This topic is in the 2009 archive. You can read it and like its posts here. It is read only.
Archive 20095D2 low ISO poor v 1Ds3
p.20 #15D2 low ISO poor v 1Ds3
http://cyberphotographer.com/5d2/alex5d2posterised_lowrez.jpg
http://cyberphotographer.com/5d2/alex5d2posterised_crop.jpg
I think it's a different league. I would have had a result out of this shot with my 1Ds3. With my 5D2 it's trash. Pushing the shadows for the sake of maximum dynamic range is producing this kind of result in DPP time and again. You can say I shouldn't be trying to push the shadows, but that's academic, because I used to get great results with that technique and the 1Ds3. You just can't push the shadows on the 5D2 the way you can on the 1Ds3 because the 5D2 seems to be missing 2 bits or more. Bear in mind that I used the 1Ds3 professionally for a year, so I know whereof I speak. Of course, this is DPP. I don't know about other convertors. And can someone please tell me how to get back to firmware 1.0.6?
http://cyberphotographer.com/5d2/alex5d2posterised_crop.jpg
I think it's a different league. I would have had a result out of this shot with my 1Ds3. With my 5D2 it's trash. Pushing the shadows for the sake of maximum dynamic range is producing this kind of result in DPP time and again. You can say I shouldn't be trying to push the shadows, but that's academic, because I used to get great results with that technique and the 1Ds3. You just can't push the shadows on the 5D2 the way you can on the 1Ds3 because the 5D2 seems to be missing 2 bits or more. Bear in mind that I used the 1Ds3 professionally for a year, so I know whereof I speak. Of course, this is DPP. I don't know about other convertors. And can someone please tell me how to get back to firmware 1.0.6?
Feb 15, 2009 at 06:19 PM
p.20 #25D2 low ISO poor v 1Ds3
That looks exactly the same noise trait I got from the 5D, and the same thing I now no longer see on my 5D2. Can't comment on the 1DS3 in comparison unfortunately.
Feb 15, 2009 at 06:38 PM
p.20 #35D2 low ISO poor v 1Ds3
AdrianRogers wrote:
That looks exactly the same noise trait I got from the 5D, and the same thing I now no longer see on my 5D2. Can't comment on the 1DS3 in comparison unfortunately.
Hotdamn! I just checked the front of my 5D mark ii and there is no mark ii logo! Just kidding ;-) I really want to try 1.0.6 to see if it suffers from the same problem. This really looks like 12 bit of old, not the luxurious 14 bits that I am used to.
That looks exactly the same noise trait I got from the 5D, and the same thing I now no longer see on my 5D2. Can't comment on the 1DS3 in comparison unfortunately.
Hotdamn! I just checked the front of my 5D mark ii and there is no mark ii logo! Just kidding ;-) I really want to try 1.0.6 to see if it suffers from the same problem. This really looks like 12 bit of old, not the luxurious 14 bits that I am used to.
Feb 15, 2009 at 06:41 PM
p.20 #45D2 low ISO poor v 1Ds3
izanaki wrote:
The 1Ds3 is a wonderful camera that produces excellent files with an almost "magical" quality. The 5D2 also produces nice files, but in a different league IMO. The fact that both cams share 21MP's doesn't mean they share the same IQ (appearantly). I prefer the 1Ds3 files over the 5D2 files any time
According to DPreview's review of the 1Ds Mark III, its RAW headroom has about 11 stops of exposure latitude where the 5DII has just over 10 stops of exposure headroom. I will be picking up a 1DsIII when its successor comes out. I can't wait to own a 1Ds body to compliment my classic 5D 
according to DPR ISO3200 also has more DR than ISO100 sooooo.......
yeah. suuurre....
Daan B wrote
I have both the 5D2 and 1Ds3... I will do some comparison tests when I have the time... but right now my feeling is that the 1Ds3 has the edge at lower ISO's and the 5D2 has the edge at higher ISO's (ISO1600/3200). Besides that, the 1Ds3 contains more shadow detail. Shadows in 5D2 files are easily clipped and contain nothing more than noise and other ugly pollution when you open them in PP. Even at low ISO's. Not so with the 1Ds3...The 1Ds3 is a wonderful camera that produces excellent files with an almost "magical" quality. The 5D2 also produces nice files, but in a different league IMO. The fact that both cams share 21MP's doesn't mean they share the same IQ (appearantly). I prefer the 1Ds3 files over the 5D2 files any time
according to DPR ISO3200 also has more DR than ISO100 sooooo.......
yeah. suuurre....
