Intel Larrabee

[powerarmour]

But who would have thought of a intel GPU into a Sony game box? That must have only happened because Sony is cash strapped (for this kind of expense) and Intel needs a graphic win and momentum, and will acquire the expertise with this deal. That will be even weirder if the PS4 ends up having a IBM CPU (cell) and a Intel GPU. Crazy interesting times!.

Well it was already rumoured, so not really a shock I guess :-

If Larrabee is going in a console, it'll likely be in the PS4 according to the rumours

But now the Larrabee is in the PS4, I don't think there will be any need for the Cell also, or if the Cell is in there it'll only be for backwards compatibility purposes only. If they do keep the Cell it would be a bizarre hybrid of PowerPC & x86 processing, I don't think Intel would be comfortable with that..!, surely it would much better from a minimalist design/cost/programming POV to run the whole OS/system/graphics on a single processing architecture and just fill the machine up with ram.

For Sony, Intel is a dream hardware partner, it couldn't really get much better, provided they are not bullied about too much. Could be more in this also like cheaper deals/timed exclusivity for their Vaio PC line etc.

By the time the PS4 is launched this will be 2nd or even 3rd gen Larrabee architecture so performance could be anything, pure 1080P real time raytracing a definite possibility.

The other news you can read into this, is how this will affect PC gaming...?, with an 'Intel' console available it'll either be a kiss of death, or a bright new level of compatibility/convergence between the two platforms (provided you've got a 'Larrabee' in your PC of course).

...but how that will roll with MS?

MS are probably praying that AMD doesn't go under before they've finished making the hardware on the new Xbox...

Playstation coders must be very happy, as intel usually releases very good tools and coding resources.

Indeed, how the worm turns, from one of the hardest consoles to develop for ever, to what looks like will be the easiest...

[Tabajara]

The other news you can read into this, is how this will affect PC gaming...?, with an 'Intel' console available it'll either be a kiss of death, or a bright new level of compatibility/convergence between the two platforms (provided you've got a 'Larrabee' in your PC of course).

MS won't help with Larrabee drivers or PS4 game ports, that's for sure.

...but how that will roll with MS?

MS are probably praying that AMD doesn't go under before they've finished making the hardware on the new Xbox...

MS must have checked that out before commiting. AMD is kind of healthier now, with ATIs profits and with no fab expenses, and should be bailed out by the GF owners or even by MS if its nedded.

Playstation coders must be very happy, as intel usually releases very good tools and coding resources.

Indeed, how the worm turns, from one of the hardest consoles to develop for ever, to what looks like will be the easiest...

Crazy alliances that were born out of crazy market conditions... But that can arrive at very interesting destinations. This same processor might end up at the PS4 and at a Apple media box.

[Tabajara]

There is a nice speculative piece by Theo Valich at BS news. Here goes a piece:

Intel re-opens Larrabee talks with AIBs: EVGA, XFX?
11/7/2009 by: Theo Valich

Currently, there isn't a project in Intel that is pulling more resources [both human and financial] than Larrabee. Given the status of the program, we are not surprised to see top engineering dogs from CPU divisions flying between Larrabee centers, pulling everything together in order to get Larrabee out the door during 2010. Our sources still say that product introduction prior to CeBIT 2011 is unwise and risks damage to Intel's brand image, given the immature state of the project.

Then again, we'll see who will win - engineering attitude is now in the front seat. It is more than obvious how marketing had its way with Larrabee so far, and while talking crazy promises to the press or analysts ["We'll have a 32nm CPU die and 45nm Larrabee die for notebooks in May 2009" - former Intel Exec] can't cost you more than a temporary stock bump or downgrade, talking the same to eco-system who then has to invest serious money into preparing their own infrastructure - can cause a loss of several million real-world dollars. For instance, we learned that one whole marketing campaign [digital future livelihood] was created for Larrabee and ultimately used on a product line from Intel's competitors speaks volumes in what situation some AIBs found themselves in.

In order to compensate for the lost investment that angered partners who signed and invested significant resources in preparation for "Larrabee coming in first quarter 2009", Intel had to change the attitude from "we're almighty" to "we know we screwed up". In fact, in conversations with our sources, it was interesting to see the change that showed to them how Intel managed to eat a humble pie and wants to talk on equal terms.

Furbie - this is how some partners are calling nVidia's GF100/GT300/Fermi ArchitectureThis is quite significant change, since back in 2007 and 2008, we were hearing AIBs telling us of the attitude by top Intel Larrabee guys such as Patrick P. Gelsinger, Jim Woodruff and others. That attitude was that "AMD is harmless, nVidia is dead." The only two companies that signed on-board were EVGA and XFX, with several other vendors being refused due to lack of global presence. Allegedly, it was this attitude that annoyed AIB [Add-In-Board] manufacturers with some major partners refusing to commit to the program. However, in the light of Larrabee troubles and missed deadlines, Intel actually turned into a more responsible company. Intel also had one unexpected ally - according to our sources, nVidia is refusing to eat the humble pie even though they massively screwed up with GT300/Fermi/Furbie and that's angering a lot of people in their ecosystem.

The difference in attitude between companies with delayed products is quite interesting: while nVidia is now putting majority of GT200 chips in Tesla cards [due to a recent sea of orders] and leaving their GeForce partners to bleed dry, Intel has re-approached its previously signed partners and some hopefuls with a new attitude. The chip giant is now offering to bundle chipsets at "rock-bottom to free" prices with Larrabee cGPUs as a compensation for created expenses. Before you cry foul and "antitrust", this is a standard business practice of compensating for previous screwups made by one party. This is nothing unusual.

