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.36 #1Canon EOS 7D Master thread
EB-1 wrote:
I assumed it was a Spoonerism. 
EBH
Yes, just a Spoonerism. Sorry to confuse. I had read a number of posts talking about the 7D's bells and whistles. But perhaps "bissel" could be a new kind of sensor cleaning system. (Bissell being a brand of vacuum.)
Liquidstone wrote
I've heard of bells and whistles, but since English is not my native language, I can't get what "wells and bissels" mean.EBH
Yes, just a Spoonerism. Sorry to confuse. I had read a number of posts talking about the 7D's bells and whistles. But perhaps "bissel" could be a new kind of sensor cleaning system. (Bissell being a brand of vacuum.)
Sep 08, 2009 at 08:30 AM
p.36 #2Canon EOS 7D Master thread
theSuede wrote:
Do also notice that the 7D cheats even more than the other cameras with the "ISO" specification - it needs 1/3-step longer shuttertimes at all ISO's to achieve the same exposure at the same aperture with the same lens as the 50D. The 50D is at the true value of ~ISO2350 when the camera display says "ISO3200". The 7D is roughly 1/3Ev lower than that - which would give a true ISO of ~2000 or lower when the camera says "3200".
Interesting observation.
The sensor is then actually capable of ISO 80, but Canon chose to adjust exposure instead of enabling ISO extension to "L".
Do also notice that the 7D cheats even more than the other cameras with the "ISO" specification - it needs 1/3-step longer shuttertimes at all ISO's to achieve the same exposure at the same aperture with the same lens as the 50D. The 50D is at the true value of ~ISO2350 when the camera display says "ISO3200". The 7D is roughly 1/3Ev lower than that - which would give a true ISO of ~2000 or lower when the camera says "3200".
Interesting observation.
The sensor is then actually capable of ISO 80, but Canon chose to adjust exposure instead of enabling ISO extension to "L".
Sep 08, 2009 at 08:31 AM
p.36 #3Canon EOS 7D Master thread
alundeb wrote:
Interesting observation.
The sensor is then actually capable of ISO 80, but Canon chose to adjust exposure instead of enabling ISO extension to "L".
Actually, this is great. Should make it behave exactly like my 5D2s.
theSuede wrote
Do also notice that the 7D cheats even more than the other cameras with the "ISO" specification - it needs 1/3-step longer shuttertimes at all ISO's to achieve the same exposure at the same aperture with the same lens as the 50D. The 50D is at the true value of ~ISO2350 when the camera display says "ISO3200". The 7D is roughly 1/3Ev lower than that - which would give a true ISO of ~2000 or lower when the camera says "3200".The sensor is then actually capable of ISO 80, but Canon chose to adjust exposure instead of enabling ISO extension to "L".
Actually, this is great. Should make it behave exactly like my 5D2s.
Sep 08, 2009 at 09:38 AM
p.36 #4Canon EOS 7D Master thread
theSuede wrote:
Anyway, here's the comparison. All pictures scaled to the same relative scale (40D and Nikon D5000 scaled up, 50D and 7D scaled down). I chose a comparison resolution somewhere in between the 50D and the D5000.
Nikon D5000, Canon 40/50/7D (1Meg picture, 2160x1427)
Yeah, I've done similar comparos of the IR samples at different ISOs and the 7D definitely appears best (though I think 3200 is a real stretch for any APS-C camera).
On a side note, what exactly does Nikon do to the luma channel on its cameras at high ISOs? Look at the edge transitions in the D5000 sample and note how they break up, then there's that "watercolor" look that you usually don't see in Canon cameras (though they often have their own problems).
When I had a pair of D300's I noticed that even in RAW captures there seemed to be something softened in the noise pattern. I've never heard anyone confirm that Nikon cooks its RAW files, but I've always wondered...
Anyway, here's the comparison. All pictures scaled to the same relative scale (40D and Nikon D5000 scaled up, 50D and 7D scaled down). I chose a comparison resolution somewhere in between the 50D and the D5000.
Nikon D5000, Canon 40/50/7D (1Meg picture, 2160x1427)
Yeah, I've done similar comparos of the IR samples at different ISOs and the 7D definitely appears best (though I think 3200 is a real stretch for any APS-C camera).
