I'm not sure if this affects the comparison but the image to the right is brighter. Either way the real test is not viewing at 100% on your monitor, rather viewing a tangible print. If the banding shows up in a print, you've got a gripe--but if the two images are indistinguishable...well, this was all in vain.
M Vers wrote:
I'm not sure if this affects the comparison but the image to the right is brighter. Either way the real test is not viewing at 100% on your monitor, rather viewing a tangible print. If the banding shows up in a print, you've got a gripe--but if the two images are indistinguishable...well, this was all in vain.
The 5D2 seems a bit more sensitive when set at 100 iso. Not much though. I agree about this only mattering if it shows in prints, but the problem is that in my opinion it will. Don't forget that the crop I show is at minimum contrast, and only pushed 1 stop. The criss-cross pattern is showing up in all sorts of other pictures, at higher isos, under mixed light etc.. If you need to pump up contrast, or push, or solve wild colour balances, all these stress tests are going to exaggerate this effect.
brainiac wrote:
The 5D2 seems a bit more sensitive when set at 100 iso. Not much though. I agree about this only mattering if it shows in prints, but the problem is that in my opinion it will.
No offense, and I'm not judging your skill or know how, but you're opinion doesn't really account for fact. Have you actually made comparable prints or are you basing your opinion off of an assumption? I'm not asking you that to be an ass--It's a legitimate and sincere question. My point is it's much easier to see flaws on your monitor at 100% than it is on a print and IMO the print is where it matters. I am, however, curious to see more comparisons and if anyone has the ability to print on site it would be insightful to find out how they compare in that medium.
Also, can you repost the above example but push the 5DII image to the left a bit to match the 1D file? Just curious if it will make a difference.
One reason that I expect the pattern in high iso files to be visible in prints is that I can see it on my 30" monitor when the image is fitted to the screen, i.e. at print size. If I can see it on a monitor at 12" high, then I can almost certainly see it in a print at 12" high. As I have said, this pattern is going to become a nuisance for those who routinely shoot under disco lights without flash, or routinely pull data out of the shadows in pictures that demand maximum dynamic range (e.g. landscapes with sky).
brainiac wrote:
One reason that I expect the pattern in high iso files to be visible in prints is that I can see it on my 30" monitor when the image is fitted to the screen, i.e. at print size. If I can see it on a monitor at 12" high, then I can almost certainly see it in a print at 12" high. As I have said, this pattern is going to become a nuisance for those who routinely shoot under disco lights without flash, or routinely pull data out of the shadows in pictures that demand maximum dynamic range (e.g. landscapes with sky)....Show more →
FYI I have no opinion on this... (just sharing info).
Thanks for that link - I have read it before. I confirm that I am seeing exactly what digilloyd describes. This problem isn't going to stop me using a 5D2, but it is a fly in the ointment. What I really want is a D3x in a 450D body with Nikon flash and an EOS mount. I'm hoping, probably in vain, that the 1D4 will be full frame, and smaller, and lighter, or that Canon fixes this issue in firmware.
M Vers wrote:
Right, but I thought this was about low ISO's....
I see the criss-cross pattern at high isos too. It's more surprising that it's there at low isos though. At 6400 you can cut the 5 a bit of slack. At 100 iso it's disappointing.
brainiac wrote:
Thanks for that link - I have read it before. I confirm that I am seeing exactly what digilloyd describes. This problem isn't going to stop me using a 5D2, but it is a fly in the ointment. What I really want is a D3x in a 450D body with Nikon flash and an EOS mount. I'm hoping, probably in vain, that the 1D4 will be full frame, and smaller, and lighter, or that Canon fixes this issue in firmware.
I'll take one of those (except the 450D, I don't mind the larger body sizes)! The difference in the shadows is really striking if you download the RAW files and see for yourself in your RAW converter of choice.
Ok, pixel peepers. Two shots at ISO 200, one from 5D2 and one from 1Ds3. Same 24L/II lens on tripod. Knock yourself out. http://fm.jirman.com/200/
Take it or leave it - I didn't pay scientific attention to star alignment.
brainiac wrote:
In my experience the 1Ds3 can produce great results when pushed.
That might be true. My point is that expectations about noise should go out the window when pushing. There is noise down there, and every camera handles it differently. The 5DII might read things out faster/cheaper/who knows, and that might adversely affect the noise in the shadows. With proper exposure and no pushing, the noise in these new cameras should be pretty good and be as rated. But when you start doing shadow expansion, fill light, pushing, etc., you are going to expose this kind of stuff.
Also, I think noise in the shadows can be higher at low ISOs. From ejmartin's page:
"On the other hand, in lower exposure zones at low ISO, where post-amplification read noise becomes important, the read noise goes down by a bit less than a factor of two (in electrons) when the ISO doubles."
Thanks Stanj, well done. I can see a difference in lum/color/noise/sharpness. "Over the thumb" as we say here, I see a draw, if I had to pick the one with less noise I d take the 5dII.
Maybe the two cameras represent rather two choices than two qualities
Bummer. My 5D2 produces pretty clean shadows even at high ISOs...Of course proper exposure plays a greater role in image quality than does the camera used... Here's one from an event I shot last night, 5D2 at 3200 then pushed a stop just for fun:
I have used three different 5D2's the past weeks. I updated the first two from FW 1.0.6 to 1.0.7. The RAW files shot with FW 1.0.7 which I opened in LR or DPP showed lots of noise at low ISO's and magenta casts in dark areas or low-contrast areas. DPP and LR gave the same results
The third 5D2 I use hasn't been updated to FW 1.0.7, but is still on FW 1.0.6. The RAW files from this cam show way less noise (almost no noise at low ISO's, very similair to my 1Ds3) and no magenta casts until the higher ISO's (ISO3200 and up mainly).
So, there is a difference in performance between these three 5D2's. It could be related to the FW update and how that conflicts with the different RAW converters (LR or DPP) or it could be that some 5D2's have "crooked" sensors or something else is going on... You tell me....Show more →
If true, this is the most disturbing post in this entire thread.
brainiac wrote:
A little reminder of why it matters not to have a cross-hatch pattern in deep shadows at 100 iso, first with no shadow pushing, second with perhaps 2/3rds of a stop of shadow rescue:
My monitor is somewhat dim here at work, but I don't see the problem with this image.
There isn't a particular problem except that it could do with more latitude, and pulling detail out of the shadows has its limits. If I had to choose a camera for taking this shot, I would use my 1Ds3 instead of my 5D2, because deep shadows have slightly better texture. You won't see that on the web at this size, but in a decent-sized print you will.
Guys, take pity on a poor potential 5D2 buyer here. Is the consensus that:
A) All 5D2's have a significant pattern noise problem compared to the 1DS3 (whose significance may depend upon the stringency of individual user's requirements)
B) Only some 5D2's have this problem
C) It's a firmware issue
Still waiting to pull the trigger, but I'd love to see some resolution of this before doing so.
I believe A, but can't offer any info on B and C. What I will say is that the only two cameras which seem to outperform the 5D2 (and even then, only very slightly) both cost much more than twice as much as the 5D2, so go ahead and buy your 5D2, it's a fantastic camera regardless of this issue, and superb value for money compared to other 21 and 24 Mpixel cameras.
