Discussion in 'Micro 4/3 News and Rumors' started by cuervo.taylor, Jun 1, 2013.
New Camera Sensor 1000x More Sensitive Than Current Sensors
Sounds similiar to what Canon is working on. The demo they did earlier this year was insane, the portrait lit by an incense stick.
amazing heres the link:
[ame="https://www.youtube.com/watch?v=9weIR32z2ls"]Canon super-low-light sensor - YouTube[/ame]
Wonder if I can retrofit one to my OMD.
Back illuminated deep depletion CCD have quantum efficiency of >85% from 250-1100nm so how exactly does one get 1000x more sensitive than 85% ?
I see an amazing amount of references to each other on the web but I do not see an actual peer reviewed paper from a reputable journal ... pardon me if I restrain my enthusiasm.
I searched the supposed source in Nature Communications but cannot find the paper. Usually a direct reference is given ....
Maybe this? http://www.nature.com/ncomms/journal/v4/n5/full/ncomms2830.html
Two primary factors to consider are the material cross section and the responsivity - cross section means basically how effective the material "grabs" a photon and responsivity is essentially how much current or charge can be generated by a photon interaction. Think of photomultiplier tube and the cascade of electrons that can be triggered by a single photon resulting in thousands of electrons.
The third factor is how much noise the material /device imposes - not much use for a high cross section and high responsivity with similarly high noise. I'm not going to pony up $32 for the paper to see if these items are addressed - just having ~1000 responsivity is a long way from low noise ISO 256000 high dynamic range imaging sensor (which is the implication). Monolayer grphene has a very small cross section (means most photons pass right through).
Here is an example of the limited cross section of graphene and the kind of efforts needed to realize something marginally useful. http://www.nature.com/ncomms/journal/v2/n12/full/ncomms1589.html
the moon shot in that video was really sick
Rob, are you saying that claims of 1000x improvement in sensitivity are theoretically impossible?, or just with CCD technology?
Take a look at the comments HERE. It looks like what they've actually produced is a sensor 1000x more sensitive that previous graphene sensors - which doesn't say anything about how it compares to sensors in current cameras.
I noticed a reference to this new sensor tech. in a Dutch paper. I did not think about how plausible it was (but yes, often these kind of press releases are highly inaccurate). It just made me wonder whether I had any use for such a high ISO sensor. Not much, really. I now have a usable 6400 and almost never have to use it. Ok, I use primes and have f1.8 and 2 at hand. But even with zooms, anything over 25000 seems reserved for rather specific use.
For every day use, I would prefer sensor development that would cover a wider range of ISO's; starting at ISO 6 going way up to ISO 25000 would cover almost anything without reverting to dark ND filters.
It a bit like the race for higher dynamic range; by now many sensors easily have a range of 11-12 stops. Printing paper has, at best, a dynamic range of 6 or 7 stops. As a result, I see many pictures getting flat from squeezing to many dynamic range into to little printing range. Viewed on a screen it is a little better, but not much.
At some point, improvement is more theoretical then real...
Yes the other end would be priority too, how about really low iso so you can use big aperture a sunny day
Skickat från min LT26i via Tapatalk 2
I'm not saying it is impossible. Just there was more to it than responsivity.
Consider two materials one has responsivity 1.0 and cross section 1000.0 and the other materials has responsivity 1000.0 and cross section of 1.0. Under the same photon flux density both materials produce the same signal.
Unfortunately it's not really applicable for our digital sensors, as noted on DPR.
Indoor non-flash photography would benefit greatly from increased sensitivity sensors - particularly in windowless rooms.
Maybe I just shoot in odd places
Separate names with a comma.