Slashdot Mirror
The Mystery of Cell Processors
LucidBeast writes "Consumer appliances requiring more computing power Sony, IBM and Toshiba started 2001 developing "Cell"-processor that comprises of multiple processor cores and should give performance ten times of conventional processors. Now the CNN Money reports that details of the processor will be released Feb. 6-10 at the International Solid State Circuits Conference in San Francisco. Also reported by EE Times. Rumors also tell that Sonys PS3 development platform has already been shipped to some developers equipped with the cell processor."
I didn't know Apu had children. I thought the QuikiMart was his entire life.
If you wonder what these processors looks like, take a look of T101 (Terminator 2) cyberbrain ;)
Chip power, times 10
Sony, IBM, Toshiba disclose details of new processor that will run next-generation electronics.
November 29, 2004: 6:13 AM EST
TOKYO (Reuters) - IBM, Sony Corp. and Toshiba Corp. on Monday unveiled some key details on the powerful new "Cell" processor the three are jointly producing to run next-generation computers, game consoles and TVs.
Cloaked in secrecy and the object of much speculation since the three conglomerates announced the project in 2001, Cell will be 10 times more powerful than conventional chips and able to shepherd large chunks of data over broadband networks.
In a joint release, the three firms gave a glimpse of their respective plans for Cell-powered products, but were mum on technical details, which will be revealed Feb. 6-10 at the International Solid State Circuits Conference in San Francisco.
IBM (Research), Sony (Research) and Toshiba are investing billions of dollars to develop and prepare for mass production of Cell, which is a multicore semiconductor composed of several processors that work together to handle multiple tasks at the same time.
"In the future, all forms of digital content will be converged and fused onto the broadband network," Ken Kutaragi, executive deputy president and COO of Sony, said in the release. "Current PC architecture is nearing its limits."
IBM said it would start pilot production of the microprocessor at its plant in East Fishkill, N.Y., in the first half of 2005. It will use advanced 300 millimeter silicon wafers, which yield more chips per wafer than the 200 mm kind.
It also announced plans to first use the chip in a workstation it is developing with Sony, targeting the digital content and entertainment industries.
Sony said it would launch home servers and high-definition televisions powered by Cell in 2006, and reiterated plans to use the microchip to power the next-generation PlayStation game console, a working version of which will be unveiled in May.
Toshiba said it planned to launch a high-definition TV using Cell in 2006. Top of page
Always value the individual over the system. --Bruce Lee "I don't need a Sig - I have a custom 191" - me
"Consumer appliances requiring more computing power Sony, IBM and Toshiba started 2001 developing "Cell"-processor that comprises of multiple processor cores and should give performance ten times of conventional processors."
What in the hell does that sentence mean? I can handle a couple of spelling or grammatical problems, but seriously! What the fuck does that mean? Are 3 companies working together to create this Cell processor, or are there three different Cell processors...
About 10 processor cores, right?
They should have enough power to divide by zero by now, right? or is that still to "difficult"
... should give performance ten times of conventional processors.
Twenty times, after absorbing an android or two.
-Stephen
details of the processor will be released Feb. 6-10
;-)
it gives a 10 times performance gain over a normal processor, from the year 2001 of course, which will be something like a 1.3 GHz P4 or a 800 MHz Celeron, both introduced in january 2001
Now it is true that multiple core chips seem to be where everyone is headed. Even so, I'm not sure how these magical chips will "converge and fuse" digital content. Remeber that this article is A) light on details, and B) put together by a person who is vying for your dollar/pound/yen etc.
Always value the individual over the system. --Bruce Lee "I don't need a Sig - I have a custom 191" - me
When PS2 was launched, incredible specs were also touted; on delivery it ended up cheaper but not more powerful than a high-spec PC with a good video card one year later. I am afraid we might end up with another mediocre product at a reasonable price point. Sony should concentrate on portable systems integration which is where its real expertise lies.
This is not a signature.
Consumer appliances requiring more computing power Sony, IBM and Toshiba started 2001 developing "Cell"-processor that comprises of multiple processor cores and should give performance ten times of conventional processors.
Methinks LucidBeast needs to brush up on the lucidity!
