False Detail vs Real detail. MPs and AA filters and crispy lines

Not taking one side or another in this, but the lens certainly plays a role.
Which one, Zeiss or Sigma?
36 mp finds flaws in even the best lenses if photographic technique is good.
You suggesting it is Ming's technique that does not bring the best out of the camera?

--
- sergey
I've whacked up the clarity of the D800. You tell me which one has more detail.





481803692fb647a6b6732e18441e6b7f.jpg
 
I know which one I'd rather be lugging around.
Indeed. And spite of our technobabble, for most people IQ has been 'good enough' in cameras from smart phones to FF for years.
For most photos that most people take and in the way most people display their photos, the difference in IQ between a cell phone and FF will have zero effect in terms of the "success" of the photo.
 
Not taking one side or another in this, but the lens certainly plays a role.
Which one, Zeiss or Sigma?
The 14-42 on the mFT camera.
I was responding to how D800 has much more detail than Foveon. The mFT just happened to be in the same frame.
Ah -- my bad! Yes, for that crop, Foveon compared well to the D800 with regards to resolution.
You're not looking close enough.

 
Last edited:
Not taking one side or another in this, but the lens certainly plays a role.
Which one, Zeiss or Sigma?
36 mp finds flaws in even the best lenses if photographic technique is good.
You suggesting it is Ming's technique that does not bring the best out of the camera?
I've whacked up the clarity of the D800. You tell me which one has more detail.

481803692fb647a6b6732e18441e6b7f.jpg
The differences in detail are still the same, and, to my eyes, not all that different. However, the D800 rendering, both before and after, is more pleasing to my eyes, and not by a little. That said, if both files were optimally processed, I think they'd be more than "close enough".
 
Not taking one side or another in this, but the lens certainly plays a role.
Which one, Zeiss or Sigma?
36 mp finds flaws in even the best lenses if photographic technique is good.
You suggesting it is Ming's technique that does not bring the best out of the camera?
I've whacked up the clarity of the D800. You tell me which one has more detail.

481803692fb647a6b6732e18441e6b7f.jpg
The differences in detail are still the same, and, to my eyes, not all that different. However, the D800 rendering, both before and after, is more pleasing to my eyes, and not by a little. That said, if both files were optimally processed, I think they'd be more than "close enough".
Hmmm. I see quite a bit more detail in the D800 shot. That said, the Foveon is far closer than, say, the m4/3rds camera.

On the other hand, Foveon tends to make the gaps and in the brick unrealistically big (for example) which is a type of alias artifact ... though it could be Sigma processing. It does this to roof tiles and pretty much any straight line that is near the edge of resolution.

Looks like detail to those not really looking.
 
Not taking one side or another in this, but the lens certainly plays a role.
Which one, Zeiss or Sigma?
36 mp finds flaws in even the best lenses if photographic technique is good.
You suggesting it is Ming's technique that does not bring the best out of the camera?
I've whacked up the clarity of the D800. You tell me which one has more detail.

481803692fb647a6b6732e18441e6b7f.jpg
The differences in detail are still the same, and, to my eyes, not all that different. However, the D800 rendering, both before and after, is more pleasing to my eyes, and not by a little. That said, if both files were optimally processed, I think they'd be more than "close enough".
Hmmm. I see quite a bit more detail in the D800 shot. That said, the Foveon is far closer than, say, the m4/3rds camera.

On the other hand, Foveon tends to make the gaps and in the brick unrealistically big (for example) which is a type of alias artifact ... though it could be Sigma processing. It does this to roof tiles and pretty much any straight line that is near the edge of resolution.

Looks like detail to those not really looking.
Here's the way I like to think about it. Let's say there were a sign on the wall with writing on it. If that writing were just as the limits of my ability to be able to read it, then I think I would be able to read it on the Foveon and D800 crops. If not, then if the text were a wee bit larger on the one I couldn't read, then I would be able to read it. At least, that's how I'm seeing it.
 
Not taking one side or another in this, but the lens certainly plays a role.
Which one, Zeiss or Sigma?
36 mp finds flaws in even the best lenses if photographic technique is good.
You suggesting it is Ming's technique that does not bring the best out of the camera?

--
- sergey
I've whacked up the clarity of the D800. You tell me which one has more detail.

481803692fb647a6b6732e18441e6b7f.jpg
That just does not look right to me at all. But fine, if this is how you see it. The D800 (as you show it) just does not have that cool transparency in detail Foveon is known for. As I said, it is just not the same.

I will come up with crops to show later, if this thread does not max out by then. D800 is more dr (it's clear), it makes files that are easily controllable, with Sigma you either get it right or you don't. But as far as detail goes, Merrills are not that far away in comparison, and in most scenarios they produce very pleasing, crispy, and detailed results.

--
- sergey
 
Last edited:
Lets try this again... Which one has more detail. The left image or the right?

D810 UPSAMPLED to the 63 MP ORF file.

D810 UPSAMPLED to the 63 MP ORF file.

Or how about this one:

683e57f9df754e03b721d9bad6a0e61d.jpg


Of course Sergey, I dont expect a straight answer out of a person such as yourself... I feel you may be allergic to it. But you hove no problems with Moire or Aliasing.

To be clear in the left image above the EM5 mark 2 absolutely trumps the artifact riddled D810 file... I mean destroys it... I mean it makes it look like a horrible mess. Or do you disagree? Do you feel the D810 file is a clean as a whistle, showing nothing but accurate colour splotches in a black and white print?
In my opinion, the EM5II photos are both the better photos. However, in the bottom frame, they are very close. In the top frame, the obvious color moire is a huge distraction for the D810 photo. However, I have to wonder if that could not be handled better with another RAW converter and that they would look significantly closer.
 
