Crop sensors better than full frame for macro photography?

[Hillsilly]

....At this point I feel the duty to summarize my learnings from your posts...

That's a good summary. Overall, it is difficult to argue that crop sensors will produce a better image than a full frame sensor. But given that we're on page 3 with vocal proponents on each side and the "Full Frame vs Crop Sensor" topic is up to page 13, you could take that to mean that a crop sensor will produce excellent results and that you will probably only notice a difference if comparing two images side by side. In the real world, where an image stands on its own, a crop sensor produces images that are more than acceptable to everyone.

 

[Marsu42]

I suspect what Marsu42 is referring to is the decrease apparent aperture, the light loss you get as magnification increases.

Indeed, but it's not "apparent" but for real, unlike Nikon Canon just chose to hide the fact that a f2.8 lens @1:1 doesn't get f2.8 light through anymore... that's why the shutter speed gets longer like by black magic. I just read up about it again, it's called "effective f-stop": http://www.cambridgeincolour.com/tutorials/macro-lenses.htm

There's no 'magic number' for that, there's no 'macro mechanism'

Imho there is, it's the "extension" that only a real macro lens has.

At 0.2x mag you lose ~0.67-stops of light, at 0.5x mag you lose ~1.33-stops, and at 1x you lose ~2-stops.

Ah, thanks, that were exactly the numbers I was looking for. Now I'd only need to calculate how much effective f-stop difference there is for crop vs. ff for some scenes, but what you wrote indicates ff still has a slight advantage, though to a much smaller degree than at non-macro distances. But *plus* you still need a smaller aperture to get a usable dof, so that means even higher iso on ff for the same framing.

That, the dr loss at the higher iso for ff, the working distance & swivel screen still make me use the 60d

 

[CarlTN]

Insult by implication is still insult.

What the hell? Um, Dr. John…you of all people should know you’re guilty of the above, especially in my case (but in the case of most people you get on your high horse with on here), if not outright insults…many many times over on this forum. So let’s call a spade a spade. People in glass houses…Mr. "I'm so smart I will make a cartoon character out of you in pink tights, but no that is not an insult, because I said so, and I own this website".

First off, your viewfinder is showing you the DoF of ~f/2.5-2.8 with the stock viewfinder. You only see the true DoF of faster lenses if you have installed an appropriate focusing screen (such as the Eg-S).

But while I know the technical reasons for what you should be seeing, I really have no idea what you're actually seeing… If I had to guess, you're comparing the same framing - being further away with APS-C means deeper DoF.

I know the reason too, and it's not all that technical. The reason I'd need a focusing screen, is because my pupil is much smaller than the sensor. Sure there is more to it than that (the path reflected through the prism and magnified by the eyepiece, etc)...but that's the gist of it.

In any case, my original point is something you have done your best to ignore. I was not arguing so much that equal framing with the two sensor sizes would produce superior results with a crop sensor. I was arguing that you get more effective magnification due to the smaller pixels of the crop sensor...WITH THE SAME LENS and same distance to the subject. I.E., regardless of whether the framing is "ideal" with either sensor size...just given the same lens...one that perhaps cannot focus on a subject that is touching the front lens element...but rather needs inches or a foot or three distance from the element.

 

[faustino]

==> The advantage of crop camera would be only in pixel count, which is a questionable advantage.

Questionable how? Not from the standpoint of putting pixels on the subject. 50MP vs. 22 is quite a difference.

I agree, it is not that much questionable. I changed my mind about this specific point, and changed my previous post accordingly.

There is only one consideration about pixel count that I believe should be taken in account:
at higher ISO, considering current sensor technology, lower pixel count sensors produce better images than higher pixel count sensors.

 

[wsheldon]

==> The advantage of crop camera would be only in pixel count, which is a questionable advantage.

Questionable how? Not from the standpoint of putting pixels on the subject. 50MP vs. 22 is quite a difference.

I agree, it is not that much questionable. I changed my mind about this specific point, and changed my previous post accordingly.

There is only one consideration about pixel count that I believe should be taken in account:
at higher ISO, considering current sensor technology, lower pixel count sensors produce better images than higher pixel count sensors.