Ah so that is how they work :)
As cunning as a fox, which has just been appointed professor of cunning at Oxford University. http://www.kinlan.co
These multi core and multi processor systems can be a bugger to program for because handling concurrrency in a way that doesnt cause deadlocking is a major pain in the ass.
One of the better ways is to model out the program in CSP (or a variant thereof) and then write in a specially designed language like Occam (developed for the original transputer, but ported now to x86). These give you code that cannot deadlock or livelock or suffer from resource starvation without needing any of the complex and buggy hacks you see in things like the Linux kernel. And the Linux kernel only has to deal with a few processors... scalling to a few thousand processors in C would require a programmer of insane genius or the implimentation of effectivly a new language on top of C to handle the problems caused.
So, what language do developers use to target this? Is it something elegant designed for the problem at hand?
Beep beep.
With all that computing power, imagine how much of a benifit it would have to distributed computing projects such as Folding@Home, Seti@home, distributed.net,etc. Mind you, it would take re-writing a new client from scratch, but imagine how much of a benifit it could have! This will be depenedent though, on how quickly Operating Systems like Linux and Microsoft can adapt to this new processor.
Sounds like Playstation3 vs. XBox2 will look like a battle between a Terminator T1000 and Clippy.
Signature deleted by lameness filter.
But UNC's Zimmons has his doubts. "I believe that while theoretically having a large number of transistors enables teraflops-class performance, the PS3 [Playstation 3] will not be able to deliver this kind of power to the consumer," quoted from /. referenced article.
Zimmons talks the details.
Now I'm the grandest Tiger in the Jungle!
I currently work at a game studio here in Melbourne Australia and we're looking at next gen stuff (currently we develop xbox, ps2, PC games). Anyway, today at a meeting, one of the senior developers told our group that 4 had been selected to go to a little show and tell by IBM/Sony in Melbourne, where some of the secrets of the "Cell" processor would be demonstrated/explained to the group. Apparently we were only able to get 4 spots at this event.
So I'm exicited looks like the tech in just around the corner and so are the multi-core platforms (like XBOX2 and PS3).... yay!
It reads ok if you are not a native english. Try to bablefish it a few times and it might come out ok.
And the "cell" is an architecture. so they build the common blocks together and each compagny slaps togeter some other blocks to make it's own appliance.
You can always read the fine article.....
in 2000 sony had a series of tv spots trumpeting the "play station 9" for the year 2078, with things like "electronic spores that tapped directly into a person's adrenal gland, improved retinal scanning, a mind control system, holographic surround vision, and telepathic personal music" -wikipedia blurb
here's a link to the video of the ad
well, with the exorbitant processing demands of the ps3 that this article suggests, it's almost like they are on track to deliver what they promise!
intellectual property law is philosophically incoherent. it is your moral duty to ignore it or sabotage it
...the current computer architecture is nearing its limits yes, but it has no relationship to the content. A modern processor is very well capable of decoding HDTV content, probably encode too if you can accept less than super compression.
Of course, I see where it is going, I assume these Cell chips will be used to control hardware encoders/decoders with hard real-time limits (i.e. no frame skips and such crap). Taking the best of "dumb" hardware players of today, combined with the multitasking and flexibility of general computers.
But it is still a computer in drag. If anything, this seems more like a "retro" trend of the past, when you had active NICs/HDD controllers/whatnot with processors of their own. Now it is back with Cells instead. Just like terminals, we're coming full circle.
Kjella
Live today, because you never know what tomorrow brings
The article mentions that the Cell CPU will be included in a HDTV from the year 2006. Anyone know what such a powerful CPU is doing in a TV ?
or take a look at the 2002 patent application:
http://theinquirer.net/?article=19941
Cell Processor-Based Workstation Prototype
The companies expect that a one rack Cell processor-based workstation will reach a performance of 16 teraflops or trillions of floating point calculations per second.
Cell Processor Unveiled
IBM, Sony Corporation, and Toshiba Corporation today unveiled for the first time some of the key concepts of the highly-anticipated advanced microprocessor, code-named Cell, they are jointly developing for next-generation computing applications, as well as digital consumer electronics.
