Reading mode: Light Dark

Canon sensors in 7D and 60D behind the curve?

Go to last post
M

mike_2008

Veteran Member
Messages
5,635
Reaction score
11
Location
IT
Some new DXOmark data for the sony A55 has just been released, and it makes for an interesting comparison with the canon 7D and 60D.

Of course these are data from the RAW output of the cameras, so the jpeg's will reflect also the quality of the jpeg conversion engine for each model, but if you shoot raw it's perhaps quite useful to compare.

I was particularly struck by the fact that both the more expensive 60D and 7D have a full stop lower colour depth and dynamic range compared with the much cheaper A55, whilst they all have very similar noise features. It looks like the new sensor in the A55, which is also shared by the much more expensive nikon D7000 and pentax K5 is a winner. How will canon respond?

 
Dxo measurements?
Digitally Xtremely Odd measurement...
 
It's a level playing field of the raw sensor performance. Not always representative of the final jpeg performance, and clearly ignores handling, AF, AE etc., but within it's limits as a comparison tool I don't really see what's odd about it, as long as all cameras are treated the same way.
photofan1986 said:
Dxo measurements?
Digitally Xtremely Odd measurement...
Click to expand...
 
mike_2008 said:
Some new DXOmark data for the sony A55 has just been released, and it makes for an interesting comparison with the canon 7D and 60D.
Click to expand...
And as usual completely bananas. The DxO measurements one sidedly favor - bigger sensors, doctored RAW files (such as the black value clipping most Nikon sensors use) and 12 bit depth sensors.
--
regards
Karl Günter Wünsch
 
mike_2008 said:
It's a level playing field of the raw sensor performance.
Click to expand...
No it isn't as the measurements don't get the true raw sensor performance, they get the RAW files which have been processed by the first stage of the camera engine - and they have seeded their measurements in such a way to favor certain manipulations or architectures.

--
regards
Karl Günter Wünsch
 
This is all just plain odd hype. I was with Canon, I am with Canon (60D will come today), and I will most probably stay with Canon. And why? Because I do own some nice lens/flash.

Look - there is no problem to admit, that Sony/Nikon might be better with recent models. But - those models will get trumped or at least aligned to by next Canon release, most probably 600D.

... and in the meantime, I am going to shoot using 60D, which is better, than my recent 450D.

... and life goes on and on :-)

Cheers,
Petr
 
They use the raw files which are available to us, the users, which is what counts in the end, surely. Other levels of data are inaccessible and therefore irrelevant. What counts is what the camera actually produces.
Karl Gnter Wnsch said:
mike_2008 said:
It's a level playing field of the raw sensor performance.
Click to expand...
No it isn't as the measurements don't get the true raw sensor performance, they get the RAW files which have been processed by the first stage of the camera engine - and they have seeded their measurements in such a way to favor certain manipulations or architectures.

--
regards
Karl Günter Wünsch
Click to expand...
 
pekr said:
This is all just plain odd hype. I was with Canon, I am with Canon (60D will come today), and I will most probably stay with Canon. And why? Because I do own some nice lens/flash.

Look - there is no problem to admit, that Sony/Nikon might be better with recent models. But - those models will get trumped or at least aligned to by next Canon release, most probably 600D.

... and in the meantime, I am going to shoot using 60D, which is better, than my recent 450D.

... and life goes on and on :-)
Click to expand...
Right on the money. This will push canon to produce a better sensor next time.
 
mike_2008 said:
They use the raw files which are available to us, the users, which is what counts in the end, surely. Other levels of data are inaccessible and therefore irrelevant. What counts is what the camera actually produces.
Click to expand...
Following this criteria, it would be fine by you a camera that half-cooked the raw data to improve -take for instance- the high iso output.

But it wouldn't be fine by me. I'd still want the high iso ouptut to be improved because of the hardware rather than software -so to speak-. That's to say, an improved ability of the sensor to capture available (scarce) light.

--
http://jaimsthesweetspot.wordpress.com/
 
mike_2008 said:
They use the raw files which are available to us, the users, which is what counts in the end, surely. Other levels of data are inaccessible and therefore irrelevant.
Click to expand...
In the context of what DxO tries to achieve this would be the only valid test.
mike_2008 said:
What counts is what the camera actually produces.
Click to expand...
And that's something which DxO blatantly ignores - there is proof that their assessments favor larger sensors over smaller sensors (simply because they are bigger), they favor 12 over 14 A/D conversion because they don't normalize their measurements before comparison and the higher resolving A/D converter gets penalized and they favor cooked (such as black value clipped or noise reduced) RAW files. All in all a very shoddy performance of someone who claims to be scientific sound - which they are far from!

