Revolutionary new sensor technologies on the way?

[sderdiarian]

I found both articles interesting, each describing potentially watershed advances in sensor performance and affordability. We'll see how much of this actually trickles down in the next few years, but thought I'd share these here, being an mFT (and 4/3's) user.

http://www.sciencedaily.com/releases/2013/06/130612133101.htm

http://www.sciencedaily.com/releases/2013/05/130530094624.htm

Thanks go to 43Rumors for bringing an interesting site to my attention.

 

[Aleo Veuliah]

It is time to replace the usual components of sensors, only this way they could achieve significant improvements.

Liked to read both articles, and I am also interested on the Panasonic and Fuji organic sensor technology recently announced, that we will see it on 2014.

 

[s_grins]

Thanks for posting!!!

This is very promising development that can revolutionize not only canera sensors, but all semiconductor industry.

Best regards

S.

 

[Detail Man]

Aleo Veuliah wrote:

It is time to replace the usual components of sensors, only this way they could achieve significant improvements.

Liked to read both articles, and I am also interested on the Panasonic and Fuji organic sensor technology recently announced, that we will see it on 2014.

We should all insist that our image-sensors are free-range, free from hormones and anti-biotics !

 

[timedrun]

I am feeling all defensive - why do,we need this new tech? Whats wrong with whatvwe have now? But when I date the digi cameras I use the oldest is 3 years. So so much for the nostalgia. Bring it on.

 

[eques]

Thanks for posting, this sounds very promising!

1000 times more sensitivity means about 10 EV, 5 times about 2,2 EV, right?

Peter.

 

[systemBuilder]

I worked for a MEMs start-up. Like many start-ups, it eventually became a shut-down. The problem with any sort of MEMs device is manufacturability : there is NONE, for 98% of all the things that can be done in the lab, there is no way to manufacture them. Mass producing a 14M sensor is about 13,999,999 times harder than doing a 1-pixel sensor in some lab where "anything goes".

Don't hold your breath!

 

[PC Wheeler]

"Revolutionary new xxxxx technologies" are always promising at this stage of the game. But statements like this

May 30, 2013 — Cameras fitted with a new revolutionary sensor will soon be able to take clear and sharp photos in dim conditions, thanks to a new image sensor invented at Nanyang Technological University (NTU).

need a liberally dose of Salt.

 

[Great Bustard]

sderdiarian wrote:

I found both articles interesting, each describing potentially watershed advances in sensor performance and affordability. We'll see how much of this actually trickles down in the next few years, but thought I'd share these here, being an mFT (and 4/3's) user.

http://www.sciencedaily.com/releases/2013/06/130612133101.htm

http://www.sciencedaily.com/releases/2013/05/130530094624.htm

Thanks go to 43Rumors for bringing an interesting site to my attention.

Modern sensors have a QE of around 50% (at least in the green channel). That means that they record half the light that falls on them. So, the first link's claim of 5x more sensitive and the second link's claim of 1000x more sensitive seem, well, problematic, to say the least.

However, in very low light, the read noise (the noise added by the sensor and supporting hardware) becomes significant. Currently, pixels on modern sensors have a read noise of 2 electrons at the higher ISOs. So, when the light is so low that 8 photons fall on a pixel, on average, the read noise matches the photon noise. For less light, the read noise is dominant.

Thus, there is significant room for improvement in read noise, which has the added benefit of increasing the DR. However, the flip side is that the photon noise will still be ridiculously high with low photon counts, so even with huge improvements in read noise, the photo will still be very noisy in low light.

So, I'm not sure what they are referring to when they say "sensitivity".

 

[Detail Man]

Great Bustard wrote:

sderdiarian wrote:

I found both articles interesting, each describing potentially watershed advances in sensor performance and affordability. We'll see how much of this actually trickles down in the next few years, but thought I'd share these here, being an mFT (and 4/3's) user.

Regarding:

http://www.sciencedaily.com/releases/2013/06/130612133101.htm

The Nature article abstract is here:

http://www.nature.com/nnano/journal/vaop/ncurrent/full/nnano.2013.100.html

This site allows a bit more of a peek into the first page of the article:

http://www.readcube.com/articles/10.1038/nnano.2013.100

... where the authors merely state that present results are "comparable" with Silicon detectors:

... the photoresponsivity of previously demonstrated
mono- and multilayer transistors is relatively low, with the first
monolayer MoS2 phototransistors exhibiting a photoresponsivity
of 7.5 mA / W, which is comparable to graphene-based
devices with a similar geometry (6.1 mA / W). Recently reported
multilayer MoS2 devices show higher photoresponsivities, on
the order of 100 mA / W, which is comparable to silicon-based
photodetectors.

Regarding:

http://www.sciencedaily.com/releases/2013/05/130530094624.htm

Joseph S Wisniewski says: "Debunked years ago" here:

http://www.dpreview.com/forums/post/51561544

Thanks go to 43Rumors for bringing an interesting site to my attention.

Yes, thanks a lot, 43 Rumors, for not having a discriminating brain in their heads ...

Modern sensors have a QE of around 50% (at least in the green channel). That means that they record half the light that falls on them. So, the first link's claim of 5x more sensitive and the second link's claim of 1000x more sensitive seem, well, problematic, to say the least.

However, in very low light, the read noise (the noise added by the sensor and supporting hardware) becomes significant. Currently, pixels on modern sensors have a read noise of 2 electrons at the higher ISOs. So, when the light is so low that 8 photons fall on a pixel, on average, the read noise matches the photon noise. For less light, the read noise is dominant.

