This topic is in the 2009 archive. You can read it and like its posts here. It is read only.
Archive 2009Canon EOS 7D Master thread
p.28 #1Canon EOS 7D Master thread
michael49 wrote:
http://fakechuckwestfall.files.wordpress.com/2009/01/galnikon.jpg
.........................
I agree. The moment you take advertising dollars from a company whose products you review, your 'non-bias' view goes right out the door, no matter how hard you try.
droopy1592 wrote
Yep, no bias or agenda whatsoeverhttp://fakechuckwestfall.files.wordpress.com/2009/01/galnikon.jpg
.........................I agree. The moment you take advertising dollars from a company whose products you review, your 'non-bias' view goes right out the door, no matter how hard you try.
Sep 04, 2009 at 08:38 PM
p.28 #2Canon EOS 7D Master thread
As far as AF of the 7D is concerned, I'm very surprised and puzzled at RG's choice of AI servo settings. Just posted this at DPR:
http://forums.dpreview.com/forums/read.asp?forum=1019&message=32924302
http://forums.dpreview.com/forums/read.asp?forum=1019&message=32924302
Sep 04, 2009 at 08:53 PM
p.28 #3Canon EOS 7D Master thread
Hmm. Don't know if RG has any agenda.
Despite his criticism of the 1DIII, it's his camera of choice (or at least was).
Seems like his PP workflow just sucks
Anyway. If you guys are in a pixel-peeping mood, check out this 40D vs 7D comparison. ISO-800/NR-off:
http://www.imaging-resource.com/PRODS/E40D/E40DLL0807XNR.HTM
http://www.imaging-resource.com/PRODS/E7D/E7DLL008007XNR.HTM
This test seems to confirm my observations from yesterday: pixel-level noise is the same on the 40D and the 7D. And considering the extra megapixels, this leaves rooms for better image level noise on the 7D.
I think when the dust settles it's going to turn out that the 7D sensor is the best 1.6x sensor Canon has made.
If you are a potential 7D buyer, though, you don't want to look at the 5DII files .
Despite his criticism of the 1DIII, it's his camera of choice (or at least was).
Seems like his PP workflow just sucks
Anyway. If you guys are in a pixel-peeping mood, check out this 40D vs 7D comparison. ISO-800/NR-off:
http://www.imaging-resource.com/PRODS/E40D/E40DLL0807XNR.HTM
http://www.imaging-resource.com/PRODS/E7D/E7DLL008007XNR.HTM
This test seems to confirm my observations from yesterday: pixel-level noise is the same on the 40D and the 7D. And considering the extra megapixels, this leaves rooms for better image level noise on the 7D.
I think when the dust settles it's going to turn out that the 7D sensor is the best 1.6x sensor Canon has made.
If you are a potential 7D buyer, though, you don't want to look at the 5DII files
.Sep 04, 2009 at 08:54 PM
p.28 #4Canon EOS 7D Master thread
garyvot wrote:
I've worked with RG on a professional basis, and while we have not spoken in a couple of years, I do not for a minute believe that he has any kind of anti-Canon agenda. He's a professional journalist by training, and his articles have always been much better researched and more objective in tone than some other sites I could name.
The heart of man and the bottom of the sea are unfathomable. --African Proverb
Liquidstone wrote:
As far as AF of the 7D is concerned, I'm very surprised and puzzled at RG's choice of AI servo settings. Just posted this at DPR:
http://forums.dpreview.com/forums/read.asp?forum=1019&message=32924302
Pretty much confirms it.
I've worked with RG on a professional basis, and while we have not spoken in a couple of years, I do not for a minute believe that he has any kind of anti-Canon agenda. He's a professional journalist by training, and his articles have always been much better researched and more objective in tone than some other sites I could name.
The heart of man and the bottom of the sea are unfathomable. --African Proverb
Liquidstone wrote:
As far as AF of the 7D is concerned, I'm very surprised and puzzled at RG's choice of AI servo settings. Just posted this at DPR:
http://forums.dpreview.com/forums/read.asp?forum=1019&message=32924302
Pretty much confirms it.
Sep 04, 2009 at 09:03 PM
p.28 #5Canon EOS 7D Master thread
I think when the dust settles it's going to turn out that the 7D sensor is the best 1.6x sensor Canon has made.
