Slashdot Mirror
Intel Kills Consumer Larrabee Plans
An anonymous reader tips news that Intel has canceled plans for a consumer version of their long-awaited and oft-delayed Larrabee chip, opting instead to use it as a development platform product. From VentureBeat:
"'Larrabee silicon and software development are behind where we had hoped to be at this point in the project,' said Nick Knuppfler, a spokesman for Intel in Santa Clara, Calif. 'Larrabee will not be a consumer product.' In other words, it’s not entirely dead. It’s mostly dead. Instead of launching the chip in the consumer market, it will make it available as a software development platform for both internal and external developers. Those developers can use it to develop software that can run in high-performance computers. But Knuppfler said that Intel will continue to work on stand-alone graphics chip designs. He said the company would have more to say about that in 2010."
No consumer version means this will turn into another i860. I guess ATI will remain the only viable competitor to NVIDIA then.
So they intend to take a product, who's chief advantage was that it could run old x86 code, and only sell it people who are designing new software? Am I the only one that sees a problem with this?
A nicer way of saying:
Uhm, guys, remember how we were supposed to ship a year ago and said recently we will ship a year from now? Well, add 5 to that now...but we will provide and totally kick ass, promise.
One that hath name thou can not otter
In case you've forgotten what a Larrabee was (like I had), it was Intel's planned graphics / vector processing chip, competing with nVidia and AMD / ATI graphics systems. Here's the Wikipedia article.
Bill Stewart
New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
Hmm... I think Intel's plan is for Larrabee GPU's to launch at the same time as Duke Nukem Forever! :)
http://www.brightsideofnews.com/news/2009/12/2/intel-larrabee-finally-hits-1tflops---27x-faster-than-nvidia-gt200!.aspx
way faster than amd's or nvid's hottest....
Sarah O'Connor? Would she be the mother of Jonny O'Connor from County Cork, who'll lead humanity to victory in the war against the machines and the English?
So the next mini, low end imac and 13" macbook's will be stuck with shit video and the mac pro will start at $3000 with 6 core cpus.
Will apple move to amd just to get better video in low end systems?
I spent most of internship in intel arguing with people hyping larabee as the 2nd coming of jesus that it would never happen... And now i can finally say HAH!
Apple already dropped GMA for low end stuff, they're using GeForce 9400M instead. They're also using Radeons on most iMac models.
I'm not sure what you're referring to. Macbook and Macbook Pros are configured with Nvidia 9400M or 9600M chipsets. They may not be powerful but at least they are dedicated graphics solutions. Far superior to Intel Integrated graphics, and they provide working hardware acceleration for H.264/VC-1. The Intel G45 chipset does so - but only with MPC-HC - not for commercial blu-ray playback - and it had some corruption last I checked.
This is being mis-reported or mis-communicated by Intel, I believe.
The first version of Larrabee silicon isn't going to consumers, that's all.
From the consumer's perspective, it's a delay. Yet to be seen if it's fatal.
Otherwise, who'd want to use it to develop software?
I'm not dead yet!"
Maybe it's just resting?
Stunned?
Pining for the fjords?
I'll show myself out.
Mod me down with all of your hatred and your journey towards the dark side will be complete!
Intel has shown real commitment to supporting their video hardware on Linux with full time staff employed to produce high quality open source drivers in addition to providing open specifications for (most) of their contemporary hardware. Unfortunately this hardware provides only limited 3D acceleration. I was hoping that Larrabee would conflate these two and provide vendor supported, open, high performance accelerated 3D for Linux.
So much for that happening anytime soon...
I can't understand why Intel cedes the GPU market to it's competitors. Have I been getting duped into paying hundreds while everyone else gets free GPUs? People are paying good money for these chips, right? NVidia's got Playstation 3 and Apple. ATI got the 360. Intel has nothing the the discrete GPU market at all. Why? What blocker within Intel prevents them from taking a piece of that pie?
Maw! Fire up the karma burner!
I think the announcement of the 48-core Intel 'Bangalore' chip just recently is not a coincidence.
When I first read about the Larrabee chip, I thought the decision to make it a cache coherent SMP chip to be simply insane - architectures like that are very difficult to scale, as the inter-core chatter scales roughly as the factorial of the number of cores. Remember how Larrabee was designed around a really wide 1024-bit ring bus? I bet that's required because otherwise the cores would spend all of their time trying to synchronize between each other.