True, but we're talking about macro here where tripods and low ISO are the order of the day. At ISO 100 I'm very happy with the IQ of the 15MPx covering the subject from my 50D at 100mm, and the extra working distance compared to my 6D is extremely helpful when shooting critters and flowers.

 

[faustino]

==> The advantage of crop camera would be only in pixel count, which is a questionable advantage.

Questionable how? Not from the standpoint of putting pixels on the subject. 50MP vs. 22 is quite a difference.

I agree, it is not that much questionable. I changed my mind about this specific point, and changed my previous post accordingly.

There is only one consideration about pixel count that I believe should be taken in account:
at higher ISO, considering current sensor technology, lower pixel count sensors produce better images than higher pixel count sensors.

True, but we're talking about macro here where tripods and low ISO are the order of the day. At ISO 100 I'm very happy with the IQ of the 15MPx covering the subject from my 50D at 100mm, and the extra working distance compared to my 6D is extremely helpful when shooting critters and flowers.

I agree about low ISO being the order of the day, also because, most probably, flashes are engaged beside good tripods.

My understanding is that the advantage of todays camera crop sensors is only pixel density (beside price, weight, and swivel screens). Right?

 

[Loswr]

==> The advantage of crop camera would be only in pixel count, which is a questionable advantage.

Questionable how? Not from the standpoint of putting pixels on the subject. 50MP vs. 22 is quite a difference.

Questionable based on whether or not you need those MP for your desired output. An 8 MP image (current FF cropped to the APS-C FoV) is fine for prints of up to 16x24"/A2 size, and overkill for web uploads.

Of course, if you get your jollies by viewing images at 100% on your display, then more MP is a big advantage. But if that's the case, you'd better have the largest, highest-resolution display money can buy, or else you're not getting the most benefit from those extra pixels. :

As PBD pointed out above, most people don't need those extra MP. Not that having them is a bad thing, but it's really about the output.

First off, your viewfinder is showing you the DoF of ~f/2.5-2.8 with the stock viewfinder. You only see the true DoF of faster lenses if you have installed an appropriate focusing screen (such as the Eg-S).

But while I know the technical reasons for what you should be seeing, I really have no idea what you're actually seeing… If I had to guess, you're comparing the same framing - being further away with APS-C means deeper DoF.

I know the reason too, and it's not all that technical. The reason I'd need a focusing screen, is because my pupil is much smaller than the sensor. Sure there is more to it than that (the path reflected through the prism and magnified by the eyepiece, etc)...but that's the gist of it.

The reason you don't see the true DoF of fast lenses is that the stock focus screen is laser microetched to give a brighter viewfinder with slow lenses (presumably because most camera kits come with a zoom that goes to f/5.6, and even the higher end FF kits come with an f/4 lens). The consequence of that laser microetching is that the true DoF of lenses faster than f/4 is not seen, although the effect is really only noticable as you get faster than f/2.8. Another case of 'you can't have your cake and eat it, too.'

In any case, my original point is something you have done your best to ignore. I was not arguing so much that equal framing with the two sensor sizes would produce superior results with a crop sensor. I was arguing that you get more effective magnification due to the smaller pixels of the crop sensor...WITH THE SAME LENS and same distance to the subject. I.E., regardless of whether the framing is "ideal" with either sensor size...just given the same lens...one that perhaps cannot focus on a subject that is touching the front lens element...but rather needs inches or a foot or three distance from the element.

Your original point is something that you're conveniently forgetting. So, I'll remind you: your point was that a crop sensor is better because you get deeper DoF. In your first post in this thread you stated, "...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 fact, the person who first mentioned the advantage of APS-C putting more pixels on target was…me. Initially in a post quoting myself from another thread, then directly in response to you. But now you're rewriting history in your own mind, and claiming credit for the idea? That doesn't work so well when everyone can just read the earlier posts in this thread to see that you're wrong.

The fact remains that you were completely wrong in your statement that a crop sensor has an advantage in terms of deeper DoF under the conditions you describe (same lens, same distance). As least now I understand why the SuperCarl graphic upsets you so much - you're 'manly' enough that the idea of wearing pink tights is anathema to you…you're just not man enough to admit that you're wrong, at least in this case.