Specifically, the companies confirmed that Cell is a multicore chip comprising a 64-bit Power processor core and multiple synergistic processor cores capable of massive floating point processing. Cell is optimized for compute-intensive workloads and broadband rich media applications, including computer entertainment, movies and other forms of digital content.
Other highlights of the Cell processor design include: -- Multi-thread, multicore architecture. -- Supports multiple operating systems at the same time. -- Substantial bus bandwidth to/from main memory, as well as companion chips. -- Flexible on-chip I/O (input/output) interface. -- Real-time resource management system for real-time applications. -- On-chip hardware in support of security system for intellectual property protection. -- Implemented in 90 nanometer (nm) silicon-on-insulator (SOI) technology. Additionally, Cell uses custom circuit design to increase overall performance, while supporting precise processor clock control to enable power savings.
IBM, Sony Group and Toshiba will disclose more details about Cell in four technical papers scheduled for presentation at the International Solid State Circuits Conference. "Less than four years ago, we embarked on an ambitious collaborative effort with Sony Group and Toshiba to create a highly-integrated microprocessor designed to overcome imminent transistor scaling, power and performance limitations in conventional technologies," said Dr. John E. Kelly III, senior vice president, IBM. "Today, we're revealing just a sampling of what we believe makes the innovative Cell processor a premiere open platform for next-generation computing and entertainment products." "Massive and rich content, like multi-channel HD broadcasting programs as well as mega-pixel digital still/movie images captured by high-resolution CCD/CMOS imagers, require huge amount of media processing in real-time. In the future, all forms of digital content will be converged and fused onto the broadband network, and will start to explode," said Ken Kutaragi, executive deputy president and COO, Sony Corporation, and president and Group CEO, Sony Computer Entertainment Inc. "To access and/or browse sea of content freely in real-time, more sophisticated GUI within the 3D world will become the 'key' in the future. Current PC architecture is nearing its limits, in both processing power and bus bandwidth, for handling such rich applications." "The progressive breakdown of barriers between personal computers and digital consumer electronics requires dramatic enhancements in the capabilities and performance of consumer electronics. The Cell processor meets these requirements with a multi-processor architecture/design and a structure able to support high-level media processing. Development of this unsurpassed, high-performance processor is well under way, carried forward by dedicated teamwork and state-of-the-art expertise from Toshiba, Sony Group and IBM," said Mr. Masashi Muromachi, Corporate Vice President of Toshiba Corporation and President & CEO of Toshiba's Semiconductor Company. "Today's announcement shows the substantial progress that has been made in this joint program. Cell will substantially enhance the performance of broadband-empowered consumer applications, raise the user-friendliness of services realized through these applications, and facilitate the use of information-rich media and comm
I don't think that a game console needs such a so sophisticated and so powerful CPU, for important reasons:
-Real-time 3d graphics of cinematic quality will always be too slow for general purpose CPUs.
-developing a game with AI that needs ten times the power of todays CPUs will take many man years and may not be that welcomed by the console audience.
-It's very difficult to do multithreaded apps, and the difficulty rises exponentially with the number of threads.
So what exactly would the be role of the CELL processor in PS3?
It would make much more sense if:
-Sony developed a platform that can move insanely great amount of graphics around, with the ability to do real-time raytracing, rather than providing so much general-purpose processing power.
-Sony developed a graphics architecture that could really be parallelised, so instead of bringing out a totally new console, they could just up the graphics spec by adding more chips. They could save millions of dollars from developing and advertising the new console.
People said x86 was end-of-the-line when the pentium first came out. That was a heckuva long time ago now. Although I, personally, would like to see a shift toward more modern platforms, I doubt anything much has changed with those claims of x86 dying.
After playing games on a PC with a mouse, playing on a console is frustrating, not fun. And, I can not imagine playing a game that puts console based players up against PC players (unless you handicap the PC's hardware). The console players would be out of luck.
InnerWeb
Freud might say that Intelligent Design is religion's ID.
The Cell processor is going to rule!
After all, look how accurate Sony's hype about the PS2 was:
The PS2 will be able to render 75 million lit, shaded polygons per second!