Feb 16, 2009 at 03:06 AM
p.20 #55D2 low ISO poor v 1Ds3
brainiac wrote:
Hotdamn! I just checked the front of my 5D mark ii and there is no mark ii logo! Just kidding ;-) I really want to try 1.0.6 to see if it suffers from the same problem. This really looks like 12 bit of old, not the luxurious 14 bits that I am used to.
none of the really give more than 12bit anyway
how about putting on lens cap, covering eye piece going into pitch black room room, shutter at 1/8000th and take 3 pics at ISO100 and three at ISO200 (HTP on all that other stuff turned off) and then post the raw somewhere, do it for your dsiii and 5dmkii.
then can measure the read noise and see if it really is worse than on your dsiii at low iso or if the 5dmkii is worse than my 5dmkii at low iso.
AdrianRogers wrote
That looks exactly the same noise trait I got from the 5D, and the same thing I now no longer see on my 5D2. Can't comment on the 1DS3 in comparison unfortunately.none of the really give more than 12bit anyway
how about putting on lens cap, covering eye piece going into pitch black room room, shutter at 1/8000th and take 3 pics at ISO100 and three at ISO200 (HTP on all that other stuff turned off) and then post the raw somewhere, do it for your dsiii and 5dmkii.
then can measure the read noise and see if it really is worse than on your dsiii at low iso or if the 5dmkii is worse than my 5dmkii at low iso.
Feb 16, 2009 at 03:11 AM
p.20 #65D2 low ISO poor v 1Ds3
Unfortunately I have sold the 1Ds3, as I said in the OP, so that'll not work.
Feb 16, 2009 at 05:30 AM
p.20 #75D2 low ISO poor v 1Ds3
brainiac wrote:
Unfortunately I have sold the 1Ds3, as I said in the OP, so that'll not work.
In any case, you could compare your 5D2 to skibum's 5D2 this way to see wether there is a difference in read noise between the cams...
Unfortunately I have sold the 1Ds3, as I said in the OP, so that'll not work.
In any case, you could compare your 5D2 to skibum's 5D2 this way to see wether there is a difference in read noise between the cams...
Feb 16, 2009 at 05:42 AM
p.20 #85D2 low ISO poor v 1Ds3
I'l chime in too for a 3rd comparison. Still on the release firmware.
Feb 16, 2009 at 07:24 AM
p.20 #95D2 low ISO poor v 1Ds3
Where are you, Adrian? I am in central London. If you are near, perhaps we could meet for a pint and a test.
Feb 16, 2009 at 07:26 AM
p.20 #105D2 low ISO poor v 1Ds3
I'm in Manchester but my other half lives in Westminster, though no plans to be down there for at least the following month. Shall happily shoot you an email when i next plan to be down though. Pubs and cameras? Hah!
Feb 16, 2009 at 07:34 AM
p.20 #115D2 low ISO poor v 1Ds3
AdrianRogers wrote:
Pubs and cameras? Hah!
A winning combination... I look forward to it.
Pubs and cameras? Hah!
A winning combination... I look forward to it.
Feb 16, 2009 at 07:41 AM
p.20 #125D2 low ISO poor v 1Ds3
AdrianRogers wrote:
I'l chime in too for a 3rd comparison. Still on the release firmware.
That will be interesting. I hope you guys are willing to share your findings with us
I'l chime in too for a 3rd comparison. Still on the release firmware.
That will be interesting. I hope you guys are willing to share your findings with us
Feb 16, 2009 at 08:26 AM
p.20 #145D2 low ISO poor v 1Ds3
I exchanged two emails with Chuck Westfall about this. I even pointed him to the threads on here about this. He chalked it up to the RAW converter:
"This sounds like more of an issue with the Adobe Camera Raw software, because there are no such issues when raw images from either the 1Ds Mark III or the 5D Mark II are processed in Canon's Digital Photo Professional software.
