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 :)

Meet Charlie..

Jannard said:
2/3" Scarlet has 3.3u pixels, Mysterium-X S35 has 5.4u pixels and Monstro is 6.0u.

Jim
Click to expand...

Which is why I can't comprehend how 2/3" Scarlet performance, especially ISO-wise, can be comparable to the others.

But you say it is so, and you haven't disappointed me yet, so I have a savings account designated "2/3" Scarlet" all ready and waiting :)
 
ordinary english: we don't know yet.

ordinary english: we don't know yet.

Joseph Hutson said:
What exactly does this mean in ordinary english? :emote_couch:
Click to expand...

There is little you can say with certainty about image quality, knowing only the pixel size; it depends on the sensor technology, AA filter, etc. The answers about image quality will have to wait until production cameras are in use and you can examine the output.
 
Lee Saxon said:
Which is why I can't comprehend how 2/3" Scarlet performance, especially ISO-wise, can be comparable to the others.

But you say it is so, and you haven't disappointed me yet, so I have a savings account designated "2/3" Scarlet" all ready and waiting :)
Click to expand...

Don't forget the difference between the original Mysterium sensor and the Mysterium-X had nothing to do with pixel size, and everything to do with making the sensor work better. As such, smaller pixels may make a difference, but the M-X tech probably makes a much bigger difference between the 2/3" Scarlet and other 2/3" sensors we're used to.
 
Jannard said:
2/3" Scarlet has 3.3u pixels, Mysterium-X S35 has 5.4u pixels and Monstro is 6.0u.

Jim
Click to expand...
Whoops sorry guys... :blushing:
I really thought they were the same, just smaller sensor... Didn't do the math. Indeed, 2/3" sensor would probably be less than 2K if it had the same pixel size as the S35 sensor.

So that means they should perform differently in low light, right ?
 
http://www.eetimes.com/news/design/showArticle.jhtml?articleID=224000253&pgno=1

EE Times: Design News

Quantum film threatens to replace CMOS image chips



R. Colin Johnson
Page 1 of 2
EE Times
(03/22/2010 12:28 AM EDT)

PORTLAND, Ore. — Just as photographic film was mostly replaced by silicon image chips, now quantum film threats to replace the conventional CMOS image sensors in digital cameras. Made from materials similar to conventional film—a polymer with embedded particles—instead of silver grains like photographic film the embedded particles are quantum dots. Quantum films can image scenes with more pixel resolution, according to their inventors, InVisage Inc., offering four-times better sensitivity for ultra-high resolution sensors that are cheaper to manufacture.
"Many innovations are said to be revolutionary, but are really incremental changes. InVisage's quantum film, on the other hand, really is revolutionary," said Tom Hausken, director of photonics and compound semiconductors at Strategies Unlimited (Mountain View, Calif.) "Quantum dots have been a solution looking for a problem for several years, and InVisage has found a very significant problem they can solve."

According to Morry Marshall, vice president of strategic technologies at Semico Research Corp. (Phoenix), InVisage could have the next generation image sensor. "It gathers more light so you can either make a smaller image sensor for a less expensive cell phone camera, or you make a higher resolution sensor for high-end digital cameras," Marshall said. "It's a huge step forward and the market is also huge, so they will also need to overcome the problems facing any small company when trying to penetrate a large market."

The new semiconducting material was invented by Univeristy of Toronto professor Ted Sargent, who is now chief technology officer at InVisage. Sargent perfected a method of suspending lead-sulfide nanoparticles in a polymer matrix to form a new class of semiconducting polymer that Invisage has spent the last three years integrating into a standard CMOS process. Now it can paint quantum film atop a low-cost wafer that has the electrode array for super-dense high-pixel-count images, but without any of the expensive CMOS photodetectors that make up the bulk of conventional digital camera sensors.

"Our quantum film replaces the silicon used for image capture, but what we have really created here is a new semiconductor material," said Jess Lee, InVisage president and CEO. "Our quantum film even looks like photographic film—an opaque black material that we deposit right on the top layer of our image chip."
 
is not 4x better than best current silicon

is not 4x better than best current silicon

San Toro said:
http://www.eetimes.com/news/design/showArticle.jhtml?articleID=224000253&pgno=1

...offering four-times better sensitivity...
Click to expand...

