Sign in with
Sign up | Sign in
Your question

Is Intel pulling a Fermi?

Last response: in CPUs
Share
November 20, 2009 12:07:40 PM

We all saw earlier nVidia showing a card that wasnt there, even have pictures of its CEO holding up a fake.
Earlier this week, Intel showed off what some believed to be LRB. Some even wrote about it as such
http://www.theregister.co.uk/2009/11/17/sc09_rattner_ke...
While others took the time to see the others pulling a "mistake"
http://www.computerworld.com/s/article/9140949/Intel_to...
Its not hard to see someone was pulling something or other

More about : intel pulling fermi

a b à CPUs
November 20, 2009 12:33:49 PM

Fermi vs Larrabee...which one will fail the hardest?
November 20, 2009 12:40:23 PM

Better question, where are either?
Both showing other stuff for promoting their "real" solutions?
It appears the numbers Id pulled earlier show that LRB, if it performs as this "Polaris" model does, will be 3+ times better than a 280.
Knowing that Fermis DP is 5.8 times as good as the 285, it looks like Fermi at this point, and thats with Fermi and their disappointing clocks.
Related resources
November 20, 2009 12:59:45 PM

Whatd really be funny is, if they both keep waiting it out, and showing other things besides what they really want to sell, is if ATI continues forging ahead and creates a killer gpgpu model with their current arch, and could just be that surprise coming in January as rumors go
a b à CPUs
November 20, 2009 1:29:11 PM

Wait did I read this part correct :-

"On the SGEMM single precision, dense matrix multiply test, Rattner showed Larrabee running at a peak of 417 gigaflops with half of its cores activated (presumably the 80-core processor the company was showing off last year); and with all of the cores turned on, it was able to hit 805 gigaflops. As the keynote was winding down, Rattner told the techies to overclock it, and was able to push a single Larrabee chip up to just over 1 teraflops, which is the design goal for the initial Larrabee co-processors."

So an overclocked larrabee with all cores on can do 1 teraflop single precision? Thats only lke 5x less than the 5970 at stock? The interesting part being it seems Larrabee is having heat issues and that is why they are running it with half cores?

Lol this is going nowhere. Did I read something about $10bn spent on Larrabee so far? What a staggering waste of cash on a total lemon...this is worse than itanic by far.
a b à CPUs
November 20, 2009 1:30:45 PM

Where are the double precision numbers?

Also...who is going to bother with this lol? intel actually expects people to retrain and change the way it's been done for many years...for a slower product?

Fermi did kill Larrabee, it's over for intel in that regard.
November 20, 2009 1:58:58 PM

I'll find the figures, but LRBs SP isnt that far from their DP figures.
Either way, youre right, but, part of their "showing" for whatever this really is, is that itll shut down certain parts and cool, while re-intiate other parts that are cooled from non use.
Seems to me, a few years back, some thought something at rest was a waste, and certainly foolish to use it as a promotional feature, to me, this looks the same.
Also, Ive seen info where LRBs scaling starts to fall off at over 32 cores, so its no wonder they "showed" only 40 working, as anything higher wouldve shown poor scaling, which is another LRB "feature", its great scaling abilities.
Looks like theres alot to do, and did I mention drivers?
a b à CPUs
November 20, 2009 5:41:52 PM

That wasn't LRB, that was a research prototype CPU. Intel has been working with several other vendors in creating prototype multi-core processors.
a b à CPUs
November 20, 2009 5:48:58 PM

jennyh said:
Fermi vs Larrabee...which one will fail the hardest?

Maybe neither of the two will fail. Why do you presume failure? Give us some valid reasons please... because without valid reason your comment is unreasonable.


jennyh said:
Wait did I read this part correct :-

"On the SGEMM single precision, dense matrix multiply test, Rattner showed Larrabee running at a peak of 417 gigaflops with half of its cores activated (presumably the 80-core processor the company was showing off last year); and with all of the cores turned on, it was able to hit 805 gigaflops. As the keynote was winding down, Rattner told the techies to overclock it, and was able to push a single Larrabee chip up to just over 1 teraflops, which is the design goal for the initial Larrabee co-processors."