 

[Loswr]

I suspect what Marsu42 is referring to is the decrease apparent aperture, the light loss you get as magnification increases.

Indeed, but it's not "apparent" but for real, unlike Nikon Canon just chose to hide the fact that a f2.8 lens @1:1 doesn't get f2.8 light through anymore... that's why the shutter speed gets longer like by black magic. I just read up about it again, it's called "effective f-stop": http://www.cambridgeincolour.com/tutorials/macro-lenses.htm

What's your point? 'Effective f-stop' and 'apparent aperture' mean the same thing. The light loss is real, but the physical aperture is not changing size. You can't say f/2.8 becomes f/5.6 at 1:1, any more than you can say that an EF 100mm lens becomes a 160mm lens when you mount it on a crop body.

If you're using a Nikon camera in Av mode and go to 1:1 magnification, the shutter speed doesn't change and you get shots underexposed by 2 stops? I didn't know that, but if true I don't think that makes sense. An autoexposure mode should give me a metered exposure, period. Therefore, I'd expect it to compensate for the light lost at 1:1 by altering the shutter speed (and/or ISO, if that was set to Auto).

Also, I'm not sure why you state, "Canon just chose to hide the fact that a f2.8 lens @1:1 doesn't get f2.8 light through anymore." That's a pretty bold and unsupported accusation. Unless you think they're trying to hide that imformation in plain sight…such as by printing it right in the instruction manual for the lens. :

 

[mackguyver]

I suspect what Marsu42 is referring to is the decrease apparent aperture, the light loss you get as magnification increases.

Indeed, but it's not "apparent" but for real, unlike Nikon Canon just chose to hide the fact that a f2.8 lens @1:1 doesn't get f2.8 light through anymore... that's why the shutter speed gets longer like by black magic. I just read up about it again, it's called "effective f-stop": http://www.cambridgeincolour.com/tutorials/macro-lenses.htm

What's your point…'effective f-stop' and 'apparent aperture' mean the same thing. The light loss is real, but the physical aperture is not changing size. You can't say f/2.8 becomes f/5.6 at 1:1, any more than you can say that an EF 100mm lens becomes a 160mm lens when you mount it on a crop body.

If you're using a Nikon camera in Av mode and go to 1:1 magnification, the shutter speed doesn't change and you get shots underexposed by 2 stops? I didn't know that, but if true I don't think that makes sense. An autoexposure mode should give me a metered exposure, period. Therefore, I'd expect it to compensate for the light lost at 1:1 by altering the shutter speed (and/or ISO, if that was set to Auto).

Also, I'm not sure why you state, "Canon just chose to hide the fact that a f2.8 lens @1:1 doesn't get f2.8 light through anymore." That's a pretty bold and unsupported accusation. Unless you think they're trying to hide that imformation in plain sight…such as by printing it right in the instruction manual for the lens. :

Neuro, I remember reading that there is a difference in the way Nikon displays the light loss vs. Canon (in one of Scott Kelby's books, I think), but at the end of the day, exposure is exposure, and the light loss (and the need for high ISOs when shooting in natural light) is the main reason I think that FF is better than crop for macro, at least for my needs.

 

[Marsu42]

0
If you're using a Nikon camera in Av mode and go to 1:1 magnification, the shutter speed doesn't change you get shots underexposed by 2 stops?

Nope, afaik they simply show the effective f-stop, and then you know why the shutter speed suddenly gets longer.

0
Also, I'm not sure why you state, "Canon just chose to hide the fact that a f2.8 lens @1:1 doesn't get f2.8 light through anymore." That's a pretty bold and unsupported accusation. Unless you think they're trying to hide that imformation in plain sight…such as by printing it right in the instruction manual for the lens.

I'm not trying to pick a fight, if you need some mudslinging go talk to some well-known contributor who responds to that. But I'm as a simple Canon beginner didn't understand (and even realize) the problem at macro distances, and only wondered why I ran into such low shutter or high iso values. Yes, rtfm, but still.

