[cali92rs]

Really? let's see you prove it...

I can easily tell the difference...one of the cameras placed two females in the frame, and one placed one man in the frame...
I want the format that will cause the women to pop up

[normhead]

Let me expand my above explanation a bit further , when expanding to a 30x 20 print a K-5 image needs 2x enlargement to get to 9000 pixels , a d800 images needs expanding 1.2 times to get to 9000 pixels the difference is .8 of a pixel or 0026. of an inch. The resolving power of the human eye has been estimated at .00178 from a distance of 6 inches. What are the odds that a human being can differentiate from 6 feet? They'll barely be able to do it at 6 inches, if they have excellent eyesight, and that's on a 30x20 print. Every step down from there and APS-c and FF become even more indistinguishable, for the same picture printed at the same size.

[rawr]

In both theory and practice FF has much better resolution than APS-C

That simply isn't true (viz 'if you wish to maintain the same depth of field, larger sensor sizes do not necessarily have a resolution advantage') under all circumstances.

You also continue to mix up lens resolution and sensor performance.

[Digitalis]

You also continue to mix up lens resolution and sensor performance.

this is exactly my point, lenses don't have fixed resolution - only sensors do.

I actually think it's the bottom one, assuming either one is FF

Actually they are both full frame, I was doing some sensor cleaning in the field on my Leica M9 with the 50mm f/1.4 Summilux pre-ASPH - one of the problems of FF sensors in the increased area for dust bunnies to adhere to, the 645D is even worse..

---

Last edited by Digitalis; 10-18-2012 at 06:01 PM.

[normhead]

Ah ha, I knew it you sneak..

[ElJamoquio]

That simply isn't true (viz 'if you wish to maintain the same depth of field, larger sensor sizes do not necessarily have a resolution advantage') under all circumstances.

You also continue to mix up lens resolution and sensor performance.

Do I.

You referenced an article that implies the edge performance of FF is worse than that of APS-C.

I'd like to see an actual lens comparison where that is true, given the same picture for both formats. I haven't so far.

You've said larger sensors do not necessarily have a resolution advantage under all circumstances. For which circumstances does APS-C have better resolution than FF?

[ElJamoquio]

this is exactly my point, lenses don't have fixed resolution - only sensors do.

Lenses have a resolution that can be measured. I've been quite specific in my vocabulary of 'lens resolution' and 'sensor megapixels'.

[Digitalis]

You've said larger sensors do not necessarily have a resolution advantage under all circumstances. For which circumstances does APS-C have better resolution than FF?

The differences between APS-C and FF are smaller than you think, to put it into perspective a DX format sensor has 28.28mm* diagonal vs FF having a 42.27mm Diagonal I see a bigger difference between FF and 645 which has a 70mm diagonal. Even with both images printed at 8X10 I can spot a difference between 35mm and 645 - but at that print size I can't tell a difference between 645 and 67.

Lenses have a resolution that can be measured.

Of course they do, but these days we rely on digital sensors to measure it and the sensors have finite resolution - not to mention many cameras that have AA filters which has an impact on lens performance as well. But I have yet to see any digital sensor that clearly out resolves a well designed lens when used below the diffraction limit. Speaking of which the diffraction limit is more of a problem on FF sensors - because you have to stop down more. An extreme example of this would be when I need to use my 8X10 view camera - to get more or less the whole landscape in focus I have to stop down to f/64** without using tilt or shift - though with tilt and shift I can often get away with using f/22 which by coincidence happens to be the optimal aperture for a majority of large format lenses.


IMHO if there is a difference between FF and APS-C It is so trivial I really couldn't bring myself to care about it. You can count Photons all you want ElJamoquio, I'll be out taking pictures.


*There seems to be some disagreement on the actual dimensions of a "standard" DX sensor, some sources say 15.8mm X 22.3mm - but here i'm using Kodak's DX standard as 15.69mm X 23.53mm

**some of the older long lenses I have for 8X10 have f/ stops as high as f/192 - or as people who value truth above all else: hello and goodbye resolution.

