Alan, coming in with all negative comments on (poor re-sized samples; not true non-NR samples; OBVIOUS NR on the sly; cheating; fake performance) over various threads (and even on your opening thread in your linked DPreview link) does not endear you to anyone. Masking it with 1 line of praise after a load full of accusations does not clear the air.
You could be more positive about what you are seeing or adopt a wait as see approach and wait for official respected (in your terms) reviews and samples. No one is forcing you to buy a K5.
I've already seen some ppl on the Nikon sub-forum saying that they have the D7K (overseas copy or whatever). You should go make the same type of remarks in that forum.
sigh...
here are ALL the post I have made in pentax forum for the past 2 weeks. Please read them through because I just did, trying to understand why you think the way you do. Cmmon can you honestly said I have a lot of negative things to say about Pentax. All I was trying to do is to educate a FACT that Pentax has NR on its raw files ISO3200 and above. I have been REALLY polite and sometimes I wonder if it is worth it because no matter what I say, people are still jumping at me.
I did not even dare to post the 14eV DR question here because you guys will jump on me like a pack of wolves. Thus I posted in dpreview. And you take that against me??? common, look at the replies in dpreview. My concern is justified and there are some pretty constructive responses there.
edit: ok, not all of you are jumping at me, but some of you are. If I have been impolite in this last post, I think you guys will understand why.
------------
"have you a sample of good high ISO performance with fine low contrast detail like fabric?
A lot of samples I see is just web sized prints with zero detail."
----------------------
but at least DXO does detect the existence of NR in pentax raw files due to the different noise distribution.
--------------------
there is OBVIOUS NR going on with the ISO3200 pic. and maybe even the ISO1600. I think it is true that Pentax does NR at raw file level so you can NEVER turn of NR completely. Which is bad.
Look at the fine detail at the blue regions in the top right corner.
-------------------------
Sorry, top LEFT... can never tell left and right
You can check you ISO80 and ISO100 pics and you can see some sort of consistent pattern in the blue region which is probably there is real life.
This pattern however is gone at the ISO3200 pic even though the pic is still very noiseless. So to me there is NR even though you set it to off.
I think pentax do force NR on the raw files so we need to consider that when comparing pics between different cameras.
Not that I am saying K-5 high ISO is poor. Even considering that, I still think that it is the BEST high ISO APS-C camera out there.
But some saying that it is better than the D700 is a probably not true. But it looks pretty close to what I get from my D700.
------------------------------
BTW, the DXOmark results for K-5 just came out and they agreed that ISO3200 and above has forced NR as seen by the SNR curve.
The high ISO noise performance looks similar to nikon D90 actually. But the STUNNING results is how good the k-5 is at ISO 80 which I feel is FAR FAR more useful than ISO3200.
The dynamic range is a mind boggling 14ev. I think that is D3x range.... wow....
-------------------------------
actually if you looked at the SNR curves, the k-5 is only slightly better than the D90 before the NR of k-5 kicked in.
I think the amazing thing about the k-5 is not the high iso but how good the numbers are at IS80, especially dynamic range which is 14 eV !
-------------------------
I think that thread was on high ISO performance right? Most of the hype was on high ISO performance and DXOMARK actually confirm that 2-yr old FF is still better in terms of high ISO performance.
DXOMARK shows that k-5 wins in terms of its ISO80 performance which I feel is more impt than high ISO.
--------------------------------
DXOMARK already confirm that pentax cameras do NR at 3200 and above so the SNR results are not accurate, ie artificially inflated. ie you have to look at ISO1600 and below for the actual performance.
There are a LOT of aritcles on this so I am surprised that people still don't know about it. Even DXOMARK explicitly mention the NR by using the term smoothing in the SNR curves.
That is a minus is what pentax does, which is forced NR on its raw files for ISO3200 and above. That is why I am skeptical of all the super clean high ISO pics people are posting claiming it is without NR when obiously you can see NR artifacts at 100% crop.
Still the k-5 sensor is REALLY REVOLUTIONARY. can't wait for a FF version of that sensor.
-------------------------
?? There is nothing negative about it PROVIDED you KNOW it is being done. If someone buys a k-5 thinking it will give high ISO performance as good as D700, he will be disappointed. Simple as that.
I was just trying to educate someone who did not know about pentax forced NR. TsammyC DID think that the k-5 performs as well as the D700 at high ISOs which is WRONG.
Yes I am a D700 user (and previously owned and used s3pro, KM7D, olympus E1 E330, samsung NX10) and I am looking for a 2nd body to mount a UWA lens for my travels. I am used to D700 high ISO performance so I was very happy to see a lighter APS-C sensor looking so good.
you guys need to be less defensive about brands. I keep saying good things about pentax k-5 but all you guys jump on are the bad stuff. I am thinking of using the k-5 side by side my D700 cuz I think it is really that good and lighter.
-----------------------
I was trying to be polite by not ruffling the sensitive pentax feathers....
To me, forced NR always kills details lah... so it is a minus to me. Other people may find that acceptable if the details lost is not significant.
-------------------------