Do people really believe the $8K camera wouldn't have improved circuitry and read out noise compared to the $2.7K camera? There has to be some advantage to owning the 1Ds III and small things like the ability to push the image a bit harder due to improved noise floor, less electrical noise etc would make sense. I can't believe Canon wouldn't offer improved hardware where possible for the 1 series.
I'm having troubles with DPP. I got the pictures taken, I did them at 1/60th with the lens cap on, but when I have them in DPP, I crank up the brightness all the way but the photo is still totally black, I don't see a hint of noise.
So I did these in PS CS4. I had to turn up the exposure on all of them to +4.
Will Patterson wrote:
I'm having troubles with DPP. I got the pictures taken, I did them at 1/60th with the lens cap on, but when I have them in DPP, I crank up the brightness all the way but the photo is still totally black, I don't see a hint of noise.
The total black is normal... just open the RAW file in PS (directly from DPP) and adjust the levels by pushing it to the left until you reach 25. That way the pattern becomes more noticeble. For fair comparison I think it is best if we all do this test the same way
Pixel Perfect wrote:
Do people really believe the $8K camera wouldn't have improved circuitry and read out noise compared to the $2.7K camera? There has to be some advantage to owning the 1Ds III and small things like the ability to push the image a bit harder due to improved noise floor, less electrical noise etc would make sense. I can't believe Canon wouldn't offer improved hardware where possible for the 1 series.
Could be talking nonsense of course.
FWIW, my 40D files after the dpp+2/ PS 25 adjustment look far better than brainiac's 5D2 files, especially at ISO 200. No criss-cross, just some horisontal banding. Still not up to Daan's 1Ds3 of course. I don't want to add confusion to the ongoing comparison, so I won't post pictures from 40D unless called for. We could ask dpreview to include this procedure in their test, though
Will Patterson wrote:
So I did these in PS CS4. I had to turn up the exposure on all of them to +4.
Thanks for the effort Will, but I think uploading images processed in a completely different way only serves to confuse the issue. The file will be black at +2 in DPP. The patterns emerge once you apply the levels in Photoshop.
Pixel Perfect wrote:
Do people really believe the $8K camera wouldn't have improved circuitry and read out noise compared to the $2.7K camera? There has to be some advantage to owning the 1Ds III and small things like the ability to push the image a bit harder due to improved noise floor, less electrical noise etc would make sense. I can't believe Canon wouldn't offer improved hardware where possible for the 1 series.
Canon stated that the 5D2 had better image quality than the 1Ds3. The statement is still on the Canon web site.
I have just tested with High ISO Noise Reduction switched on, and Long Exposure Noise Reduction switched on, and it makes no difference.
brainiac wrote:
Thanks for the effort Will, but I think uploading images processed in a completely different way only serves to confuse the issue. The file will be black at +2 in DPP. The patterns emerge once you apply the levels in Photoshop.
FWIW, side by side, the photos look the same in DPP or CS4.
But, I'm not sure what you mean when you say 'apply the levels in Photoshop.' DPP is only letting me open the file as a TIFF in Photoshop and I hate that because you lose all those controls that are normally there in PS's raw editor (ACR).
And in any case, Canon SLR support is telling me to send my camera in after looking at the photos. So I'll be doing that on Monday. I'd be willing to bet it comes back no better if not worse.
Will Patterson wrote:
But, I'm not sure what you mean when you say 'apply the levels in Photoshop.' DPP is only letting me open the file as a TIFF in Photoshop and I hate that because you lose all those controls that are normally there in PS's raw editor (ACR).
Well I think we just wanted a method with the least possible variation for the sake of fair comparison. If people adjust lots of controls it becomes harder to see whether there is variation between the cameras, which is what we're looking for.
>And in any case, Canon SLR support is telling me to send my camera in after looking at the photos. So I'll be doing that on Monday. I'd be willing to bet it comes back no better if not worse.
I have to say that both of my canon cameras 20D and 5D show the same type of pattern when pushed ...
if I slide the darks and shadows curves out past 50 it makes it much more pronounced but I am seeing the same thing in both cameras... I am starting to think this is common in all canon cameras... I will see if I can get a hold of a 1Ds to test with also... so far this does not seem out of the normal to my testing...
BTW my test is this
body cap on,
Manual settings,
1/50s exposure
raw image
iso 100
in lightroom 2.2
open raw, set exposure to +2 or greater +4 is max
set tone curve settings
all to +60
I would love for someone with a nikon to do this test also...
jcrawford wrote:
here are a couple of shots taken with my 5D and 20D both lenscaps on at 100iso
Can't confirm with the 5D Mk1 yet (haven't tried to make my photos expose their weak points...so far my cameras point out my weak points), but a few shots I took with my 40D tonight, under optimal lighting (umbrella, background set up so on so forth) that were properly exposed, then pushed 2 stops (from ISO100) didn't show this. I also tried it with a shot that was under exposed by 2 stops, then pushed it 2 stops. Still nothing (again this is a 40D, and not your camera).