There should be some truth inside... And the change of tone of BS news towards nVidia is interesting in itself.

As was seldom written in some other plaice, the whole shebang can be found after the jump:
http://www.brightsideofnews.com/news/20 ... c-xfx.aspx

[powerarmour]

what-s-intel-up-to-v2-t63445-590.html#p633598

Cancelled...

[powerarmour]

More...

Intel Cancels Larrabee Retail Products, Larrabee Project Lives On :-

http://www.anandtech.com/weblog/showpost.aspx?i=659

[Tabajara]

Arstechnica's take:

If the fact that Larrabee is "launching" not as a GPU, but as a kind of multicore graphics demo unit, sounds like a cancellation to you, that's because it kinda sorta is. It's not a cancellation in the sense that Intel is throwing in the towel on discrete graphics, because that's definitely not the case. The company reiterated that it still plans to launch a many-core discrete graphics product, but won't be saying anything about that future product until sometime next year. Whatever it is, it won't be the hardware/software combination that it previously announced, and that we described in our coverage of Intel's big August 2008 Larrabee reveal. It will be something else, and Intel wouldn't even characterize the relationship of that future something to the current Larrabee product.

The main issue behind the delay, it appears, was the hardware. That's not surprising, because Larrabee is a big, complex part, and it's quite a departure from anything that Intel has done. The hardware delay would have resulted in a software delay, and if Intel were to launch Larrabee with an immature software stack then it would be roadkill as a GPU.

More at:
http://arstechnica.com/hardware/news/20 ... n-2010.ars

[Tabajara]

Happy new year everyone!

Rumor: Sony considered Larrabee for PS4, will reconsider for PS5, PS6

[Tabajara]

There is an interesting article at Ars about the demise of Cell, well worth a read:
http://arstechnica.com/hardware/news/20 ... s-cell.ars

At a point in the article it compares the Cell and Larrabee, with some interesting conclusions; even Fusion is mentioned.

The differences between Intel's future CPU/Larrabee hybrid and IBM's Cell may seem small, but they're critical. Cell's smaller floating-point cores are not general-purpose—they're specialized and they implement their own instruction set. These small cores also don't have cache coherency and a real virtual memory implementation. Rather, they have "local store" pools of programmer-managed memory that make them a huge pain to program for. So in terms of programmability, the difference between Cell and a CPU/Larrabee hybrid is night and day.

OK, it's not exactly night-and-day; it's more like "night and early evening," because there are still significant hurdles to actually designing software that takes full advantage of higher levels of parallelism. So while it may be easier to write code for a heterogeneous x86 CMP, the higher-level challenge of breaking an application up into parallelizable tasks still remains.

As for comparisons between Cell and what AMD is doing, this is more difficult, because AMD isn't dropping many hints about what it has in store for the final stages of its CPU/GPU "Fusion" plan. I've suggested that they won't go the CPU/Larrabee route described above, but that remains to be seen.

Speaking of Larrabee, the oft-floated rumor that Intel's upcoming GPU will form the brains of Sony's PlayStation 4 hinges in part on the idea that Sony and IBM will abandon the Cell, and not double down on it. Now that we know IBM is ditching Cell, does this mean Larrabee is a shoo-in? Not quite. This removes one of the objections to Sony using Larrabee, but an even bigger objection remains: even in 2011, which is the earliest that we could expect a PS4, Larrabee still may not be as good at graphics as discrete GPUs from NVIDIA and AMD/ATI. There's also an outside possibility that Sony could just use one or two of the last version of Cell in its console, and let Toshiba do the fabrication.

[powerarmour]

Intel Kills Larrabee GPU, Will Not Bring a Discrete Graphics Product to Market :-

http://www.anandtech.com/show/3738/inte ... -to-market

Fusion CPU's it is then...

[Tabajara]

Larrabee wasn't killed, it was delayed:

We will not bring a discrete graphics product to market, at least in the short-term. As we said in December, we missed some key product milestones. Upon further assessment, and as mentioned above, we are focused on processor graphics, and we believe media/HD video and mobile computing are the most important areas to focus on moving forward.

Charlie wrote about Larrabee the other day. I'll quote his conclusion:

The new Larrabee 2 is what the old Larrabee 3 was meant to be, but retargeted to what a high end GPU will be in 2012. Instead of chasing hoping to get the next three generations out less late than the last, Intel just scrapped the timeline, and went back to the drawing board. Specs changed, timelines changed, and all sorts of details were rethought.

Larrabee 3's converged pipeline ISA is now going to be the only ISA the public ever sees. Compiler tricks and dual code path binaries are no longer necessary. When people get their hands on a Larrabee, be it the lucky few who can talk their way into a robins egg blue Larrabee 1, or the new one, the coding path will be set.

In the end, Larrabee was effectively delayed for 3 years, but a lot was learned on both the software and hardware sides. Those who declare the project dead, or say Intel ran away with it's tail between it's legs either have an axe to grind or simply don't understand what happened. Intel is creating a new paradigm, changing the graphics game from the ground up, not just making a GPU. It is a slow and long term process.

The whole converged pipeline CPU/GPU is going to be where both AMD and Intel eventually end up, but their paths are going to diverge near term. Any company that is not aiming for that goal is not going to be in the CPU or GPU game in 3 years because neither will exist as such. Then again, they will have a lot of company in the graveyard of those who fought Moore's law. Two years.

I don't know if two years equals "short term", but this seems to be intel's strategy. I don't think this articles are contracditory, but that none is "right", both writers speculate a lot and their conclusions are complementary. The truth must be somewhere between them.

The title "Intel Kills Larrabee GPU, Will Not Bring a Discrete Graphics Product to Market" is just wrong, though.