So an overclocked larrabee with all cores on can do 1 teraflop single precision? Thats only lke 5x less than the 5970 at stock? The interesting part being it seems Larrabee is having heat issues and that is why they are running it with half cores?

Lol this is going nowhere. Did I read something about $10bn spent on Larrabee so far? What a staggering waste of cash on a total lemon...this is worse than itanic by far.

Assumptions, assumptions, assumptions. So what makes you think Larrabee has heat issues? You state that the processors was running at half cores activated and then assume that the only plausible reason would be that it is having heat issues (when it could theoretically be a whole slew of plausible reasons). You don't have enough information to draw such conclusions. Furthermore you end by claiming that an unreleased product is similar to a released product when we don't know how the unreleased product will perform. You're relying on blind faith (or blind hope) of a failure.


jennyh said:
Where are the double precision numbers?

Also...who is going to bother with this lol? intel actually expects people to retrain and change the way it's been done for many years...for a slower product?

Fermi did kill Larrabee, it's over for intel in that regard.

Larrabee is MIMD therefore the Double Precision numbers won't be far off the Single Precision numbers. You end your comments with a rather peculiar statement. You claim that unreleased product A has killed unreleased product B. Can you please elaborate as to how unreleased products are capable of such things?


Thank You.
November 20, 2009 6:23:07 PM

And its to the point of this thread.
Both are unreleased, yet both companies keep "showing" its capabilities.
If they keep waiting, ATI could slip in with something here, and more later, which from what Im reading wouldnt be a huge diversion from their current HW, which is superior to both actually, sans the ECC requirements, but thats only a portion of this market requiring the ECC, tho a lucrative part of it.
The numbers Ive seen "leaked" by both companies, going by this, the other info we talked about earlier Elmo, and what nVidia is claiming, is that nVidia has a slight lead at this point.
Now, LRB Im sure has some polishing to do, but not alot, as we all agree, Intels process is tried and true, and not alot more there to do, beides some drivers etc
meanwhile, at 5,8 x that of their old HW, nVidia has also clock problems, which is rumored at 20%.
I know this talk is as phoney and speculative as these "showings", but using what weve been allowed to "see", or "leaked". it does appear Fermi is ahead at this point

"For instance, if only a certain number of cores are needed for a job, they could run while the other cores are idle. When the cores being used start to heat up, they could be shut down and cool idle cores could take over running the job. "
http://www.computerworld.com/s/article/9140949/Intel_to...

To me this sounds much like SMT in ways, and LRB has been said to do anything its required to do, as its totally non fixed function, which has its drawbacks in certain scenarios, same for SMT, when the app is asking too much, it just wont work, or in LRBs case, either will downclock, or just be slower.
Thats the design, not approach, excluding the non FF part.
The approach is being touted as user/dev friendly, but again, theyve writted several new languages for it, so at this point, its another CUDA to me.
a b à CPUs
November 20, 2009 6:54:13 PM

JAYDEEJOHN said:
And its to the point of this thread.
Both are unreleased, yet both companies keep "showing" its capabilities.
If they keep waiting, ATI could slip in with something here, and more later, which from what Im reading wouldnt be a huge diversion from their current HW, which is superior to both actually, sans the ECC requirements, but thats only a portion of this market requiring the ECC, tho a lucrative part of it.
The numbers Ive seen "leaked" by both companies, going by this, the other info we talked about earlier Elmo, and what nVidia is claiming, is that nVidia has a slight lead at this point.
Now, LRB Im sure has some polishing to do, but not alot, as we all agree, Intels process is tried and true, and not alot more there to do, beides some drivers etc
meanwhile, at 5,8 x that of their old HW, nVidia has also clock problems, which is rumored at 20%.
I know this talk is as phoney and speculative as these "showings", but using what weve been allowed to "see", or "leaked". it does appear Fermi is ahead at this point

"For instance, if only a certain number of cores are needed for a job, they could run while the other cores are idle. When the cores being used start to heat up, they could be shut down and cool idle cores could take over running the job. "
http://www.computerworld.com/s/article/9140949/Intel_to...