On a side note, what exactly does Nikon do to the luma channel on its cameras at high ISOs? Look at the edge transitions in the D5000 sample and note how they break up, then there's that "watercolor" look that you usually don't see in Canon cameras (though they often have their own problems).
When I had a pair of D300's I noticed that even in RAW captures there seemed to be something softened in the noise pattern. I've never heard anyone confirm that Nikon cooks its RAW files, but I've always wondered...
Sep 08, 2009 at 09:57 AM
p.36 #5Canon EOS 7D Master thread
theSuede wrote:
Do also notice that the 7D cheats even more than the other cameras with the "ISO" specification - it needs 1/3-step longer shuttertimes at all ISO's to achieve the same exposure at the same aperture with the same lens as the 50D. The 50D is at the true value of ~ISO2350 when the camera display says "ISO3200". The 7D is roughly 1/3Ev lower than that - which would give a true ISO of ~2000 or lower when the camera says "3200".
I find that very odd, given how much Canon made of the fact that they had - with the 40D and Mk III - aligned their cameras' "sensitivity" ratings to the CIPA standard.
Do also notice that the 7D cheats even more than the other cameras with the "ISO" specification - it needs 1/3-step longer shuttertimes at all ISO's to achieve the same exposure at the same aperture with the same lens as the 50D. The 50D is at the true value of ~ISO2350 when the camera display says "ISO3200". The 7D is roughly 1/3Ev lower than that - which would give a true ISO of ~2000 or lower when the camera says "3200".
I find that very odd, given how much Canon made of the fact that they had - with the 40D and Mk III - aligned their cameras' "sensitivity" ratings to the CIPA standard.
Sep 08, 2009 at 10:00 AM
p.36 #6Canon EOS 7D Master thread
thw2 wrote:
Similarly, the 10 MP 40D (released in 2007) is better than 8 MP 30D (released in 2006).
I'm not sure about that. Never did side-by-side testing, having owned the two at different times. But I recall that my 30D had some latitude for pulling up shadows with less noise than on my 40D - especially at ISO 400 and up. You had to hit it right on the 40.
Similarly, the 10 MP 40D (released in 2007) is better than 8 MP 30D (released in 2006).
I'm not sure about that. Never did side-by-side testing, having owned the two at different times. But I recall that my 30D had some latitude for pulling up shadows with less noise than on my 40D - especially at ISO 400 and up. You had to hit it right on the 40.
Sep 08, 2009 at 10:30 AM
p.36 #7Canon EOS 7D Master thread
EOS20 wrote:
I've seen some Australian prices on the 7D and they are:
$2360 for the body only.
$3060 with the 18-135 IS lens.
$3300 with the 18-200 IS lens.
$3350 with the 15-85 IS lens.
Looks like it will be available early/mid October.
Big price jump on the lens kits. That body only price is actually more affordable than I thought it would be and I think you can get it for under $2,000 from a grey importer.
I've seen some Australian prices on the 7D and they are:
$2360 for the body only.
$3060 with the 18-135 IS lens.
$3300 with the 18-200 IS lens.
$3350 with the 15-85 IS lens.
Looks like it will be available early/mid October.
Big price jump on the lens kits. That body only price is actually more affordable than I thought it would be and I think you can get it for under $2,000 from a grey importer.
Sep 08, 2009 at 11:11 AM
p.36 #8Canon EOS 7D Master thread
L. H. Smith wrote:
The guy has a point. You apparently don't like the fact that he raised it(and others).
It's not that he raised it.
It's that he raised again.
And again.
And again.
And again.
And again.
And again.
And again.
And - well you get the picture, I imagine.
And no amount of reasoning would shake him from his view that the new focusing screen in the 7D was nothing short of a sin against God.
Believe me, it does get tired.
I have not read all of the nearly 30 pages of posts in this thread
Perhaps if you had, you'd understand something of the provocation that elicits such responses...
Maybe someone will even have demonstrated that they have what it takes to apologize for a demonstration of badgering.
