Archived posting to the Leica Users Group, 2008/01/31
[Author Prev] [Author Next] [Thread Prev] [Thread Next] [Author Index] [Topic Index] [Home] [Search]Hmmm... Somewhere along the line "pixel" has replaced "photosites" in sensor specs. Kodak and Sigma (and probably all the other sensor manufacturers) do this now. The Sigma spec, however, leaving off the X3 claim they make to compare their sensor to others, is closer to the truth when you want to know the potential resolution of the sensor. The number of resolution elements in the Sigma sensor is 2460 x 1768 or 4.6 million elements. Kodak doesn't make a 14MP sensor, but their 10 megapixel KAF-10010 has 3876 x 2584 active pixels or, more accurately, photosites. The number of resolution elements in this sensor is about half that,* or 1938 x 1292. Scaling this up to a 14 megapixel array, makes it 2713 x 1292, or 4.9 resolution elements. So the side-by-side (planar) photosite sensors and stacked-photosite sensors have about the same number of resolution elements. Looks like Sigma has a point. Pixel used to mean resolution element, now it means photosites. I wish people would stop changing the meaning of words without telling me. :-) -------------------- *It's really much more complicated that this, but since half the photosites in the Bayer array are filtered for green light they should provide the great majority of the signal that defines resolution. Regards, Dick On Jan 31, 2008, at 1:25 PM, Meino de Graaf wrote: > See http://www.steves-digicams.com/techcorner/aug_2007.html > > Regards > Meino > > Didier Ludwig wrote: >> Yes. Sigma counts R and G and B values for each pixel. This way my >> R-D1 would be a 18 megapix camera. Gorgeous! >> :-) >> Didier >>> Remember, too, this is really a 4.6 megapixel camera (2640 x >>> 1760), despite Sigma's claims. >>> Regards, >>> >>> Dick >>> >>> >>> >>> On Jan 31, 2008, at 3:29 AM, Didier Ludwig wrote: >>> >>>> Mehrdad >>>> >>>> A pocketable travel camera with a fixed wide angle lens is >>>> interesting in any case, but there are some issues with the >>>> Sigma DP-1: >>>> >>>> - slow lens: f4 is disappointing for such a concept >>>> - partially outdated specs since the camera has been announced >>>> one and a half year ago (sept 2006) >>>> - and unfortunately, Sigma could not take much advantage in terms >>>> of high ISO behaviour out of this foveon sensor, which is also >>>> used in the already released Sigma SD14 dslr. There's visible >>>> noise already at ISO200, see http://tinyurl.com/33fh8t >>>> (frontside of speakers desk), but my impression is it's slightly >>>> better than the Ricoh GR- D. The fact it does not offer higher >>>> ISO than 800 looks like Sigma admits 1600 would be useless. >>>> >>>> The main advantage of the foveon sensor seems to be the better >>>> dynamic range. Btw this sensor has not DX format, but is smaller >>>> (1.7 crop factor not 1.5) >>>> >>>> Practical tests, handling and 1:1 comparisons with the GR-D II >>>> will tell us more. Meanwhile, I pack my R-D1 with a CV wide >>>> >>>> Didier >>>> >>>> >>>> >>>>> is the sigma DP-1 camera >>>>> http://www.dpreview.com/news/0801/08013108sigmadp1.asp >>>>> dx size cmos in a pocket camera!! >> _______________________________________________ >> Leica Users Group. >> See http://leica-users.org/mailman/listinfo/lug for more information > > _______________________________________________ > Leica Users Group. > See http://leica-users.org/mailman/listinfo/lug for more information