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 :)

New project...

Ha, somebody hadn't offered me that job. Anyway, I came up with a new way of doing that from writing it (designs just assemble themselves in my head) but seriously, the layered sensor is good as long as you can apply suitable low light and HDR technologies to it (the issue might be the vertical parralel sampling can interfere with this). However, I don't want to think about it, the working out comes out too much and I have other things to deal with. They should just buy Sigma, get the technology and develop a line of consumer Sigma cameras and lens. I imagine Sigma would be.most grateful.

I'm getting a bit worried about this, starting in the lsst year, around 3 to 7 times I've had occasions where the ideas assemble themselves so fast I get a small instant headache, like I've busted a micro blood vessel or something (I've had some of those things show up on MRI in the past). So I'm cracking down on high rate geometric assembly thinking.
 
Michael J Brennan said:
Something to process and display 10k motion picture images with ease....

Mike Brennan
Click to expand...

Yoo hoo! (Sorry an Australua joke, look up Barry Humpries). I imagine they are wasting their time with their current processing stratergy. Honestly, this is probably the case for over 90% of the processing strategies out there, because they will all be over taken with something better, so most.

There are a handful of sttatergies, one is a comlkeye custom circuit, where everything has it's own subcircuit to speed things up. However you duplicate this in parallel and the circuit gets big in parallel, and if you don't know what you are doing, extra energy drain and heat.

The second is to use an array of processing elements under control of a higher order processing unit. This can be more compact etc. (Sorry brain is going off for the night).

The third is to use a series of full processors. This issuallt is slow, and very large, but more flexible. So a hybrid strategy is used. On GPU's they integrate CPU like functionality these days, and custom circuits of the previous two above to more economically perform some codecs.

However, suppose you can produce a CPU with data processing small, low powered enough and fast enough to compete with normal custom circuites above.

Now, the other stratergy is a shift in technology the processing is constructed upon. Spintronics, cascade, are some proposals. But closer to home I was reading of one mainline silicon replacement, but due to the many years and my memory, I can't remember. However, twice the speed per node size. Now, silicon is fast coming up to 7nm, where shrinkage will result in higher power consumption per extra clock. Hiwefver, I don't know what this limit on the other material is, it should be double the speed of silicon. However, now we are talking advantage with carbon nano-tubes becoming a reality.

Now, between new materials and spin tronics etc, is magnetic computing, on particular Magnetic Quantum For Automata. Quantum dot is available on other technologies too. The problem is clocking tje signal to stabilise the signal greatly slowing the circuit down, below silicon (at the moment I don't know where the speed difference lays). But the power the circuit can operate us .such smaller. Speed is not everything, as long as you can fit in enough parallel processing power. Now, size is not such an issue, it costs more and normally gets less and less manageable in the increased heat. But here, there is not much heat. Now, say an MQDCA FPGA (they have been developed) could do this. Now, if they can produce a blank slate approach, the volume increases their side, cheap deals the manufacturer's side, that can translate circuits across too. So, costs may become affordable for a switch, and a lot more than 10k possible (which is should be doable high speed). But 16k-100k/200k is the holographic future (and let's say that this can do that (but surround is 600k, and double again to get extra precision). So, a long way to go.

Now, getting a speed limited technology to process a progressively larger datarate in parallel is an issue. If the interface, for instance, outstrips the technologies ability to ingress it at least (plus you are talking about compare for compression the image/s). Now, you can take a bit in per bit of pixel in psrrallel, but.unless you plan bit level/plane compression (which most don't) its a bit useless. Interfaces will come down to multiple data channels per interface channel. However, ultimately you can move something like QDCA behind the sensor, taking pixel data directly from pixels, I'll stop there. So, then you can do more.
 
Hrvoje Simic said:
I applaud the creative thinking but you are interpreting the idea and the invention as the same thing.
Click to expand...

The invention is how the idea works. Which you are neglecting to think about.

I'm not an inventor of ideas, but of mechanisms and finale.mechanisms Hrvoje.
 
Wayne Morellini said:
Invented that. That is how I was going to do my home made medium format sensor. However, you can do it with a layer sensor. In matter of fact you can do a lit with layered sensors, like a very accurate 7 color design, for use in science, industrial and security too. I was looking at doing that.
Click to expand...

:thumbsup:
 
Anyway, the home made thing was noval technology that could be garage tested etc. But as usual, the advances in sensor technology elsewhere makes it note irrelevant. This is a lot of research just simply becomes uneconomical before it gets out the door.
 
