3 channel 36 bit 1280 X 720 low $ camera - Viper?

[Steve Nordhauser]

Coming along. I will announce it here first. The chips are tough to get right now. We have several camera designs right now that are circuit boards built up except for a big missing component. I may have something else brewing of interest here. More information while I do some testing......

[Matthew Miller]

Oh Snap!
I didn't realize that you guys were building cameras using the Altasens chips, Steve. Is that right or am I confused? If so... that's wicked good. And what's this other development? Is it edible? Don't keep us holding our breath.

[Rob Lohman]

Matthew: as I understand it they are using Micron chips in the
current camera's and might be offering an AltaSens based camera
in the near future.

[Obin Olson]

so you can run the chip in 720p without cutting the FOV down? that is really good news!

[Rob Lohman]

Obin: I think this would always be possible with a line and column
skip algorithm which almost all of these chips implement. I'm hoping
Steve is talking about their test on the rolling shutter "solution".

[Steve Nordhauser]

Rob:
You are correct for the Altasens, but 'always' is a strong word. Most CMOS sensors can do basic subsampling - skipping rows and columns to keep the same FOV with lower resolution. For Bayer color sensors, you need a complex subsampling to keep the Bayer pattern intact (drop rows and columns in pairs). Region of Interest (ROI or windowing) reduces the FOV while keeping the same resolution. ROI always maintains the Bayer pattern but you may have to adust the starting point. Then there is binning - the summing of adjacent pixels, keeping FOV, losing resolution, gaining sensitivity and only useful for monochrome sensors.

The Altasens has a 2/3 subsample - very unusual but will give a 1280x720 60fps output. I think with a unique Bayer pattern, though. This is two pixels on, one off, two on. This means that a row would be GBBGGBBG and the next row would be RGGRRGGRRG. A row is skipped so that the next row is RGGR again. I'm trying to document that now. Hey, I just did!

Rob, guessing at the new info:
Actually, rolling shutter 'solutions' are getting more murky. Obin has found that at higher clock rates, the Micron 1.3Mpix is prone to smearing - bright oversaturated areas don't get reset properly and smear across the line. Micron's official solution is to run at lower rates. The higher resolution sensors in the same architecture are supposed to improve on this (even smaller pixels and less sensitivity, though). There is also a new mask coming out that *may* improve it on the 1.3s. As I told Obin, I will swap cameras if they are better for anyone in this group who buys one now, if the new ones are better.

[Rob Lohman]

Always was in context to doing it in our own code. Can always
skip lines and columns, ofcourse. But I see I should have worded
it better.

Steve: the rolling shutter solution was your line that you had
talked to an engineer at SI and he thought of a possible solution
that you would test last week if you had the time.

[Steve Nordhauser]

Rob,
I don't have an SI-1300 around to do that testing - working too hard on the Altasens manual. It does need to be tested though. The other person doing testing for applications is swampeder (new word, I think, but with a clear meaning) than I am. On the rolling shutter thing, I just wanted to give you a heads up that there were some potentially unresolvable Micron 1.3Mpix problems that people in this group need to know about.

You are abolutely correct that upsampling and downsampling in software are much more abitrary than in hardware. There are moire and other artifact issues but scaling can be to non-column based intervals with appropriate algorithms.

[Wayne Morellini]

Cross post.

Hi guys heres the little secret I mentioned a while back, and was what I was intending to use on my original camera project last year. I was supposed to read up on it and it has been waiting so long I nearly forgot about it. The technology is Processor in Memory modules. The idea is to implement processing elements (or full arrays) inside memory, simular to clearspeed but on the internal bus of the memory modules instead, here greater speeds and wider busses are accessable, and unlike clearspeed, large chunks of memory are directly accessible, which makes it very good for what we want, and indeed it is earmarked for things like compression. I think I have found the article I read last year, I thought it was refering to making standard dram sticks for PC's using the technology for 50% more, but I am unsure now. Production was hoped to be "18 months" (august 5th 2002 article). The speed up for one was upto 25-40 times over workstation performance (potential for several hundred), and for another it was upto 1000 times (Active page) using arrays of FPGA processing elements). The idea is that if they produce pc memory modules with it, you pop it into your PC memory slot and program it and "hey presto" your 1GHz nano-itx board is capable of processing and compressing 8 mpixel SHD streams (maybe a little exageration) but you get this without (maybe) even needing a cooling fan on the main board, low powered, low cost. To make things even better (depending on what Windows API standards are now) I think Windows had an API that allowed DSP's functions in add-in cards to be transparently used in programs (simular to Direct X API calls) to accelerate them. I remember some international meeting they had for this tech, the web site for the meetings would be a good source of contacts. For somebody like Steve in SI a non PC version could be hooked up to an ARM processor to provide a simple to program alternative to FPGA design (not that the programming is as simple as C coding on clearspeed). Normally I would keep this quiet (to stop companies from interfering) until I had researched, approached and negotiated with suitable companies about the possibilities of even using samples, but because of my health this is just not going to happen quick enough. So, if somebody with technical knowledge would like to do this for us it would be most appreciated.