Why? There's a clear cause and effect here, and it's the cause that could have realised long since that enough was enough. It's not as if this was the only thread where we got to hear - ad nauseam - about how stupid Canon were, and how ignorant they were of the needs of its customer base by - possibly - having made MF harder to do.
In fact perhaps you might consider apologising for so readily jumping to conclusions about what's going on, without actually taking the time to find out...
The guy has a point. You apparently don't like the fact that he raised it(and others).
It's not that he raised it.
It's that he raised again.
And again.
And again.
And again.
And again.
And again.
And again.
And - well you get the picture, I imagine.
And no amount of reasoning would shake him from his view that the new focusing screen in the 7D was nothing short of a sin against God.
Believe me, it does get tired.
I have not read all of the nearly 30 pages of posts in this thread
Perhaps if you had, you'd understand something of the provocation that elicits such responses...
Maybe someone will even have demonstrated that they have what it takes to apologize for a demonstration of badgering.
Why? There's a clear cause and effect here, and it's the cause that could have realised long since that enough was enough. It's not as if this was the only thread where we got to hear - ad nauseam - about how stupid Canon were, and how ignorant they were of the needs of its customer base by - possibly - having made MF harder to do.
In fact perhaps you might consider apologising for so readily jumping to conclusions about what's going on, without actually taking the time to find out...
Sep 08, 2009 at 12:13 PM
p.36 #9Canon EOS 7D Master thread
Tom_W wrote:
I'm not sure about that.
I am!
Never did side-by-side testing
I did!
You had to hit it right on the 40.
I find the 40D to be way more forgiving than are my 30Ds.
I suspect though, that conversion software decisions pay a big part in that, Tom.
I'm not sure about that.
I am!
Never did side-by-side testing
I did!
You had to hit it right on the 40.
I find the 40D to be way more forgiving than are my 30Ds.
I suspect though, that conversion software decisions pay a big part in that, Tom.
Sep 08, 2009 at 12:18 PM
p.36 #10Canon EOS 7D Master thread
keithreeder wrote:
I suspect though, that conversion software decisions pay a big part in that, Tom.
Or, perhaps, the fact that I've compared both to my 5D over the years. In terms of the ability to bring up shadows, the 5D produces much more forgiving files than either body. I just don't seem to remember the difference being as significant with the 30 as I did with the 40. Maybe I just wasn't noticing that aspect.
I suspect though, that conversion software decisions pay a big part in that, Tom.
Or, perhaps, the fact that I've compared both to my 5D over the years. In terms of the ability to bring up shadows, the 5D produces much more forgiving files than either body. I just don't seem to remember the difference being as significant with the 30 as I did with the 40. Maybe I just wasn't noticing that aspect.
Sep 08, 2009 at 01:41 PM
p.36 #11Canon EOS 7D Master thread
thw2 wrote:
On a slightly different but related note, I used DXOMark to compare Canon 40D to the Nikon D200.
Canon 40D:
Color Depth = 22.1
Dynamic Range = 11.3
Low Light ISO = 703
Nikon D200:
Color Depth = 22.3
Dynamic Range = 11.5
Low Light ISO = 583
Everyone knows the D200 is a noisy bugger at high ISO and its low light ISO score supports this.
YET, DXOMark assigns an overall score of 64.2 to D200 and 'only' 63.5 to 40D. Is there a reasonable explanation of this or is this pure BIASNESS? If there is no satisfactory explanation for this, I will classify DXOMark as another extremely biased...Show more →
They seem to base overall score mostly on low ISO.
Whatever the case, don't pay attention to those figures go to the plots and set it to print for the real juice from that site.
On a slightly different but related note, I used DXOMark to compare Canon 40D to the Nikon D200.
Canon 40D:
Color Depth = 22.1
Dynamic Range = 11.3
Low Light ISO = 703
Nikon D200:
Color Depth = 22.3
Dynamic Range = 11.5
Low Light ISO = 583
Everyone knows the D200 is a noisy bugger at high ISO and its low light ISO score supports this.
