FF, APS-H, APS-C, etc advantages

Started Jun 11, 2012 | Discussions thread
tony field
Veteran MemberPosts: 5,730
Like?
Re: FF, APS-H, APS-C, etc advantages
In reply to Peter 13, Jun 12, 2012

Peter 13 wrote:

For example, I could choose a very sharp lens of a specific focal length for an APS-C shot and, if my similar angle of view lens on FF is less sharp, by your argument, the viewer might anticipate that the sharper image was FF.

Most of the time, it will be. Here is an example. An OK but not so great zoom vs. a supersharp macro lens (sharper than the 100L), equivalent shots:

http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=116&Camera=453&Sample=0&FLI=5&API=2&LensComp=335&CameraComp=474&SampleComp=0&FLIComp=0&APIComp=2

Interesting test - and needs closer examination. When I compared the 1D-IV with the 135mm F2 and the 5D-II with the 24-105 (both at F4), the centre sharpness of the 1D-IV/135 was visibly sharper than the 5D-II/24-105 when printed 8x10 for the same print view. However that is APS-H compared to FF. Seems to me that I will revisit this. Just downloaded Norman Koren test chart and ImageJ.

Sensor size is a fundamental factor for resolution. Why do you own a 4x5 system?

Since I had a 4x5 enlarger, I had to get a companion camera And... a Leica M with Tri-X did not quite cut the mustard for scenic work.

They cannot be (all of them). Equivalence is a fact, like it or not. Certain shots with FF are just impossible on crop.

Interesting observation -- I have never run into this situation - however I probably don't understand what your really mean by "impossible on crop". Certainly the sensor size may make certain image attributes better/worse for a specific image. But does that really mean "impossible"?

Yes. It is impossible to get the DOF with most (but not all) fast primes. What can get you a 35/1.4 DOF on crop? Or 50/1.2-1.4?

If you look at an arbitrary image, in what way is "equivalence" important? If your mind works in that fashion, that is a reasonable way to approach a photo session with different camera formats - and allow you to deliver very specific results.

In DOF, given the same FOV, etc. Also, the same amount of light which means about the same noise, tonality, etc. But resolutions are not equivalent.

This includes DOF and noise. Even when DOF is possible to match, close to wide open, the IQ is much higher. For example, a shot taken with the 35L at f/2.2 on FF can be almost matched by the 24LII on crop wide open, but the latter would be much softer (I own/used both).

If you are looking for true "equivalence" among the sensor sizes all of which you describe are definitely important. However, for what I shoot, that is the least of the things that interest me. I use 6 of the high speed primes and am only interested in the ability to record useful exposures at high ISO. The down-side is the lack of DOF for my theatre and dance work. I prefer the slight extra DOF from the 1D-IV compared to that of the 5D-II.

I agree that, for any given lens, the absolute IQ (i.e. sharpness) will be better on a larger format with more pixels.

No, the pixels are not such an important factor. Very often, but not always, sharpness will be better on FF even with less pixels . See this for a mathematical explanation (the formula there appears in some books and papers in optics, it is not my invention):

http://forums.dpreview.com/forums/readflat.asp?forum=1029&message=39087561

Most interesting presentation. I am going to digest that over the next few days.

Noise is a sensor characteristic which does not specifically allow you to identify sensor size.

Shot noise is a function of the total light. Since most fast lenses are about f/1.4 fast and no more, independent of the format, FF collects about 1 1/3 more light with those lenses wide open. This is a property of the format (and the available lenses). Differences in sensor characteristics like QE are important but not so critical for sensors of the same generation.

If you knew the characteristics of the lens in use, the degree of crop, and ISO, I suspect that it might be reasonable to make an informed guess as to the format of the sensor - but that would be still a difficult decision depending on image content.

What is more important are the images that you did not take. Also, how the image reacts to pp, etc.

Working with wide aperture lenses specifically for the DOF control certainly gives an advantage to the FF camera when you want to minimize DOF. That is, however, a specific artistic interpretation that has a specific technical requirement and maybe sensor size is important.

It is a myth that wide open fast primes always give you thin DOF. GB has great examples of shots at f/1.2 and f/1.4 that have from deep enough to huge DOF. He even posted landscapes at those apertures.

I have not seen these shots but obviously must be the case since you mention it.

However, IMHO, I cannot tell the format of the camera when I look at pictures on my wall.

Maybe because you are not using the format the way other people do. I can tell the format when I an behind the camera - what is possible with good IQ, what is not, and that is what matters.

I certainly cannot tell the difference between essentially full frame APS-H and FF 16x20 images on my wall. In many cases, I see a no difference between 6x7 and FF. In a small few cases, I can see no difference between 4x5 and FF. Of course, this is predicated upon the nature of the subject and the final post processing.

-- hide signature --
Reply   Reply with quote   Complain
Post (hide subjects)Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark post MMy threads
Color scheme? Blue / Yellow