Thanks for reading Tech Tips!"
I then wrote him back and told him there was a gentleman having bad issues using the DPP converter. His response:
"I am not suggesting that you simply give up on Photoshop in favor of DPP. Far from it! There are obviously multiple benefits to a Photoshop workflow that will never be matched by DPP or any other image editing program. On the other hand, it's been a reasonably well known fact since day 1 of Adobe Camera Raw that it does not render shadows from EOS Digital SLR raw images as cleanly as DPP. You can check on that with anybody who's used both programs. There is no doubt that Adobe has made great improvements to ACR over time, and by the same token so has Canon with DPP. Please understand that I am not saying that shadows in 1Ds Mark III or 5D Mark II raw images converted in DPP are completely clean. However, they are cleaner in the shadows than the same images processed in ACR 5.2/CS4. So, ultimately, what I am suggesting to you is simply this: convert the same RAW image with both apps and see for yourself. Once you've rendered a TIFF or a JPEG, then by all means continue your editing in Photoshop to get all the benefits deriving from that software.
For what it's worth, Adobe has always had access to Canon's raw image processing algorithms via the Canon Software Development Kits. However, they've chosen to go their own way, which is their prerogative, and completely understandable. Canon is in no position to tell Adobe how to design their software, just as Adobe is in no position to dictate to Canon. But the net result is that each type of raw conversion software ends up with its own look. The best I can suggest is that you try both and see which one you prefer. "
"This sounds like more of an issue with the Adobe Camera Raw software, because there are no such issues when raw images from either the 1Ds Mark III or the 5D Mark II are processed in Canon's Digital Photo Professional software.
Thanks for reading Tech Tips!"
I then wrote him back and told him there was a gentleman having bad issues using the DPP converter. His response:
"I am not suggesting that you simply give up on Photoshop in favor of DPP. Far from it! There are obviously multiple benefits to a Photoshop workflow that will never be matched by DPP or any other image editing program. On the other hand, it's been a reasonably well known fact since day 1 of Adobe Camera Raw that it does not render shadows from EOS Digital SLR raw images as cleanly as DPP. You can check on that with anybody who's used both programs. There is no doubt that Adobe has made great improvements to ACR over time, and by the same token so has Canon with DPP. Please understand that I am not saying that shadows in 1Ds Mark III or 5D Mark II raw images converted in DPP are completely clean. However, they are cleaner in the shadows than the same images processed in ACR 5.2/CS4. So, ultimately, what I am suggesting to you is simply this: convert the same RAW image with both apps and see for yourself. Once you've rendered a TIFF or a JPEG, then by all means continue your editing in Photoshop to get all the benefits deriving from that software.
For what it's worth, Adobe has always had access to Canon's raw image processing algorithms via the Canon Software Development Kits. However, they've chosen to go their own way, which is their prerogative, and completely understandable. Canon is in no position to tell Adobe how to design their software, just as Adobe is in no position to dictate to Canon. But the net result is that each type of raw conversion software ends up with its own look. The best I can suggest is that you try both and see which one you prefer. "
Feb 16, 2009 at 11:51 AM
p.20 #155D2 low ISO poor v 1Ds3
FWIW I didn't see much difference between DPP and ACR with regards to shadow noise at lower ISO's.
Feb 16, 2009 at 12:18 PM
p.20 #165D2 low ISO poor v 1Ds3
When I processed a 5DII RAW image in a discussion on another thread ( 5DII vs D3X) the DPP processed file showed less noise in the deep shadows than either ACR or C1 - this was with all noise reduction turned off, no sharpening, and exposure sliders matched to +2 stops to see the worst that could be done to the shadows.
Feb 16, 2009 at 12:44 PM
Andi Dietrich
Offline
[X]
Details
p.20 #175D2 low ISO poor v 1Ds3
brainiac wrote:
http://cyberphotographer.com/5d2/alex5d2posterised_crop.jpg
With my 5D2 it's trash.
