What to make of DxOMark scores

Started Jan 23, 2013 | Discussions thread
(unknown member) Senior Member • Posts: 1,144
Re: Nonsense

I shoot both Canon and Nikon, and I absolutely drank the Kool Aid when the D800 came out.  Don't get me wrong, the D800 is the real deal, and the details that can be pulled from shadows puts all Canon cameras to shame.

However, for high-ISO performance the D800 and 5DMIII are in a dead heat.  Download raw files--not jpeg--from Imaging Resources. You'll see a small difference in resolution, but the noise profiles are virtually indistinguishable.

I jumped ship to Nikon because of how dismally  my D7 performed at high-ISO shooting.  It was no match for my D90.  However, I have so much Canon glass that I couldn't resist picking up the T4i.

Anyway, after testing--and I only shoot raw--theT4i performs amazingly well.

There is virtually no noise through ISO800.  I have to look real hard to see noise at ISO1600, and there is very little image degradation.  At ISO3200, the image has very fine noise, but the image is still very good--Definitely usable.  ISO6400 is perhaps a step too far but usable in a pinch, whereas I would reserve ISO12800 solely for emergencies.

This is easily 1-1.5 stops better than what I saw on my 7D.  On the 7D I would very reluctantly go pass ISO1600 and even at ISO1600, I didn't like the loss in contrast and vibrancy.  For my T4i, I  shoot at ISO3200 without reservations.

So, whatever Canon is doing, the DXOMarks are not picking it up.  However, work in the field suggests otherwise.

Where the DXOMarks scores are accurate is dynamic range.  Canon still has a lot of work to do to catch up on shadow details.  I hear a lot of Canon shooters say that they don't care about pushing the shadows.  Well, for me shooting outdoors--morning,noon, or night--it is rare that I don't want to push shadows.  Nikon gives me that flexibility.  Canon does not.

brightcolours wrote:

Rick880 wrote:

husky92 wrote:

Yes, I'm still trying to pick a camera to upgrade to. The XTI has a low light DxOMark score of 664. The t4i has a low light DxO mark score of 722. So in 5 generations, it's gone up about 10%? So if I upgrade to a t4i I can bump up iso by 10% to get similar results to what I see now?

Canon has the same sensor design for ages and thus miminal improvement in these generations in my opinion.

Nonsense. Just nonsense. The XTi has way worse high ISO performance than the T4i. As I said before, the silly DXO scores are NOT about high ISO performance.

The Nikon D5200 has a DxO mark score of 1284. In the side by side comparisons here on DPreview it does appear to be about a full stop better than the 650D. That's just my subjective view looking at the samples, but it's very hard to tell.

I think it is about time for Canon to introduce new sensor (design) to catch up with competitors. Hope we have good news come February.

Still not getting it.

What is the general consensus on DxOMark?

I only look at DxOMark as a reference, not an absolute yardstick wen evaluating cameras. I currently shoot with XSi and if I go by DxOMark alone, Nikon's D5200 is a better choice than 6D if I want to replace XSi today. But I will not choose D5200 because there are other aspects of the camera that must be taken into consideration (the size of viewfinder, for example).

(repeating myself) The silly DXO ratings are NOT about high ISO performance.

Post (hide subjects) Posted by
(unknown member)
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow