Macro pics - FF or APS-C

[jebrady03]

I have debated the merits of upgrading to FF for a while now.  I think the pros outweigh the cons (other than cost) for the most part except possibly in one very important area for me, and that's macro shots.  Could someone help me to determine if my logic/research is correct?

I currently shoot with a 60D and the very good EF-S 60mm macro.  Generally for macro, you're not shooting wide open, or even close to it.  In general, I'm at f8 (range varies from f5.6-f11 most of the time) so that I get a somewhat decent depth of field.

If I were to switch to FF (either 6D or 5D3), I'd be using the 100mm macro instead (similar FOV to the 60mm macro on the 60D ~ 96mm).  To obtain a similar depth of field, would I not need a much smaller aperture?

For instance, I just used http://www.dofmaster.com/dofjs.html to come up with the following scenario:

If I were using my current setup (60D + 60mm macro), the following parameters:
f8, 6.3' distance to subject
Would yield the following depth of field:
1 ft

***I realize this is not a macro shot but I used it because "1 ft" is a nice round number with some "play" to it - an actual macro calculation would have resulted in a very small fraction of a foot making a comparison much harder.

Now, if I were to pick up say, a 5D3 and the 100mm macro, in order to achieve 1 ft of depth of field at a distance of 6.3', I'd need to shoot at roughly f14.3 (the online calculator says this would be .99 ft).

Here's the problem...  Diffraction/lens sharpness.  I'm fairly sure that even as good as the 100mm macro is (even the L), it can't possibly be as sharp at f14 as the 60mm macro is at f8, can it?

What about when I'm in a pinch and need to go to f11 with my 60mm macro on APS-C?  I'd need to go to over f20 to achieve a similar DOF with the 100mm on FF.

So, is my thinking/calculations correct in this case?  Or am I overlooking something?  Could the 100mm macro (L or not) be as good or better than the 60mm macro when obtaining a similar depth of field?  Is the smaller APS-C sensor the key here?  And finally, would I be even BETTER served (for macro) by going down to m4/3?

Thanks for any insight!
Jonathan

[witeken]

There's no such thing as a free lunch.

Crop sensors usually have smaller pixels. That means you suffer earlier from diffraction than FF sensors. That's the reason why compacts only go to f8. The only thing that you loose when you go to FF is the narrower FoV. With a 1:1 macro lens on (Canon) aps-c, you get a 1.6:1 FF equiv. magnification. 

[DJL329]

An EF-S lens has the equivalent focal length on an APS-C body as an EF lens of that focal length has on a FF body.  Therefore, a theoretical EF-S 50mm on an APS-C body would have the same FOV as an EF 50mm on FF body.  This means that your EF-S 60mm Macro is not like a 96mm lens on FF.

If you are shooting 1:1 macro, the longer the focal length, the further you will be from the subject.

The 100mm f/2.8L IS is a great lens, however the non-L is also quite good, especially since it is about half the price.  So, to answer your question: yes, you are over-thinking it.

[AdamJ]

With a 1:1 macro lens on (Canon) aps-c, you get a 1.6:1 FF equiv. magnification.

We have to be careful with semantics when describing macro with a crop sensor. With a 1:1 lens, the image on the sensor is 1:1 regardless of sensor size. Pixel density is the upside of APS-C.

Concerning the OP's calculation, it assumes the same subject distance. In reality, at 1:1 the 100mm will have a greater subject distance than the 60mm.

[neuroanatomist]

You're overlooking the fact that the smaller pixels of the 60D make it more sensitive to the effects of diffraction.  That starts at f/6.9 on the 60D, but at f/10.1 on the 5DIII. 

Personally, I prefer FF for macro.

[CharlieB]

Doesn't directly apply here, but one thing to keep in mind is that the DOF "tables" or formulas all fly out the window for magnifications greater than 1:1, and... begin to open that window they fly out of... someplace about 1:5 or so.  Just slightly, increasing disparity as magnification increases.

And another thing to remember - at macro ranges - weird things also happen depending on the lens design.  That is, the traditional modified Gauss design used for macro lenses is based on lens extension to focus.  Now we have all sorts of new designs, many using internal or rear focusing via moving elements, and those tables fly out the window even more.

And keeping on topic - I use the 7D or 5D2 for macro, depending on the whim at the time, and whichever is closest.  I don't find the 7D focus system much to any advantage, since my shots are usually very close flowers and natural things (bugs, bark, moss, etc) in the field, where even a little wisp of breeze kills the shot.

[Kernuak]

I also prefer full frame, as the image quality is better. The higher pixel density is largely counteracted by the greater effects of diffraction, due to the DLA differences, so sharpness is much better on the full frame. As I often shoot handheld (insects tend to move before I can set the tripod, except for certain times of the day), I need to push the ISO higher to get adequate shutterspeeds, with the 7D, ISO 1600 is pushing it for the high detail shots (although fine for many "normal" wildlife shots), but with my 5D MkIII I can go to ISO 6400 and still get more detail.

[witeken]

With a 1:1 macro lens on (Canon) aps-c, you get a 1.6:1 FF equiv. magnification.

We have to be careful with semantics when describing macro with a crop sensor. With a 1:1 lens, the image on the sensor is 1:1 regardless of sensor size. Pixel density is the upside of APS-C.

Concerning the OP's calculation, it assumes the same subject distance. In reality, at 1:1 the 100mm will have a greater subject distance than the 60mm.