Not so fast, you can still get out a lot of this shot
http://www.andidietrich.com/elements/fm3/U_1.jpg
Sorry, I cant belive it is that much worse than other Canon cameras, I would have to see this side by side
http://cyberphotographer.com/5d2/alex5d2posterised_crop.jpg
With my 5D2 it's trash.
Not so fast, you can still get out a lot of this shot
http://www.andidietrich.com/elements/fm3/U_1.jpg
Sorry, I cant belive it is that much worse than other Canon cameras, I would have to see this side by side
Feb 16, 2009 at 01:25 PM
p.20 #185D2 low ISO poor v 1Ds3
Andi, I'm not sure I understand your post. Is that second shot your version of Brainiac's shot? The noise is horrible in both, but worse in yours.
Feb 16, 2009 at 01:46 PM
p.20 #195D2 low ISO poor v 1Ds3
shirozina wrote:
When I processed a 5DII RAW image in a discussion on another thread ( 5DII vs D3X) the DPP processed file showed less noise in the deep shadows than either ACR or C1 - this was with all noise reduction turned off, no sharpening, and exposure sliders matched to +2 stops to see the worst that could be done to the shadows.
Ok, I did some further comparing at ISO400...
When pushing the exposure two stops, DPP gives slightly less noise in the shadows than ACR at the expense of detail when all NR/sharpening are off.
But I have a feeling that DPP always applies some NR, even when you have all the NR set to off. Maybe this explains why DPP files look somewhat "softer" than ACR files with NR/sharpening set to off.
When tweaking the NR a bit in both converters I get very similair results. So, in pratical terms there is not much difference IMO
BTW To match ACR rendering to DPP rendering I had to tone down the "blacks" slider to 1 or 0. When you leave the ACR "blacks" slider to its default position of 5 it makes matters much worse regarding shadow noise.
When I processed a 5DII RAW image in a discussion on another thread ( 5DII vs D3X) the DPP processed file showed less noise in the deep shadows than either ACR or C1 - this was with all noise reduction turned off, no sharpening, and exposure sliders matched to +2 stops to see the worst that could be done to the shadows.
Ok, I did some further comparing at ISO400...
When pushing the exposure two stops, DPP gives slightly less noise in the shadows than ACR at the expense of detail when all NR/sharpening are off.
But I have a feeling that DPP always applies some NR, even when you have all the NR set to off. Maybe this explains why DPP files look somewhat "softer" than ACR files with NR/sharpening set to off.
When tweaking the NR a bit in both converters I get very similair results. So, in pratical terms there is not much difference IMO
BTW To match ACR rendering to DPP rendering I had to tone down the "blacks" slider to 1 or 0. When you leave the ACR "blacks" slider to its default position of 5 it makes matters much worse regarding shadow noise.
Feb 16, 2009 at 02:00 PM
p.20 #205D2 low ISO poor v 1Ds3
cineski wrote:
Andi, I'm not sure I understand your post. Is that second shot your version of Brainiac's shot? The noise is horrible in both, but worse in yours.
What he is saying is that he cannot imagine any other Canon body doing any better when pushing exposure in PP--And until I see a side by side between the 1DsIII and the 5DII of the same exact shot using the same settings and under the same conditions I'd have to agree. All this crap about 'it looks worse than when I had a 1DsIII' is garbage...lets see some side by sides using multiple samples to make legitimate comparisons. Regardless, if the $2700 5DII under performs the $6500 1DsIII when it comes to shadow noise its not going to be by much and for the most part is negligible--specifically when speaking of prints
Andi, I'm not sure I understand your post. Is that second shot your version of Brainiac's shot? The noise is horrible in both, but worse in yours.
What he is saying is that he cannot imagine any other Canon body doing any better when pushing exposure in PP--And until I see a side by side between the 1DsIII and the 5DII of the same exact shot using the same settings and under the same conditions I'd have to agree. All this crap about 'it looks worse than when I had a 1DsIII' is garbage...lets see some side by sides using multiple samples to make legitimate comparisons. Regardless, if the $2700 5DII under performs the $6500 1DsIII when it comes to shadow noise its not going to be by much and for the most part is negligible--specifically when speaking of prints
Feb 16, 2009 at 02:13 PM