The PS2 will be able to run games at HDTV resolution (1280x960) out of the box with no performance loss!
We will build professional workstations out of 32 Emotion Engine chips which will be able to render movies in realtime and take over the professional graphics industry!
Since all the hype turned out to be completely 100% accurate, I'm sure we can expect the same for the PS3 / Cell Processor.
I suppose it's also possible that it will be another massively over-hyped disappointment with builtin Sony patented lameness that sucks even harder than ATRAC. But you'd have to be a real fucking cynic to believe that!
If this part reaches the promised performance, will CPUs then overtake GPUs for SIMD-type operations? Will a software implementation of OpenGL running on a Cell system top the performance of whatever NVidia and ATI are selling by then?
-jcr
The only title of honor that a tyrant can grant is "Enemy of the State."
I for one welcome our new Cell-powered overlords!
There's only one small problem with your contention - Linux /does/ scale out to at least 512 processors - hardly 'a few' - and is heading up to multiple thousands with SGI's current work.
Of course, one could argue that the Linux folks have more than one insane genius among them...
I always find it odd that so many "Nerds", people who spend their time programming in languages that demand incredibly exact syntax, can't get basic "natural language" syntax right.
There are many species of Nerd, and Coder is but one of them. Don't forget about the unwashed masses of Trekkies, role-players, script kiddies, video gamers, trench coat goths and other assorted misfits that also comprise Homo Dorkus.
It's based on the Power architecture and very possibly the Power5. Which means it will be awesome. AMD or Intel don't have anything comparable to Power5.
The x86 platform is in bigger crisis than ever, thanks to AMD's half-hearted upgrades. In order to be even marginally competitive, x86-64 would have needed much more than just 8 new registers.
Perl is for those who like to choose their own syntax. :)
Why have one or two ways to do things, when you can have eleven?
500GB of disk, 5TB of transfer, $5.95/mo
Playing first-person shooters on a console is frustrating, since the control is designed for a keyboard and mouse. Similarly, playing a typical console 3D platformer like Wind Waker on a PC will be equally frustrating since the control is designed for console pads.
It's the nature of the beast. And until someone sorts out the control differences, PC vs. console play will always fall short for one of the two.
Ita erat quando hic adveni.
I like the picture more than the text. don't you have alink that includes the picture?
First thing that a game console should get is a mouse and keyboard standard with the joystick.
Why choose a keyboard designed to put letters on a piece of paper? Why not deliver a unique keypad with dynamic reprogrammable key labels to provide a tailored input device to the game your playing? I would rather see "Thrust", "Reload", "Switch", "Comm", etc depending on which game I'm playing rather than qwerty...
It's all about the bandwidth now. The cache(s) and path(s) to memory should be the most fascinating aspects to this processor. Speed is nothing without data to process.
"It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
This does not surprise me in the least. A Prescott processor has 125 million transistors, a Motorola 68000 had 68000 transistors. Yet the Prescott is not 1838 times more productive on a per clock-cycle basis. Admittedly, some of those Prescott transistors go to cache, superscalar magic, creating long fast pipes to achieve the GHz and implementing nifty MMX features. Even so, fabbing a 68k in 90 nm would create a tight little processor that is not 1800 times slower than the Prescott.
Thus, one can imagine creating a tighter core processor design with a budget of a million transistors each (15 times the original 68k budget) with a few million for L1 cache and another million for glue and then place 20 of them on a single die. Add optical interconnects and that new optical-to-silicon technology invented recently (for multiple channels of GHz I/O to feed all those cores) and you have yourself a powerful little processor.
The point is that with a budget of 125 million transistors, designers can do more than create a bloated single-core CISC processor.
Two wrongs don't make a right, but three lefts do.
Considering the companies involved, and the devices that they want to put the chip in, I'm really tempted to say that the Cell is nothing more than the biggest effort we've ever seen to get a DRM (trusted computing) CPU and associated parts on to the market. Obviously, this scares the bejesus out of me, since it would mean that these Cell devices would effectively be mod-proof; systems like Xbox Live already keep cheaters away, so this seems to be an attempt to stop modding alltogether. So, I have to ask: how is this going to benefit me, the consumer? If Live already gets rid of possible cheaters, how does stopping me from modding my box altogether help me?