If you are a potential 7D buyer, though, you don't want to look at the 5DII files
.I just hope that sensor, or a slightly improved version comes to the 60D. I don't need all of the features, may try video a few times, but I just get tired of processing all of my images for pattern noise on the 50D
Sep 04, 2009 at 09:26 PM
p.28 #6Canon EOS 7D Master thread
For me my 5DII, 5D and a 7D would be a nice field 3sum. I do look forward to more tests and actual hands on use.
Sep 04, 2009 at 09:42 PM
p.28 #7Canon EOS 7D Master thread
Nowhere Man wrote:
.........................I agree. The moment you take advertising dollars from a company whose products you review, your 'non-bias' view goes right out the door, no matter how hard you try.
Last time I checked, RG doesn't do reviews on his site, and his advertising policy is posted publicly for all to read. I'm pretty sure he would accept Canon's advertising dollars if they were to offer them.
BTW, that image you linked to is hosted at the FakeChuckWestfall site. Not exactly the paragon of unbiased reporting, now is it?
I don't really need to defend RG, so this will be my last comment on this.
michael49 wrote
droopy1592 wrote
Yep, no bias or agenda whatsoever .........................Last time I checked, RG doesn't do reviews on his site, and his advertising policy is posted publicly for all to read. I'm pretty sure he would accept Canon's advertising dollars if they were to offer them.
BTW, that image you linked to is hosted at the FakeChuckWestfall site. Not exactly the paragon of unbiased reporting, now is it?
I don't really need to defend RG, so this will be my last comment on this.
Sep 04, 2009 at 09:45 PM
p.28 #8Canon EOS 7D Master thread
garyvot wrote:
.........................I agree. The moment you take advertising dollars from a company whose products you review, your 'non-bias' view goes right out the door, no matter how hard you try.Last time I checked, RG doesn't do reviews on his site, and his advertising policy is posted publicly for all to read. I'm pretty sure he would accept Canon's advertising dollars if they were to offer them.
BTW, that image you linked to is hosted at the FakeChuckWestfall site. Not exactly the paragon of unbiased reporting, now is it?
I don't really need to defend RG, so this will be my last comment on this.
When RG was going over the 1dmkiii issues and when 50D and 5Dmkii were released his page looked exactly the same. I remember seeing that very same photo on FCW site and not believing, then going to check myself. FCW may be a satire site, but that snapshot is 100% honest. For about a month or two that site was covered in Nikon ads, just like that photo. There is no exxaggerating about that one. I think the snapshot caused RG to rethink his "unbiased" stance when balancing ads (brand) and ad revenue as to not lose readers. Regardless of whether or not he does reviews, he's is definitely influential.
Nowhere Man wrote
michael49 wrote
droopy1592 wrote
Yep, no bias or agenda whatsoever .........................BTW, that image you linked to is hosted at the FakeChuckWestfall site. Not exactly the paragon of unbiased reporting, now is it?
I don't really need to defend RG, so this will be my last comment on this.
When RG was going over the 1dmkiii issues and when 50D and 5Dmkii were released his page looked exactly the same. I remember seeing that very same photo on FCW site and not believing, then going to check myself. FCW may be a satire site, but that snapshot is 100% honest. For about a month or two that site was covered in Nikon ads, just like that photo. There is no exxaggerating about that one. I think the snapshot caused RG to rethink his "unbiased" stance when balancing ads (brand) and ad revenue as to not lose readers. Regardless of whether or not he does reviews, he's is definitely influential.
Sep 04, 2009 at 10:10 PM
p.28 #9Canon EOS 7D Master thread
jorkata wrote:
If you are a potential 7D buyer, though, you don't want to look at the 5DII files .
Why not? Because then you might feel really smug about the lack of banding in the 7D files vs. the 5DII files?
If you are a potential 7D buyer, though, you don't want to look at the 5DII files
.Why not? Because then you might feel really smug about the lack of banding in the 7D files vs. the 5DII files?
