CineLink(tm) is security software for DLP digital projection. The technology was created by TI. Check out this product page (search in the page for CineLink):

http://www.dlp.com/about_dlp/about_d..._dlpcinema.asp

Ben, do you need more RAW footage to test with? I am not sure what I have posted on this site for download..let me know if you you need it...do you think we could use your bayer filter in our capture software? maybe for some sort of trade ;) or??

it will work on RAW files right? so we could use it as we convert from RAW data on disk into a format like avi or others?

Obin, I'm using some of the images you've posted here, and I'm getting some of your images via Jason. :) I've also been working with footage from other cameras to be sure that I'm not over-customizing the software.

I've tested it on both RGB and RAW images, and once my camera gets here (tomorrow? thursday?) I'll finish up the last remaining things I want to do with it.

I'd be happy to pass along my algorithms as long as your software winds up being 100% free. If you charge for it, we'll have to work something out.

Actually, if you could shoot a test in front of a greenscreen (flourescent green posterboard from the grocery store is fine -- don't worry about the lighting too much) we could see how my software fares with keying... all I need are a couple RAW 16-bit B&W bayer images.

Jason,

What I'm saying is that even if we do spline interpolation for the G channel, we still have the logic-based de-zippering going on. So we'll have a beautiful spline interpolated G channel (with zippering) which we'll then etch over with linear patterns like we do now. The result will be the exact same amount of gridding... Having written a few keying algorithms, I don't think a post-blur will be absolutely necessary, but it will depend on how poorly the shooter has calibrated the G1 and G2 gain. Getting the green gains lined up is key with linBayer...

Yah, the gains on the camera,

that brings up a good point.

Once you have the G1 and G2 gains balanced, do you have to adjust them for each scene, or is it a set-once and forget type thing?

That I don't know yet. Steve? Obin?

What about it?

Yes, programming it can be. In future implementations of direct X, the on chip co-processors are to be combined combined and upgraded into a more general purpose CPU for graphics with logic branching.

Rob, your slow GPU transfer rate to memory. AGP works on extreme transfer rates with block transfers. How is this transfer being done? I can understand that some ways might be slow, but there should be a fast way to write to card memory and then transfer out in large blocks to main memory (individual word access by the main board CPU would be slow).

I had a look at your developement Blog, about the machine code, that seems to be a classic example of how much C can slow you down, but it seems a little bit large.

Nice assembler link, thanks.

The problem with AGP is that it is very fast but only on one way, from system memory to Graphic card, not from Graphic card to the motherboard's memory.It isn't a symmetrical bus.
This has been solved with the arrival of PCI-E.

If someone here is interested about what I'm saying,the new graphic cards from Nvidia require two molex power connectors and need around 100 watts at full speed.So a P4 3.2 GHZ with a GeForce 6800 sum up to 200 watts, add a Raid 3 or 0 or 1+0 or what you want and you may end with a 300 watts power requirement.

http://support.intel.com/support/pro.../CS-007983.htm

Pentium-M 2 GHZ needs 21 watts and it hasn't got the same horsepower as a P4 2 ghz...

So I can't still understand why people keep saying a laptop needs more energy than an Eden system, unless you were going to use a 1 GHZ Pentium-M...(which isn't so more powerfull than an Eden)

Juan: thank you. Finally someone with some common sense. Pentium M is the way to go. It has the horsepower, and it doesn't draw too much power. The GPU is an interesting post-production possibility, but in no way should we rely on it for an in-camera system.

Just use a laptop. If you absolutely have to put everything in one box, take the laptop out of its case, ditch the LCD, and cram it all into a box.

Hi there peeps,



Kinda lost track of this thread, and just recently picked it up again. Exciting to see how things are going! Def learned a lot too, by reading Les' and the two Robs' posts (just to name a few).

Now, noticed the wavelet section over at the Wiki page and that reminded me of an old link:

http://www.maven.de/code/wavelet.zip

I'm not the person to judge if this is something that might jumpstart you guys into implementing something similar. But just maybe it does.