If these assumptions are right, I don't like where this is going.
For FUCK'S sake, would it be hard to spell GRAMMAR without an E? Everytime some moron thinks he's being witty, he spells grammar like some sort of foetal-alcohol syndrome survivor.
i am just wondering if multicore will increase the performance of the games? does it work like the nvidia sli? i am not sure about games today if they can be fully multithreaded (if so, please do point it out). will they run let say geometric calculations on a core and ai engine on another core, etc.? just a thought because there may be a reduction of performance for single threaded apps if they would increase the total performance of the chip but decrease the performance per core.
Live your life each day as if it was your last.
1) If you don't know when to flip a burger you should not cook without adult supervision. 2) Can you now sue spatula makers if you get e-colli when you set it to well-done?
Buddy of mine made tools to do this using FORTRAN, although it did still have its limits.
Seriously - I don't know the intricacies of compiler design but I do know he won the obfuscated c contest several years ago and now works on multiprocessing tools for some very high end uses (like rocket motor simulations for NASA) - all in c. Last time I asked about the project he wasn't using gcc for it because gcc lacks certain libraries he needs (or something like that) - but it is still c.
I would say linux or no, there are many people of "insane genius" addressing this issue. It's not like we didn't know this was coming a decade ago... I mean, just look at transputer. Everything old really is becoming new again.
Note for the non-German user: a Handy is the common German term for a cell phone. I'm UKanian, so call the thing a mobile phone.
From http://journal.pcvsconsole.com/?thread=9240:
Say, for instance, that a website wanted to stream a TV signal to you in their new improved format DivY. They could send you a cell that contained the program instructions for decoding the DivY stream into a regular TV picture. Then they send you the DivY-endoded picture stream. This would work if you had a PS3 or if you had a digital TV, or even if you had a powerful enough PDA - assuming their design followed the new standard.
And you thought the broadcast flag was bad. When these folks talk about the PC platform being dead, they're not just talking about the silicon. They are talking about end-to-end computing, where you control your own hardware and software. This is a power grab, and a big one.
Our sources say the extra power of the new Cell processors will first be put to use running a grammar check on this post.
or is that still to "difficult"
TOO/0 = TO
D6 63 0D 70 89 81 BB 8E 7B 7C 5F 5D 54 EA AB 73
"-Real-time 3d graphics of cinematic quality will always be too slow for general purpose CPUs."
Says who? Seriously. At one point people had a hard time seeing 3D graphics! I can remember the awe that doom created. And some of the latest games come DARN close.
"-developing a game with AI that needs ten times the power of todays CPUs will take many man years and may not be that welcomed by the console audience."
Again, quite often decisions on a AI's design come down to cpu power. There are some cpu intensive stuff such as pathfinding that can use more power.
"-It's very difficult to do multithreaded apps, and the difficulty rises exponentially with the number of threads."
Yes and no. I do multi-threaded apps today. However a look at the past may help you out. The Amiga had multiple custom chips, and design philosophy could change to handle multiple cpu's. Seperate out the graphics engine, the physics engine, the OS, etc etc and you can use multiple cpu's easily. Will things use the full potential? Good programing will balance the load on the cpu's to do so. Even handing off sections of things to different threads when needed. Should be interesting.
"So what exactly would the be role of the CELL processor in PS3?"
More power! Perhaps seperating out video, sound, 3D etc.
"It would make much more sense if:
-Sony developed a platform that can move insanely great amount of graphics around, with the ability to do real-time raytracing, rather than providing so much general-purpose processing power."
Actually no. The insane amounts of CPU power we are discussing here can do a lot of this, and assuming that sony will showoff only the new cpu with the old graphics chip would be a mistake. Look for both.
"-Sony developed a graphics architecture that could really be parallelised, so instead of bringing out a totally new console, they could just up the graphics spec by adding more chips. They could save millions of dollars from developing and advertising the new console."
And lose millions of dollars by selling folks one system rather then 3. Also games are made to the lowest common denominator in the console market.