So, Larrabee is effectively cancelled, but only a day or two before Intel announced an almost identical sounding part without cache-coherence! It sounds to me like they've given up on the 100% x86 compatibility, and realised that a chip with some extra instructions around explicit software controlled memory synchronization and message passing would scale way better. Without cache coherence, a "many core" chip is basically just an independent unit repeated over and over, so scalability should be almost infinite, and wouldn't require design changes for different sizes. That sounds like a much better match for a graphics processor.
While Intel kept their cards relatively close to their chest, from all of the presentations I've seen, no first-gen Larrabee chip could scale beyond 24 cores even with a 1024 bit bus, while the new Bangalore chip starts at 48 cores. There's no public info on how many lanes Bangalore has in its on-chip bus but based on the bandwidth of its 80 core experimental predecessor, I'm guessing it's either 32-bit or 64-bit (per core).
I like the Fusion concept, and feel that Intel will ultimately be forced to imitate it as well. Their abandonment of Larrabee is consistent with that. Hell, I even hope that Scorpius will become the foundation for Nintendo's Wii-2 or Wii-HD.
i3/i5 cut off nvidia and the low end cpus have gma build in and apple likely will put i3 in the mini and stick it with carp video at $800 as well.
Wow... thanks for your insight! Should have known Intel would be logical even about their failures, and roll them over to something that has a chance of applicability. The only thing I wish they would do is skip the 64-bit crap and make 128-bit architectures that are compatible with both 32- and 64-bit predecessors. It would ease the development of new applications since the life time of 128-bit archs would be decades as opposed to developing all 64-bit apps to only have 128-bit archs appear in 5-10 years.
I'm not sure if you're trolling or not, but 64-bit memory capacity is not "twice" as big as 32-bit, it's 4.3 billion times as big. That's more than just 5 to 10 years of Moore's law, that's more like 50 years. Physical bus widths have nothing to do with architecture bitness either, there are memory buses for 64-bit architectures that only have a few pins.
sounds more like gma video good on paper but the divers / performance is just not there.
Ah, I thought you were talking about their current rather than future offerings.
Many people really don't care about their graphics card. If you don't do games, an Intel chipset graphics unit works fine. It accelerates the shiny interface in Windows 7 and everything is nice and responsive. For business uses, this is plenty.
Ok well if you do care about games, then you want a discreet graphics solution. Integrated solutions will just never do well. Big reason is memory. You can make your card as fast as you like, if it shares system memory it is severely bottlenecked. Graphics cards needs their own dedicated high speed memory to perform well.
As such I just don't see ATi having a slightly better integrated solution as something people will care much about. The bigger question is who makes the better CPUs that that is firmly in Intel's arena. Their CPUs are faster, and can be lower power. So regardless of if you want a power saving app or a performance solution, they've got a good chip.
AMD really has to get their chips up to snuff before they'll start competing with Intel more. They don't have to beat Intel at everything, but they need to have at least one area they are better for and they really don't seem to. Also they need to do better with chipsets and motherboards. A big advantage Intel has with regards to the reseller market is that they do their own solutions. Intel will sell you a CPU, chipset and motherboard and they all work together well. OEMs like this, cuts down on supply chain problems and problems of vendors blaming each other when there's trouble.
This has also historically been a weakpoint for AMD. I remember when their Athlons came out and there was no question, they beat the P3's price/performance ratio. They were the kings of the hill. I bought one... and returned it two weeks later. The reason? Chipsets. I could not get a chipset that would work with my GeForce 256 properly. They had poor regulation of the AGP signal and it just wouldn't work. Bought an Intel chip/board and it worked flawlessly the first time.
So when AMD has a good CPU/chipset/mobo combo and CPUs competitive with Intel in at least one arena, I think maybe they'll make gains. Until then, I think they'll mainly be relegated to "cheap brands" and to enthusiast BYO systems.
"mostly dead". maybe Miracle Max has a cure!
I have a 4 year CS degree and I can tell you with certainty that that blogger is full of shit. The problems that are already parallelizable, are easily multithreaded with current technology. The problems with serial dependency, are not, and never will be, easily multithreaded.