Keep it up!

Xtof

Juan, about the AGP thing, thanks, thats good. We can allways target cheap low end, low powered, versions of PCI express 3D cards, and integrated chipsets, they should have more than enough power. That cloud does have a silver lining.

At the moment the via processors are for the cheapest, lowest powered raw 720p alternative. The base level. When they are upgraded (maybe when we get 1080 cameras) they might do 1080. Plus if you want to put it into a handheld case, or a small sholder case, you will have to get a small laptop break it apart, recable it (to fit new shape), maybe replace the laptop drive (if it is slow)and the power hungry LCD (shortly this should not be a problem), or recable it as an external monitor.

Re-edit
--------------------
OK I'm sold on the laptop idea (except for battery life). It is expensive, but if you factor in the display as a external monitor it is not so expensive. Basically you could even extend the LCD cabling (risky, get an expert to do it) to use as an external monitor in a case. Now to make the smallest 720p system we could use the small VIA Antuar based laptops, but none would have Gigbe yet.

The Windows Tablets are expensive, in part, due to (this was a year or so ago) MS charging something like $100-200 each for the OS.

Ben, if you can get that USB camera working smoothly, let us know, I know where I want to use it.
----------------------

But people can use whatever they want on their systems to suit their needs. For compressing, 3 chip, 1080+ etc you are going to need even more than a Pent M. If we can find a way to ring out performance from the integrated 3D GPU than we could still use the Pent M, at no extra cost.

Thanks.

<<<-- Originally posted by Ben Syverson : The smallest tablet is the Sony VAIO U70. But the micro HD inside will never keep up with the data.
-->>>

I think, Transmedia might have done a smaller one that is being sold, called the qbit or something.

Actually the VIA handheld gaming platform (they also did tablets) has a screen and buttons, and support for 1GHZ processor (when somebody uses it). That could be broken down, using the buttons for recording.

If you want to get a better drive, attach a external box to the back with drive and extra battery/camera. You could use it sharp cam style, or lengthways with a small wideangle mirror angled to veiw the screen.

A 2Ghz Pentium M is very fast when you turn off all the lower-power settings. A 1.6Ghz Pentium M was out-doing many 2.4Ghz P4 Desktop systems on AE benchmarks, Cinebench, etc., so at 2Ghz, I'd expect even better performance.

can you put a P M on a standard micro atx or itx board?

FedEx came with the 8inch touch screen ! looks very nice! I think it will work very well for this camera and capture software

I have serious doubts as to whether a 1ghz processor from VIA (whose processors don't perform quite as well as Intel's at the same mhz) can handle the data being shoved at it.

Maybe there really isn't that much processing going on, and it's 90% what chipset and hard drive setup you're using. Rob, do we have any hope of running your software on a slower-than-normal 1ghz machine?

I have no experience with VIA's I/O chipsets and their real-world performance. Anybody?

I doubt it. My software runs at about 85-90% CPU utilization on my laptop, running a ~2 GHz Athlon XP-M. I think the Eden CPU could be suitable for handling a pre-compressed signal, but I'm not sure about this much data handling. On the other hand, a lot of the performance issues may have to do with memory bus speed. To really tell, we'd need someone to build one and try it out.

@Obin: "can you put a P M on a standard micro atx or itx board?"

Probably, but that defeats the whole purpose of the Pentium M architecture. The whole idea is that you have the Pentium M, which was designed to talk to specific chipsets such as the ICH4-M.

Also, why put a Pentium M in a micro atx board when a laptop motherboard has it built-in -- and is already slimmer, uses lower-power components, and has a batter system pre-built? I guess if you simply must have the camera be all-in-one, you're going to have to find some way to cram a motherboard into your camera housing. It's going to be about equally difficult with a micro/nano-ITX and a laptop's innards.

At the risk of repeating myself ad nauseum, I say keep it simple -- a camera connected to a laptop. If you want to upgrade the sensor, just get a new camera. Want to upgrade the processor or laptop? Go for it. Putting it all in one box discourages these kinds of changes and upgrades.