YET, DXOMark assigns an overall score of 64.2 to D200 and 'only' 63.5 to 40D. Is there a reasonable explanation of this or is this pure BIASNESS? If there is no satisfactory explanation for this, I will classify DXOMark as another extremely biased...Show more →
They seem to base overall score mostly on low ISO.
Whatever the case, don't pay attention to those figures go to the plots and set it to print for the real juice from that site.
Sep 08, 2009 at 03:28 PM
p.36 #12Canon EOS 7D Master thread
7D:
blackpoint 2047, full-well 15303, StdDev blackframe (mask area only choice to measure for now) 5.41, DR =11.26 'per pixel' or 11.82 in 8.2MP 20D/30D normalization
table:
KEY
Camera:
DR 'per pixel', DRn normalized to 8.2MP, ISO at which max DR (but not necessarily max 18% gray SNR) occurs, pattern banding comment at ISO of highest DR (high pattern banding may reduced real-world effective DR)
5D2:
11.24, 11.9, ISO100, quite major pattern banding
7D:
11.26, 11.8, ISO200, appears to be very low pattern banding
50D (typical copy tested):
11.07, 11.5, ISO200, quite major pattern banding
40D (exceptionally good copy tested):
11.32, 11.5, ISO200, relatively moderate pattern banding
20D (exceptionally poor copy tested):
10.56, 10.6, ISO100, not characterized but prob intermediate
Looking at the screen in 1:1 view the 40D will seem to have the best DR, followed by the 7D, the 5D2 would be right in there if not for the noticeably worse pattern banding. The 50D is getting a little worse, but again basically the same aside for the worse pattern banding as with the 5D2. The 20D was far enough behind to be realistically noticeable in usage.
At the print or equally re-scaled image view the 7D should have the best DR since it almost matches the 5D2 but has less pattern banding; next are probably the 5D2 and 40D, the 40D less pattern banding maybe makes up for the 1/3+ stop worse measured difference
; then follows the 50D about 1/3+ stop behind in measured from the 40D but with worse pattern banding and then the 20D noticeably farther behind finishes it off.
The 5D2 should have noticeably better SNR measurements at 18% than the APS-C cams at all ISO. it is possible the 7D might have a tiny bit better and the 20D a tiny bit worse SNR at 18%, after normalization, than the other APS-C cams. Although this is just guessing based on sensor size and mentioned technology.
MAIN POINT:
Does not appear to be anything at all to worry about with the 7D DR compared to the older, lower MP cameras from Canon, it measures 1/3+ stop better at the print level and with less banding than the 50D, yeah no worries at all.
Since it has better microlenses and photodiodes one would imagine that it can't do worse for SNR than the other APS-C cams either at print level.
Despite all the fussing it seems quite likely, although this is tentative, to have the best IQ, even ignoring MP advantage, of any Canon APS-C camera yet, even if the improvement may in many cases be not wildly noticeable other than when pulling very deep shadows at low ISO or in areas of very low light at high ISO.
I.E. don't worry about it one way or the other, other than you should have less banding issues. Main thing to see is if the AF actually does much better or not.
blackpoint 2047, full-well 15303, StdDev blackframe (mask area only choice to measure for now) 5.41, DR =11.26 'per pixel' or 11.82 in 8.2MP 20D/30D normalization
table:
KEY
Camera:
DR 'per pixel', DRn normalized to 8.2MP, ISO at which max DR (but not necessarily max 18% gray SNR) occurs, pattern banding comment at ISO of highest DR (high pattern banding may reduced real-world effective DR)
5D2:
11.24, 11.9, ISO100, quite major pattern banding
7D:
11.26, 11.8, ISO200, appears to be very low pattern banding
50D (typical copy tested):
11.07, 11.5, ISO200, quite major pattern banding
40D (exceptionally good copy tested):
11.32, 11.5, ISO200, relatively moderate pattern banding
20D (exceptionally poor copy tested):
10.56, 10.6, ISO100, not characterized but prob intermediate
Looking at the screen in 1:1 view the 40D will seem to have the best DR, followed by the 7D, the 5D2 would be right in there if not for the noticeably worse pattern banding. The 50D is getting a little worse, but again basically the same aside for the worse pattern banding as with the 5D2. The 20D was far enough behind to be realistically noticeable in usage.