[gazonk]

The differences between APS-C and FF are smaller than you think, to put it into perspective a DX format sensor has 28.28mm diagonal vs FF having a 42.27mm Diagonal I see a bigger difference between FF and 645 which has a 75mm diagonal.

645 film has a 70 mm diagonal, not 75 (the actual format is 56x42, not 60x45). And 645D's diagonal is only 55 mm.

I wonder what the sensor size of the next 645D will be!

[Digitalis]

645 film has a 70 mm diagonal, not 75 (the actual format is 56x42, not 60x45). And 645D's diagonal is only 55 mm.

Thanks for the clarification ( I corrected it) I have had a lot on my mind lately.

[ogl]

645 film has a 70 mm diagonal, not 75 (the actual format is 56x42, not 60x45). And 645D's diagonal is only 55 mm.

I wonder what the sensor size of the next 645D will be!

56*41,5 mm

[normhead]

You referenced an article that implies the edge performance of FF is worse than that of APS-C.

I'd like to see an actual lens comparison where that is true, given the same picture for both formats. I haven't so far.

And I'd like to see a picture where you can tell the difference at a normal viewing distance, between an APS-c image and an FF image, which I also haven't seen. If you really wanted to see these things you could have done it, how hard can it be.

But from the photozone testing charts. For the sigma 85 1.4 on an FF camera..



From an APS-c camera, same lens



If what you're saying is you can't see the difference from a normal viewing difference then MTF mean almost nothing and you should just quit pretending like you think there is any validity to optical theory.

[ElJamoquio]

The differences between APS-C and FF are smaller than you think, to put it into perspective a DX format sensor has 28.28mm* diagonal vs FF having a 42.27mm Diagonal

? You don't think I can compare diagonals myself and decide whether 50% better linear resolution (all else the same) is worth it or not for me?

Of course they do, but these days we rely on digital sensors to measure it and the sensors have finite resolution - not to mention many cameras that have AA filters which has an impact on lens performance as well. But I have yet to see any digital sensor that clearly out resolves a well designed lens when used below the diffraction limit.

This is a bit of a different subject. If you wanted to see if you could approach a lens' limit for internal curiosity's sake, you should try testing a lens on the Q. If you'd like to try border performance make an offset adapter.

Speaking of which the diffraction limit is more of a problem on FF sensors - because you have to stop down more. An extreme example of this would be when I need to use my 8X10 view camera - to get more or less the whole landscape in focus I have to stop down to f/64** without using tilt or shift - though with tilt and shift I can often get away with using f/22 which by coincidence happens to be the optimal aperture for a majority of large format lenses.

Since the 'picture' that you're sensing is larger on larger format cameras, a constant airy disk is less important; making the penalty at a higher stop basically go away.

IMHO if there is a difference between FF and APS-C It is so trivial I really couldn't bring myself to care about it. You can count Photons all you want ElJamoquio, I'll be out taking pictures.

Great. I prefer to know the limits of my tools. In the meantime it hasn't been enough of a difference in actual pictures for me to switch tools, but I also do a lot of panoramas.

**some of the older long lenses I have for 8X10 have f/ stops as high as f/192 - or as people who value truth above all else: hello and goodbye resolution.

You should do the calcs for f/192 (or something more reasonable, like f/64 or so) on 8x10... it might not be all that bad.

[normhead]

It's become an argument about numbers, not about actual pictures, a fellow can only take so much of that. Honestly, I don't know what you can see and can't see. But, at some point I'm going to find out, not looking at numbers, out there in the real world, because really, that's all that counts. Until someone has done some real world comparisons, they got nothing. The question isn't is there a difference, the question is , is that difference significant in any meaningful way.

[cali92rs]

And I'd like to see a picture where you can tell the difference at a normal viewing distance, between an APS-c image and an FF image, which I also haven't seen. If you really wanted to see these things you could have done it, how hard can it be.

But from the photozone testing charts. For the sigma 85 1.4 on an FF camera..



From an APS-c camera, same lens



If what you're saying is you can't see the difference from a normal viewing difference then MTF mean almost nothing and you should just quit pretending like you think there is any validity to optical theory.

Looks like 2 different lenses to me, top chart being 50mm, bottom 85mm...