Lenses sharper on FF

[J.R.]

Hi CR folks,

I've found that my lenses are sharper on FF as compared to APS-C cameras ... This is particularly true for the 100-400L and the 50 f/1.4. Is it because the FF is more forgiving and the higher pixel density of the APS-C highlights the problems of the lenses more? Just curious ...

Cheers and a happy weekend ... J.R.

 

[Loswr]

For a given output size, the image from the smaller APS-C sensor must be enlarged more and thus appears softer. A higher pixel density gives APS-C higher spatial resolution, but in practice the images from a FF sensor do look sharper.

 

[J.R.]

For a given output size, the image from the smaller APS-C sensor must be enlarged more and thus appears softer. A higher pixel density gives APS-C higher spatial resolution, but in practice the images from a FF sensor do look sharper.

Thanks, but I am have been making the comparison on screen viewing pixels at 100% (not sure if I can call that output). Something seems to be different in sharpness on the APS-C and FF, I'm not able to figure out exactly what.

 

[ajfotofilmagem]

If your comparison uses the same lens on full frame and APS-C (EF 100-400L), the image will be sharper on full frame. But ... If you use EF-S lenses like high quality 55-250mm STM, may have sharper picture on APS-C, compared to 100-400mm on full frame for the same angle of view. EF lenses, when used on full frame camera in wide open aperture, usually sharper in the center, but has the least sharp corners, compared with the same lens on APS-C. If the comparison is made at high ISO, full frame has a greater advantage in sharpness compared with APS-C.

 

[Loswr]

The enlargement still applies when viewing at 100%.

 

[gshocked]

Hi all,

Wouldn't it depend on the lens and the body combo?
And yes I agree at high ISO it's a different ball game...FF would win hands down...?

 

[ajfotofilmagem]

Hi all,
Wouldn't it depend on the lens and the body combo?
And yes I agree at high ISO it's a different ball game...FF would win hands down...?

Yes, sharpness depends on the combination "camera + lens + image processing." However, the same EF lens when used on full frame camera usually has more sharpness in the center, due to the larger diameter of the pixel full frame sensor. On the other hand, the corners of the image sharpness may be lower in full frame, by using the edges of the glass lens, which usually produce less sharp images. An APS-C camera does not use the edges of the glass EF lens, and does not suffer much loss of sharpness in the corners of the image as the camera full frame.

 

[candc]

You have to have a really sharp lens for it to be close to as sharp on a crop body compared to ff. The crop sensor has a higher pixel density and will out resolve a lot of lenses. The link below compares crops of a 200 f/2 on a 6d and a 70d which are both 20mp the 6d might be a bit better in the center but I don't see a significant difference.

http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=458&Camera=819&Sample=0&FLI=0&API=0&LensComp=458&CameraComp=845&SampleComp=0&FLIComp=0&APIComp=0

 

[sdsr]

You have to have a really sharp lens for it to be close to as sharp on a crop body compared to ff. The crop sensor has a higher pixel density and will out resolve a lot of lenses. The link below compares crops of a 200 f/2 on a 6d and a 70d which are both 20mp the 6d might be a bit better in the center but I don't see a significant difference.

http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=458&Camera=819&Sample=0&FLI=0&API=0&LensComp=458&CameraComp=845&SampleComp=0&FLIComp=0&APIComp=0

In real life those two combinations may produce results that look the same, but in this particular chart comparison I can see a difference, which is, perhaps oddly, most apparent in the corners, where the 6D wins all the way down to f8 (at which point I stopped), even though the difference narrows as you go. (Obviously, unlike more than a few FF lenses this lens doesn't have corners that benefit from being cropped off.)

 

[candc]

You might be right, the center is so close its hard to tell the difference. If it's hard to see a difference in 100% crops of test charts then I think its safe to say that you won't see a difference at normal viewing size and distance.

 

[rs]

The enlargement still applies when viewing at 100%.

+1

J.R. - Viewing at 100% simply means a 1:1 mapping of pixels from the sensor to pixels on your monitor. It's a useful tool to see how detailed your image is by making sure none of the pixels are merged together when outputted to screen, but that's about it.

If the number of megapixels or the dpi of the monitor were to change, the size of the details at 1:1 or 100% would vary. Therefore stating it's at 100% is no way to quantify enlargement.

Enlargement is merely the ratio between the size of the image projected by the back of the lens (onto a sensor or film), compared to the output size (monitor, print etc). If you view a low quality picture on a 4" phone display it might appear fine. Enlarge it more to fill a 30" monitor, and any flaws within that image are more visible, or enlarged.

Polaroids are a good example of an output with no enlargement, whereas a smartphone photo printed as a large poster is the complete opposite. More MP don't mean less enlargement is needed for a particular output size. It's just the same as scanning a smaller section of film at higher res. If the detail isn't there in the first place (due to the lens, AF or whatever), more data in more pixels don't help.

 

[Skywise]

You might be right, the center is so close its hard to tell the difference. If it's hard to see a difference in 100% crops of test charts then I think its safe to say that you won't see a difference at normal viewing size and distance.

I find that interesting too. That might be because the crop sensor is shorter so you're not really reaching the "corners" of the lens on a crop as you are with the FF. (with a non EF-S lens).

In general I found my EF-S lenses (10-22 and the 17-55 f2.8) on my T4i were incredibly sharp. (I had one picture I took at Disneyworld with the 10-22 from the 2nd floor of the train station looking over mainstreet and towards the castle and with a crowd shot and, unbeknowest to me at the time, there was a show going on at the castle with the princes and princesses and I lucked into a shot where they were perfectly posed as if I had planned it that way. I didn't even know until a relative went to make a large print of it and the printer pointed out how cool the shot was!) (Yeah... yeah... I planned it that way!) But I also find my 16-35 f2.8II ion my 6D to be just as sharp. But I bet if I had taken that shot with the 6D and the 16-35 I could've zoomed in much better onto the stage and gotten a better cropped image then I was able to with the APS-C sensor. (EG, not at all!)