Am I oversharpening
in [Digital Cameras]
|
From: JPS on 12 Jun 2005 10:46 In message <1118529983.251922.197830(a)f14g2000cwb.googlegroups.com>, george_preddy(a)yahoo.com wrote: > > >Tetractys wrote: >> George Preddy wrote: >> >> > In general, if you are sharpening at all then >> > you are oversharpening. >> >> This is an untrue statement. Sharpening is part >> of any proper digital workflow. > >"Bayer" images require artificial sharpening, digital does not. Bayer and Sigma are both digital. You've never seen X3F RAW data, have you? Hint: SPP does significant sharpening, and gets a boost from the aliasing. -- <>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS(a)no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
From: george_preddy on 12 Jun 2005 19:04 Celtic Boar wrote: > Please take a look at the attached link. I am still trying to get the hang > of this Unsharp Mask Thing. Are these oversharpened. > > Canon 20D - Raw - 75-300 IS Zoom Focus rings exist because blurry can't be fixed properly after the shutter closes. If your camera's images aren't sharp enough and you think you need to sharpen them, enhancing edge contrast without regard to lens optics will only make it look worse, all things considered. You've discovered one of the many reasons so many top photographers reject (Bayer) digital as too low end, even the most expensive stuff, which is every bit as bad as the cheaper stuff. In plenty of ways, worse.
From: KennyJr on 12 Jun 2005 20:51 In article <1118617451.162970.96070(a)g43g2000cwa.googlegroups.com>, george_preddy(a)yahoo.com says... > You've discovered one of the many reasons so many top photographers > reject (Bayer) digital as too low end, even the most expensive stuff, > which is every bit as bad as the cheaper stuff. In plenty of ways, > worse. > I've spent the last several days reading up on the foveon image sensor and I've got to say that I like the idea. One pixel site reading all three colors. It's a good idea with a lot of promise. I've also spent the last several days looking at pictures from Sigma SD9 and SD10 cameras. So far I haven't been impressed. I don't know if the problem is with the sensor or with the firmware in the camera. Hopefully Foveon and/or Sigma will work the bugs out of the technology. I think once they work the bugs out and start offering lower prices on cameras with foveon sensors they'll start cutting into the market now dominated by cameras using Bayer sensors. Until that time though I'll stick with cameras that use the Bayer sensor. ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==---- http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups ----= East and West-Coast Server Farms - Total Privacy via Encryption =----
From: Ray Fischer on 13 Jun 2005 00:56 KennyJr <kennyjr(a)NOSPAM.floodcity.net> wrote: >In article <1118617451.162970.96070(a)g43g2000cwa.googlegroups.com>, >george_preddy(a)yahoo.com says... >> You've discovered one of the many reasons so many top photographers >> reject (Bayer) digital as too low end, even the most expensive stuff, >> which is every bit as bad as the cheaper stuff. In plenty of ways, >> worse. >> > >I've spent the last several days reading up on the foveon image sensor >and I've got to say that I like the idea. One pixel site reading all >three colors. It's a good idea with a lot of promise. > >I've also spent the last several days looking at pictures from Sigma SD9 >and SD10 cameras. So far I haven't been impressed. I don't know if the >problem is with the sensor or with the firmware in the camera. The problem is that the different layers don't discriminate very well between the different colors. The result is that you have three channels that have almost the same content and the software has to magnify very small differences. One effect is that colors can shift in rather odd ways. > Hopefully >Foveon and/or Sigma will work the bugs out of the technology. It's been a while. Progress hasn't been impressive. >I think once they work the bugs out and start offering lower prices on >cameras with foveon sensors they'll start cutting into the market now >dominated by cameras using Bayer sensors. I think that they'll get squashed by the big players. -- Ray Fischer rfischer(a)sonic.net
From: Ken Tough on 13 Jun 2005 10:23
KennyJr <kennyjr(a)NOSPAM.floodcity.net> wrote: >I've spent the last several days reading up on the foveon image sensor >and I've got to say that I like the idea. One pixel site reading all >three colors. It's a good idea with a lot of promise. From an engineering point-of-view, I don't agree. I know this has been rehashed ad infinitem, but spatially there are fewer sampling sites per unit area on the foveon sensor. The Bayer technique cleverly uses the colour sensors separately to increase the density of luminosity sampling. Using the properties of real- world objects/images this is an excellent compromise. If you think about it, given the same number of R+G+B sensors, whether the RGB are stacked vertically or not, there is the same number of, say, red sensors per unit area for either the foveon or the Bayer. So if there is a (missed) red transition between red sensors on Bayer, it will equally fall between foveon sensors (or be averaged between two, meaning it is missed). Where the Bayer gains luminosity resolution from its technique, the foveon only gains registration between the colours in its technique. If real- world images were colour noise with a 1 pixel/sensor spatial frequency, then the foveon would win out. But that's not the case, as the principles of JPG compression clearly show! Real-world objects and images have generally predictable properties to their colour distribution; it isn't just random noise. So the vast (9x) gain in luminosity resolution on the Bayer by far outweighs the opposing gain in terms of colour registration. This engineering benefit also has to be countered by the difficulty in making the sensor, and the compromises due to the poor optical results in making a stacked sensor. (Perfecting one aspect of a picture is not worth it if it reduces the quality of another aspect by almost an order of magnitude). All-in-all, it accounts for the fact that foveon sensors are far more expensive to build for a given quality of image, and have a trivial market share and will most likely disappear in the next 5 years. >I've also spent the last several days looking at pictures from Sigma SD9 >and SD10 cameras. So far I haven't been impressed. I don't know if the >problem is with the sensor or with the firmware in the camera. Hopefully >Foveon and/or Sigma will work the bugs out of the technology. > >I think once they work the bugs out and start offering lower prices on >cameras with foveon sensors they'll start cutting into the market now >dominated by cameras using Bayer sensors. Until that time though I'll >stick with cameras that use the Bayer sensor. -- Ken Tough |