--
regards
Karl Günter Wünsch
 
mike_2008 said:
Right on the money. This will push canon to produce a better sensor next time.
Click to expand...
Wrong, all that might happen would be sensors that perform better in their metric but not in the real world application - so would you rather have a sensor that performs noise reduction before the RAW is saved or which disqualifies itself from certain applications because it clips the black level - just to look good in DxO measurements while in real life the results are worse than the previous generation. That would be the consequence if DxO were to become the metric by which the camera quality is measured. So you can have excellent (in terms of DxO) sensors but mediocre to abysmal real life performance...
--
regards
Karl Günter Wünsch
 
mike_2008 said:
Of course these are data from the RAW output of the cameras, so the jpeg's will reflect also the quality of the jpeg conversion engine for each model, but if you shoot raw it's perhaps quite useful to compare.
Click to expand...
I shoot raw and do all my own processing. Ultimately, the quality of my prints are much more a function of my ability to process my images than the differences between sensors as measured by DX0. These comparisons are academically interesting but I have never seen any demonstration that the differences typically measured by DX0 have a significant difference in the final result, a print or image on a display.
--
Leon
http://web.me.com/leonwittwer/landscapes.htm
 
mike_2008 said:
They use the raw files which are available to us, the users, which is what counts in the end, surely. Other levels of data are inaccessible and therefore irrelevant. What counts is what the camera actually produces.
Karl Gnter Wnsch said:
mike_2008 said:
It's a level playing field of the raw sensor performance.
Click to expand...
No it isn't as the measurements don't get the true raw sensor performance, they get the RAW files which have been processed by the first stage of the camera engine - and they have seeded their measurements in such a way to favor certain manipulations or architectures.

--
regards
Karl Günter Wünsch
Click to expand...
Click to expand...
I agree it's the pictures people get from their camera files, that really mean something.

All the Engineering stuff some think is so important, really means nothing to most buyers. ;-)
 
Until I can see comparison images that show how differences in DXO scores actually effect the image quality, I will continue to be pretty unimpressed with their metric.

You cant tell me that the D90 is that much better than the D300, for example.

To many people (not saying you're one of them) throw the DXOmark scores around, and have no clue as to what they mean.

--
Canon 7D
FujiFilm F20

http://www.pbase.com/timothyo
 
DxO is fine if you never shoot pics and only compare specs. But wait! Even then there's a problem. The 7D and the 60D have the same sensor, but DxO lists different specs, like the size of the photo sites for each one.

Hmmm.....

--
-------------------------------------------------
'Hit Refresh if pix do not appear. Flaky ISP at work.'

 
If DxO and the point you are trying to make is so correct, I should be seeing an exodus of Canon and Nikon users towards the Sony camp. But heck no, that is not what is happening. This is what has been happening:

-Walkman was swallowed up whole by Apple's iPod
-Playstation was beaten by Nintendo
-Bravia is edged out by Samsung

So who will reasonably expect Sony to come up with anything better in an area where it has not even been a major player, when it can barely sustain its lead on segments it used to rule? Gosh even in the point-and-shoot category, Lumix made by Panasonic (which used to be regarded as a second-rate manufacturer to Sony) has offerings that Sony cannot match.

DxO can continue with their ratings for all they care. For as long as the most number of compelling photographs are shot using Canon and Nikon, that is the only benchmark that would ever matter.

--
Noogy
"Photography is therapeutic."
http://www.pbase.com/joshcruzphotos
 
dmanthree said:
DxO is fine if you never shoot pics and only compare specs. But wait! Even then there's a problem. The 7D and the 60D have the same sensor, but DxO lists different specs, like the size of the photo sites for each one.

Hmmm.....
Click to expand...
Exactly - and the 550D they have different again!

DxO have no credibility in my eyes if they can't even get the pixel size of the sensors they are 'testing' correct!

And, to the OP - the D90 scores exactly the same as the Alpha 55. So, if your willing to believe DxO, then Sony has made no progress what-so-ever since the D90 2 years ago - LOL :-D
 
That pretty much sums up the kind of company Sony is :-)
schmegg said:
dmanthree said:
DxO is fine if you never shoot pics and only compare specs. But wait! Even then there's a problem. The 7D and the 60D have the same sensor, but DxO lists different specs, like the size of the photo sites for each one.

Hmmm.....
Click to expand...
Exactly - and the 550D they have different again!

DxO have no credibility in my eyes if they can't even get the pixel size of the sensors they are 'testing' correct!

And, to the OP - the D90 scores exactly the same as the Alpha 55. So, if your willing to believe DxO, then Sony has made no progress what-so-ever since the D90 2 years ago - LOL :-D
Click to expand...
--
Noogy
"Photography is therapeutic."
http://www.pbase.com/joshcruzphotos
 
schmegg said:
dmanthree wrote:

And, to the OP - the D90 scores exactly the same as the Alpha 55. So, if your willing to believe DxO, then Sony has made no progress what-so-ever since the D90 2 years ago - LOL :-D
Click to expand...
What does it say about current crop of canon sensors that they don't even approach the D90 or A55? Worth bearing in mind that the A580, K5 and D7000 all have this new sensor and may well be even slightly better performing than the A55 which gains 10 fps with full AF and PDAF in video mode at the expense of 30% of light. My point is, how will canon catch up?
 
You must log in or register to reply here.

Similar threads

P
Canon 7D colors blue and red
Replies
5
Views
693
alex_virt
A
B
  • Question Question
Best digital RAW files for Kodachrome 25/64 emulation: highlight roll-off, tonal behavior, and senso
2
Replies
33
Views
2K
zotynimon
Z
M
Canon R5 or Leica SL2 for Architectural Photography
Replies
6
Views
2K
Micchi
M
X
Canon 7D MKII Outperformed By T3i?
Replies
6
Views
136
3Percent
3Percent
JimKasson
Noise, Dynamic Range, Cropping, and Print Size
2
Replies
24
Views
5K
JimKasson
JimKasson

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top