What is perhaps most interesting is the ability to have an OS written for cooperation so that processors could be introduced into the network. Or, having different OSes cooperate on a task should the OS designers wish to follow open standards. The Cell is designed for networking and cooperation, this is what is important.
-Real-time 3d graphics of cinematic quality will always be too slow for general purpose CPUs.
But Sony hasn't said that the Cell is going to be used for graphics. It's the CPU, not the GPU.
-developing a game with AI that needs ten times the power of todays CPUs will take many man years and may not be that welcomed by the console audience.
Irrelevant. But CPU power iis used by other expensive things, like inverse kinematics and physics. And let's not ignore the benefits of letting developers use languages that aren't so low-level as C.
-It's very difficult to do multithreaded apps, and the difficulty rises exponentially with the number of threads.
For traditional threaded applications, yes, but not if you use a sane method of handling large numbers of processes, like CSP (the book on which, BTW, was published 20 years ago).
Yes, you are correct, however leonig linked that patent 2.5 hours ago in reply to my post.
And i would liked to see the picture instead of the text.
however 2 points for effort.
What people seem to be neglecting is the fact that the thrust of the cell processors is to speed up floating point -- but leaves integer performance alone. Thus, "normal" processing tasks won't be any faster at all but those that require floating point (digital media, scientific) will indeed improve. Mostly, I'd think this affects video and audio card makers since it makes much of what they do redundant. So the potential I see is bringing down overall system costs -- not exceeding mainstream performance; there is a cost as well since it will require changing the software tool chain. So I'd say to expect to see cell-based solutions in "niche" markets for the short and mid term. Desktops and servers will likely continue to rely on proven designs (including multi-core) for sometime to come.
but what do i know, i'm just a model.
the beowolf clusters jokes?
I just wonder how much heat these cell processors will produce. Will they require heatsinks 10 times larger than we currently have?
"Hey look at my brand new computer."
"Man! This thing is gigantic! I thought it used very tiny CPU's"
"They are. 90% of the case is for the heatsink."
"Oh..."
Looks like a 21st century transputer.
Sony originally promised the PS2 could render 75M simple polygons per second, but also said the geometry engine's limit was 36M polygons per second. This figure is accurate, but like all such numbers in the graphics industry, it is achievable only in a single-function demo app. Such figures are useful only for comparing the raw performance of different designs.
m l
t ml
Sony never claimed the PS2 could support HDTV resolution. The company was very clear about the limited frame-buffer memory on the Graphics Synthesizer chip.
Sony did, in fact, make a multiprocessor PS2-based workstation, the GSCube, which combined 16 complete PS2-compatible subsystems. The "Graphics Synthesizer I-32" chip used in this system had a 32MB HDTV-sized frame buffer, leading me to speculate at the time (August 2000) that Sony would soon introduce an HD-capable PS2. They could have, but they never did. I can't get 'em all right, I guess. They demonstrated to me (personally) this system rendering scenes from the Final Fantasy movie in real time, so that wasn't hype either.
Here are the Microprocessor Report articles I wrote at the time (subscribers only):
http://www.mdronline.com/mpr/h/19990419/130501.ht
http://www.mdronline.com/mpr/h/2000/0821/143402.h
. png
Now wouldn't that be something? Computers with a cooling tower.
will this finally put an end to moore's law, but in the opposite direction one might expect?
When you're afraid to download music illegally in your own home, then the terrorists have won!
Clive Sinclair and Sinclair Research worked on
Wafer Scale Integration in 1983. http://www.electromagnetism.demon.co.uk/z005.htm
At first glance, this cell processor thing seems
about the same, i.e. use the whole wafer of circuits, instead of cutting it to individual processors.
Well, dual core Opteron's are really close to the store selves. I can see dual core Athlon64s being produced much earlier than the Cell chip. And if dual core versions work well(which eventually requires that programmers start using threads/processess effectively) then scaling to 4-core Athlons will be quite easy. Once you write "multithreaded" code, going from 2 to 4 or 8 or 128 threads can be quite natural, depending on the task at hand. I wonder whether a Cell chip, made up from many (slow) cores can beat a dual or quad core Athlon64 whose cores are very, very fast by todays standards. Remember that a dual-core cheap can be much easier to use effectively.