Here are some links.

The only three that seemed to be aimed at intergration into memory modules is Diva, FlexRam, and Activepage (The FPGA solution), but I don't know which one is the standard PC module.. I have only been able to skim the documents due to health, so I don't know exactly the details, and haven't even looked up all the websites for each competing version.

www.isi.edu/stories/31.html
http://www.wired.com/news/technology...,54294,00.html

http://216.239.57.104/search?q=cache...pages%22&hl=en
http://216.239.57.104/search?q=cache...pages%22&hl=en
http://216.239.57.104/search?q=cache...pages%22&hl=en

This site was useful finding academic papers:

citeseer.ist.psu.edu/kang99flexram.html
citeseer.ist.psu.edu/context/100858/181563

[Laurence Maher]

Say guys, a friend of mine just emailed me an announcement of a relatively new codec from apple called the "H.264/AVC". Is this something we could use?

http://www.apple.com/macosx/tiger/h264.html

Claims to be some greatly sophistocated codec that can do scaleable HD. I don't know really what to look for that much, but one of you guys probably does. Check it out and tell us what you think.

[Rob Scott]

Laurence Maher wrote:
... a relatively new codec from apple called the "H.264/AVC"

The catch in a lot of these codecs is the 8-bit depth issue. For workflows that target 8-bit only, many of the new "delivery" codecs like this (MPEG4, etc.) could be useful.

For a workflow that requires 10+ bits throughout, we'll need codecs that are designed for production, not delivery.

I did run across this H.264 SourceForge project. It looks like it got started in March 2003 and didn't get anywhere.

[Wayne Morellini]

<<<-- Originally posted by Laurence Maher : Hey wayne, ya, . . . duh . . . you already posted the fuel type. Not eating my wheaties. Hope your health is good. -->>>

Thanks Laurence.

My Health has picked up surprisingly, looks like it is not going to be long term, I can get around for a few hours now without getiing stuffed, and I can think reasonably clearly before that.


<<<-- Originally posted by Steve Nordhauser : Rob:
The Altasens has a 2/3 subsample - very unusual but will give a 1280x720 60fps output. I think with a unique Bayer pattern, though. This is two pixels on, one off, two on. This means that a row would be GBBGGBBG and the next row would be RGGRRGGRRG. A row is skipped so that the next row is RGGR again. I'm trying to document that now. Hey, I just did!

There is also a new mask coming out that *may* improve it on the 1.3s. As I told Obin, I will swap cameras if they are better for anyone in this group who buys one now, if the new ones are better. -->>>

That is really excellent Steve, thanks for all your support you are giving us.

About the Bayer pattern above, I am worried if there is too many of them NLE companies might not want to support them.

I think binning is the best, but it needs a 3 chip camera for that.

I read the excellent tutorials that you have on your SI website, I'm surprised you mentioned the cathode ray tube memory device, pretty obscue nowdays. What about a page describing how to read the specs of cameras and cmos sensors, I'm a bit lost at the meaning of some of the terms when reading the data sheets (thoug I know what dark current and Quatum efficiency means now)?

Thanks.

<<<-- Originally posted by Jason Rodriguez : Both Ikegami, JVC, and Kinetta are going to be using the Altasens 3560 in their cameras. How come they don't have any problems with rolling shutter artifacting?

In fact I saw the 3560 and some other cameras built on the 3530 at NAB, and again, there were no problems with rolling shutters. I know there must be solution here because these are manufacturers using the same chips, and aren't having any dificulties with slanted lines, etc., because these types of artifacts wouldn't be acceptable to their production-bases markets.