I don't know where berk came from, but I know in the Advanced Dungeons and Dragons 2nd Edition campaign setting "Planescape" they were calling people a berk a lot, and it meant pretty much somebody who was a dip-shit.
This was like 13 years ago or so, so I'm going from memory here.
brainiac wrote:
[Canon stated that the 5D2 had better image quality than the 1Ds3. The statement is still on the Canon web site.
Yes, but... I think you might be pushing that "promise" a bit too far.
"Betterness" is composed of more than one parameter. They did not write that "the noise produced when you push the exposure by three stops is better on the 5D II than on the 1Ds3" or whatever we are comparing to. Let's say, just for argument, that the 5D II has a larger dynamic range (I have no idea if it does), higher resolution, and "better color rendition" (whatever that might mean), but that it shows a bit more noise banding when pushed than the other camera. Which is "better?"
Searching for another analogy, we imagine two versions of a car - the old one and the new one. The old one does 0 to 60 in 6 seconds and gets 25 mpg and needs an oil change every 7500 miles; the new once does 0 to 60 in 5.9 seconds and gets 35 mpg and needs an oil change every 5000 miles. Which has "better performance?"
I think that, in the end, your crusade to determine whether the 5D II generates more banding in noise when pushed to extreme degrees will be interesting, whether it turns out that this is a real thing or an artifact of testing. But I also think that a few things are a bit over the top here:
1. The "Canon claimed the 5D II is better" thing isn't working for me.
2. The "problem" seems to occur only if we go looking for it, and in many cases if we do pretty unusual things in exposure and in post.
3. The same "problem" seems to have been present all along in other cameras. For example, when I went back to some old 5D files and did the "push the shadows 3 stops" thing I can see it there. Others report that they see it in other cameras.
4. It isn't an issue in what we might call "normal shooting," and I haven't found it to be an issue even when I do shadow recovery in post in photos with very large dynamic range.
Again, I don't have a problem with trying to understand the nature of shadow noise at the low limits of the sensors, but I think that some lowering of the excitement level is in order. I'm starting to read posts elsewhere along the lines of "now that everyone says that the 5D II has severe noise banding problems..." That is not a good thing.
danmitchell wrote:
Yes, but... I think you might be pushing that "promise" a bit too far.
"Betterness" is composed of more than one parameter. They did not write that "the noise produced when you push the exposure by three stops is better on the 5D II than on the 1Ds3" or whatever we are comparing to. Let's say, just for argument, that the 5D II has a larger dynamic range (I have no idea if it does), higher resolution, and "better color rendition" (whatever that might mean), but that it shows a bit more noise banding when pushed than the other camera. Which is "better?"
Searching for another analogy, we imagine two versions of a car - the old one and the new one. The old one does 0 to 60 in 6 seconds and gets 25 mpg and needs an oil change every 7500 miles; the new once does 0 to 60 in 5.9 seconds and gets 35 mpg and needs an oil change every 5000 miles. Which has "better performance?"
I think that, in the end, your crusade to determine whether the 5D II generates more banding in noise when pushed to extreme degrees will be interesting, whether it turns out that this is a real thing or an artifact of testing. But I also think that a few things are a bit over the top here:
1. The "Canon claimed the 5D II is better" thing isn't working for me.
2. The "problem" seems to occur only if we go looking for it, and in many cases if we do pretty unusual things in exposure and in post.
3. The same "problem" seems to have been present all along in other cameras. For example, when I went back to some old 5D files and did the "push the shadows 3 stops" thing I can see it there. Others report that they see it in other cameras.
4. It isn't an issue in what we might call "normal shooting," and I haven't found it to be an issue even when I do shadow recovery in post in photos with very large dynamic range.
Again, I don't have a problem with trying to understand the nature of shadow noise at the low limits of the sensors, but I think that some lowering of the excitement level is in order. I'm starting to read posts elsewhere along the lines of "now that everyone says that the 5D II has severe noise banding problems..." That is not a good thing....Show more →
Well, first of all, it's not meant to be a crusade - I just wanted to see if others were seeing what I'm seeing, and to get to the bottom of whether this is a feature of the camera model, or an anomaly. I've stated a few times that this is on the edge of performance.
I think it's good that people are warned that this camera may have noise pattern problems under some circumstances, and those circumstances have been spelled out pretty clearly here, so people can make their own minds up, preferably by testing the camera themselves before buying to see if this is going to be a problem for their kind of shooting. Nothing is being said here to mislead. Canon is going to sell a lot of cameras regardless of anything written here, and if sales suffer because people don't like gauzy noise in the shadows, that's fair isn't it? Why should we be in hock to Canon's PR department? The camera has surprisingly large scale noise in shadows at low isos. As far as I'm concerned, that's a fact, and if people want to get hysterical about it, that isn't going to stop me speaking the truth I see.
