thedigitalbean wrote:
At the risk of repeating of myself, I'll say that comparing the RAW files, the 5D2 is definitely NOT 1 stop better than the 7D, I put the 5D2's advantage at between 1/3 to 2/3 stop. Furthermore, the 7D has visibly lower noise than the 50D at all ISOs 800 and higher.
Sounds too good to be true
Did you also notice that the 7D exposures were different, so we are effectively comparing the 7D at ISO640 to the 5DII at ISO800 here?
To be fair, we should subtract 1/3 stop from the 7D's noise performance, at similar shutter speeds.
alundeb wrote:
Sounds too good to be true
Did you also notice that the 7D exposures were different, so we are effectively comparing the 7D at ISO640 to the 5DII at ISO800 here?
To be fair, we should subtract 1/3 stop from the 7D's noise performance, at similar shutter speeds.
Yea just noticed that. That is somewhat disturbing, not sure what is going on there.
alundeb wrote:
Sounds too good to be true
Did you also notice that the 7D exposures were different, so we are effectively comparing the 7D at ISO640 to the 5DII at ISO800 here?
To be fair, we should subtract 1/3 stop from the 7D's noise performance, at similar shutter speeds.
As skibum points out one should be careful to distinguish shot noise from read noise. I'll admit that I've been looking at the deep shadow noise more than noise in the highlights. I think the 5D2's shot noise advantage is closer to the 2/3 stop mark and maybe even a stop. However the read noise difference is what is considerably less.
From my estimates of the system gain, the 7D has considerably larger full well capacity than the 50D, even per pixel. All other things being equal, this would show up as native ISO 80 capability. Canon may have seen two benefits form adjusting away from the true ISO sensitivity:
1) They didn't have to make an ISO "L" expansion to allow for this extra light to be captured.
2) It does give very pretty noise characteristics in non-scientific tests, without sacrificing DR at nominal ISO 100.
thedigitalbean wrote:
As skibum points out one should be careful to distinguish shot noise from read noise. I'll admit that I've been looking at the deep shadow noise more than noise in the highlights. I think the 5D2's shot noise advantage is closer to the 2/3 stop mark and maybe even a stop. However the read noise difference is what is considerably less.
This is music to my ears, as I do care about shadow noise
alundeb wrote:
Sounds too good to be true
Did you also notice that the 7D exposures were different, so we are effectively comparing the 7D at ISO640 to the 5DII at ISO800 here?
To be fair, we should subtract 1/3 stop from the 7D's noise performance, at similar shutter speeds.
I noticed similar when comparing the 5D2 to the original 5D - the 5D Classic exposes slightly brighter. Maybe 1/6 to 1/3 stop brighter for the same exact settings indicating that it is slightly more sensitive at a given ISO.
JohnR84740 wrote:
I too am impressed by the specs of the 7D, and will probably pick one up. But the thing that is so ridiculous is to see the reviews that are already posted on retailers' websites. I mean, how is it possible for these folks to be posting reviews of something they have never even seen, let alone touched? Granted that there are a few who have test models, but lets get real!
thedigitalbean wrote:
I have been scrutnizing the files to try to find an answer to this myself but its hard to say. Comparing the 5D2 @ ISO 6400 vs. the 7D @ ISO 3200, the 7D is definitely cleaner, so the advantage to the 5D2 is definitely < 1 stop.
The difference becomes minimal at lower ISOs however. My guess would be the 5D2 is cleaner by 1/3 to 2/3 stops (maximum) depending on ISO. Most of the difference I see is in the highlights (which I find very odd). What is making the comparisons difficult for me is that the difference in color and contrast (which is most likely because the 7D hasn't been fully calibrated yet in my ACR beta, and they probably used different lenses). Whats interesting is comparing both at ISO 12800, though the 7D has more noise in the highlights, it has virtually no pattern noise, where as the 5D2 has distinct pattern noise....Show more →
Very interesting...so this puts it around the ISO performance of the 5D/1DIII at stops lower than 3200, possibly within 1/3-1/4 stop(?). I'm really looking forward to seeing production model performance in conjunction with fully compatible PP software.
I too am concerned about the difference between camera specified ISO and the "real" iso based on standards. All of the iso tests are suspect unless actual iso sensitivity is accounted for. I have a 7d on order and this is only issue see remaining to get resolved in my mind (or I cancel the order). If you look at DXOmark iso sensitivities for various Canon cameras, there seems to be trend for recent bodies to actually be well under ISO standards at higher ISO's.
john660 wrote:
I'm interested to know who's planning on buying this 7D right away... or are most thinking they'll wait to see how it actually performs like I am?