So again, if this the way the chip is built, and these are the manufacturers successfully using it, then there must be solution somewhere (and it's not in loosing the high frame-rates of the chip, because Kinetta is going to use the 3560, and they are going up to 60fps without dropping frames). -->>>

I think I heard something about them having on chip buffering so they could perform global type reads.


<<<-- Originally posted by Anhar Miah : Thats last Century (literally),

Thats old tech, if you really want some thing more advanced you should look up about trapping into the engery from the vacum of free space, just google MEG (or Motionless Electromagnetic Generator)

Its FREE unlimted amounts of Energy to power anything really, (still mostly theory though, which *some* working prototype


=--===----=--=--=-=-=====-==-
----=--=-=-=---=-=-=-==-=-====
-=-=-==-----==-=-=-=-=-=-=-=-=

Anhar Miah -->>>

Hmm, I know (but how many years), but seriously you mentioned some working prototypes, can you give examples? I once had an idea for Physical Vacume Energy Sail, I wanted to use in a childrens movie "Johanes Berge and The Space Pirates), and had a few ideas flash around my mind for a method to extract free energy, but with all this camera stuff I've forgotten it :). Of course you know that it has all been done before, I have met some interesting people in my life and know a few things. We even have a local excess energy type engine thing here, and even though I Haven't met the guys myself (as far as I know) I have multiple idependant freinds that know them. The idea sounds a bit simular (no detailed info yet) to the device invented in the 70's in the States, can't remember his name, I have had plans in the past to it, hold it, with modern technology (rare Earth magnets etc) it should definetly have power for a camera? Does anybody have engineering expertese in electrical to give it a go if I could find plans? To state it plainly, I can take statements for or against these devices with a grain of salt, just remaining objective. And before anybody debates this, one of my best freinds is a top young engineer in the nation (actually our UNI also produced the top chip designer in America), he is really negative on the potentiual of the idea, but I would not trust his judgment on the issue, to much flawed logic blindly trusting in the 1% science does know (until the last 10-15 years, they were scratching the surface, now it sounds like a episode of Star Trek). I am waiting for them to go below the quantum level to proper analysis of the structure and properties of space itself, the substance we are written on! By the way, what is the code at the end of your post. Well, bring it on!

[Wayne Morellini]

OK, furthur omissions to my list of potential camera types.

Power supplies. PC Motherboards run off of AC to multiple line DC power supplies, but batteries are DC. How can we do this with the minium of fuss (without lossing too much power), as it would cost too much to make a multiple voltage line DC power supply for the MB, or does anybody know of cheap DC PC power supplies for ATX/BTX and ITX?

Can anybody discuss this?

Getting dissy, have to quit now, see you latter.

Thanks

Wayne.

[Anhar Miah]

For Wayne Morellini

This is a very good site:

http://jnaudin.free.fr/meg/meg.htm

It does have a US Patent, so it is coming along nicely .

And yes it could diffinately be used in Camcorders (needs a some work though).

If you go to yahoo forums and join into the MEG society they have a kinda "builders club" thing and there are lots of helpful fellow "builders" if you want to try.

I'm just passively keeping n eye on these projects, they are very interesting an i feel i break will come soon!!

P.S That crytic code was just fake, but i love the whole crypto scene ---

"Whats that you say?! he broke our 1024 bit random seed dual Laronze quad layered level seven secure line......Hmm.....What!!? hes 12 years old......F*%$£&%!!!"

[Steve Nordhauser]

Wayne:
I think you hit on a good topic for a non-programmer to research. Here is a start:

http://www.powerstream.com/DC_PC.htm
http://www.opussolutions.com/150watt.html

There are DC input power supplies for PCs. This is a better solution than battery->inverter to AC->computer PS back to DC.

To help on the battery sizing, take the power requirements (let's say 200W average (recording and just viewfinder). The two power supplies aren't perfect so maybe 75% efficient. So about 270W in (this is all guess work). Let's say you run on 24V in so you can use a nice scooter charger for $75 (like a Soneil). To run one hour, you need 24V @about 12Ah. Again, typical scooter sizes. Battery choices are sealed lead acid (SLA- either UPS batteries or Hawkers- better for deep cycle), NiCd, NiMh, LiOn. Size and weight go down, cost goes up as you go along the list. You can get bigger batteries or parallel smaller ones for longer run times. Might be better to swap batteries than lug the max.