FYI - good high iso b&w from foveon sensor

Eric T said:
Recall that foveon pixels each have blue, green and red channels. Now randomize the order of the three color channels on each pixel. That is, blue channel would be on top in 1/3 of all pixels on average, red on top for 1/3, and green on top for 1/3.
Click to expand...

not sure this would work because according to one of the articles posted, http://www.foveon.com/article.php?a=67, the layers are ordered based on the finding that red, green, and blue penetrate silicone to different depths. so i don't think it's possible to randomize for all 3 layers. you could do it on the top layer but for example you can't put blue on the second or third levels because the blue light won't penetrate the silicone that far. but good try.
 
dfatty said:
not sure this would work because according to one of the articles posted, http://www.foveon.com/article.php?a=67, the layers are ordered based on the finding that red, green, and blue penetrate silicone to different depths. so i don't think it's possible to randomize for all 3 layers. you could do it on the top layer but for example you can't put blue on the second or third levels because the blue light won't penetrate the silicone that far. but good try.
Click to expand...

I couldn't find the article that states that light penetration varies with the color of the channel. Not sure that is the case. Even if it is true, it is a matter of degree. Amplification of the signal may vary for each channel in data processing (which already occurs with the X3 sensor).
The biggest problem I see with my proposed new generation of Foveon sensor is that the data processing required for each image would be HUGE. None of today's cameras could handle it.
 
Took a selfie today at 6400 and underexposed by a stop so effectively 12800! :eek:

Did all the things recommended and I have to say I like the look in black and white ... very 'noir' and although there is a little banding I'm fairly impressed. Definitely gives the little Merrill another string for its bow! :)


med_U5265I1371101919.SEQ.0.jpg
 
As matter of interest this how that same image looks when you open it in SPP as the original raw file. :eek:


SDIM0447p_zpsd4c24f9d.jpg
 
lynnb said:
Keith, I'm impressed with that processing!
Click to expand...


It's mainly a matter of removing all the bad information and concentrating on the good stuff that's left. The in camera processing engine combined with SPP really mutilates the files at these high ISOs.
 
:eek:

impressive...

i´ve done myself some tests, but i´m sticking to the 800 iso limit...although using the blue and a little green chanels really makes a huge difference...

BTW i have the ef-140 dg flash and it works really well!
 
Eric T said:
I couldn't find the article that states that light penetration varies with the color of the channel. Not sure that is the case. Even if it is true, it is a matter of degree.
Click to expand...

Yes, that is the case. Silicon (not silicone) is transparent to IR, red penetrates deeply, green not so much, blue not well at all. When I was in grad school, I had a professor (who earned his PhD in Carver Mead's lab) who encouraged me to investigate and design a similar imager as a course project, which I did. As I was using a commodity CMOS process, I was limited to a 2-layer stack instead of 3. Sadly, my lab partner was unwilling to commit to the testing of the IC which would have to continue on into the next semester, so we never fabbed the chip.

Using the penetration depth vs. wavelength characteristics of silicon means less control over the spectral width of the bandpasses (varying the layer thicknesses), and more importantly the steepness of the cutoffs. The penetration depth is a statistical distribution, not a "435 nm light penetrates exactly 0.35 micron, no more no less" exact value. I'm guessing that there's more to RAW conversion with the Foveon than just separating the colors; there's probably some post processing to better separate the primaries, but that's just a guess based on my planning for what I would have had to do for my 2-color version.

Sadly, these days I'm not involved in IC fab and this design project was getting on 10 years ago. Some of the specifics have faded, and of course the X3 is almost certainly more sophisticated than my design and with a process that can be tuned to their specific needs. However, I'm pretty sure ( insert winky smiley here) that there aren't color filters between the layers as in color film, and so randomizing which color is on top simply isn't an option using the Foveon process. For that matter, there really isn't going to be a silicon-based process that picks up blue photons at the bottom of a 3-pixel stack. Even without any color filter, those photons simply won't penetrate that deep.
And a thin silicon layer on top won't pick up many red photons, so the detection efficiency will be terrible.
 
Just discovered this thread. Very exciting ! Will try this out. I just wondered- have ben using Iridient instead of SPP. Does this work as well re: the B&W Monochrome for high ISO ? I'm not sure I'm having much success doing this. Only, I prefer Iridient for color work with the Sigma DP2M as it is quick , and I don't see the "beach ball" slow processing at all, compared to the SPP.
 
I don't have problems w/ spp giving me the beach ball...

In terms of iridient, so long as u have ability to control all three RGB channels separately it should work.

Gary
 
Here's something to report-
Picture of my dog, at ISO2000, converted, and noise corrected in Iridient. It took some exploring to figure out the options, but very pleased...


p1056780095-5.jpg
 
GaryLH said:
Yep...

It really shows what is really happening at the pixel level. The blue layer being the top color pixel needs very little signal amplification compared to the green or red layer below it. To get the equivelent signal amplification of the top most layer, each additional layer needs more amplification, thus introducing more noise w/ each additional layer.
Click to expand...

Old post Gary, but not quite right for the Foveon F13, F20 pre-Merrill sensors. If one opens their raw histogram in RawDigger, the levels are reasonably similar.

The Merrill sensor does get about aX2 multiplier in the red channel during conversion to RGB and the raw levels are indeed about half in the raw histogram.

GaryLH said:
If this had been a pure monochrom sensor, I would suspect one would get clean iso up to 6400.

Gary
Click to expand...

Yes, and the Foveon F13, F20 sensors also do well with high ISO at large apertures. Probably F21 (Quattro) too but I've never used one.
 
Last edited by a moderator:
xparUSA said:
The Merrill sensor does get about a X2 multiplication in the red channel during conversion to RGB and the raw levels are indeed about half in the raw histogram.
Click to expand...

Oops, I meant " ... and the red channel raw level is indeed about half in the raw histogram."
 
You must log in or register to reply here.
Back
Top