Sep 04, 2009 at 10:20 PM
p.28 #10Canon EOS 7D Master thread
droopy1592 wrote:
.........................I agree. The moment you take advertising dollars from a company whose products you review, your 'non-bias' view goes right out the door, no matter how hard you try.Last time I checked, RG doesn't do reviews on his site, and his advertising policy is posted publicly for all to read. I'm pretty sure he would accept Canon's advertising dollars if they were to offer them.
BTW, that image you linked to is hosted at the FakeChuckWestfall site. Not exactly the paragon of unbiased reporting, now is it?
I don't really need to defend RG, so this will be my last comment on this.When RG was going over the 1dmkiii issues and when 50D and 5Dmkii were released his page looked exactly the same. I remember seeing that very same photo on FCW site and not believing, then going to check myself. FCW may be a satire site, but that snapshot is 100% honest. For about a month or two that site was covered in Nikon ads, just like that photo. There is no exxaggerating about that one. I think the snapshot caused RG to rethink his "unbiased" stance when balancing ads (brand) and ad revenue as to not lose readers. Regardless of...Show more →
I wasn't putting down RG, but merely stating that I don't believe you can run a business that is 100% non-biased, when you have manufacturers advertise like that. I too remember seeing Nikon ads all over his website.
I agree in that no matter what, he has some influence on things, whether it be people that follow his site, or the camera manufacturers themselves that read his reviews and follow his procedures for testing cameras.
garyvot wrote
Nowhere Man wrote
michael49 wrote
droopy1592 wrote
Yep, no bias or agenda whatsoever .........................BTW, that image you linked to is hosted at the FakeChuckWestfall site. Not exactly the paragon of unbiased reporting, now is it?
I don't really need to defend RG, so this will be my last comment on this.
I wasn't putting down RG, but merely stating that I don't believe you can run a business that is 100% non-biased, when you have manufacturers advertise like that. I too remember seeing Nikon ads all over his website.
I agree in that no matter what, he has some influence on things, whether it be people that follow his site, or the camera manufacturers themselves that read his reviews and follow his procedures for testing cameras.
Sep 04, 2009 at 10:25 PM
p.28 #11Canon EOS 7D Master thread
I think it's simply stating the obvious - the man accepts advertizing from Nikon. How does that affect his reviews of Canon products? It's hard to imagine it encourages impartialty. Ultimately, nobody knows but he. Nevertherless, it's a legitimate concern.
Sep 04, 2009 at 11:23 PM
p.28 #12Canon EOS 7D Master thread
Yohan Pamudji wrote:
.Why not? Because then you might feel really smug about the lack of banding in the 7D files vs. the 5DII files?
5D2 files aren't the holy grail of IQ. Sure, they have 21MP resolution (because that is what counts right?)... but they also have banding at ISO3200 (and probably even at lower ISO's too) 
jorkata wrote
If you are a potential 7D buyer, though, you don't want to look at the 5DII files .5D2 files aren't the holy grail of IQ. Sure, they have 21MP resolution (because that is what counts right?
)... but they also have banding at ISO3200 (and probably even at lower ISO's too) Sep 05, 2009 at 01:42 AM
p.28 #13Canon EOS 7D Master thread
garyvot wrote:
From which samples, RG's? I don't think Rob's RAW workflow is tuned to minimize noise, as I get better results with my 5D2s. A better comparison may be the Imaging-Resource samples. While it's hard to judge definitively from in-camera JPEGs, it seems that the 7D is an improvement over the 50D. When you factor in the high resolution vs. cameras like the 1D3, noise control seems surprisingly good.
I don't think anyone should expect this camera to be a noise king in absolute terms, but within the range of APS-C cameras, it looks promising. If these samples hold up, good...Show more →
I also spent some time comparing the 7D and 50D samples on imaging-resource. I've had the 50D pretty much as long as it's been out and I'd say that the 50D samples are a pretty fair representation of it's noise levels. On a pure noise level I don't see any significant difference between the 7D and 50D. This was also my impression from the 40D (which I've used before, but not extensively) to the 50D... so, in my opinion, Canon has pretty much pulled off the same trick: increase the MP keep noise *about* the same. Personally, I'd rather they leave the resolution be and improve the noise, but that's not the point I'm after. So, FWIW, from a 50D user, noise is no worse then the 50D... but that's not what prompted my long winded post tonight....
