I have an example of how amazingly low shadow noise is on the K-5 in my review:
http://enticingthelight.com/2010/12/01/review-pentax-k-5/#raw For those who don't want to click through, here is the evidence (warning, full-rez files): Image as shot: http://enticingthelight.com/wp-content/uploads/2010/12/IMGP2690.jpg Image recovered in ACR: http://enticingthelight.com/wp-content/uploads/2010/12/IMGP2690-recovered.jpg Same scene shot with proper settings: http://enticingthelight.com/wp-content/uploads/2010/12/IMGP2687.jpg Sony should get a Nobel prize for this sensor :-) —M. \/\/o/\/\ --> http://WorldOfMiserere.com http://EnticingTheLight.com A Quest for Photographic Enlightenment On 20 May 2012 16:23, Paul Stenquist <pnstenqu...@comcast.net> wrote: > You won't notice any difference in noise between the K5 and K20D at ISO 200 > in flat light. But when I have to use fill in conversion or brighten shadows > with dodging, I find that those actions will generate significant noise in > the K20 or K7 image but not in the K5 pic. And the few extra megapixels do > appear to add some detail resolution in K5 images. I think I can see it in > car pics that I shoot off tripods, but I haven't done any side by side, same > day tests. I'm not into pixel peeping. But I do know I'd never go back to the > K7 or K20. > Paul > > On May 20, 2012, at 12:08 PM, Bruce Walker wrote: > >> In the "Something to think about." thread I opined that the D800E was >> likely to be in my upward growth path for more useable resolution in >> the type of studio shooting I'm doing lately. A few kind PDMLers >> suggested that the K-5 might give me what I'm looking for and sent me >> some RAW and high-rez JPEGs to compare against. Thank you very much, >> Paul, Larry and Boris! >> >> I pulled all these images into Lightroom, made a gallery of them and >> some of my best in-studio (untouched) raws, closely examined eyes and >> eyebrows in full-body and head and shoulder portraits, and here's what >> I concluded. >> >> - at ISO 80 (K-5) and 100 (K-5 & K20D), the noise (or complete lack >> of) is indistinguishable between them. >> >> - in all cases, in full-body shots eyebrows are indistinct (read: >> fairly blurry smudges). No diff between K-5 and K20D. >> >> - in head & shoulders portraits, eyes and brows are sharp and >> well-resolved and it's very hard to say which is better, but I think >> the K-5 may have a very slight edge over the K20D. >> >> - the lenses being used make more difference than the two bodies. (No >> great surprise here.) And Boris's Sigma (whatever it is) is *sweet!* >> >> - I'll get more resolution improvement by simply using a tripod or >> monopod to shoot with rather then upgrading to a K-5. >> >> >> I also grabbed a few D800 head & shoulders portrait images from >> DPreview and compared. It's pretty clear that there's a large >> improvement in resolution, but it's also hard to see by how much. I'm >> convinced that the D800 shots were all done with a tripod, whereas all >> of mine and the loaners were hand-held. There is not an order of >> magnitude difference in resolution. There were no full-body, f8 or >> above, studio lighting D800 shots, so there was nothing for me to >> compare there. >> >> Final conclusion: for my work, K-5 isn't going to help much, if at >> all. Jury is still out on if D800E would really shake my world either. >> I need to investigate further -- probably rent one. I do have an >> acquaintance with one; maybe I can borrow that. >> >> The good news for me I don't feel so much disadvantaged by my >> 2008-vintage kit as I was beginning to. I'm still in the ballgame. :-) >> >> -- >> -bmw >> >> -- >> PDML Pentax-Discuss Mail List >> PDML@pdml.net >> http://pdml.net/mailman/listinfo/pdml_pdml.net >> to UNSUBSCRIBE from the PDML, please visit the link directly above and >> follow the directions. > > > -- > PDML Pentax-Discuss Mail List > PDML@pdml.net > http://pdml.net/mailman/listinfo/pdml_pdml.net > to UNSUBSCRIBE from the PDML, please visit the link directly above and follow > the directions. -- PDML Pentax-Discuss Mail List PDML@pdml.net http://pdml.net/mailman/listinfo/pdml_pdml.net to UNSUBSCRIBE from the PDML, please visit the link directly above and follow the directions.