Rendering graphics is already done, because its easy to split the task of rendering a bunch of pixels into pixel-sized chunks. Each small thread can read from the same shared memory (the scene graph and textures, etc.) and write to a distinct location (its one pixel in the frame buffer).
Encoding video using motion-compensation techniques (basically all modern video codecs) will never be satisfactorily parallelizable because the best bang/bitrate can only be achieved when frames are processed serially. Frames need to be processed as a whole to optimize for panning and other full-scene motion, and the results of the previous frame's motion analysis is typically needed to compute the next delta. You can break the processing up into multiple threads easily enough, but you miss out on opportunities to make the output more efficient or better looking.
When Mr. PseudoScience blogger can parallelize the video encoding problem without so many dependencies that its essentially a serial process, then he should get some credit, not before then.
I suspect the Intel and PowerVR partnership may have have something to do with no consumer Larrabee plans. This partnership already has resulted in the 3100ce and PowervR has been working on some 1080p media accelerators. Larrabee does use a lot of power for the level of performance it would offer as a 3D chipset perhaps Intel and PowerVR have came with with something that does not use 160watts.
My wife and I play wow but most users prefer to use a wii or ps3 if they want to play games.
Its frustrating and I agree that the intel chipsets and integrated chips (not true video cards) put desktops 5 - 6 years behind and piss off game developers forcing them to port only to consoles.
The netbook phenomena shows this trend for slim boring graphics that are cheap cheap and uh cheap.
Most game developers have left the pc as a result due to angry kids whose parents get a nice i945 graphics chipset computer for them and they wonder why Crysis is a slide show.
http://saveie6.com/
Sounds like a character played by Rodney Dangerfield in a teen grope movie.
... and then they built the supercollider.
The problem is, a many-core non cache-coherent x86-like system isn't particularly interesting. The big advantage of Larrabee was that you could treat it like a normal SMP system, including (presumably) running standard multithreaded C code on it. Once you have to deal with memory synchronization explicitly, Larrabee starts to look a lot more (from a programming standpoint) like Fermi, Cypress or whatever other Nvidia/ATI GPUs are out at the time.
There's nothing magic about x86/AMD64 in the HPC world. It's attractive because it is cheap and has good performance. Clusters can, have been, and still are built using POWER and other architectures.
but at least they are dedicated graphics solutions
Actually, the 9400m is not. It uses system memory but does a much better job then Intel. It also acts as the memory controller and does system IO. The reason for the parent's comments is that all future Intel CPUs will have integrated memory controllers (like the i7 and i5) and an integrated GPU. Performance will suck but it will make for cheap systems. This will make it difficult for system builders to make a low end system with good graphic performance as the market for such systems will be small. The smaller market will reduce the quality/performance of available parts for those system builders - one of which is Apple.
The problem is, a many-core non cache-coherent x86-like system isn't particularly interesting. The big advantage of Larrabee was that you could treat it like a normal SMP system, including (presumably) running standard multithreaded C code on it. Once you have to deal with memory synchronization explicitly, Larrabee starts to look a lot more (from a programming standpoint) like Fermi, Cypress or whatever other Nvidia/ATI GPUs are out at the time.
There's nothing magic about x86/AMD64 in the HPC world. It's attractive because it is cheap and has good performance. Clusters can, have been, and still are built using POWER and other architectures.
But for "embarrassingly parallel" problems, which are the target application for these chips, cache coherence is often not necessary, and simply imposes a design burden. There are lots of problems where it's better to have 1000x the performance than 1/2 the developer time.
It may not even involve less development time: Others have pointed out that the Unix "fork" mechanism combined with "copy-on-write" at the memory page level would also work, and wouldn't require cache coherency. Similarly, any existing code designed for message-passing supercomputers would work out of the box, with only a recompile using a new library. Developers just have to start thinking in terms of "many processes" instead of "many threads".
I suspect that in the long term (decades), cache coherency will simply not scale, and most computers will use explicit message-passing internally, even at the single processor level. The transition has already started: most new servers are NUMA systems, where there's a concept of "near" and "far" memory visible to the software, and most of the real heavy lifting in PCs are done by the GPUs, some of which do not have complete cache coherency across all cores.
"The new macbook pro, now with AMD... and only 3 hours of battery"
Somehow I think AMD still has a few things to learn about mobiles, and that's the mac's main market.