@Wayne: "At the moment the via processors are for the cheapest, lowest powered raw 720p alternative."

Wayne, as we were discussing on other threads, I don't think the VIA systems can handle raw 720 @ 24fps. At least not yet. Maybe they'll come out with a kickin' 2ghz system at some point, and we can take another look. But for now, their 1ghz fanless nano boards aren't even out yet and they will be underpowered for our purposes.

Miniture PC's
 

To Wayne Morellini:

Have any of you guys considered miniture PC?

These would be way better than a laptop, take a look:


http://www.littlepc.com/


"Rugged LCDs

Industrial Grade NEMA Rated LCD Monitors with Touch Screen Options. Designs include Panel, Rack, Open Frame & High Bright Sized from 6.4” to 21” with options and custom designs available."

AND

"LittlePCs

Small pc computers with the power of a desktop PC in the palm of your hand.

Available in P4, P3, Fanless and expansion slot configurations. PCMCIA, Multi-LAN & Compact Flash Custom options available."



Thanks hopes this helps!

@Anhar: "These would be way better than a laptop, take a look"

They say they have a "Pentium 4 Fanless" model, but it just seems to be a 2ghz Celeron. I have no idea whether 2ghz of Celeron is enough -- maybe it is.

I guess I missed how these would be better than a laptop -- you still have to provide 12VDC current, and you still need to add a display that can handle DC power.

These things aren't light, either -- around 6 pounds. Add in a battery and monitor, and you're at around 8 or 9 pounds.

LittlePC - crap - I can build a pc with the same specs for 1/2 the money they want $1400 for the bottom line unit with NO pci slot! I can build a microatx with pci P4 2.8ghz 256ram 400gb raid sata disks 20gb system disk case PS etc for $760

seems more and more this is what I need to do...

Aren't good P M laptops kinda $$$$??

Mine was under $1000...

micro pc
 

Obin, do those micro motherboards for the P4 support HT and 800 Mhz FSB yet?

It sucks that the P4 Procs use so many watts!
-Les

guys, I just made a call about that nano pc board that was posted a while back... that thing is THE board its P M @ 1.6ghz has 2 SATA 1 pci-x and is 5inch x 4.5inch in size!!! PERFECT! I am going to be on the list of pre-sale testing for the board!

pci-x = 1080p 10bit!!!!! Altasens!

I guess maybe a cheap shuttle will do for now untill that board is out..This will give me time to design the camera case and have software made for it and find all the small stuff

In case it helps anyone, here's a motherboard finder page.

Obin,

Do you still have a link to that motherboard (or was it the motherboard I posted? :-|)

Huffyuv gives me an amazing speed of 80 fps for a 720x380 RGB source on an Athlon XP 2000.

So I guess on a P4 or similar, 2 GHZ and up, it should give at least 24 fps for a 1280x720.


A question to anybody who knows about the topic:

What could happen if I store Luma info with 10-16 bit depth and U and V with 8 bit depth?

What could be the final bit depht when converted back to RGB??

crossed post,sorry.



Huffyuv gives me an amazing speed of 80 fps for a 720x380 RGB source on an Athlon XP 2000.

So I guess on a P4 or similar, 2 GHZ and up, it should give at least 24 fps for a 1280x720.


So Rob, could you be able to parse data and feed huffyuv with (there is a variant of it called VBLE which uses YV12, which is ideal for the Bayer structure) three seperate planes of G, R and B??
Remember that the seperate planes of a Bayer Pattern will have the same structure of a YV12 source, which is full Green/Luma and 1/4 Red and Blue.

Well here is one possible FPGA IPcore for Steve.
It has a Huffman encoder/decoder and other things.

http://www.opencores.org/cvsweb.shtm...ssion_systems/

Juan, is that 8bit only? we really need 10bit for capture/compression - then we can bump down for 8bit editing later on...if we need anyway...