Simply giving the effective f-stop rather than the nominal value makes more sense to me, and everything else is "hiding" not in a moral sense, but as a loss of potential information.

 

[Loswr]

0
If you're using a Nikon camera in Av mode and go to 1:1 magnification, the shutter speed doesn't change you get shots underexposed by 2 stops?

Nope, afaik they simply show the effective f-stop, and then you know why the shutter speed suddenly gets longer.

So, you'd prefer that the camera report an aperture that is not the one you selected, whereas I'd prefer that the camera show the selected aperture and adjust the exposure to compensate.

It's not a moral issue, which calling it 'hiding' implies. Rather, Canon and Nikon have simply chosen different ways to represent the loss of light at high magnification.

As a side note, that must mean a Nikon camera will display an aperture value that is smaller than the lens is physically capable of achieving. So if we're going to make it a moral issue, Canon is hiding something, and Nikon is lying about something else.

There's no 'magic number' for that, there's no 'macro mechanism'

Imho there is, it's the "extension" that only a real macro lens has.

The EF-S 18-55mm f/3.5-5.6 IS STM has a maximum magnification of 0.36x, and at that magnification it is losing ~1 stop of light based on the narrower apparent aperture (or smaller effective f-stop, if you prefer that terminology). I wouldn't call the Rebel kit lens a 'real macro lens'...

That, the dr loss at the higher iso for ff, the working distance & swivel screen still make me use the 60d

You seem to be ignoring the fact that the APS-C sensor has less DR than the FF sensor, at all ISO settings.

ISO 800 on the 6D has the same DR as ISO 100 on the 60D - a 3-stop advantage for the FF sensor. At ISO 400 and above on the 60D, the 6D has approximately the same DR at a 2-stop higher ISO setting. Where is the DR advantage of the APS-C sensor?

Swivel screen? Canon's free EOS Remote app allows remote live view from the 6D, and may be more useful than a swivel screen. Granted, the 6D is the only Canon FF camera with Wi-Fi (so far)…but there's also CamRanger for the others.

As for the working distance advantage, that's the same argument for why APS-C is better when you're focal length limited in telephoto situations. It's only an advantage if you require more than ~8 MP for your output (i.e. you're printing larger than 16x24"/A2) and you're shooting at relatively low ISO. For normal and small prints, web use, or if you have to boost the ISO, the cropped FF image will be as good or better.

 

[Loswr]

Overall, it is difficult to argue that crop sensors will produce a better image than a full frame sensor.

From the standpoint of IQ, yes. However…I'd argue that a crop sensor camera that one can afford will produce much better images than the full frame camera that one cannot afford.

IMO, the only real and significant advantage of the APS-C sensor is that crop bodies are cheaper. In many cases and for many people, that advantage trumps all other considerations.

 

[Marsu42]

ISO 800 on the 6D has the same DR as ISO 100 on the 60D - a 3-stop advantage for the FF sensor.

Ah, you're comparing a 2012 camera to a 2010 camera. Let me counter this: The 7d has more dr at iso 800 than the 5dc, giving crop a solid 2-stop advantage over the FF sensor :-> ... the dr argument is just a side note, it just happens to add up with the fact that the ff advantage is less than usual for macro.

Swivel screen? Canon's free EOS Remote app allows remote live view from the 6D, and may be more useful than a swivel screen.

Yeah, right, *may* ... but to anyone doing serious macro tripod work it isn't, but I understand you missing out on this because your cameras don't have it so you never really tried. Wifi+Phone requires another gadget and draws power, and unless someone reverse engineers the protocol you cannot use focus peaking nor focus stacking with it. Nice tourist selfie feature mind you or for photo journalism, but certainly not for macro.

As for the working distance advantage, that's the same argument for why APS-C is better when you're focal length limited in telephoto situations. It's only an advantage if you require more than ~8 MP for your output (i.e. you're printing larger than 16x24"/A2) and you're shooting at relatively low ISO. For normal and small prints, web use, or if you have to boost the ISO, the cropped FF image will be as good or better.

I don't think so, first because cropping even the crop shot is usual if limited by the 1:1 mag, so you'll end up with even less mp on ff - I share your opinion for tele photography, but not for macro. Plus as discussed the iso advantage of ff is a bit less than usual because of the effective f-stop and shallower dof (same framing).