At the print or equally re-scaled image view the 7D should have the best DR since it almost matches the 5D2 but has less pattern banding; next are probably the 5D2 and 40D, the 40D less pattern banding maybe makes up for the 1/3+ stop worse measured difference
; then follows the 50D about 1/3+ stop behind in measured from the 40D but with worse pattern banding and then the 20D noticeably farther behind finishes it off.The 5D2 should have noticeably better SNR measurements at 18% than the APS-C cams at all ISO. it is possible the 7D might have a tiny bit better and the 20D a tiny bit worse SNR at 18%, after normalization, than the other APS-C cams. Although this is just guessing based on sensor size and mentioned technology.
MAIN POINT:
Does not appear to be anything at all to worry about with the 7D DR compared to the older, lower MP cameras from Canon, it measures 1/3+ stop better at the print level and with less banding than the 50D, yeah no worries at all.
Since it has better microlenses and photodiodes one would imagine that it can't do worse for SNR than the other APS-C cams either at print level.
Despite all the fussing it seems quite likely, although this is tentative, to have the best IQ, even ignoring MP advantage, of any Canon APS-C camera yet, even if the improvement may in many cases be not wildly noticeable other than when pulling very deep shadows at low ISO or in areas of very low light at high ISO.
I.E. don't worry about it one way or the other, other than you should have less banding issues. Main thing to see is if the AF actually does much better or not.
Sep 08, 2009 at 03:37 PM
p.36 #14Canon EOS 7D Master thread
I don't mean to slosh around the open can of worms here, but I have an Ef-S screen in my 40D, an Ee-S screen in my 5D, and a Katz-eye screen in my 350D. I probably autofocus at least 90% of the time but these screens make an unbelievable difference the other 10%. I also think it is a very large mistake that the 7D screen, by all pictures I've seen, does not have the Canon "S" screen as an option.
I am very frightened that this is the beginning of a trend at Canon, and that future releases will not feature the quite easy and simple screen change design of the current cameras.
I am very frightened that this is the beginning of a trend at Canon, and that future releases will not feature the quite easy and simple screen change design of the current cameras.
Sep 08, 2009 at 03:44 PM
brainiac
Offline
[X]
Details
p.36 #15Canon EOS 7D Master thread
Liquidstone wrote:
If someone who is a Nikon loyalist at DPR is biased against 21 MP sensors, that bias should also work against the D3x with its even denser pixels? This seems to be a paradox to me.
I agree that the D3x is also punished in DPR's broken method of comparison. However, it has generally been the case in recent years that Canon cameras have had more pixels than their Nikon equivalents. The 1Ds3 came out very slightly before the D3, and the D700 is still at 12 Mpixels compared to the 5D2's 21. The 50D and 500D are up to 15 Mpixel, and soon the 7D will be at 18. Thus per pixel noise comparisons between similarly priced cameras from Nikon and Canon will generally favour Nikon. But I agree that DPR's prejudice against high resolution sensors, and its prejudice in favour of Nikon may well be two separate but convenient trends.
If someone who is a Nikon loyalist at DPR is biased against 21 MP sensors, that bias should also work against the D3x with its even denser pixels? This seems to be a paradox to me.
I agree that the D3x is also punished in DPR's broken method of comparison. However, it has generally been the case in recent years that Canon cameras have had more pixels than their Nikon equivalents. The 1Ds3 came out very slightly before the D3, and the D700 is still at 12 Mpixels compared to the 5D2's 21. The 50D and 500D are up to 15 Mpixel, and soon the 7D will be at 18. Thus per pixel noise comparisons between similarly priced cameras from Nikon and Canon will generally favour Nikon. But I agree that DPR's prejudice against high resolution sensors, and its prejudice in favour of Nikon may well be two separate but convenient trends.
Sep 08, 2009 at 04:10 PM
p.36 #16Canon EOS 7D Master thread
brainiac wrote:
However, it has generally been the case in recent years that Canon cameras have had more pixels than their Nikon equivalents.
The 5D classic should be "favored" by DPR's methods. The D3x should suffer from the same.