Yes, it is actually 1:1. But in FF equivalent terms, Canon aps-c gives you a 1.6:1 magnification. Just like you also talk in equiv. focal length, DoF etc. This is just the same. If you want the same picture on FF, you'll need a 1.6:1 macro lens .
 
http://en.wikipedia.org/wiki/Macro_photography

[neuroanatomist]

With a 1:1 macro lens on (Canon) aps-c, you get a 1.6:1 FF equiv. magnification.

We have to be careful with semantics when describing macro with a crop sensor. With a 1:1 lens, the image on the sensor is 1:1 regardless of sensor size. Pixel density is the upside of APS-C.

Concerning the OP's calculation, it assumes the same subject distance. In reality, at 1:1 the 100mm will have a greater subject distance than the 60mm.

Yes, it is actually 1:1. But in FF equivalent terms, Canon aps-c gives you a 1.6:1 magnification. Just like you also talk in equiv. focal length, DoF etc. This is just the same. If you want the same picture on FF, you'll need a 1.6:1 macro lens .
 
http://en.wikipedia.org/wiki/Macro_photography

Ummm...no, it's pixel density as stated, not sensor size.  Compare the 5DII to the 20D, FF and APS-C with the same pixel density - does the 20D have a 'magnification' advantage?  No.

[SpecialGregg]

An EF-S lens has the equivalent focal length on an APS-C body as an EF lens of that focal length has on a FF body.  Therefore, a theoretical EF-S 50mm on an APS-C body would have the same FOV as an EF 50mm on FF body.  This means that your EF-S 60mm Macro is not like a 96mm lens on FF.

If you are shooting 1:1 macro, the longer the focal length, the further you will be from the subject.

The 100mm f/2.8L IS is a great lens, however the non-L is also quite good, especially since it is about half the price.  So, to answer your question: yes, you are over-thinking it.

This is NOT true. I don't know why some people still confuse this. That would make the 17-55mm an ultra-wide lens. It's not, it's for all intents and purposes the equivalant of the 24-70mm L. The 10-22mm is NOT wider than the 16-35mm L, it's for all intents and purposes it's equivalant. You DO have to take the crop-factor into consideration for EF-S lenses.

[rj79in]

You're overlooking the fact that the smaller pixels of the 60D make it more sensitive to the effects of diffraction.  That starts at f/6.9 on the 60D, but at f/10.1 on the 5DIII. 

Personally, I prefer FF for macro.

Just curious Neuro whether for the same framing, will the APS-C at f/6.9 give a much deeper field than the FF at f/6.9? I understand that the DOF will be deeper with the APS-C but then as compared to the FF, at what point (for the same framing) will diffraction become an issue?

[neuroanatomist]

An EF-S lens has the equivalent focal length on an APS-C body as an EF lens of that focal length has on a FF body.  Therefore, a theoretical EF-S 50mm on an APS-C body would have the same FOV as an EF 50mm on FF body.  This means that your EF-S 60mm Macro is not like a 96mm lens on FF.

...you are over-thinking it.

He may be overthinking, but you are flat out wrong.  Focal length is an intrinsic property of the lens, regardless of the sensor.  EF-S lenses, EF lenses, medium format lenses, focal length is focal length, period.  A 60mm lens on APS-C gives the FoV of a 96mm lens on FF.

[marinien]

You're overlooking the fact that the smaller pixels of the 60D make it more sensitive to the effects of diffraction.  That starts at f/6.9 on the 60D, but at f/10.1 on the 5DIII. 

Personally, I prefer FF for macro.

Just curious Neuro whether for the same framing, will the APS-C at f/6.9 give a much deeper field than the FF at f/6.9? I understand that the DOF will be deeper with the APS-C but then as compared to the FF, at what point (for the same framing) will diffraction become an issue?

For the same framing, APS-C at f/6.9 is like FF at f/11. Weather it's much deeper or not depends on other variables, one of which is the definition of "much".

[neuroanatomist]

Just curious Neuro whether for the same framing, will the APS-C at f/6.9 give a much deeper field than the FF at f/6.9? I understand that the DOF will be deeper with the APS-C but then as compared to the FF, at what point (for the same framing) will diffraction become an issue?

For the same framing, APS-C will give the DoF of 1.6x the aperture (= 1.3-stops) on FF, meaning there's no big difference in diffraction for the same DoF, with current sensors.  So, f/7.1 on APS-C gives the same DoF as f/11 on FF (for the same framing with the same focal length, because you're further from the subject with APS-C).  My 18 MP 7D starts to be affected by diffraction at f/6.9 (close enough to f/7.1), and my 18 MP 1D X at f/11.  The slightly higher pixel density of the 5DIII means f/10 or so, but the tradeoff in better high ISO noise (useful for a faster shutter because many macro subjects move, either biologically or from wind) more than makes up for that, IMO B

[jebrady03]

Thank you everyone for your input!

From what I can gather by assimilating all of the responses, I was mostly right, but failed to take into account the pixel size factor which, with smaller pixels (which when talking about the cameras I mentioned, applies to my 60D), diffraction sets in earlier - thereby compromising the APS-C shots earlier than the FF shots.  Due to the image quality of FF, the 100mm L, and the ISO advantage, apparently FF is simply better, correct?

I would be interested to see a comparison (perhaps one shot w/ zero cropping along with a 100% crop of the same image) of a current APS-C with the 60mm macro to a FF (perhaps the 5D2 since they're of a similar sensor generation with the 100L to compare image quality - if anyone is willing/able to oblige.

Thanks again for all of the responses!
Jonathan