To me this sounds much like SMT in ways, and LRB has been said to do anything its required to do, as its totally non fixed function, which has its drawbacks in certain scenarios, same for SMT, when the app is asking too much, it just wont work, or in LRBs case, either will downclock, or just be slower.
Thats the design, not approach, excluding the non FF part.
The approach is being touted as user/dev friendly, but again, theyve writted several new languages for it, so at this point, its another CUDA to me.

But that research chip is not Larrabee. I see the articles in this thread pointing to an Intel research chip, and somehow, it's failings being attributed to Larrabee (which is not the research chip in question).
a b à CPUs
November 20, 2009 6:59:16 PM

The 80 Core prototype is an old project: http://www.bit-tech.net/news/hardware/2007/01/18/intel_...

Dating back to 2007.

Quote:
Intel's research team has managed to successfully produce a prototype 80-core Tera-Scale processor that uses less energy than the company's current flagship Core 2 Extreme QX6700 quad-core processor.

The prototype was built so that the chip giant's researchers could investigate the best way to make such a large number of processing cores communicate with each other. This was in addition to researching new architectural techniques and core designs.

The chip, dubbed the Tera-Scale Teraflop Prototype, is just for research purposes and lacks a lot of necessary functionality at the moment. However, R&D Technology Strategist Manny Vara said that the company will be able to produce 80-core chips en masse in five to eight years.

Currently, the prototype chip consumes less than 100W of power, which is less than the 130W consumed by the quad-core QX6700. Of course, the prototype currently lacks some key functionality, which could potentially throw the power consumption characteristics out of proportion, but it's an impressive feat nonetheless.

Vara added that although there are many more cores on the Tera-Scale prototype, they're a different type of core than the ones used in today's microprocessors. "The new ones will be much simpler. You break the core's tasks into pieces and each task can be assigned to a core. Even if the cores are simpler and slower, you have a lot more of them so you have more performance."

Today's microprocessor cores are very flexible, while Intel believes that tomorrow's microprocessor cores will be much more specialised, but of course, there will be many more of these simpler cores. AMD's Fusion project appears to be going down the route of scaling what we've already got, while Intel is moving towards what would be a more flexible approach to energy efficiency.

Before you get too excited though, this is all on paper at the moment; the real war of the cores won't be decided until both companies have released their respective massively multi-core processing architectures in a few years time.
November 20, 2009 7:03:37 PM

On the SGEMM single precision, dense matrix multiply test, Rattner showed Larrabee running at a peak of 417 gigaflops with half of its cores activated (presumably the 80-core processor the company was showing off last year); and with all of the cores turned on, it was able to hit 805 gigaflops. As the keynote was winding down, Rattner told the techies to overclock it, and was able to push a single Larrabee chip up to just over 1 teraflops, which is the design goal for the initial Larrabee co-processors.
http://www.theregister.co.uk/2009/11/17/sc09_rattner_ke...
November 20, 2009 7:10:36 PM

These tests are showing hard numbers tho/ The question is, will this be LRBs numbers?
a b à CPUs
November 20, 2009 7:13:59 PM

JAYDEEJOHN said:
On the SGEMM single precision, dense matrix multiply test, Rattner showed Larrabee running at a peak of 417 gigaflops with half of its cores activated (presumably the 80-core processor the company was showing off last year); and with all of the cores turned on, it was able to hit 805 gigaflops. As the keynote was winding down, Rattner told the techies to overclock it, and was able to push a single Larrabee chip up to just over 1 teraflops, which is the design goal for the initial Larrabee co-processors.
http://www.theregister.co.uk/2009/11/17/sc09_rattner_ke...

Oh I see the mistake there...

Quote:
(presumably the 80-core processor the company was showing off last year)

He is assuming that Larrabee is the Tera-Scale processor Intel was working on last year (which it isn't) the two are separate projects.

The performance seems to be right where it should be. Although RV870 can theoretically hit up to 2.7TFLOPs, in practice it can only hit up to 1.3TFLOPs (you can view that here: http://www.beyond3d.com/content/reviews/53/1).


And much like Larrabee different mathematical workloads show different performance figures as you can see in the picture above.

If Larrabee hits those numbers it will be more powerful than Fermi (based on the nVIDIA white paper here: http://www.hardocp.com/article/2009/09/30/nvidias_fermi...).
a b à CPUs
November 20, 2009 7:26:56 PM

Ok so it's Polaris that was being demo'd here, not larrabee.

