Originally posted by normhead And the K-3 is about a stop less than the K-5 or K-1 or K-P. You're picking on the weak link.
The K-3 was delayed 6 months while they tried to get the noise and cross talk under control, and they still never made it back to K-5 spec.
Originally posted by luftfluss People say this, but personally I've not found it to be the case. When I downsize a 24MP K-3 II pic to match my 16MP K-5 II, I find the noise performance pretty much the same.
Viewed side-by-side at 100% reproduction, I'd agree that my K-3 images are just a little bit noisier than those from my K-5... But down-sampled to the same physical dimensions, I don't see a significant difference between the two (not saying it isn't there - just that I don't see it as significant).
Member @kh1234567890 did some interesting analysis into K-5 and K-3 images using RawDigger (see
this post and responses from me over the next couple of pages). It seems there's mandatory noise reduction applied to the K-5's raw files from ISO 1250 - which is interesting, because I recall our site review of the K-3 mentioning that it was a bit noisier than the K-5 after ISO 800. The K-3 also applies mandatory noise reduction to raw files, but not until ISO 3200.
Going back to the Canon SL1, the reason I was fairly impressed (not "blown away", understand... just "pleasantly surprised") was that it made a decent fist of things when compared to the much higher-end K-3. I hadn't expected that at all...