511f705c1f594fdab9659daf3ff7ac94.jpg


Why do the bricks in the Foveon shot look so much fatter than the shots in the D800 image? Also, what is that purple stuff on the Foveon curtains?
 
I will come up with crops to show later, if this thread does not max out by then. D800 is more dr (it's clear), it makes files that are easily controllable, with Sigma you either get it right or you don't. But as far as detail goes, Merrills are not that far away in comparison, and in most scenarios they produce very pleasing, crispy, and detailed results.
You have both cameras don't you? I'd be pleased to see equivalent shots of a detailed something from both. Post and I'd be pleased to discuss.
 
511f705c1f594fdab9659daf3ff7ac94.jpg


Why do the bricks in the Foveon shot look so much fatter than the shots in the D800 image?
Hardly. Also, do not forget you are looking at D800E with absolutely the best lens on it. Your statement that D800 images will have so much more detail in them is just not true.
Also, what is that purple stuff on the Foveon curtains?
They lack red, it is very distinctive Merrill look.

--
- sergey
 
You have both cameras don't you? I'd be pleased to see equivalent shots of a detailed something from both. Post and I'd be pleased to discuss.
I downloaded them from elsewhere, some like sample RAW files, those should be enough. But why not, might try out images on the same scene just as well.
 
511f705c1f594fdab9659daf3ff7ac94.jpg


Why do the bricks in the Foveon shot look so much fatter than the shots in the D800 image?
Hardly. Also, do not forget you are looking at D800E with absolutely the best lens on it. Your statement that D800 images will have so much more detail in them is just not true.
Also, what is that purple stuff on the Foveon curtains?
They lack red, it is very distinctive Merrill look.


The D800 gives similar results because that is the camera with which I first compared the Merrill. However, we really should stop posting in this thread as this is hugely off topic.

Give me a nudge if you decide to post a side by side comparison at full res and we can have a chat.
 
511f705c1f594fdab9659daf3ff7ac94.jpg


Why do the bricks in the Foveon shot look so much fatter than the shots in the D800 image? Also, what is that purple stuff on the Foveon curtains?
Bricks look the same to me, just greater contrast in the shadows between bricks in the Foveon photo. As for the curtains, the color in the Foveon photo looks off all the way around, but, yes, the Nikon photo is better in that regard.
 
Doesn't matter how many pixel the sensor has...

Demosaicing will introduce false information because it's a little bit guess work (Bayer sensor).
I think that's part of the issue. Bayer image processing is still using algorithms designed to maximise the apparent resolution, on the basis that the sensor hasn't enough. With the newer high resolution sensors it makes more sense to aim to minimise artefacts, on the basis that the sensor has too much of those.
The problem with a less aggressive demosaic algorithm is that it will tend to reduce detail (real and false) and increase blur at the same time it tends to reduces artifacts. The trade-off is illustrated by looking at the raw and jpeg versions of the D810 studio shot. There are confounding processing differences, of course, but nonetheless such a comparison is a convenient way to get a sense of the trade-off inherent in a more aggressive demosaicing algorithm (ACR's) vs. a less aggressive one (Nikon's).



Raw version on left. JPEG version on right.

Raw version on left. JPEG version on right.
 
Doesn't matter how many pixel the sensor has...

Demosaicing will introduce false information because it's a little bit guess work (Bayer sensor).
I think that's part of the issue. Bayer image processing is still using algorithms designed to maximise the apparent resolution, on the basis that the sensor hasn't enough. With the newer high resolution sensors it makes more sense to aim to minimise artefacts, on the basis that the sensor has too much of those.
The problem with a less aggressive demosaic algorithm is that it will tend to reduce detail (real and false) and increase blur at the same time it tends to reduces artifacts. The trade-off is illustrated by looking at the raw and jpeg versions of the D810 studio shot. There are confounding processing differences, of course, but nonetheless such a comparison is a convenient way to get a sense of the trade-off inherent in a more aggressive demosaicing algorithm (ACR's) vs. a less aggressive one (Nikon's).

Raw version on left. JPEG version on right.

Raw version on left. JPEG version on right.
Absolutely spot-on. The point is, if you don't want the whole resolution, then why not have the less aggressive demosaicking and avoid at least some of the more objectionable artefacts (you'll never get rid of them all, which is why no-AA-filter is such a bad idea.

--
Bob
Eating people is wrong.
Michael Flanders.
 
Last edited:
True. The bottom shot is very similar, but there are still plenty of clues the mark 2 is describing a fraction more detail.

i asked the designers at work their thoughts and it was 50 50. It seems camera companies understand the risks of artifacts are with the trade off for pixel level detail. However I know which one I prefer I am just astonished at how few people notice.

just interesting.
 
Funny how all the 'no AA' fans have suddenly swapped sides, isn't it?
It was predictable, the same way as a Sony quality sensor suddenly made the old 12 mp seem not so good. Previously, the old 10 and 12 mp sensor were totally competitive with everything around them ...
I still keep one of those 12mp sensors around and was tempted to keep and e3 ( however it has too many focus issues). Rather like film they all have their own qualities especially with regards to subtle colour differences. I do wonder if the push for higher ISO and pixel detail is doing something to colour, Kirk tuck sometimes waxes lyrical about older cameras, and the e1 has remained a favorite for many.
 

Keyboard shortcuts

Back
Top