Regardless of whether DPR's test protocols are effective or not, we can't conclude that there is brand bias based on the method alone, though there could be some sort of bias for a brand based on other factors.
Edited on Sep 08, 2009 at 04:40 PM · View previous versions
However, it has generally been the case in recent years that Canon cameras have had more pixels than their Nikon equivalents.
The 5D classic should be "favored" by DPR's methods. The D3x should suffer from the same.
Regardless of whether DPR's test protocols are effective or not, we can't conclude that there is brand bias based on the method alone, though there could be some sort of bias for a brand based on other factors.
Edited on Sep 08, 2009 at 04:40 PM · View previous versions
Sep 08, 2009 at 04:30 PM
p.36 #17Canon EOS 7D Master thread
Liquidstone wrote:
The 5D classic should be "favored by DPR's methods." The D3x should suffer from the same.
Regardless of whether DPR's test protocols are effective or not, we can't conclude that there is brand bias based on the method alone, though there could be some sort bias for a brand based on other factors.
they did once (semi-)jokingly say that they eagerly await the day canon totally drops the ball on the sensor and they can say FAIL!
They have seemed to have some curious interpretations of the amount of detail visible at times. And they tend to say the detail level is very significant if they say the nikon has lots visible and tend to downplay it as not having any practical significance when some other brand shows lots of detail. Not a huge bais, but I really think I have seen some.
They also compare using default bootup settings and older canon cams did no chroma NR at default setting and nikons did and they would always bash the canons for having chroma noise.
And they, have at times, ragged on non-Nikons for using NR while NEVER mentioning how the Nikon curves sometimes magically arc down and give a higher ISO better SNR than the one before.
And they have been prone to harp on slightly lower 50D per pixel sharpness (using debatably focused portions of images) while ignoring per image detail and yet praising higher MP nikons for great per image detail.
I mean it is not too extreme, but it does seem like they like nikon and some of that unintentionally seeps out (and they also seem to have a chip on their should still from teh M3 and early incident with canon).
brainiac wrote
However, it has generally been the case in recent years that Canon cameras have had more pixels than their Nikon equivalents.Regardless of whether DPR's test protocols are effective or not, we can't conclude that there is brand bias based on the method alone, though there could be some sort bias for a brand based on other factors.
they did once (semi-)jokingly say that they eagerly await the day canon totally drops the ball on the sensor and they can say FAIL!
They have seemed to have some curious interpretations of the amount of detail visible at times. And they tend to say the detail level is very significant if they say the nikon has lots visible and tend to downplay it as not having any practical significance when some other brand shows lots of detail. Not a huge bais, but I really think I have seen some.
They also compare using default bootup settings and older canon cams did no chroma NR at default setting and nikons did and they would always bash the canons for having chroma noise.
And they, have at times, ragged on non-Nikons for using NR while NEVER mentioning how the Nikon curves sometimes magically arc down and give a higher ISO better SNR than the one before
.And they have been prone to harp on slightly lower 50D per pixel sharpness (using debatably focused portions of images) while ignoring per image detail and yet praising higher MP nikons for great per image detail.
I mean it is not too extreme, but it does seem like they like nikon and some of that unintentionally seeps out (and they also seem to have a chip on their should still from teh M3 and early incident with canon).
Sep 08, 2009 at 04:38 PM
p.36 #18Canon EOS 7D Master thread
skibum5 wrote:
I.E. don't worry about it one way or the other, other than you should have less banding issues. Main thing to see is if the AF actually does much better or not.
Thanks for the quantitative analysis and comparisons. I'm a laggard, who will be upgrading from a 20D.
+1 on the AF functionality/performance! That's the real game-breaker for me.
Jeff
I.E. don't worry about it one way or the other, other than you should have less banding issues. Main thing to see is if the AF actually does much better or not.
Thanks for the quantitative analysis and comparisons. I'm a laggard, who will be upgrading from a 20D.
+1 on the AF functionality/performance! That's the real game-breaker for me.
