Full Frame vs Crop

I

IanPhillips

Messages
4
Name
Ian
Edit My Images
No
I have had a 7d and recently bought the 7d MKii - I also upgraded my lenses to the 500mm L IS USM Canon and the 70-200 2.8

On both cameras the images were never really spot on sharp even when placing the camera in a super stable platform and testing with hig shutter speeds of 1/1000

I have now tested both lenses on the 5d MKiii and both come out super sharp.


So - in summary, are L series tele lenses not suited to crop sensors??
 
Sounds more like you need to calibrate the camera to the lens.
 
I'm only guessing as have none of the kit you're talking about. If you're viewing at 100% for each body, the 7d images are more magnified than they are on the 5Diii... aren't they?
 
DemiLion said:
Sounds more like you need to calibrate the camera to the lens.
Click to expand...
This. I use the 500L on 7D and 1D MK IV crop bodies and never had a sharpness issue. When I bought the 400 5.6L many years ago I did have to micro adjust the camera (7D) to the lens.
 
GeeJay57 said:
I'm only guessing as have none of the kit you're talking about. If you're viewing at 100% for each body, the 7d images are more magnified than they are on the 5Diii... aren't they?
Click to expand...
Yeh, it's a 1.6 crop.
 
My tests on the images were on the pc to check quality, not the camera. Shot in RAW I used MFA and just can't get either crop camera to get anywhere near as sharp.
 
IanPhillips said:
My tests on the images were on the pc to check quality, not the camera. Shot in RAW I used MFA and just can't get either crop camera to get anywhere near as sharp.
Click to expand...

Never had a issue with either version of the crop sensor (1.6x or 1.3x) on the 20D, 70D or 1DMKII, taken some crackingly sharp images, as mention, you need to calibrate you lens and camera, this sometimes happens.
 
I will give it a go again on the 7d and see how I get on. I use the MFA settings in the camera and now I have proved the lenses are pin sharp hopefully I can sort this out. Oddly using the Sigma 10-20mm 3.5 on the 7d is pin sharp hence why I was unsure.
 
Hi marksweb

In the Canon camera under custom functions is an option to calibrate the lens to camera. It is done in micro adjustments, but you need really good light and some fixed targets, solid platform for the camera. This is how I do it so maybe someone else will offer up an idea.
 
If the lenses are performing optimally on both cameras, the full-frame camera will always show better sharpness than a cropper (and greater dynamic range, and higher ISO performance).

It's basic physics (lens MTF theory). Smaller sensor needs greater enlargement for same size output; therefore higher lens resolution is required to show the same level of detail; fact of optics is that when resolution goes up, image contrast goes down (the two axes of an MTF graph); it's image contrast that contributes most to perceived sharpness.

Or more simply, bigger is better.
 
Yes, but you should still be able to get sharp results from an APS-C camera fitted with a FF lens and the OP does say that his 10-20mm pictures look sharp which leaves the question of why the L's are less sharp.

The last APS-C camera I had was a 20D and I was very happy with the sharpness and I'd expect a more modern camera to be able to produce a sharp image when fitted with a sharp lens but I wonder what magnification the OP is viewing at?

I've never used a high mp count Canon body but I'm pretty sure I've read in reviews that things don't always look lovely at 100% but that wouldn't explain why the 10-20 pictures look sharp and the L's don't.

As previously stated smaller formats need sharper lenses to account for the higher magnification of the final image so if taken to the extreme a good APS-C lens may well produce a sharper image on an APS-C body than a FF lens which produces a perfectly acceptably sharp image on a FF camera.

If the lenses and camera are calibrated for best effect and the 10-20mm is sharper than the L's on the camera then the result is what it is. Maybe the lens test charts will confirm what the OP is seeing or maybe there's another answer. Time to Google lens test charts maybe.

Just looked at APS-C results on a review site and the 10-20mm seems sharper in the centre of the frame than the 70-200mm f2.8 USM L. However the FF lens is sharper away from the centre. The site didn't have the 500mm listed under APS-C tests. I don't know how significant the differences in sharpness in the test results are when the pictures are viewed by eye.
 