Intel; 5th December 2009, for immediate release:
Intel Corp. today announced the release of their new FirstPost processor(tm), known internally as FristPsot and FrostyPiss. This new processor will let you post first on any web internet board. Demos of this processor's achievements have been given showing astounding performance on sites such as Slashdot.
The FirstPost processor(tm) will be available in Q1 2010. Sorry, 2014.
It's likely that Apple will have to use discrete graphics on all but the lowest-of-the-low (a theoretical $799 MacBook) in order to not regress graphically. NVIDIA GT240 could be an option as a discrete replacement for the integrated 9400M.
It will require motherboard redesigns, but the CPU will force that anyway. The Intel I/O hub for the new systems is quite small, so there should be room.
However Apple have regressed graphically in the past (Radeon 9550M -> Intel 2006 rubbish integrated graphics). It wouldn't fit in well with OpenCL and all that stuff that Apple harp on about though.
The ATI patents were not included in the recent settlement.
I am not going to reply to you again as Anonymous Coward on Slasdhot. Commenting on my blog requries a wordpress logon. It would be great if you got a Slashdot user name. This will enable people to follow you.
Best Regards
Foredecker
Jibe!
Hi APK - thanks for the posts. I see them.
A few things.
Also, did you have a chance to get signed into Microsoft Connect and apply for a connect relationship with the Windows networking team?
Thanks,
Foredecker
Jibe!
You're an AC, but your use of "we" is interesting in the grammar analysis realm. You are clearly posting from a subjective point of view and your post is worthy of note.
I think your opinion is interesting, though I don't share it.
Help stamp out iliturcy.
Trolls thrive on attention. Don't feed them and they go away. Validate them with responses and they will swarm you as if they were ducks and you were unwrapping loaves of bread.
This is pretty basic online stuff. You have been sheltered, haven't you?
Help stamp out iliturcy.
Ya - I know. But someting in me couldn't resist. I suppose its character flaw... :)
Jibe!
Dud - you are so the APK guy. Its pretty hilarious that you are posting as two differenet people. Note, I wasn't critizing your (or APK's) spelling or grammer - but the overall polimic and disjoint style. I'm a bad speller... I live with it :)
Jibe!
My appologies for not being more specific: I should have said: "I'm not going to reply lengthly replies to here on Slasdhot."
About supplying email alias to MSFT people: It's just the wrong thing - on many lelves - for me to post the email address of MSFT employees on any public forum.
The most effective way for you to get in touch with teams at MSFT and to have an actionable conversation with them is to use the Microsoft connect site.
Also, i belive I said in another post somewhere that I enabled no-login comments on my blog. So, you don't need a wordpress login to comment.
Ok, its fine that you want to protect your IP address and not be tracked. That's cool.
I'll continue to post longer repsonses on my blog, and short ones here.
Best Regards -Foredecker
Jibe!
I still think you and the other Anonymous Coward are the same guy.
Yup, you caught me red handed. I'm a frequent mis-typo-ist. I live with it...
Again, I didn't criticize your spelling or grammar. Its your polemic, and now very argumentative and picayune style.
It’s almost as if you are trying to win some court room trial. I told you in my original post that I sent email about this internally and to be patient in waiting for an answer.
Despite your pushy and obnoxious approach - I think you have a really interesting question and raise some good points. I also see that your motives are generally helpful in nature.
So, really - you are spending way to much time sending me these long re-re-re-iterated replies. I get it, you don't like that the '0' address is going from the Windows hosts file implementation. It’s your time, but its not helping me help you.
-Foredecker
Jibe!
Hi APK,
Im unclear on what you are arguing about. Im not arguing with you and Im not asking you to apologize for anything.
Of course, larger files take longer to load.
In any case, like I said before. I appreciate your questions and points about the host file. They are interesting and Im working on finding an answer. Again, please be patient.
At this point, I dont know why the 0 thing was removed from the hosts file parser. Maybe it was an oversight, maybe there was a good reason for it. Maybe you are right and it needs to be supported again.
On another note, I have a few questions for you.
You mention the proper channels. Again, I encourage you to use Microsoft Connect. Thats the proper channels. Im only doing this because its interesting. Slashdot anonymous posts are most definitely not the proper channels.
Thanks
- Foredecker
Jibe!
No worries APK. Its good to meet you as well. Its your IP address to control.
Jibe!