denoir wrote:
The M9 shot was exposed by a 1/3rd stop more in that shot if one goes by exposure values. There's a difference in dynamic range between the cameras and you have to remember that the different RAW developer do their bit on the "plain" image.
Ah, I forgot that you were using IDC for the NEX-C3 and LR3 for the M9. That explains the difference. Highlight recovery is still a forte of the ACR engine.
denoir wrote:
That was surprising and disappointing. Given that the M9 is not particularly good at this I had expected the C3 to be a lot better. It looks that like with the M9 one has to expose to preserve the highlights when shooting the C3, but even more so.
It would not be surprised if the results will be much better as soon as LR supports the raws. I find that the highlights recover very nicely and easily on the NEX-3.
douglasf13 wrote:
Ah, I forgot that you were using IDC for the NEX-C3 and LR3 for the M9. That explains the difference. Highlight recovery is still a forte of the ACR engine.
I tried both the Sony RAW software and Lightroom for my NEX 5, and LR is miles better. So I expect that to carry over for the C3, the pictures of which should look commensurately better yet...
denoir wrote:
Alright, time for some test results. As the NEX-C3 is supposed to serve as my backup camera for an M9, the comparisons are with an M9.
The lens used in all tests was the Zeiss 25/2.8 Biogon ZM. It holds the distinction of being the lens with highest recorded resolving power (400 lp/mm, MTF10) so it outresolves any existing sensor - a full frame sensor of over 500 megapixel would be required to make full use of the resolution of the lens.
What would happen if we reduced the C3 image size to match the M9 FOV? I.e could you use the C3 as the equivalent of a tele converter, or are you better off just cropping an M9 file:
Close, but no cigar. The C3 image was reduced to an 8 megapixel image to match the field of view and it isn't quite there. So perhaps 16 megapixel is a bit of an overkill, at least as far as resolution goes.
Latitude
Dynamic range in digital cameras is a complex issue that to a large degree will depend on the RAW development. So a more useful thing and more simple thing to look at is the latitude - i.e how much can you push or pull the exposure. The test scene is the same as above but I took one shot that was 2 stops over exposed and one that was 2 stops under exposed with each camera. The idea is to see how good it is at recovery.
Shadows
The M9 is quite good at recovering shadows and my hopes were that the C3 with its newer sensors could match it.
Quite good results. It looks like the C3 is a match for the M9 in that respect.
Highlights
I had very high hopes for the C3 for this. The M9 is terrible at recovering highlights. Given that my 5DII is quite good at it, I was expecting a lot for a more modern CMOS sensor.
That was surprising and disappointing. Given that the M9 is not particularly good at this I had expected the C3 to be a lot better. It looks that like with the M9 one has to expose to preserve the highlights when shooting the C3, but even more so. ...Show more →
ulrikft2 wrote:
Weird, my nex5 is far better than that actually.
sony raw converter...
also, they may have given the 16MP sensor a different AA filter. i'd be curious how the comparison looks with ACR (or even better RPP) and optimal sharpening for each to compensate for the AA inequality.
Luka it would be great to see a test with corresponding focal lengths, if you are shooting 18mm on m9 shoot 12mm on nex. i think the results will come out more evenly matched.
AhamB is right regarding the asymmetry. Luka, your first pixel pattern is incorrect. Red and Blue channels are never in the same row. You are assuming that light leakage happens linearly. Leakage goes into all the neighboring pixels in all directions. ie. take a 3x3 pixel grid, and the leakage in the center pixel will affect, to varying extent, the neighboring 8 pixels. And this mix produces shifts.
An easy way to decipher this phenomenon - for example, figure out what are the major components of cyan (i.e. additive mix of RGB components), from this mix you can roughly identify where the impurity is coming from. Given that white balance (gains by the charge amplifier for each color) is fixed, it has to be the non uniform pixel response and color addition which will create this uneven color response instead of white. Even going out from the center to the edge of the array, there will be subtle shifts in color. What this is suggesting is that the light rays are bending at different angles (which is expected), and hence the pixel response is different. This is the reason why micro lenses are gradually shifted as they move away from the center of the optical array.