It's all a matter of programming, really. I wonder how many years will it take for the programmers to catch up and start writing really efficient parallel code. Intel has already paved the road somewhat with its HyperThreading technology so I suppose that some expertise is already available.
P.
The cell chips are interesting, but from the description they seem to be more of a variant of the current multi-core chips that Intel, Sun, HP, IBM, et. al, are doing. It sounds like they have the capability of putting more than one type of CPU on the same silicon, which is different, but it is still a variant of existing designs. It doesn't sound like they will be putting more than a hand full of cores on their chips. Sun is already talking about 8 cores on their new Throughput Computing chip line now, and more in the future. Sun claims that their Throughput computing chips will ultimately be 30x more powerful than what is out there today.
If you want to look at something really different, check out the Mathstar FPOA chips. Right now they can put up to 400 1Ghz processors / devices in different mixes on a single programmable chip. There will probably be a lot of applications that this will be a better match for than a multi-core IBM Power based chip, although in fairness they are probably targeted at different applications. On the other hand, maybe the FPOA will be the "IBM PC" of the multicore chip world. When the IBM PC first came out it was laughed at by the IT shops as being so small and limited, that it wasn't a "real computer". But the people who had PCs didn't have to wait weeks for the IT Department to rewrite their program, run their report, or crunch numbers for them on the mainframe. The PC changed the world. I wonder what the FPOA will do?
You could always buy a console-like pad for the pc.
Ewige Blumenkraft.
Here are some articles on the cell processor by Paul Murphy (he"s the author of The Unix Guide to Defenestration).
m l
m l
m l
m l
Among other things he predicts it will replace x86 as the preferred platform for Linux, and says that it plus a somewhat similar chip from Sun will crush Wintel.
http://www.linuxinsider.com/perl/story/34548.ht
http://www.linuxinsider.com/perl/story/34707.ht
http://www.linuxinsider.com/perl/story/34885.ht
http://www.linuxinsider.com/perl/story/34994.ht
I thought everyone knew how to use the 3 sea shells.
Slashdot = -1 Redundant, Asperger, kdawson FUD, Libertarian, and Linux
I stopped reading right here:
multiple synergistic processor cores capable of massive...
Exactly how are the cores "synergistic"? What the hell does that mean?
Typical marketspeak. No, we haven't just stuck a bunch of cores together. They're synergistic!
Real-time 3d graphics of cinematic quality will always be too slow for general purpose CPUs
Most Graphics processing units (GPU's) do very little of the actual work you see in the renering pipeline. Most, at best, push some triangles through. Some of the newer ones have capabilities to do more advanced work.
developing a game with AI that needs ten times the power of todays CPUs will take many man years and may not be that welcomed by the console audience. However physics systems -- blowing things up, interactive buildings, terrain, etc. are things console gamers eat up, and these things are performed in the simulation tree in the game engine by the main CPU, and not by the video card, in the vast majority of cases (all current cases that I know of).
Additionally, watch X-Play for a few minutes, listen to them complain about the AI in console games, and then say that gamers don't want a better AI.
It's very difficult to do multithreaded apps, and the difficulty rises exponentially with the number of threads.
Incorrect.
Each thread is a separate program which should be isolated from other programs except for input and output, exactly as every Unix program is already typically designed.
The complexity of the system varies linearly, and only linearly, with the number of resources that must be synchronized. This is because the threads should not be interdependent.
You can eliminate even this linear encroachment in many cases by using circular buffers and limiting the number of writers.
Sony developed a platform that can move insanely great amount of graphics around, with the ability to do real-time raytracing, rather than providing so much general-purpose processing power.
Well here's the issue. I can raytrace fine, in real time, moderately complex objects at 320x240 even on relatively low-end hardware. It's not that difficult, it doesn't take as much processor power as most people are throwing at the problem.
The thing is, polygons are a ton easier to work with in the physics engine, in the AI engine, in the collision engine, when doing hit tests, etc. The graphics processor does not do all of this work in any modern game that currently exists, and aren't even remotely close to doing so on video hardware that currently costs more than many people's computers from nVidia and ATI.