Maybe this has been covered already (though I didn't see it) but for me, the huge striking difference (on the imaging-resource site) between the 7D and the 50D samples was the NR muddiness/smearing... it's significantly reduced (almost gone!) in the 7D shots. If you compare the fabric swatches (or whatever they are) on the upper left of the sample shots (even at low ISO), the 50D looks smeary by comparison. The patterns in the fabric are crisp and well defined in the 7D shots, and decidedly not so in the 50D shots.
It's so striking I thought that maybe they switched lenses or JPEG compression or something. Digging through exif it appears that everything that can be the same is the same. Did I miss something?
Here's the crazier thing... it's better then the 5Dm2! Of course, the 5Dm2 beats the 7D handily in terms of noise (again, even at low ISO). But the 5Dm2, to my eye, shows much more smearing. Some review said that canon had dropped the default saturation levels on the 7D (and you can see in the comparison that the 7D is less saturated then the 5Dm2 or 50D). Could this be it?
I don't think so... When I compared to the 1dm3 (bold comparison, I know), I see that (finally) the NR smearing I'm seeing in the 5dm2 and 50D is gone, but the 1dm3 is clearly more saturated then the 7D shot.
So what gives? Is it a change in NR / in-camera processing which is similar to (or at least emulating) the look from the 1 series? Am I seeing things? Is it too late for a post like this?
Andrew
musclepics wrote
From those samples, the 7D looks to be about 1 stop nosier than what I get from my 1D MkIII at higher ISOs. In fact the ISO800 sample pic looks like what I get at ISO3200 from the 1D MkIII, so about 2 stops behind there (yes even when resized to the same resolution).I don't think anyone should expect this camera to be a noise king in absolute terms, but within the range of APS-C cameras, it looks promising. If these samples hold up, good...Show more →
I also spent some time comparing the 7D and 50D samples on imaging-resource. I've had the 50D pretty much as long as it's been out and I'd say that the 50D samples are a pretty fair representation of it's noise levels. On a pure noise level I don't see any significant difference between the 7D and 50D. This was also my impression from the 40D (which I've used before, but not extensively) to the 50D... so, in my opinion, Canon has pretty much pulled off the same trick: increase the MP keep noise *about* the same. Personally, I'd rather they leave the resolution be and improve the noise, but that's not the point I'm after. So, FWIW, from a 50D user, noise is no worse then the 50D... but that's not what prompted my long winded post tonight....
Maybe this has been covered already (though I didn't see it) but for me, the huge striking difference (on the imaging-resource site) between the 7D and the 50D samples was the NR muddiness/smearing... it's significantly reduced (almost gone!) in the 7D shots. If you compare the fabric swatches (or whatever they are) on the upper left of the sample shots (even at low ISO), the 50D looks smeary by comparison. The patterns in the fabric are crisp and well defined in the 7D shots, and decidedly not so in the 50D shots.
It's so striking I thought that maybe they switched lenses or JPEG compression or something. Digging through exif it appears that everything that can be the same is the same. Did I miss something?
Here's the crazier thing... it's better then the 5Dm2! Of course, the 5Dm2 beats the 7D handily in terms of noise (again, even at low ISO). But the 5Dm2, to my eye, shows much more smearing. Some review said that canon had dropped the default saturation levels on the 7D (and you can see in the comparison that the 7D is less saturated then the 5Dm2 or 50D). Could this be it?
I don't think so... When I compared to the 1dm3 (bold comparison, I know), I see that (finally) the NR smearing I'm seeing in the 5dm2 and 50D is gone, but the 1dm3 is clearly more saturated then the 7D shot.
So what gives? Is it a change in NR / in-camera processing which is similar to (or at least emulating) the look from the 1 series? Am I seeing things? Is it too late for a post like this?
Andrew
Sep 05, 2009 at 01:50 AM
p.28 #14Canon EOS 7D Master thread
Andrew Krend wrote:
Am I seeing things?
You are seeing/comparing in-camera JPEG engines at work. Fine if you shoot JPEG's, not relevant if you shoot only RAW.
I would like to see 5D2, 50D and 7D RAW files framed exactly the same and with the same exposure (shoot at multiple ISO settings)... Imaging Resource comes to mind
Am I seeing things?