Thus, there is significant room for improvement in read noise, which has the added benefit of increasing the DR. However, the flip side is that the photon noise will still be ridiculously high with low photon counts, so even with huge improvements in read noise, the photo will still be very noisy in low light.

So, I'm not sure what they are referring to when they say "sensitivity".

 

[Jorginho]

Yes, thanks a lot, 43 Rumors, for not having a discriminating brain in their heads ...

It is called "43 rumors" and not "43 facts" for a reason.

 

[Detail Man]

Jorginho wrote:

Yes, thanks a lot, 43 Rumors, for not having a discriminating brain in their heads ...

It is called "43 rumors" and not "43 facts" for a reason.

Maybe they picked "43" because they assumed that it equals the average IQ of their audience ?

 

[Joe Pineapples]

The 1000x was with respect to the previous graphene sensor, but it was mis-quoted by a lazy journalist and spread all over the internet, even an IEEE magazine...

J.

 

[Joe Pineapples]

Detail Man wrote:

Aleo Veuliah wrote:

It is time to replace the usual components of sensors, only this way they could achieve significant improvements.

Liked to read both articles, and I am also interested on the Panasonic and Fuji organic sensor technology recently announced, that we will see it on 2014.

We should all insist that our image-sensors are free-range, free from hormones and anti-biotics !

What you thought was dust turns out to be sensor-poop!

J.

 

[oeoek]

I find very little left to wish for with my present sensors. Bigger dynamic range would only result in flatter pictures, like quite a few HDR pictures already show. Right now I have ISO 6400 in a usable quality, but I find myself using 3200 at most. A few stops more might be nice, but not that crucial to me. A non-forced low ISO might be more interesting to me (really low, like ISO 6 or ISO 9), but is probably way to specialistic for common use (just like extreme high ISO's).

All in all I feel that good enough is pretty much reached, the rest is just for tech lovers, not for photography.

 

[Jorginho]

Detail Man wrote:

Jorginho wrote:

Yes, thanks a lot, 43 Rumors, for not having a discriminating brain in their heads ...

It is called "43 rumors" and not "43 facts" for a reason.

Maybe they picked "43" because they assumed that it equals the average IQ of their audience ?

I guess if so, you are not aware of it ;-)

 

[Pedagydusz]

Most people who write "I am not interested in high ISO", just are not aware of what it would mean to have really high sensitivity at very good quality.

Would you not be interested in taking indoors pictures of people, runnig children, pets, whatever, without even bothering about "the ISO"? Just setting the exposure the way you want it (i.e.. aperture and shutter, to take the desired DoF and/or motion-freeze settings) and getting it always?

Not to mention, for those who want that, things like indoor sports, shows, lectures, etc.?

It may never be possible, but wouldn't it be nice?

 

[bgalb]

I suppose you haven't heard of the Linked Quantum Sensor. Every Qbit cluster is a pixel. Effectively 100% efficient at the image record level. Nearly zero read noise. HUGE megapixel potential. All random noise is absorbed in the absolute zero heat sink.

Quantum entanglement allows an image recorded on Mars to be instantly present in the linked sensor on Earth. It's going to be BIG.

Record your images on the roll and they are instantly present on your home recording system. No cell or WiFi necessary. The biggest current problem is to miniaturize the absolute zero control system and to develop compact storage that can store and transfer gigapixel images at high speed. We are looking at a system that the average bear could lift and point (with difficulty) in the next 10 years or so. Not much worse than current SLR's :-D

 

[sderdiarian]

Pedagydusz wrote:

Most people who write "I am not interested in high ISO", just are not aware of what it would mean to have really high sensitivity at very good quality.

Would you not be interested in taking indoors pictures of people, runnig children, pets, whatever, without even bothering about "the ISO"? Just setting the exposure the way you want it (i.e.. aperture and shutter, to take the desired DoF and/or motion-freeze settings) and getting it always?

Not to mention, for those who want that, things like indoor sports, shows, lectures, etc.?

It may never be possible, but wouldn't it be nice?

--
Antonio
http://ferrer.smugmug.com/

I agree, it will be nice and I'm sure we'll get there soon enough.

I enjoy photography as a hobby, as I have for over 30 years, and anything that further reduces the technical hurdles between what I see/compose and the resulting image will always be appreciated, naysayers aside (there will always be naysayers and sloggers, as I'm sure the first tool using hominid encountered).

Not being in the sensor design field (as a landscape architect I try to stick to design of public and private living environments, where I am in my comfort zone ;-) ), I found both articles intriguing. I also appreciated the responses that provided links to counter points that helped advance my understanding, limited though it naturally may be.

As for the classless response to my OP by DM, really not worthy of a response. This behavior unfortunately happens too often on this forum and I'm sure sends the more reasonable among us to other sites to have a civil dialogue relating to photography.

Unfortunately, I'm not "reasonable", so I'm here to stay!

On with Donna and my vacation on Martha's Vineyard, the weather has been fabulous and the range of nature sanctuaries and wonderfully kept period homes in Edgartown a photo enthusiast's dream.

--
Sailin' Steve

 

[Detail Man]

sderdiarian wrote:

As for the classless response to my OP by DM, really not worthy of a response. This behavior unfortunately happens too often on this forum and I'm sure sends the more reasonable among us to other sites to have a civil dialogue relating to photography.

Unfortunately, I'm not "reasonable", so I'm here to stay!

Sailin' Steve

It appears that my taking around 15 minutes using a search engine to discover a paucity of substance to such quasi-journalistic displays may not have been the kind of information that enchants the imagination. I suppose that "43 rumors" may perform a valuable service, after all ?

The idea of pursuing tangles of thought to what may result in unravelments can be troublesome.

From ignorance our comfort flows, the only wretched are the wise.
- Matthew Prior