Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Could Dec 3rd = Black Silicon?

Tom Lowe

Tom Lowe

Well-known member
Joined
Jan 5, 2007
Messages
8,520
Reaction score
1
Points
0
Jim says Dec 3rd will make Nov 13th look humble by comparison. How could that be? Many of us suspect some type of sensor breakthrough. But could it actually be the mythical Black Silicon? :ph34r: :w00t:

Could we see a Monstro capable of shooting ISO 200,000?!? :gun:
 
Reminds me of the monolith in Kubricks' 2001.

Yeah, I don't think that Jim is just hyping us up.

If he says that it's important, then something is really going on. :biggrin:

Let's wait and see...

Black Silicon coul really be something.


P.S. These forums are like a movie. :)
 
Black silicon is at least two years from becoming a functional reality. It takes significant capital and time to equip the foundries for new tech like that.
 
I don't know what's worse, Jim teasing, or Tom and David talking about light sensitivity that would turn a slow, mediocre 5.6 lens into a low-light, night lens. That would be too cool.

David, for some fun musing, if Red created a sensor that was highly sensitive to light (I don't know what the number would be, but make up some reasonable numbers)...

...what would that mean to your cinematography? How would it impact it?
 
It would make blackout material really expensive
 
It may all well be, but by fixing one thing, you create other problems. There are no known optical ND filters that would achieve good quality filtration at ND 10!

But, it would be a nice option for dedicated night vision sensor.
 
I for one am not as interested in the extreme end of light sensitivity unless it's for specific purposes. In general, what I seek is a sensor that allows me to increase the dynamic range when needed, with very little noise tradeoff. In other words give me more to work with utilizing the same lenses we have now, with the same tools used for controlling light. For example, if I'm doing a night shoot on a city street and I can't bring in a huge lighting package, I'd like the ability to change the ISO to 800 and still have a relatively clean image with highlights holding and shadows with clean detail. The more we can push the image without degrading it to the point of not being useable, the closer we get to the holy grail. I would be thrilled with a sensor rated at 500. That combined with Superspeeds and a decent lighting package would get me quite far I would think.
 
I was thinking the same thing Tom...either way, any "breakthrough" for the digital realm coming from RED is exciting. Sensitivity would be key at this point in the game.
 
Sensitivity and DR, for sure.

It's funny, I was looking over the dynamic range numbers for Canon and Nikon DSLRs the other day. They have barely changed in 5 years! The low-end models have like 11 stops, and the high-end recent cameras have like 11.5 stops. That is one area where massive improvements can be made, for sure.
 
Tom said:
But could it actually be the mythical Black Silicon? :ph34r: :w00t:

Could we see a Monstro capable of shooting ISO 200,000?!? :gun:
Click to expand...

I hate to rain on the parade, but *no* technology, not even Black Silicon, can improve sensitivity by even *one* stop. Current technology, e.g. cheap digicams, already detect 70% of the photons that hit the sensor. A 1/3 stop improvement would put them at 100%, and after that there is no further improvement possible because every single photon is being counted. You can't see light that isn't there. That last 1/3 stop will be very difficult, though, because it's caused by the color filter itself.

As all cameras have approached the same physical limits for QE, read noise has become the principal factor in low light photography. The D3 only increased sensitivity by 15% over the 1Ds Mark III and that's considered a significant jump. That is why RED's larger sensor sizes are so smart: since the limits of quantum efficiency are almost reached, increasing the sensor is the only way to increase total light gathering power, assuming aperture goes up with it. Just going to FF35 increases the number of photons by over a stop.

Luckily, there is lots of room for improvement in read noise and no theoretical boundaries like sensitivity. Of course, there are tons of practical limitations, which is why sensor designers have such a hard job. My point is that if Black Silicon will have any significant low light benefit, it will be due to the reduction of read noise, not an increase in sensitivity.

Dec 3rd could also be about full well capacity, some other dynamic range technology, frame rate related, or not related to sensors at all. I'm looking forward to it.
 
Lexicon said:
Here's some information about BS and sensitivity.
Click to expand...

SiOnyx said:
This phenomenon results in external quantum efficiencies of 10,000% and optical detectors with 100x improvement in photoresponse
Click to expand...

Their talk of 10,000% QE and 100x improvement is only referring to multiple electrons per photon. Counting the same photon 10 times doesn't mean you have 10 photons: the true sensitivity is still the same. However, the extra information will likely be useful for improving read noise.

100% QE is not just a good idea, it's the law.
 
Maybe I had better lower expectations here... significantly.

Jim
 
You must log in or register to reply here.
Back
Top