Ugly Hedgehog - Photography Forum
Home Active Topics Newest Pictures Search Login Register
Main Photography Discussion
DXO Mark Score
Page <prev 2 of 2
Apr 11, 2017 00:02:49   #
CHOLLY Loc: THE FLORIDA PANHANDLE!
 
^^^Yes it is.

One sample is NOT enough to give a statistically significant result.... or even representative data.

Reply
Apr 11, 2017 01:31:55   #
MtnMan Loc: ID
 
rook2c4 wrote:
I don't have much faith in their published scores either. DXO Mark has a tendency to give higher scores to lenses from manufacturers they do most business with. Purely coincidence? I think not.


I agree. The camera manufacturers generally make better lenses. Especially Nikon and Canon.

I don't know that DXOMark "does business" with any of the camera manufacturers other than to solicit test samples.

Reply
Apr 11, 2017 06:25:16   #
selmslie Loc: Fernandina Beach, FL, USA
 
Uuglypher wrote:
... As you can imagine, I found that response from DXO to be rather surprising, given the common phenomenon of "performance variance" that plagues a wide variety of silicon-based semiconductor devices (including our cameras' photosensors) ...

That's not surprising at all considering the nature of quality control and the precision of the engineering and testing performed by the manufacturers.

I have never heard of any "common phenomenon of performance variance" affecting camera sensors although there are some slight variations affecting mechanical components such as lenses.

Any performance variance you might be finding in your personal testing is more likely the result of not doing your tests under controlled laboratory conditions.

PS: I googled 'image sensor performance' and came across this scholarly gem, Fundamentals of Image Sensor Performance from 2011. In the summary it states, "This paper described how modern image sensors work. ... It also gave a description of many of the common metrics used when comparing the performance of different image sensors. An analysis was performed using a sample of state of the art image sensors. The analysis showed that for dark current and dynamic range, no significant difference could be seen, although the sample size was small for both CCD and CMOS populations. ..."

If anything has changed since then it has probably been an improvement but how much better can you get than "no signficant difference."

Reply
Page <prev 2 of 2
If you want to reply, then register here. Registration is free and your account is created instantly, so you can post right away.
Main Photography Discussion
UglyHedgehog.com - Forum
Copyright 2011-2024 Ugly Hedgehog, Inc.