PS: I have given up on telling people that latitude and dynamic range are not the same, especially for digital sensors!
Jim: To my knowledge, there is no asymmetry with regards to Red and Blue being different in the different sides of Bayer array. It will create on-board processing nightmares. As it exists, it is 2:1:1 for G:R:B and the red pixels are evenly spaced throughout the array as blue pixels.
Yes, it'll be interesting if the 24mpx rumors pan out, I personally think it won't be that high.
I imagine though that since the Nex7 will be a flagship camera for Sony, it might showcase leading edge performance of a new generation of sensors like the 16mpx D7000/K5 sensor did.
Ajay C wrote:
AhamB is right regarding the asymmetry. Luka, your first pixel pattern is incorrect. Red and Blue channels are never in the same row. You are assuming that light leakage happens linearly. Leakage goes into all the neighboring pixels in all directions. ie. take a 3x3 pixel grid, and the leakage in the center pixel will affect, to varying extent, the neighboring 8 pixels. And this mix produces shifts.
I know, that's why I provided the second image. The first one was just intended as a simplified illustration.
An easy way to decipher this phenomenon - for example, figure out what are the major components of cyan (i.e. additive mix of RGB components), from this mix you can roughly identify where the impurity is coming from. Given that white balance (gains by the charge amplifier for each color) is fixed, it has to be the non uniform pixel response and color addition which will create this uneven color response instead of white. Even going out from the center to the edge of the array, there will be subtle shifts in color. What this is suggesting is that the light rays are bending at different angles (which is expected), and hence the pixel response is different. This is the reason why micro lenses are gradually shifted as they move away from the center of the optical array....Show more →
All true, but that is just explaining why there is a color shift, not why there is a different shift on the left and right side. The angles are identical - the only difference is if it is coming from the left side. The Bayer grid, as you confirmed is uniform so there could be no left/right side asymmetry.
The big question remains - why would there be a difference between the left and the right side of the sensor when everything is symmetrical and uniform?
Or to illustrate it, if we have points A & B at a distance d from the center of the frame, why is there a magenta cast at A and a cyan at B when the chief ray angles should be identical:
FlyPenFly wrote:
I imagine though that since the Nex7 will be a flagship camera for Sony, it might showcase leading edge performance of a new generation of sensors like the 16mpx D7000/K5 sensor did.
When the C3 was announced it was widely speculated that it would have the A55/D7000/K5 sensor. I'm not sure if it's the case, but it's a Sony and it's 16 mpx.
denoir wrote:
When the C3 was announced it was widely speculated that it would have the A55/D7000/K5 sensor. I'm not sure if it's the case, but it's a Sony and it's 16 mpx.
Sony claimed, upon announcing the C3, that the sensor is "brand new." Who knows what that means, but, my guess that it is the same sensor tuned differently with different support electronics to aid in power consumption. Any knowledgable ideas on this from anyone?
What makes it unlikely that it's the same one is that it doesn't provide support for ISO below 200.
I've done some high ISO shots now (I'll post some later) and although the images are 'clean' at ISO 1600, they have been subjected to pretty extreme noise reduction so the loss of detail is very noticeable. AFIK people were praising the K5 for amazing high ISO performance.. On the other hand that might have been due to aggressive noise reduction as well.
ulrikft2 wrote:
Weird, my nex5 is far better than that actually.
sebboh wrote:
also, they may have given the 16MP sensor a different AA filter. i'd be curious how the comparison looks with ACR (or even better RPP) and optimal sharpening for each to compensate for the AA inequality.
Morfeus wrote:
It would not be surprised if the results will be much better as soon as LR supports the raws. I find that the highlights recover very nicely and easily on the NEX-3.
Guys, I think you are misinterpreting the tests. You think it looks worse than it should, but it doesn't. It's just that it is being compared to an M9 which is in a completely different league. There is no small format camera on the market that comes even close to the per pixel sharpness and micro contrast the M9 produces. The comparison is as uneven as if one was comparing the M9 to a medium format digital camera. If the MFD image was resized to the 18 mpixels of the M9 then the latter would look like crap in comparison.
If a more reasonable comparison is made the NEX-C3 performs quite well. Here's a 100% crop of it compared to the Canon 7D (50 Summilux ASPH on the C3 and Zeiss 50/2 Makro Planar on the 7D):
As you can see the C3 does quite well and is a bit more crisp (probably due to weaker AA filter). Sebboh commented a few pages back that he found the NEX-5 to be slightly better than the 7D and that's what we're seeing here. That's impressive as the 7D is a camera that costs more than three times as much as the NEX-C3.
That is a fair comparison unlike the one against the M9.
kahren wrote:
Luka it would be great to see a test with corresponding focal lengths, if you are shooting 18mm on m9 shoot 12mm on nex. i think the results will come out more evenly matched.
denoir wrote:
Is it better at anything compared to ACR?
i've heard some people argue that color shows more subtlety with the sony converter. i've not fiddled with it enough to have an opinion on that, but it's possible as color is not one of acr's strong points with sony files.
denoir wrote:
What makes it unlikely that it's the same one is that it doesn't provide support for ISO below 200.
I've done some high ISO shots now (I'll post some later) and although the images are 'clean' at ISO 1600, they have been subjected to pretty extreme noise reduction so the loss of detail is very noticeable. AFIK people were praising the K5 for amazing high ISO performance.. On the other hand that might have been due to aggressive noise reduction as well.
That NR look is another product if IDC. You'll really need to wait until LR3 supports the C3 to make any real pixel-level comparisons...or you could try RPP.
denoir wrote:
If a more reasonable comparison is made the NEX-C3 performs quite well. Here's a 100% crop of it compared to the Canon 7D (50 Summilux ASPH on the C3 and Zeiss 50/2 Makro Planar on the 7D):
As you can see the C3 does quite well and is a bit more crisp (probably due to weaker AA filter). Sebboh commented a few pages back that he found the NEX-5 to be slightly better than the 7D and that's what we're seeing here. That's impressive as the 7D is a camera that costs more than three times as much as the NEX-C3.
interesting to see the differences in leica vs zeiss rendering here in addition to the camera differences.
sebboh wrote:
i've heard some people argue that color shows more subtlety with the sony converter. i've not fiddled with it enough to have an opinion on that, but it's possible as color is not one of acr's strong points with sony files.
Hmm, it's probably because I don't know how to use the software properly yet but I'm finding the colors to be perhaps the weakest point - they seem sort of muddy. I have not quite figured out how to get them to work - if I boost contrast I typically start blowing highlights and shadows.
douglasf13 wrote:
That NR look is another product if IDC. You'll really need to wait until LR3 supports the C3 to make any real pixel-level comparisons...or you could try RPP.
I could.. I have it installed on my laptop.. I'm not sure if the C3 is supported yet though.
sebboh wrote:
interesting to see the differences in leica vs zeiss rendering here in addition to the camera differences.
Yes, but can you tell what differences the sensor is responsible for and what can be attributed to the lens? I think that according to the MTF charts the 50 MP should have a bit higher micro contrast in the fine detail, but then again that's exactly what an AA filter kills.
I'm seeing higher edge contrast in the C3 shot, which would correspond to high micro contrast in the lower spatial frequencies 10-20 lp/mm, but again I'm not sure if it is the sensor or the lens. My guess would be the sensor. If somebody had shown me those crops and asked me which one was the Leica and which the Zeiss, without telling me that the cameras were different then I would have guessed wrong.
