while i continue to be a loyal Canon user, Canon lacks behind Nikon in picture quality, why is that?

[Miri]

Even their large mega pixel camera outshines Canon offerings. Canon just seems to do better in marketing 😞

[photonut]

That's much too broad of a statement as you neglected to provide any details as to which types and model cameras you are comparing. 

[khooby]

Well for a start, why do all their sensors lag so far behind Nikon's? I'm referencing DxOMark measurements here. While DxOMark might not be the most accurate guys around, there still shouldn't be that much of a disparity.

[jfo]

What would be a "good" answer to these kind of questions? 

In the whole Canon vs. Nikon (aka 'the civil war'), most photographers working professionally have figured out that each brand has their pluses and minuses.  There is no calculation or measurement to fully quantify how one is "better" than the other.

I'm pretty sure Canon won't chime in to give you answers because you have to understand that this is the essence of making a product: create something that fits market needs with features and a price point acceptable to generate a profit. 

P.S.  I've never heard complaints about picture quality in the Canon vs Nikon debate before, usually only mentions of different color tints.  Where/when is this a practical issue other than in scientific "tests"?  Seems almost like a 'straw man' argument for Nikon vs. Canon.

[JELPhotography]

over the course of time, one is up, then the other

the biggest thing that sold me on canon was the human-computer-interface - or "ergonomics" - or however you want to call it

canon is still tops in that dept

i know that's an extremely subjective assessment, but as long as canon's images still "wow" me (like they have since my t2i days - through the 7D and now with the 1dx) - i see no reason to jump ship

also, have you ever had to take a camera in for service? have you experienced their customer relations?

WOW

i've seen nikon people respond on forums indicating nikon could take some lessons from canon in that dept

as long as canon keeps me happy.. i'm fine right where i am

[killswitch]

As much as I love Canon bodies and lenses, I wish and only wish they'd improve on capturing shadow detials a little better.  Tried lifting shadows on Nikon D600 and I was just blown away with all the details it capture plus better noise control in dark areas.

[bvukich]

I don't think anyone would turn down a little extra DR, but I think overall the features on Canon bodies are better than Nikon. Plus Canon glass is better than Nikon except for a few missing lenses (e.g. 12-24/2.8).

[David_Hull]

Forget the terms, IQ and Image Quality as they are meaningless.  There is nothing wrong with the quality of the images produced with any Canon camera that I have seen assuming it is properly used and used within it's limitations.  This same thing can be said for any camera.  You would be at a complete loss to walk through a glary of images shot by competent photographers and identify which was shot by Canon and which originated from a Nikon capture device.

The real issue is sensor dynamic range or more correctly "camera" dynamic range since the culprits in this crime are not entirely in the sensor itself.  Where Canon lags behind Nikon (actually Sony who makes the sensors in question) is in terms of read noise at low SNR which primarily impacts the shadows of the photographs making it more difficult to lift shadows from Canon equipment in post than it is to do so with Nikon bodies that are e/w Sony sensors.  Additionally the shadow noise of the typical Canon camera seems to exhibit repetitive patterns in it's shadow noise that the other guys do not which further reduces the DR beyond what you see in the DxO plots.

The root cause of this (many believe) lies in Canon's system architectural choices.  While Nikon/Sony have chosen to implement a distributed A to D conversion regime on the sensor itself, Canon have chosen to do this off-chip using a wide band ADC which provides lower performance in terms of the number of effective bits.  Sony gets 14 bits (stops) of DR while Canon (by virtue of their architectural choices) seems limited to about 12.5 bits (stops).  This shows up in the DxO DR curve and in all those lovely artistic shots of Bar-B-Q lids and backs of lens caps that the banding crowd loves to flaunt about.

Most people who post about this stuff really don't understand it, they confuse the pattern noise with the DR shown by the DxO curves.  They are not really the same.  The 'banding" most likely comes about due to slight variations in the gain of the various analog signal paths coming out of the sensor to the external ADC array.

IMO Canon needs to address this somehow if for no other reason than competitiveness and corporate pride, I suspect that they will at some point.  In the mean time there is really no issue in terms of what you call "picture quality" that cannot be worked around using time honored photographic techniques.

[bvukich]

---

@David_Hull wrote:

Sony gets 14 bits (stops) of DR while Canon (by virtue of their architectural choices) seems limited to about 12.5 bits (stops).

---

I totally agree with your post, but I'm compelled to point how much the quoted portion annoys me (aimed not at you, but DXO).

Getting a claimed 14.4 stops out of a 14 bit sensor is impossible, and merely an artifact of their normalization process (down to 8MP).  I'm not debating that the D800 has less DR than the 5D3/1Dx, I accept that it is higher.  And most reasonable people realize that DXO is biased towards Nikon and "adjusts" their methodology to favor their preferd brand, but touting comical metrics isn't doing anyone any good.

Second, and something you didn't touch on, but still bothersome to me; is their claim of better high ISO performance of the D800/D600 than the 5D3 or 1Dx.  Their DR conclusion is correct, if slightly distorted in magnitude, whereas their high ISO conclusion is clearly, without a doubt, wrong.  Heck, ISO6400 on either of the Canons looks as clean or cleaner than ISO1600 on either of the Nikons.  Biased much?

[David_Hull]

bvukich --

I tend to agree WRT DxO.  I don't know if they are biased or not but I can't figure out how you can get 14.4 bits out of a 14 bit converter.  I have been working with ADC's for the last 20 years and have never seen anything like that :-).  However, to me the bigger issue with the current Canons is the pattern noise which comes about (as far as I can guess) from the analog muxing to the off chip converters.  I think this is more problematic than the 2 stops less DR.  My 5DIII has less of this than my 5DII did, though so it DOES appear that they worked on it.  For anyone worried about this, in the case of teh 5DIII, it cleans up nicely with the Nik NR tool.

As for the top end (high ISO end) the same DR curve pretty much shows that the Nikon/Sony sensor and the Canon sensor are equivalent as far as I can tell.  One of the guys on DP Review keeps track of the quantum efficiency of these things and I believe his latest measurement shows the 5DIII QE to be 49% compared to the D800 at 57% (or something like that).  This is pretty much a wash as far as I am concerned.  It is a big improvement for Canon though given that the 5DII was at 33%.