Originally posted by Mystic well, that particular test show that kx is better
you win
It's not about winning -
I too really wanted the K-r to be an improvement over the K-x -
which means that Pentax are improving their IQ with each subsequent offering -
but unfortunately wishful thinking cannot be a substitute for actual results and direct comparison.
Fortunately for us at least the K-r is not any worse than the K-x for any real practical usage -
but to say the K-r IQ is better than the K-x is really just in the realms of "wishful thinking"
- don't worry, this is not meant to be a dig at you -
I am very prone to this myself -
that is until I actually took the time to examine the comparable images posted out there.
Again I emphasize that the IQ is close - but unfortunately the K-r is not better than the K-x from any comparison I have made (and I'll be the first to admit I probably have not found all the comparable images out there).....
But to show shots taken side-by-side, in the same conditions and even time-frame -
we can use devorama's thread
K-r vs. K-x ISO samples (or return of the ISO flower!)
and his direct comparison images - ref:
All sizes | high-key_ISO_elevator | Flickr - Photo Sharing! All sizes | low-key_ISO_elevator | Flickr - Photo Sharing! All sizes | grey_ISO_elevator | Flickr - Photo Sharing!
I've taken the 100% crops of the two cited ISOs from each of those series:
high-key:
Low-key:
Grey:
Like I said is close enough that it probably will not make any practical difference in my kind of photography -
but unfortunately much as I would like to wish it - the K-r's IQ is not better than the K-x - if anything in the High-key and Grey elevator series the K-x seems to have less noise and retain more detail. In the low-key there just isn't enough detail to judge well - but the K-x does have less noise.
But it is so close that probably different samples may give different results - however you now have Imaging-Resource.com and devorama's results - it's still a very small and statistically insignificant sample
- but unfortunately both sets do seem to favor the K-x albeit very, very close - the fact is from these limited and statistically insignificant results the K-r is not better than the K-x.