Why are intel demo'ing a test chip and not larrabee then? If anything that is worse. Wasn't larrabee supposed to be out Q4 2009?
a b à CPUs
November 20, 2009 7:29:52 PM

jennyh said:
Ok so it's Polaris that was being demo'd here, not larrabee.

Why are intel demo'ing a test chip and not larrabee then? If anything that is worse. Wasn't larrabee supposed to be out Q4 2009?

In the Register article they name the chip Larrabee but Intel's 80 Core prototype chip for HPC wasn't and hasn't been known to be Larrabee (but rather Tera-Scale).

As for when Larrabee will be released.. sometime in 2010. I think they're working to perfect their manufacturing process to get the most out of Larrabee.
November 20, 2009 7:31:43 PM

Capable, even in these synthetics, and deliverable are 2 different things, and is why Nehalem is also being mentioned in some discussions regarding LRB.
Plus theres the loss as well, in BW or whatever else.
Nehalem cant deliver what its said it can, but does well.
The numbers are closer than we think, and from what Ive read, Fermi is ahead.
Its approach as well as capacity, and why I used those other numbers in my previous thread, as those are real world, where thruput, latencies are all being done on the same app.
Going from those numbers, and then reading the nVidia white paper puts a whole new look upon Fermi IMO
a b à CPUs
November 20, 2009 7:42:04 PM

JAYDEEJOHN said:
Capable, even in these synthetics, and deliverable are 2 different things, and is why Nehalem is also being mentioned in some discussions regarding LRB.
Plus theres the loss as well, in BW or whatever else.
Nehalem cant deliver what its said it can, but does well.
The numbers are closer than we think, and from what Ive read, Fermi is ahead.
Its approach as well as capacity, and why I used those other numbers in my previous thread, as those are real world, where thruput, latencies are all being done on the same app.
Going from those numbers, and then reading the nVidia white paper puts a whole new look upon Fermi IMO

The RV870 test I showed above was a synthetic test (much like Larrabee is being run on synthetics right now).

In Single Precision Fermi is theoretically capable of around 1.25TFLOPs- 1.7TFLOPs. The lower figure is taking into account their CEOs Claimed Double Precision Figure of around: 625GFLOPs and both taking and not taking into account the current clock speed issues that have been described here: http://www.semiaccurate.com/2009/11/16/fermi-massively-....

I have been mentioning this often (under the pseudo name z3r0c00l here: http://forum.x c p u s.com/nvidia/18565-gf100-real-world-technologies.html) but others are starting to take notice as well: http://www.brightsideofnews.com/news/2009/11/17/nvidia-...
Quote:
According to specifications based on NV100 A2 silicon [subject to change], C2050 will deliver 520 GFLOPS of IEEE 754-2008 Dual Precision format and 1.040 TFLOPS of single precision. C2070 stands a bit better, 630 GLOPS of Dual-Precision and 1.26 TFLOPS in Single Precision.


These are all theoretical numbers (not real world). It has always been the case that real world numbers end up lower than theoretical numbers.

Fermi doesn't look that great IMHO (well Double Precision wise it looks nice but as a consumer computational device.. it looks dreadful).
a b à CPUs
November 20, 2009 7:54:57 PM

I'm assuming this 80-core test chip is actually more powerful than Larrabee is tbh.

Otherwise, they'd have demo'd larrabee.


I mean come on - this is supposed to be an imminent release and we basically havent seen anything about it at all?

Either it's underpowered or it's nowhere near release, and the same goes for Fermi too.
November 20, 2009 8:03:00 PM

Ive read those numbers, and as Ive said, it comes down to approach now.
We know the gpu numbers work, as theyve been done before, but we dont know that LRBs will, as the chart I showed in the other thread shows the fall off above 32 cores.
Also, we dont know how well the communication is, or thruput will be, but we do on the gpus.
We dont know how well the SW will work on LRB, but again, we do with the gpus.
Weve already deducted any loss from Fermi/280 or whatever 5870 (tho the x2 would simply double those numbers and leave them all in the dust). but we havnt with LRB, which is only taking capability in thought.
Thats what Im saying here.
a b à CPUs
November 20, 2009 8:06:36 PM

jennyh said:
Ok so it's Polaris that was being demo'd here, not larrabee.