Jeff
Sep 08, 2009 at 05:00 PM
p.36 #19Canon EOS 7D Master thread
garyvot wrote:
When I had a pair of D300's I noticed that even in RAW captures there seemed to be something softened in the noise pattern. I've never heard anyone confirm that Nikon cooks its RAW files, but I've always wondered...
If we assume DXOMark plots are scientifically measured, then we can deduce something from them.
Consider the various plots for Nikon D3, D3X, D90 and Canon 40D at the pixel level:
(a) DR of D3X and 40D have different trends. The 40D DR plot is a curve that seemingly plateaus at low ISO while the D3X is a straight line.
It's been suggested that this difference is due to different ADC (analog to digital) architecture. The straight line DR plot is a signature of Sony's architecture which integrates onto the sensor chip an analog to digital converter in every column.
(b) Whereas the SNR 18% curve for D3X and 40D are straight lines, that for the D90 has a slight kink at ISO 6400. This is not difficult to understand 'cos ISO 6400 for D90 is fake.
Now, what really gives away the fact that Nikon cooks some of their RAW data is the DR curves for various cameras. Whereas D3X and 40D DR plots are smooth (either straight line or curve), the same plots for the D90 and D3 are full of kinks. This tells us the D90 and D3 RAW files have been cooked in-camera.
The D3X RAW data is by far the most impressive. It is uncooked and has better DR than 1Ds3 and 40D at all ISO. Interestingly enough, 1Ds3 has the same DR as 40D at the pixel level.
When I had a pair of D300's I noticed that even in RAW captures there seemed to be something softened in the noise pattern. I've never heard anyone confirm that Nikon cooks its RAW files, but I've always wondered...
If we assume DXOMark plots are scientifically measured, then we can deduce something from them.
Consider the various plots for Nikon D3, D3X, D90 and Canon 40D at the pixel level:
(a) DR of D3X and 40D have different trends. The 40D DR plot is a curve that seemingly plateaus at low ISO while the D3X is a straight line.
It's been suggested that this difference is due to different ADC (analog to digital) architecture. The straight line DR plot is a signature of Sony's architecture which integrates onto the sensor chip an analog to digital converter in every column.
(b) Whereas the SNR 18% curve for D3X and 40D are straight lines, that for the D90 has a slight kink at ISO 6400. This is not difficult to understand 'cos ISO 6400 for D90 is fake.
Now, what really gives away the fact that Nikon cooks some of their RAW data is the DR curves for various cameras. Whereas D3X and 40D DR plots are smooth (either straight line or curve), the same plots for the D90 and D3 are full of kinks. This tells us the D90 and D3 RAW files have been cooked in-camera.
The D3X RAW data is by far the most impressive. It is uncooked and has better DR than 1Ds3 and 40D at all ISO. Interestingly enough, 1Ds3 has the same DR as 40D at the pixel level.
Sep 08, 2009 at 05:03 PM
p.36 #20Canon EOS 7D Master thread
skibum5 wrote:
they did once (semi-)jokingly say that they eagerly await the day canon totally drops the ball on the sensor and they can say FAIL!
They have seemed to have some curious interpretations of the amount of detail visible at times. And they tend to say the detail level is very significant if they say the nikon has lots visible and tend to downplay it as not having any practical significance when some other brand shows lots of detail. Not a huge bais, but I really think I have seen some.
They also compare using default bootup settings and older canon cams did...Show more →
Yes, their bias is all there for the world to see.
This is why I never pay too much attention to reviews from that site. Same goes for Rob Galbraith's reviews. DXOMark is OK if we ignore their assigned sensor scores.
they did once (semi-)jokingly say that they eagerly await the day canon totally drops the ball on the sensor and they can say FAIL!
They have seemed to have some curious interpretations of the amount of detail visible at times. And they tend to say the detail level is very significant if they say the nikon has lots visible and tend to downplay it as not having any practical significance when some other brand shows lots of detail. Not a huge bais, but I really think I have seen some.
They also compare using default bootup settings and older canon cams did...Show more →
Yes, their bias is all there for the world to see.
This is why I never pay too much attention to reviews from that site. Same goes for Rob Galbraith's reviews. DXOMark is OK if we ignore their assigned sensor scores.
Sep 08, 2009 at 05:10 PM