You are seeing/comparing in-camera JPEG engines at work. Fine if you shoot JPEG's, not relevant if you shoot only RAW.
I would like to see 5D2, 50D and 7D RAW files framed exactly the same and with the same exposure (shoot at multiple ISO settings)... Imaging Resource comes to mind
Sep 05, 2009 at 01:58 AM
brainiac
Offline
[X]
Details
p.28 #15Canon EOS 7D Master thread
Tom_W wrote:
Perhaps an alternative testing/comparison method could be shown to Phil Askey. His tests, while not perfect, have evolved in a mostly positive way since he began the web site.
I sent dpr a very polite email about this issue, detailing why I thought their noise tests were thoroughly misleading to the average photographer. I never received an acknowledgement, let alone a response. I believe that someone at DPR is a Nikon loyalist and so is quite happy with their noise comparison, since it biases noise comparisons heavily in favour of 12 Mpixels (D3, D700) against 21 (1Ds3, 5D2).
Perhaps an alternative testing/comparison method could be shown to Phil Askey. His tests, while not perfect, have evolved in a mostly positive way since he began the web site.
I sent dpr a very polite email about this issue, detailing why I thought their noise tests were thoroughly misleading to the average photographer. I never received an acknowledgement, let alone a response. I believe that someone at DPR is a Nikon loyalist and so is quite happy with their noise comparison, since it biases noise comparisons heavily in favour of 12 Mpixels (D3, D700) against 21 (1Ds3, 5D2).
Sep 05, 2009 at 04:11 AM
brainiac
Offline
[X]
Details
p.28 #16Canon EOS 7D Master thread
Imagemaster wrote:
...as long as those subjects are in the central 1/3rd of the frame.Well yeah, that is the general idea when tracking fast-moving subjects. You have no time to worry about composition when trying to track a fast-flying bird...
Nonsense, composition is exactly what you are worrying about when you ensure that your central AF sensors are over the bird; the problem is that central composition is the only one the camera allows. What is more, for most photographers tracking small and fast-moving objects through the sky is very unlike typical shooting conditions. Most photographers shoot a range of subjects, and in more typical conditions the best AF systems are considerably worse at doing their job than they are at BIF. Current AF systems are heavily biased towards BIF/sport and thoroughly obstructive when it comes to general photography. Fortunately Panasonic, with its nearly adequate contrast-detect AF, and Sony with its tap-here-to-focus technology look set to bring AF forward to the 19th century.
...nor do you have the skill to MF on such a subject.
Sure, bif and sport photography did not exist before AF was invented in the 80's.
brainiac wrote
Imagemaster wrote
I would bet that any DSLR's AF would be better than your MF on fast-moving subjects.Nonsense, composition is exactly what you are worrying about when you ensure that your central AF sensors are over the bird; the problem is that central composition is the only one the camera allows. What is more, for most photographers tracking small and fast-moving objects through the sky is very unlike typical shooting conditions. Most photographers shoot a range of subjects, and in more typical conditions the best AF systems are considerably worse at doing their job than they are at BIF. Current AF systems are heavily biased towards BIF/sport and thoroughly obstructive when it comes to general photography. Fortunately Panasonic, with its nearly adequate contrast-detect AF, and Sony with its tap-here-to-focus technology look set to bring AF forward to the 19th century.
...nor do you have the skill to MF on such a subject.
Sure, bif and sport photography did not exist before AF was invented in the 80's.
Sep 05, 2009 at 04:19 AM
brainiac
Offline
[X]
Details
p.28 #17Canon EOS 7D Master thread
thw2 wrote:
Actually DXOMark takes the opposite stand of DPReview with regards to pixel density. You can read it
here:
"Contrary to conventional wisdom, higher resolution actually compensates for noise"
That's all very well, although false, but the problem at DXO is that they load the per pixel comparisons by default. Very few people seem to understand why an 18Mp sensor can give you less noisy pictures EVEN WHEN it has more per pixel noise than a 10Mp sensor. That is why I think DXO is just as misleading as DPR. By default they should load the per square inch of print result, instead of the per pixel result. Then those blissfully ignorant of this issue wouldn't be misled.
