.

As an example, if we compare full frame vs. 4/3 sensors:
- full frame is roughly four times larger than 4/3 (measuring the surface), thus has virtually a two stop advantage in terms of light gathering capacity;
- anyhow, if with four thirds sensor we need to close the aperture to f/8, then on a full frame sensor we would need to close the aperture at f16 to get the same DoF, thus loosing two stop of light (offsetting completely the larger sensor advantage).
I am not sure if the second point is correct. The comparison shall be done considering different lenses on the two systems:
- if we use a 50mm macro on the 4/3 sensor, for correct comparison (to get the same angle of view), a 100mm lens should be used on the full frame (I know there would be slight differences in the final image due to the different aspect ratio).

I am skeptical about the above argument, anyhow I am unable to tell why it would be wrong.
What do you think, is really a crop sensor better than full frame for macro photography? Or is the full frame better? In the latter case, can you explain why?

Thanks!
Fausto

I agree, for macro photography, a crop sensor has the edge.  Unless of course, you're a fan of focus stacking 10 or more images...which I'm not.  A full frame sensor necessarily requires more focus stacking, unless you want your macro images with shallow depth of field...in which case either FF or crop are ok for that.  I don't profess to be a macro expert, though...but I do know that a macro image I shot with an effective 270mm focal length at f/20 on a crop camera, would have needed to be at f/45 or smaller on a full frame sensor, and would have provided 1.6x less magnification to boot.

Let's just be honest and say that generally, full frame sensors have an advantage for medium or wide angle, non macro photography (the advantages there are vast and rarely disputed).  Full frames can also be useful for longer distance telephoto photography (and can be vital for it in low light).  But under optimum conditions, good light, and with the "perfect" lens, a crop sensor can and usually does have a reach advantage.  And for macro with good or controlled light (and especially with stationary subjects and tripods or other mounting devices), I honestly see a disadvantage with a full frame sensor, over a crop sensor (unless perhaps the very poor performing 12 MP crop sensor that was recently discontinued).

[LetTheRightLensIn]

It's sort of a wash. The APS-C cameras often have higher density but you can get that back with an extension tube and if you take things really far DOF and so on become tricky so it's not quite like birding where you can always easily go for more and more reach.

I don't have time to get into it but the short of it is that is basically comes out to be a wash and you can use APS-C or FF DSLR equally well for macro stuff except in the very rare occasion so long as you have a few cheap extension tubes (and sometimes those won't even be needed) and if you have the MPE then it's really pretty meaningles between the two other than in very particular and tricky scenarios when you are really maxing out an dhave the ability and scenario that lets you max out without messing up.

The size vs mag vs diffraction vs noise and so on all just kinda balances out for macro, again don't have time to get into it, maybe someone has.

For stuff like distant birds though the aps-c cams with high density are hard to beat with FF though.