Huffman IPcore or Huffyuv?

is there a 10 bit open source realtime lossless codec available?

May be Macintosh world, but PC?


About the IPcore I don't know.

Huffyuv is 8 bit, but it is the fastest LOSSLESS OPEN SOURCE codec available.
It won't be so difficult for someone with enough assembler knowledge to adapt it to 10 bits if that is the case although I know it would take sometime.


Also take note that Huffy doesn't make use of SSE, just MMX.So there is a lot of speed improvement available...


BTW VBLE gives me a speed of above 90 fps...(oh sorry it is 8 bit too!!! =S ..... just kidding :) but the speed is real)

Here is also another COMPLETE FPGA project.

Huffman encoding.

http://www.ece.cmu.edu/~ee545/f98/swingers/index.html

Rob we should have a working Version of CineLink today... give me your thoughts on the convert software?

I had thought about adapting it to 16-bit since that would simply amount to doubling the size of the standard data type. If you were storing 10- or 12-bit data, it would simply compress out the extra zeroes.
Please correct me if I'm wrong, but isn't SSE strictly floating-point? If HuffYUV works strictly using integer math, SSE won't help in that case.
Possibly, especially if we have a very efficient Bayer filter.

Problem is, the Bayer filter expands the data by 3x immediately; HuffYUV provides a 2x - 2.5x compression, so the data has expanded, reducing frame rate to disk. Plus, HuffYUV is 4:2:2, so you've already lost some data.

This application would be developed and released under the GPL, perhaps hosted at SourceForge. The understanding would be that none of the code (except code you write yourself) would be used in a closed-source application.

Version 1
A command-line and/or very simple Windows GUI that does:

• Reads a clip in either your "raw" format and my "raw" format into a standard internal representation
• Applies a selected Bayer filter to each frame
• Converts the clip to 16-bit TIFF and QuickTime

Each of us would take pieces of the project. For example, your team might write

• the "reader" for your particular format
• one Bayer algorithm
• QuickTime output

I might write

• the "reader" for my own format
• another Bayer algorithm
• 16-bit TIFF output

... and so on. (BTW, Rob Lohman and I have been working on an open format that you might want to use also.)

Version 2

• The GUI would be based on GTK+ (free) or Qt ($1,500 for Windows) for cross-platform support.
• Realtime playback of clips

I'll correct you.SSE is integer and floating point.
If I'm not wrong SSE is integer and SSE2 is floating point.But I'm not sure.It is usually called iSSE (to avoid confusions)

I've never talked about storing the De-Bayered data ,but the RAW BAYER data.
It is extremely inefficient to de-bayer for storage purposes.
So there is no problem to store the seperate three color components as if they were YV12 compressed with Huffy.
IMHO.

OK, I stand corrected. I've used MMX, but I've barely started getting familiar with SSE.
Sorry, my bad. That sounds like a pretty good idea. I'll look into that when I have a few extra minutes :-)

Why not this, it is well known, is very stable and easy to use.

http://www.wxwindows.org/

It runs on almost every available OS..

I've heard good things about it, and I'm not sure why I hadn't considered it for this project. I'll look into that one too. Thanks!
(Easy to use = A Good Thing)

Could someone tell me anything about the SI-3170-CL ????
It is supposed to give 2048x1536 @ 30 fps.

Why nobody talks about it?

Sounds good,

BTW, do you know what type of bayer algorithm you're going to incorporate? Variable Gradients, spline-based, Optimal recovery, etc.?

One thing I might mention is the inclusion of some sort of "blur" or filtering effect that can be variably applied (variable from none to a decent amount depending on the image, not really a gaussian blur, but some sort of anti-aliasing type blurring filter so it doesn't just create overall "softness" like a gaussian blur does, but it does soften to reduce edge artifacting or other bayer-artifacts). IMHO I think Bayer processing can really be a weak link in this whole system because that's where you get your final image from. If you don't have a good bayer algorithm, then you can have all the bit-depth you want, but you're going to be fighting all those nasty bayer artifacts that can be quite distracting, and just scream out the fact that you shot on digital.

