A. looks more detailed, contrastier (not a word but you get my meaning), and the out of focus background looks smoother.
But in the end, I'd never bother looking at either of those photos long enough to be able to pick the difference. And as far as I can tell, those differences are pretty close to irrelevant to the taking of a good photograph. Because do you know what the best photo I've seen recently was? A shot of a 5-year-old running through a loaded clothes-line on a sunny afternoon that was taken with a Holga. A cheap, plastic, almost-ancient Holga.
And it's realising the fact that that's the best photo I've seen recently, that reminds me that DOF characteristics, absolute sharpness, peak contrast (and so on and so forth) simply don't matter as much as taking an interesting photo in the first place.
Nov 09, 2008 at 08:56 AM
Andre Labonte Offline • • • • • • Upload & Sell: Off
Mark Kenfield wrote:
A. looks more detailed, contrastier (not a word but you get my meaning), and the out of focus background looks smoother.
But in the end, I'd never bother looking at either of those photos long enough to be able to pick the difference. And as far as I can tell, those differences are pretty close to irrelevant to the taking of a good photograph. Because do you know what the best photo I've seen recently was? A shot of a 5-year-old running through a loaded clothes-line on a sunny afternoon that was taken with a Holga. A cheap, plastic, almost-ancient Holga.
And it's realising the fact that that's the best photo I've seen recently, that reminds me that DOF characteristics, absolute sharpness, peak contrast (and so on and so forth) simply don't matter as much as taking an interesting photo in the first place....Show more →
And those differences in such an unscientific test could be the result of a user-biased test.
Sam Bennett wrote:
There is indeed a lot of variables, which is why I stated this test is not ideal, but I think galenapass summed things up pretty well. I'm intentionally not showing full-resolution images or even crops of the images since that would put the DX version obviously at a very serious disadvantage - but this is not a resolution test. At some point I'll do this again with a D300 - I'd like to get one eventually.
An interesting example to do would be to use the 24-70/2.8 at 47mm and f/2.8 on the DX and 70mm f/5 on the FX - I think in that case the difference would be much less pronounced. All lenses that can open past f/2 or so have fundamental issues like birefringence that you just can't "design out", so that area will always be problematic. If you're already using high-quality zooms on your DX and FX gear, I wouldn't worry about this much. The FX will likely be a bit sharper, but not dramatically so....Show more →
In that case, I think the thread title is a little misleading. In any case... I'll butt out and get back to shooting.
Sam Bennett wrote:
Well, scientist, put your money where your mouth is and give us a better designed test and prove your point.
Actually use a DX camera to test the theory? I understand your reasoning, but come on, this test is ridiculous. It's not FX vs. DX at all. It's lens vs. lens (both wide open) on FX with one in low-resolution crop mode.
MHofmann wrote:
Actually use a DX camera to test the theory? I understand your reasoning, but come on, this test is ridiculous. It's not FX vs. DX at all. It's lens vs. lens (both wide open) on FX with one in low-resolution crop mode.
Once I have access to a D300, I will. But judging from what I'm seeing already, I doubt it will make much a difference. It doesn't change the characteristics of the lenses being used, which is where the changes are really seen - the birefringence, the overall contrast, etc. The sharpness/detail is really only one part of the equation.
The thing I'd really like to see is a comparison with a more "modern" prime. Once I get my 50mm f/1.4G I'll redo the test - hopefully I can borrow a friends D300 at the same time. The G will certainly be better, but again... wide-open any fast prime is at a huge disadvantage, regardless of the quality of the lens.
In any case, Andre has a point to prove. So he should give a test of his own design to prove it. So far he's just ranted about theory and expects us to trust him at his word and reputation as a scientist. Visual, practical proof means much more to me than theory since I use my gear (shock!) for taking photos.
OK, using an FX in DX mode does not in the least compare a DX sensor to an FX sensor.
I have both the D700 and D300. My observation is that when you shoot the same subject with the same effective FL and similar/same DOF, the edge will go to the FX simply because the color is almost always stronger/more vibrant that those from the DX sensor.
As for everything else I have seen, they are the same and the obvious optics you talk about are not any different, in fact the detail offered in a DX shot versus the DX "mode" of the FX sensor, is far better. That is, the D300 at an effective 50mm versus a DX mode shot from the D700 at an effective 50mm shows the D300 shot is far superior.
So, not quite sure what your point is in doing this comparison other than to say using the full sensor versus a part of the sensor will give better results. Am I missing something?
DavidWEGS wrote:
Oh, one more thing. What is a test shot at f1.4 on a lens like that? Useless, totally.
That lens is cr*p compared to the 24-70/2.8.
So would you say that if someone has the choice between shooting at f/1.4 on that lens on a DX body and f/2.8 on the 24-70/2.8 on FX, they should shoot with the 24-70?
Nov 10, 2008 at 10:56 AM
Andre Labonte Offline • • • • • • Upload & Sell: Off
Sam Bennett wrote:
Once I have access to a D300, I will. But judging from what I'm seeing already, I doubt it will make much a difference. It doesn't change the characteristics of the lenses being used, which is where the changes are really seen - the birefringence, the overall contrast, etc. The sharpness/detail is really only one part of the equation.