I'm just waiting for good things about the AF.
Pretty much all the examples images I've seen so far haven't really impressed from an AF point of view, but Romy Ocon is doing some birding tests (as we speak, I think) and they'll be influential (Romy, stick a taped-pins 1.4x TC and a 100-400mm on there at some point, please!!)
Assuming that Romy's testing is as surprising - and in the same way! - as the IQ tests the guys here have been sharing, then it's going to be a hard camera to resist.
alundeb wrote:
To be fair, we should subtract 1/3 stop from the 7D's noise performance, at similar shutter speeds.
Not suggesting that this is what's going on here, but body to body variations between the same models can be just as big/small [delete as necessary].
My 40D is - supposedly - aligned to the CIPA standard, and I run it at + 0.3 EC as a matter of course, as a starting point: one of my 30Ds is on 0 EC to give the same shutter speed at a given exposure and aperture, and the other 30D needs to be 0.3 under for the same shutter speed.
So there's a third of a stop between the 30Ds, but it doesn't really hurt.
Having had another look at IR, it seems that the 7D is exposing like the 40D.
Rick1000 wrote:
I too am concerned about the difference between camera specified ISO and the "real" iso based on standards. All of the iso tests are suspect unless actual iso sensitivity is accounted for. I have a 7d on order and this is only issue see remaining to get resolved in my mind (or I cancel the order). If you look at DXOmark iso sensitivities for various Canon cameras, there seems to be trend for recent bodies to actually be well under ISO standards at higher ISO's.
In the real world it won't make any difference. One half stop better/worse noise performance doesn't make any difference. An extra half stop of dynamic range doesn't matter.
Carry on.
Jeff Nolten wrote:
I found an Apple page that says the 2 GHz G5 Dual can play full HD at 24 fps if it has the better video card with 128 MB of ram. I have a dead G5 with that card so I will swap it and try. Interestingly, the Apple page says the Intel based Mac requires at least 3 GHz dual core to play. Knew there was a reason I preferred Power PC.
PowerPC's suck compare to the new intels, and the new intels suck compared to the new PC's with faster video cards for less money
alundeb wrote:
Sounds too good to be true
Did you also notice that the 7D exposures were different, so we are effectively comparing the 7D at ISO640 to the 5DII at ISO800 here?
To be fair, we should subtract 1/3 stop from the 7D's noise performance, at similar shutter speeds.
I would say from image examples, the 5D2 is at LEAST 1 stop better than the 7D for high ISO noise. My 1DMkIII is about about 1-2 stops better than the 7D, and about 1/2 stop better than the 5D2. I've posted a few comparisons, and can post more. ISO3200 coming from my MkIII is much cleaner than ISO1600 from the 7D (even when the 7D has be resized to 10MP). The 5D2 is much closer to the MkIII than the 7D.
musclepics wrote:
I would say from image examples, the 5D2 is at LEAST 1 stop better than the 7D for high ISO noise. My 1DMkIII is about about 1-2 stops better than the 7D, and about 1/2 stop better than the 5D2. I've posted a few comparisons, and can post more.
That doesn't match other findings at either pixel level (where the differences seem too dramatic) or print level where there is no way the 1D3 with a much smaller sensor should do 1/2 stop better SNR than the 5D2.
dcmiller wrote:
In the real world it won't make any difference. One half stop better/worse noise performance doesn't make any difference. An extra half stop of dynamic range doesn't matter.
Carry on.
True, the real thing we need to focus on is whether the 7D AF is truly up to speed or not.
For someone just wanting a bit more than a 50D it doesn't matter and the 7D is surely good for everyone else I think the AF is really the make or break as to whether it would be a fine buy or second body or pointless.
musclepics wrote:
I would say from image examples, the 5D2 is at LEAST 1 stop better than the 7D for high ISO noise. My 1DMkIII is about about 1-2 stops better than the 7D, and about 1/2 stop better than the 5D2. I've posted a few comparisons, and can post more. ISO3200 coming from my MkIII is much cleaner than ISO1600 from the 7D (even when the 7D has be resized to 10MP). The 5D2 is much closer to the MkIII than the 7D.
