Cell Chip Coming To the PC Via a PCI Express Card

arcticstoat writes with an excerpt from Custom PC: "After developing a brand new CPU architecture from the ground-up, you'd expect that Toshiba, Sony and IBM would have more uses for the Cell architecture than the PlayStation 3, and Toshiba has been quick to make use of the architecture's HD video transcoding abilities in its new Qosimo laptops. However, Leadtek is now taking Toshiba's efforts a step further by putting the chip onto a PCI-E card for desktop PCs. The WinFast PxVC1100 is based on Toshiba's SpursEngine SE1000 processor, which is a cut-down version of the Cell chip. The SpursEngine chip features four SPEs (synergistic processing elements) based on 128-bit RISC cores, along with H.264 and MPEG-2 codecs, but it doesn't contain its own CPU as the chip in the PS3 does. The chip is capable of encoding and decoding H.264, MPEG-2 and MPEG-4 video streams in hardware."

mythtv apps

[pak9rabid]

this + mythtv = interesting possibilities

Two things

[iamwhoiamtoday]

#1: Is there going to be a Mac Version? I would love to put this in my Apple Tower, I have 3 PCI-E x16 slots sitting around doing nothing. #2: When is this actually going to come out? I mean, I keep reading things on "fantastic pieces of tech" and they either never come out, or they come out everyone forgets about them. Anyone know what this should retail for, or if software can even take advantage of it yet?

Re:Two things

[somersault]

Any PCIe card is a 'mac version' just as much as it is a 'PC version' - perhaps you mean will there be drivers or a developer API for the Mac - the good thing is that a lot of Linux geeks will be wanting this (probably good for University research projects), and if there is Linux support then basically you will already have OSX support.

The interesting question is, what are you planning to do with it that you can't already do fast enough with a multicore CPU, GPU or physics type add in card? Or do you just want this because it's there? I'm not meaning to criticize especially, I tend to waste a lot of money on gadgets myself..

Does it run ... ?

[sergstesh]

The mandatory "does it run Linux ?" boils down to "do they provide enough documentation to write drivers for it ?".

I RTFA, but I didn't find an answer in it.

Re:Why bother?

[tonytnnt]

Hardware encoding acceleration h.264 isn't easy to do on GPUs as I recall. Your source video file isn't really meant to be worked in parallel, so a serial approach (like this) should work better. At least from what I've been reading/told, which is mostly related to transcoding rather than pure encoding. Someone else might be able to enlighten us more (hopefully a dev from x264 maybe?)

50/50.

[fuzzyfuzzyfungus]

The fate of this device hinges pretty much exclusively on the quality of its software and documentation. If all you get is some gaudy half-broken-and-all-ugly fixed purpose video encode decode app(in the fine tradition of graphics card shovelware, remember the bad old days when the card vendor was responsible for the driver?) then this thing is dead in the water. A few will sell to Netflix pirates looking to rip and encode 3 times as much video as they could ever watch, instead of just twice as much; but that'll be about it.

If it has good general purpose support(I'd really prefer that this mean "good documentation" and properlinux support; but I suspect a proprietary sdk would do alright as well) then it could be a killer in certain lower end computing scenarios. Since the cell is produced in nontrivial bulk, and this thing is only about 1/2 the complexity of a full cell(does that mean that this card is "spursengine on the half-cell?) it should be cheap, cheap, cheap compared to FPGA boards or custom ASICs for such purposes as the cell architecture is useful.

I hope the do the right thing, and get rewarded(and I hope so, surely somebody looking to sell computational hardware would see the virtues of making it as useful as possible for as many customers as possible?); but if they don't, I suspect that they'd be lucky to do as well as physX, and will probably do worse.

Re:I can see a use for it....

[fuzzyfuzzyfungus]

Just hope that they expose the card's power in a nice way. Documentation and/or SDK so that your in-house geek and/or the next version of $EDIT_SUITE can silently harness the power of the coprocessor? Instant win.

Attempting to integrate Leadtek l33tripZ SE (Now with the crushing power of the "buggy, ill-defined, good enough for consumers" h.246 profile in hardware! Totally Vista compatible(32 bit systems only, when run as administrator during waxing moon)) into a professional workflow? World of pain.

So, yeah, do linux geeks, I mean... yourself a favor and tell Leadtek that your outfit will totally buy them by the crate if documentation is good. ;)

Re:How Not to Build a Multicore Processor

[91degrees]

I wouldn't be so sure. Dedicated hardware is typically a lot cheaper than a general purpose CPU unless the tasks you want to do are extremely general. GPUs work very well with a simpler SIMD approach, and this can be extended to raytracing. It's an approach that works well for a lot of big number crunching tasks.

For more general purpose work, MIMD is useful. I have to wonder why Cell didn't take more cues from the Transputer. From what I've read, The Cell seems to be based on the idea of running multiple threads in parallel and having one core handle each thread. Always seemed rather inefficient. Seems that a better idea would be to package up the processes into a number of very short tasks, and assign each task to the next free core. This will, of course, require a totally different software architecture.

Re:See also, Mercury Computers

[Lance Cooper]

After looking briefly at this, it is an entirely different beast. It is a full Cell processor, with 8 SPE's and a PPE and 5GB of RAM, as well as Flash memory that allows it to boot Cell Linux. It's really a system on a PCI-E board, rather then an accelerator board.