The thing I'd really like to see is a comparison with a more "modern" prime. Once I get my 50mm f/1.4G I'll redo the test - hopefully I can borrow a friends D300 at the same time. The G will certainly be better, but again... wide-open any fast prime is at a huge disadvantage, regardless of the quality of the lens.
In any case, Andre has a point to prove. So he should give a test of his own design to prove it. So far he's just ranted about theory and expects us to trust him at his word and reputation as a scientist. Visual, practical proof means much more to me than theory since I use my gear (shock!) for taking photos....Show more →
I have nothing to prove. The laws of physics are the laws of physics because generations of scientists and engineers have already done the experiments and proven the lows. The burden of proff lies with the person making claims contrary to the laws of physics. Any peer-reviewed journal and/or court of law would agree.
Your impressions as a user and your user-biased test are NOT proof. Do a proper study. The burden is on you!
So in other words, you have no practical examples to back up your claim. That's what I figured. This is just another demonstration of people over-intellectualizing what is actually creative, highly subjective act. So, in lieu of you demonstrating your claim I'm going to have to trust my own experience on this one.
Nov 10, 2008 at 12:14 PM
Andre Labonte Offline • • • • • • Upload & Sell: Off
Sam Bennett wrote:
So in other words, you have no practical examples to back up your claim. That's what I figured. This is just another demonstration of people over-intellectualizing what is actually creative, highly subjective act. So, in lieu of you demonstrating your claim I'm going to have to trust my own experience on this one.
I am talking about erroneous claims made by you and others about FX vs DX resolution etc. This has nothing to do with the subjective art aspects of photography. The burden of proof still lies with you as you are the one making the erroneous claims and still have failed to provide any data to support your claims other than your "say so".
The laws of physics are my proof ... what's yours? Your "say-so"? That's laughable. Have a good time in you self-deluded world.
Sam Bennett wrote:
So would you say that if someone has the choice between shooting at f/1.4 on that lens on a DX body and f/2.8 on the 24-70/2.8 on FX, they should shoot with the 24-70?
No, just that the two do not compare and especially at that aperture. Your test does not make any sense to me, sorry.
Andre Labonte wrote:
I am talking about erroneous claims made by you and others about FX vs DX resolution etc.
I never said anything about resolution. If I was talking about resolution, I would have included full rez crops - but that would have been completely asinine in this case given the very obvious differences in resolution between the two photographs. You initial response to my statement was harping on sharpness when comparing using a lens at f/1.4 and an "equivalent" lens stopped down to f/2.5. This is not exactly the same thing, but it shows what I'm talking about - the f/1.4 lens is exhibiting obvious optical problems that wouldn't be problems if the lens (or a focal length equivalent prime of similar quality) was stopped down to f/2.5. This is pretty plain to see with the same lens, which looks significantly better in every way when stopped down to f/2.8 - the closest I can get. Not just in sharpness, but in every optical measure.
Andre Labonte wrote:
This has nothing to do with the subjective art aspects of photography. The burden of proof still lies with you as you are the one making the erroneous claims and still have failed to provide any data to support your claims other than your "say so".
And I've provided my "proof", with caveats stated plainly for all to see. People who haved viewed this photos and noted the distinct differences, even at this modest resolution are obviously seeing something that you're not.
Andre Labonte wrote:
The laws of physics are my proof ... what's yours? Your "say-so"? That's laughable. Have a good time in you self-deluded world.
My "proof" is real life and my experience as an well-respected photographer. You haven't posted a single photograph to support what you're saying. I have no doubt that some of what you're claiming makes sense on paper in perfect optical situations and so on. But the real world is much different. Lens manufacturers make design compromises. There are the problems that arise when using dSLRs vs. film (birefringence being the most obvious issue here with wide apertures), etc. You can keep touting science all you want, but ultimately your rationale is in your head, and that's it.
DavidWEGS wrote:
No, just that the two do not compare and especially at that aperture. Your test does not make any sense to me, sorry.
Yes, they don't compare at that aperture, and that's the point. If you're a DX shooter and you want X amount of DoF, you're forced to use the prime in a manner that is far from optimal.
Sam Bennett wrote:
Yes, they don't compare at that aperture, and that's the point. If you're a DX shooter and you want X amount of DoF, you're forced to use the prime in a manner that is far from optimal.
I don't just mean they don't compare at that aperture, I mean the glass used is not comparable in optical quality. If you shot the D300 with the 24-70 on there at F2.8 and then used that same lens on the D700 (given that you must keep the distance to subject, and equivalent FL the same) where your image fills the same visible area, I don't see any quantifiable (theoretic or practical) difference.
Another thing regarding this is the fact that though there is apparently greater DOF with a crop sensor, that is only true because you adjust your lens or distance to subject to obtain the same FOV. The actual DOF is technically greater on the FX sensor than a DX.
So, all else being equal, if you take a D300 with a 24-70; at 30mm (for the sake of choosing an FL to compare) and make a shot from say 10'; of subject A; at an aperture of F2.8, then take the D700+24-70; at 45mm (to be comparable in FOV to the D300 shot) and make another shot of the same subject from 10'; at an aperture of F2.8, then we have a comparable test that will show....
Try that and see what the real world difference is. You may be surprised.
......Enjoy... 