Sony could waste its money trying to provide this feature through graphics hardware, as you suggest, or they could continue to buy 3rd party video hardware (NEC PowerVR) from a company that makes its entire business out of graphics processors.
So far as moving data around, PowerPC used to be pretty good at that compared to x86's, it used to be far beyond x86's.
Sony developed a graphics architecture that could really be parallelised, so instead of bringing out a totally new console, they could just up the graphics spec by adding more chips. They could save millions of dollars from developing and advertising the new console
The big issue here is without multithreading or a multiprocessor core, this would be meaningless.
Lets say you have a physics engine. This needs to do some transforms. If the physics engine runs on the same thread the game runs on, then you have a problem because every time that you do a transform (and you'll be doing a lot of them), your game stops while it waits on the card. If you've got a really slow bus (like an AGP bus, or even a PCIx bus), you could be waiting a long time for the data to come back.
And so you need multiple threads.
If the arguement is the graphics chips could do the physics, etc. then essentially you're still multithreading, because you're going to send a program to the graphics chip, and it's going to run it.
It can be argued you could do physics with vertex shaders, etc. but most game physics are complicated enough you really need to be in code to do it.
The big thing with parallel graphics chips is that there aren't a lot of methods of making it work. either you divide the screen in half and use
If your code is acting bloated, and is running rather slow, it's likely and predicted that some loops you will unroll.
games.slashdot.org/comments.pl?0 942070
sid=131085&cid=1
"On-chip hardware in support of security system for intellectual property protection."
The PS-2 and the XBOX have very different graphic pipelines. The PS-2 has less complex graphics hardware but the CPU is more powerful because it has extra vector processing units. At least one of the vector units is asynchronous. The XBOX puts a lot more hardware in the graphics and has a "normal" i86 CPU.
It is likely that the Cell will be more like the PS-2 and have multiple CPUs and keep the graphics hardware simple. IBM is already in the multi-core CPU game. One of the advantages of this design is that the game programmer can use the enhanced CPU power do implement whatever algorithms they want, rather then be forced to use whatever is hard wired in the graphics pipeline. This is one of the reasons that graphics cards now have user programmerable GPUs. Flexability is important.
It is also likely that the Cell CPUs will be loosely coupled and communicate using an internal on chip network. This is not like current 2 or 4 core CPU designs. Games can make use of this divison of labor rather efficiently. Each CPU can have ir's own dedicated memory, which eases the memory bandwith problems. This is one way the chip can be "10" times more powerful. Look at the MIT RAW architecure or earlier work on one the WARP architecture (at USC?).
This is what I was trying to do with Chuck Moore, back in 2000... No one would believe us. Or want anything to do with it. Now it's the hot new tech!
Reported here.
Ask Chuck Moore About 25X, Forth And So On From April 2001.
Chuck Moore Holds Forth
I am always doing that which I can not do, in order that I may learn how to do it. - Pablo Picasso
if this is the next gen of the powerpc chip it's probably safe to bet it's only a matter of time before imacs and powerbooks are running these chips...
with 2 of the dominant game consoles and one large computer maker onboard how long before we see powerpc desktops competing with x86... better yet, how long before windows longhorn for powerpc ships (probably 2012)
Get your torrents...
bingo
If the hardware itself supports network applications, with perhaps not even a traditional OS in the middle, say goodbye to any thoughts you once had about being empowered by owning a general purpose computer. Forget about cars with their hoods welded shut. This is more like being sold a toolbox full of tools that's welded shut. Inside, the toolbox will build little nifty machines to do their master's bidding, as long as you keep pumping coins in the slot. "Can I watch this? Can I play this? Pleeeaaze???"
What's with all this "Cell" bullshit? It's just a FREAKING VIDEOGAME CONSOLE. If this new space-age technology they're developping were as revolutionary has they are trying to make us believe it will be, they wouldn't waste on in a console system, at least not right away.
http://graphics.cs.uni-sb.de/~sidapohl/egoshooter/
replacing it with NEW Folger's Crystals! (lets see if they notice the difference)