Wayne Morellini said:
They should just buy Sigma, get the technology and develop a line of consumer Sigma cameras and lens. I imagine Sigma would be.most grateful.
Click to expand...
Sigma had 333 million dollars of sales in 2014. Not sure what would be the valuation. Do you think RED is big enough to swallow that sized a prey?
 
Then buy their camera and sensor business.
 
Sabyasachi Patra said:
Sigma had 333 million dollars of sales in 2014. Not sure what would be the valuation. Do you think RED is big enough to swallow that sized a prey?
Click to expand...

Leveraged, I don't see why not.

On a side note, and probably unrelated, Elon Musk is using the name Red Dragon for the Mars launches...
 
Last edited:
Hmm, trademark might not apply on Mars :)

Seriously, they are going to have to address laws in consumer space travel, and setteling, also maritime, and cartage law. Occasionally much too much power is given to the wrong people.

Much too much power is given at times to potentially the wrong people.
 
How far does trademarks go these days? They used to get very particular in the old days, even on things remotely worded similar. Having Axoim advertised here, and the Axion canwea is very confusing, as with Silicon image and silicon imaging, and so other thing around here that has same name, different industry. Can you imagine how Steve would react in the old days.
 
Wayne Morellini said:
How far does trademarks go these days? They used to get very particular in the old days, even on things remotely worded similar. Having Axoim advertised here, and the Axion canwea is very confusing, as with Silicon image and silicon imaging, and so other thing around here that has same name, different industry. Can you imagine how Steve would react in the old days.
Click to expand...
Not sure about laws, but at least with Google advertising, as long as the tradermark wasn't in the same industry, it's fair game. (A bit, may get flagged, and have to prove not in same industry).
 
I can't remember, but there were issues so they expanded industry categories, but I forget if they were there before. I'm OK with somebody say have a unique name trademarked accriss all industries (as is the potentually multiple industry nature of business today). But not on common descriptive of functional words, like color, or digging. Then only the unique branding art work associated in combination with the word should be patentable. Makes a good logical division. And no trademark should be able to be picked up after expiry by another company for like 10 years, maybe five. Stops a lot of trademark stealing, and standover tactics the day the trademark expires because somebody forgot to pay it in the chain of command. The company csnbe notified and given x months grace, but the new payment backdated on the end if the old payment. The company past that would not legally (by new law) be allowed to use the trademark without renewing it, or pronouncing it common. As usual the artwork, including sentences, in a trademark would he subject to common copyright law. Which they should do anyway, with trademark purely applied to the unuque words/nonunuque words/artwork combination.
 
Jim, we know you are having a good time changing the world. Just checking for a pulse on the new project. Are we 2016 close? Or 2017 close?
 
Yes. And Jim's been too quite, with Jarred and Brent doing all the work and getting a lack of sleep and going a bit spinout over in the update thead (see Jarred's recent post on what outsiders must think). Man I could have cracked a few jokes about the funny things happening there, but I keep to myself.

So me thinks he is doing something in the background, and something might be coming.
 
Speculation running amok

Speculation running amok

Wayne Morellini said:
Yes. And Jim's been too quite, with Jarred and Brent doing all the work and getting a lack of sleep and going a bit spinout over in the update thead (see Jarred's recent post on what outsiders must think). Man I could have cracked a few jokes about the funny things happening there, but I keep to myself.

So me thinks he is doing something in the background, and something might be coming.
Click to expand...

I have speculated as well, that October could be host to a dynamic RED and Jim announcement! How swell would that be?
 
HypeVR?

HypeVR?

Came across this in another forum. Followed the link, and saw Jarred's name in the thank you for this project.

Volumetric video! http://hypevr.com/ Shot with 14 RED Dragons!

Wonder if JJ is involved with this? No names attached to the website, but the teaser video is incredible. Exactly as I imagined VR would turn into.

https://www.facebook.com/hypevr/videos/1097692456965324/
 
Too many secrets

Too many secrets

I never really got past the "I will say that AR and VR are now officially obsolete -- Jim". A bit of equivocation - as his team secretly works on the competition crusher? Or is his market read literal?
 
Holy cow... I crossed that quote before, but misread it. I thought he had said that he thought AR/VR were nonstarters.

This makes a lot more sense. And now I'm convinced that Jim Jannard is working on a volumetric video capture system. This makes his Lytros comment make sense. Convinced. And I wonder if HypeVR is his!!

HypeVR is the best volumetric video capture system out there, bar none. Nothing compares to it. I've been studying the development of this stuff for the better part of 3 years now.

If so, Jim has done it again. Made 2 dreams of mine come true. I just wish I could afford his creations!
 
You must log in or register to reply here.
Back
Top