I don't think so. It may be a revolutionary new material, but the best silicon sensors already achieve well over 50% peak quantum efficiency. See, for example http://www.kodak.com/global/plugins/acrobat/en/business/ISS/supportdocs/Quantum.pdf reporting 85% PQE.

There is no such thing as better than 100% quantum efficiency, so "4x better" needs some explanation.
 
Wired has this very interesting article on the creation of a new sensor using Quantum (vs silicon) that will allow for better light sensitivity. Created by InVisage Technologies, Quantum dot-based sensors won’t be more expensive than traditional CMOS-based sensors, which is nice to know. They should be out next year. In the headline, the author is saying in the headline that the sensor will promise wedding photos from your cameraphone.
http://www.wired.com/gadgetlab/2010/03/quantum-dot-images/
 
http://news.bbc.co.uk/1/hi/technology/8580372.stm
Page last updated at 15:11 GMT, Monday, 22 March 2010
E-mail this to a friend Printable version
'Quantum dots' to boost performance of mobile cameras


The sensor can be integrated into existing manufacturing methods
Tiny semiconductor particles known as "quantum dots" have been used in a sensor that could make for mobile phone cameras that outperform larger cousins.
A film made from these dots is more light-sensitive than existing approaches to camera sensors, according to its makers, Invisage.
That means that cameras made using the film need not be as large as some to achieve the same performance.
 
Lee Saxon said:
Well I am very pleased with this image, and think you guys are doing a great job. Maybe my expectation are just too low because I have a DSLR from 2005, but I'm satisfied.

One thing that's blowing my mind, assuming I don't epic fail at math (like usual) : This jpeg is 4mb. RC28 / 30fps = <1MB per frame and I'm sure higher quality. So that makes RC like a zillion times more efficient compression than JPEG. Pretty neat.
Click to expand...

It is VERY good - way better than anything else in many ways. However, to be fair, motion compressors are always more efficient per frame because they can look at previous frames to find redundancies and compress them down.
 
Rob Ruffo said:
However, to be fair, motion compressors are always more efficient per frame because they can look at previous frames to find redundancies and compress them down.
Click to expand...

Only if they are inter-frame compressors (i.e. temporal compression as well as spatial) - and as far as I know REDCODE is not an inter-frame compressor. Each frame is treated as its own entity.
 
Álex Montoya said:
I guess RED's modular system allows for a quantum dot sensor update if the technology really is worth it.
Click to expand...

I suppose so. After having read the article at dpreview.com re: quantum dots, it's not so exciting. Their "4x better" claim is based on comparing to a sensor which has 50% QE, and also 50% loss due to top-layer metal and other geometry factors on the sensor. Neither of these are true for modern DSLR chips (or, presumably, RED's sensors). So this is something of a straw-man comparison, in my opinion. On the other hand, it may well be 4x better than existing cell phone cameras.
 
"quantum dots" is nothing more than fancy name at this point. The technology is cool and definitely something to watch, but it's not that new. I'm not even sure if the group currently getting all the press is really technically the inventors or originators of using this tech for sensor implementations. It does seem that they are the first to commercially offer such a sensor product though. I'm very curious to see it in action, if it is indeed operational. Of all the hypothetical uses mentioned for "quantum dot" structures many years ago when they were first created, image sensors were one application mentioned by Bell Labs -- who created them. They have been pitching them for use in high-resolution imaging (medical and military, mostly) as well as for use in quantum and optical computing circuits since their inception.

The thing about quantum dots is they are a very tiny transistor that is spatially confined and they can redirect electrons along different dimensional axes. In terms of implementation, they're like carbon nanotubes or other structures that while are not hugely difficult to make in proper lab settings, would be a daunting task for mass production. The additional technology needed to really use along with them for specialized purposes has not really materialized yet. I like to bring up nanotubes as they have the ability to direct streams of electrons or photons and could serve for bridging large gaps in optical or quantum computing structures. For hypothetical computing purposes, quantum dots and nanotubes go hand in hand.

This quantum film thing is another small, but I think very important, evolutionary step in sensor technology. It's like the black silicon talk we had last year. Over time, such developments will filter through the industry and sensors will continue to improve and transition to new foundation technologies. CMOS was such an evolutionary concept on the horizon at one point too and will one day be obsoleted.

And if RED continues their current program, then I have no worries about being able to leverage new sensor technologies in the form of sensor upgrades and/or new camera brains as the new technologies present themselves.
 
Last edited:
You must log in or register to reply here.
Back
Top