For normal and small prints, web use, or if you have to boost the ISO, the cropped FF image will be as good or better.

We have to agree to disagree here unless someone comes up with actual side by side comparisons :-o

 

[Loswr]

ISO 800 on the 6D has the same DR as ISO 100 on the 60D - a 3-stop advantage for the FF sensor.

Ah, you're comparing a 2012 camera to a 2010 camera. Let me counter this: The 7d has more dr at iso 800 than the 5dc, giving crop a solid 2-stop advantage over the FF sensor :-> ... the dr argument is just a side note, it just happens to add up with the fact that the ff advantage is less than usual for macro.

Ok, let's compare the 6D from 2012 to the newer 70D from 2013 - the latter is basically the same as the 60D from a DR standpoint. Still no DR advantage for APS-C.

We have to agree to disagree here unless someone comes up with actual side by side comparisons :-o

Fair enough.

 

[AmazonOfExeter]

Hello!
Have been reading the thread with great interest, and trying to keep up with the information!

Just to add a comment in relation to Marsu42 and Neuros' thoughts on swivel screens/remote viewing for macro work (or other work too).

In favour of the swivel screen is the fact it is attached and easily manipulated, not easily lost and rapidly deployed.

In favour of a remote viewing is, if you use DSLR Controller then you have the advantage of a larger screen (dependent on the device - currently using a Sony Z1 phone to test it on, more or less a 5'' screen), in theory a very good range of distances and angles to see it from, plus the ability to completely control the camera from the screen - including focus peaking. On the downside, it is very fiddly unless you have some sort of bracket made up to carry the device (potentially reducing the flexibility of the solution) and if, like me, you tend to be trying to photograph subjects that are constantly moving, this bracket idea is essential unless you have 3 hands...! I've not really explored all that DSLR controller can do, and I haven't installed Magic Lantern (tempted but...) so I cannot compare the quality or abilities of the two systems!

Hope this helps! Thank you for the great information!

 

[CarlTN]

My post was deleted but I sent it to you privately, neuro.

I can see this argument must now end, because well...anything else I say is going to get deleted anyway.

 

[Loswr]

My post was deleted but I sent it to you privately, neuro.

I can see this argument must now end, because well...anything else I say is going to get deleted anyway.

Yes, I can see why it was deleted.

"Insults are the arguments employed by those who are in the wrong." —Jean-Jacques Rousseau

But in the "macro" world, you're taking a picture of something that is tiny. It seems to me, that the smaller the sensor, the more DOF an image of that subject is going to have. Because both sensor and subject are tiny.

Rather sad that you're still clinging to the same flawed understanding of what determines depth of field.

"To make a mistake is only an error in judgment, but to adhere to it when it is discovered shows infirmity of character." —Dale Turner

 

[CarlTN]

My post was deleted but I sent it to you privately, neuro.

I can see this argument must now end, because well...anything else I say is going to get deleted anyway.

Yes, I can see why it was deleted.

"Insults are the arguments employed by those who are in the wrong." —Jean-Jacques Rousseau

But in the "macro" world, you're taking a picture of something that is tiny. It seems to me, that the smaller the sensor, the more DOF an image of that subject is going to have. Because both sensor and subject are tiny.

Rather sad that you're still clinging to the same flawed understanding of what determines depth of field.

"To make a mistake is only an error in judgment, but to adhere to it when it is discovered shows infirmity of character." —Dale Turner

That's because I am sad, a sad little man wearing pink tights, I can't be manly like you...not on here...you're the big man on campus :

 

[pdirestajr]

Hi I've photographed products with a 7D while my 5D2 watched in disgust. I then charged the client money for this craziness. They had no idea. I sure fooled them!

 

[CarlTN]

Hi I've photographed products with a 7D while my 5D2 watched in disgust. I then charged the client money for this craziness. They had no idea. I sure fooled them!

If you have a 5D2 that is self aware, then perhaps it is a better and smarter camera than I have thought it was all this time! My 6D can't quite attain self awareness.