BTW, one more thing,

Obin, the motherboard we were looking at from Kontron (the P M model with 64-bit PCI-X), how would you hook a battery up to it if it requires an ATX-type powersupply? Can that even be done without making the mechanism a behemoth? Just curious how that might be done.

Little PC nice, case, could be made to look like a shoulder cam. If you have a look through this thread, the viper thread ands my Custom camera thread, you will find references to simular devices.

There is some miss conceptions here.

For 720p RAW, 1Ghz should be enough, using MMX, SSE, GPU, the new media processing architechure that via seems to be developing, and whatever else we can milk, maybe even simple lossless routines (Huffyuv ??). PentM claims simular power consumption at 1 Ghz, so maybe it is as good, or better.

Just because it is on an itx board, doesn't mean it can't use a laptop chipset.

As I said before, whatever board and whatever case you want, the future software is meant to support a wide varitey of configurations. If anybody wants to use a laptop great, do it. But there are other people we also want to consider and serve aswell, and we are trying to do cheap entry level and professional versions for them.

Looking at the big picture, thinking of the majority of people out there, and growing this market:

• We want people, for the project to be a sucess, rather than a red herring, that will also give involvement, supporting products, and volume.
  
  To get success we need it to appeal to the most people. For this we need it to be simple, reliable, professional, cheap and with good quality performance.
  
  The majority of people are not techy computer people. Techy people, and their circles of techy/very intelleigent freinds, often miss this piont and try to get everybody else to conform to how good they are.
  
  A big portion of people are video/film people rather than computer people, and don't want the computer hassels or the techy image, they want simple, reliable, plug and play.
  
  People who are video/film people might mock at the techy idea of carrying a laptop around instead of a easier professional END shoulder case, and not support it.
  
  Raise the cost and you loose most of the wanabees, prosumers, and low end local professionals (wedding videos etc). So we want cheap non laptop versions for these people aswell.

Earlier I was writing:
I am focusing on a few cases. A small sized allweather/underwater shoulder case (I want to give it long sexy lines simular to, but not the same, as the Olympus SHD case), nano itx (12cm*12cm) is preferable. A whacky styled (move over XL1) Indie special for upto 17cm*17cm boards (or newer simular sized Intel), the case for my camera project last year. The last one is a handheld model (upto PD150 in size, but handycam would be preferable) nano itx sized (or handheld portable PC). The allweather case can be designed so that any small pc, like the Little PC mentioned, can be slid inside, but this will defeat the unique cooling system, so it might need to be dropped. Some of the cases could be made to accomodate upto micro-atx size, but any bigger you might as well goto fullsized cases and dolly for them. But the work on these has yet to be started.

Very impressive, it is amazing what outside of Intel designs can achieve (they bought the Israeli company that developed the Pent M). If 1 Ghz VIA is 7w, smae as a 1Ghz Pent M, then we should also closely watch the Pent M as well. But the question is how does the MMX, SSE etc modes perform on the Pent M compared to the P$(4) that is supposed to have a special enhanced architecher for media processing. Does Intel get beat again?

The software isn't ready yet, we are waiting a few months, and I think the 1Ghz nano-itx were announced a week or so ago. Use a laptop now, it doesn't matter.

On a futuristic, pie in the sky, aliens abducted my XL1 type of side note. Interesting article in New Scienctist last week.

A researcher was claiming they could, one day, make a device that uses Ultrawide band technology to deliver between 100 to 1000 Gb/s that could be held in the palm oif your hand (and probably be used as a tricorder I would say). Ultra HD could barely dent that.

Well, I guess that the SI-3170-CL is the camera Steve N. was talking about (but he called it SI-3300-CL).
Anyway here is the datasheet from a German site that is offering it now!!

http://www.ehd.de/products/cmoscamer...MegaCamera.pdf

What I still can't get is: if it is the Micron sensor or not.