The DXO statement that 'higher resolution actually compensates for noise' is ambiguous too. People could take that to mean that higher resolution cameras have an advantage, which is just as untrue as the widely believed falsity that lower resolution cameras have and advantage. The real point is that pictorial noise levels are effectively independent of pixel pitch in available cameras, and that noise depends enormously more on sensor size, age, and type than on pitch.
I would prefer DXO to state that 'noise is largely independent of resolution, and mostly dependent on sensor size and age'. That would go some way toward dispelling the tired falsity that appears on almost every page of this and other discussion sites.
brainiac wrote
Don't worry - DPReview and DXO will do a great job of showing everyone how the 7D's 18mp noise is worse than other cameras even though it won't be.here:
"Contrary to conventional wisdom, higher resolution actually compensates for noise"
That's all very well, although false, but the problem at DXO is that they load the per pixel comparisons by default. Very few people seem to understand why an 18Mp sensor can give you less noisy pictures EVEN WHEN it has more per pixel noise than a 10Mp sensor. That is why I think DXO is just as misleading as DPR. By default they should load the per square inch of print result, instead of the per pixel result. Then those blissfully ignorant of this issue wouldn't be misled.
The DXO statement that 'higher resolution actually compensates for noise' is ambiguous too. People could take that to mean that higher resolution cameras have an advantage, which is just as untrue as the widely believed falsity that lower resolution cameras have and advantage. The real point is that pictorial noise levels are effectively independent of pixel pitch in available cameras, and that noise depends enormously more on sensor size, age, and type than on pitch.
I would prefer DXO to state that 'noise is largely independent of resolution, and mostly dependent on sensor size and age'. That would go some way toward dispelling the tired falsity that appears on almost every page of this and other discussion sites.
Sep 05, 2009 at 04:36 AM
p.28 #18Canon EOS 7D Master thread
brainiac wrote:
Nonsense, composition is exactly what you are worrying about when you ensure that your central AF sensors are over the bird; the problem is that central composition is the only one the camera allows.
Sorry Richard, composition (in an aesthetic sense) is the last on your mind when you're trying to keep the AF point on a BIF - if you've got the time to think abut comp when shooting a BIF, you're not shooting anything that would challenge the AF in the first place.
Sure, bif and sport photography did not exist before AF was invented in the 80's.
BIF photography of sorts existed (typically pre-focused remotely/automatically triggered high speed flash set-ups - Eric Hosking et al), but not as you see it today. Even average BIF images you can see these days would simply not have been possible then.
Nonsense, composition is exactly what you are worrying about when you ensure that your central AF sensors are over the bird; the problem is that central composition is the only one the camera allows.
Sorry Richard, composition (in an aesthetic sense) is the last on your mind when you're trying to keep the AF point on a BIF - if you've got the time to think abut comp when shooting a BIF, you're not shooting anything that would challenge the AF in the first place.
Sure, bif and sport photography did not exist before AF was invented in the 80's.
BIF photography of sorts existed (typically pre-focused remotely/automatically triggered high speed flash set-ups - Eric Hosking et al), but not as you see it today. Even average BIF images you can see these days would simply not have been possible then.
Sep 05, 2009 at 05:00 AM
p.28 #19Canon EOS 7D Master thread
In AI servo and if using AFPS, do you still need to acquire focus first with the centre AF point, or can it acquire focus with any point(s)?
Sep 05, 2009 at 05:10 AM
p.28 #20Canon EOS 7D Master thread
brainiac wrote:
I sent dpr a very polite email about this issue, detailing why I thought their noise tests were thoroughly misleading to the average photographer. I never received an acknowledgement, let alone a response. I believe that someone at DPR is a Nikon loyalist and so is quite happy with their noise comparison, since it biases noise comparisons heavily in favour of 12 Mpixels (D3, D700) against 21 (1Ds3, 5D2).
How about the D3X, DPR must be biased against that too according to your theory?
Tom_W wrote
Perhaps an alternative testing/comparison method could be shown to Phil Askey. His tests, while not perfect, have evolved in a mostly positive way since he began the web site.How about the D3X, DPR must be biased against that too according to your theory?
Sep 05, 2009 at 05:16 AM