Last edited:
Have u tried stopping down to 5.6 or so?
And you will go :O at a foveon raw file if u want sharpness _:p
 
I'm talking to the far east, what's your excuse for being awake at this time?
 
Can't sleep and I'm often awake late...
Early morning in the east isn't it?
 
I noticed a huge difference when I moved my 24-70 2.8 L from my old 7d to my a 5d2... And that was with micro adjustments to the 7d. In fact, same thing for my 50mm 1.4.... Pin sharp wide open now, could never shoot wider that 2.2 before on the 7d with it...
 
Unfortunately, as computing power and monitor size increases, we are becoming hyper-critical of things like sharpness. HoppyUK is right - it's the way optics work.

Plus you have to remember that L lenses are designed with full frame sensors in mind, not crop.
 
pingu666 said:
Can't sleep and I'm often awake late...
Early morning in the east isn't it?
Click to expand...

Sorry to hear that. They're seven hours ahead, it's a regular thing for me as it's a big market and I seem to be in demand.
 
sunnyside_up said:
I noticed a huge difference when I moved my 24-70 2.8 L from my old 7d to my a 5d2... And that was with micro adjustments to the 7d. In fact, same thing for my 50mm 1.4.... Pin sharp wide open now, could never shoot wider that 2.2 before on the 7d with it...
Click to expand...

7D vs 5D2 is a good comparison between full-frame and APS-C - same generation tech, similar pixel count, and same lens of course. Though it's not really about pixels; that plays a part of course but a high pixel counts is way less important than sensor size and lens quality.

Put it like this: if a car will accelerate from 0-60 in six seconds, you wouldn't expect it to go from 60-120 in the same time - the faster you go, the harder it gets. Same with lenses, so if you need a certain standard of sharpness at 30-lines-per-mm resolution on full-frame, on APS-C the lens has got to deliver that same standard at 48-lpmm (x1.6 crop factor) and physics doesn't allow that, so the image looks less crisp/sharp.
 
I think that APS-C reached its optimum at 16mp, cameras using that size sensor seem to get the best results. Admittedly most use the well regarded Sony sensor, no coincidence the Fuji X series also stick at 16mp.

I owned a 7D and think they overdid that at 18mp, perhaps the latest model would also have been better dropping back to 16mp.

One more consideration is that camera shake appears to be amplified as the pixel count rises. Not sure if there is any scientific proof, but I certainly noticed it
 
Last edited:
Manually focus using both crop and full frame. This will take any AF or micro-adjustment tolerances out of the equation, and show you what both bodies can do with the same lens.

If things are sharper when focussed manually then you have an AF issue.
 
tijuana taxi said:
I think that APS-C reached its optimum at 16mp, cameras using that size sensor seem to get the best results. Admittedly most use the well regarded Sony sensor, no coincidence the Fuji X series also stick at 16mp.

I owned a 7D and think they overdid that at 18mp, perhaps the latest model would also have been better dropping back to 16mp.

One more consideration is that camera shake appears to be amplified as the pixel count rises. Not sure if there is any scientific proof, but I certainly noticed it
Click to expand...

Camera shake remains the same but the higher resolution magnifies it so you can see it more - remember the fuss Nikon made about correct technique with the D800 came out.
 
Im using a 500 f4 on both FF and 7D MkII and ive got no complaints about sharpness with either body. However all of my lenses have needed positive micro adjustment on the 7D MkII whereas on the 1Dx two of them are spot on and one is positive. But thats down to just tolerance variation and the adjustment is around +3 to +5 so of no real concern.

To get accurate MFA you have to have a decent set up that you can repeat time and again. tripod, sensible distance to check focus at, good contrast target, target with some depth to it IMO is better than a flat target as this will give you front or back focusing info. Best of all you can do this for free or try focal or similar if you want a more refined, but not necessarily quicker method.
 
You must log in or register to reply here.
Back
Top