Sigma/Foveon Questions
in [Digital Cameras]
|
Prev: SI Facescape
Next: FF camera with mirrorless design
From: Chris Malcolm on 19 Apr 2010 14:38 nospam <nospam(a)nospam.invalid> wrote: > In article <MPG.2635a0118e4698e198c2a2(a)news.supernews.com>, Alfred > Molon <alfred_molon(a)yahoo.com> wrote: >> In article <180420100922201916%nospam(a)nospam.invalid>, nospam says... >> >> > wrong. every pixel captures luminance. >> >> No. In a Bayer sensor every pixel captures either the red, green or blue >> channel. This is NOT luminance. > it's one component of the luminance, with the remaining two later > calculated. >> > > To properly capture luminance at each pixel you need the full colour >> > > information at each pixel. >> > >> > wrong, as bayer has proven. >> >> No - you are wrong. > so all of the zillions of photos that very accurately reproduce the > subject all have completely bogus luminance? how can that be? Because it's not as black and white an issue as you seem to think :-) It's not a case of having perfect luminance information or completely bogus information. It's a question of exactly what proportion of theoretically perfect complete information can be captured by the two different methods. -- Chris Malcolm
From: nospam on 19 Apr 2010 15:20 In article <MPG.2636bbab4140b21698c2a4(a)news.supernews.com>, Alfred Molon <alfred_molon(a)yahoo.com> wrote: > Luminance requires all *three* colour components. If you do not capture > all three colour components at each pixel, you do not capture luminance > at each pixel. you may not capture full luminance but you do capture enough information to calculate the correct value. the system works. > > > The accuracy is not that high - there are some errors which have the > > > effect of reducing the effective resolution. > > > > it's actually very high, that's why photos look as good as they do. > > > > can you point to a unbiased test (i.e., not from foveon) that shows > > otherwise? > > Ever heard of colour aliasing (to make one example)? ever heard of luminance aliasing? by not having an anti-alias filter, aliasing with foveon worse. plus, false colour due to aliasing on bayer is very rare these days anyway.
From: nospam on 19 Apr 2010 15:22 In article <833m9nF5onU1(a)mid.individual.net>, Chris Malcolm <cam(a)holyrood.ed.ac.uk> wrote: > It's not a case of having perfect luminance information or completely > bogus information. It's a question of exactly what proportion of > theoretically perfect complete information can be captured by the two > different methods. and that's where bayer wins, by quite a bit.
From: Ray Fischer on 19 Apr 2010 22:23 Alfred Molon <alfred_molon(a)yahoo.com> wrote: >In article <180420102251244632%nospam(a)nospam.invalid>, nospam says... >> > > > > wrong. every pixel captures luminance. >> > > > >> > > > No. In a Bayer sensor every pixel captures either the red, green or blue >> > > > channel. This is NOT luminance. >> > > >> > > it's one component of the luminance, with the remaining two later >> > > calculated. >> > >> > So you agree that in a Bayer sensor the luminance is not captured in >> > each pixel. >> >> don't twist what i said. luminance is measured at every pixel, even >> though only one component is actually captured. > >Luminance requires all *three* colour components. Luminance requires no color at all. By definition is is only the brightness of the pixel. > If you do not capture >all three colour components at each pixel, you do not capture luminance >at each pixel. And yet images captured with Bayer sensors have color AND luminance information in them. Your claims contradict reality. -- Ray Fischer rfischer(a)sonic.net
From: Ray Fischer on 19 Apr 2010 22:24
nospam <nospam(a)nospam.invalid> wrote: >In article <4bcb8f32$0$1651$742ec2ed(a)news.sonic.net>, Ray Fischer ><rfischer(a)sonic.net> wrote: > >> >> >it's certainly not 1/3rd or whatever other silly math the foveon >> >> >fanbois claim. >> >> >> >> It is actually not too far from 1/3. >> > >> >it's approximately the same. both bayer and a foveon sensor with the >> >same number of pixels will resolve approximately the same, roughly >> >70-80% of nyquist. >> >> Tests show otherwise. > >which ones? Look at DPReview's discussion of Sigma's cameras. I can't provide the link right now. -- Ray Fischer rfischer(a)sonic.net |