Why are intel demo'ing a test chip and not larrabee then? If anything that is worse. Wasn't larrabee supposed to be out Q4 2009?



Wasnt Bulldozer supposed to be out Q2-3 2009?

http://blogs.zdnet.com/BTL/?p=5766
November 20, 2009 8:17:57 PM

For gaming, LRB with no FF parts wont play well, and tho it may seem an accomplishment, when needed, the wait, switch capablity of LRB is limited at best.
Regarding gpgpu usage, it may or may not work better, again, I point towards approach.
The apps listed in the register article wont be using only half of LRB, and no ocing will be possible, and even underclocking may be needed, plus any other latencies involved, vs a product thats been around for a decade, tho using a newer approach, is only to be better than what weve had before.
So, if Fermi is made to beat the 5870, it wouldnt surprise me to see higher numbers from it as well
a b à CPUs
November 20, 2009 8:22:34 PM

BadTrip said:
Wasnt Bulldozer supposed to be out Q2-3 2009?

http://blogs.zdnet.com/BTL/?p=5766


LOL - excellent comeback!! +1

It must be a full moon out tonight, as JDJ & Jenny are off on their monthly "woe is Larrabee" rant again. Makes you wonder - if it's as bad as they keep insisting it is, why are they so afraid of it??? Hmmm???????

And as for demoing unreleased products, nothing can be lower than Randy Allen's flogging Barfalona as "40% faster across a wide variety of workloads" some 9 months before it was released, and the truth (25% slower than Core2) was revealed :D ...
a b à CPUs
November 20, 2009 8:24:20 PM

Nice attempt at deflecting the discussion to long past technology :D 

But...no we're discussing Larrabee here, or more accurately, the complete lack of Larrabee.
November 20, 2009 8:27:13 PM

Gee, and all I can say is, Ive got 1.25 billion reasons why, nuff said
Anyways, if trhis is a straight translation of LRB, then these tests can apply, which gives a a framework of capability without approach, or possibly both.
Itll be interesting, and the hints are thick, as Intel said theyre closer than they thought, and it wont be as big as 80 cores, hint hint
a b à CPUs
November 20, 2009 8:27:22 PM

http://arstechnica.com/hardware/news/2008/01/larrabee-b...

Quote:
Larabee first silicon should be late this year in terms of samples and we'll start playing with it and sampling it to developers," said Otellini. "I still think we are on track for a product in late 2009, 2010 timeframe.
a b à CPUs
November 20, 2009 8:30:59 PM

jennyh said:
Nice attempt at deflecting the discussion to long past technology :D 

But...no we're discussing Larrabee here, or more accurately, the complete lack of Larrabee.


LOL - neither has Bulldozer, but that hasn't stopped you from silly comments like "Intel fanboys are very worried" or "Bulldozer will reclaim the crown" or some such :D .

Besides, I was just showing Intel is, in fact, copying AMD here by flogging unreleased products :D . Hopefully Intel won't copy AMD verbatim and release Larrabee as a total dog :bounce: 
a b à CPUs
November 20, 2009 8:32:27 PM

jennyh said:
http://arstechnica.com/hardware/news/2008/01/larrabee-b...

Quote:
Larabee first silicon should be late this year in terms of samples and we'll start playing with it and sampling it to developers," said Otellini. "I still think we are on track for a product in late 2009, 2010 timeframe.


Last I checked it is still late 2009. Sounds like he still has over a quarter time frame left.
a b à CPUs
November 20, 2009 8:36:52 PM

jennyh said:
http://arstechnica.com/hardware/news/2008/01/larrabee-b...

Quote:
Larabee first silicon should be late this year in terms of samples and we'll start playing with it and sampling it to developers," said Otellini. "I still think we are on track for a product in late 2009, 2010 timeframe.


Hmm, I just checked my calendar and guess what - it ain't 2010 yet :D .

Face it - AMD won't have anything new for the next 12+ months except 2 extra cores and a speed bump on a 45nm CPU. Big yawner there. Intel will be at 32nm, achieved "fusion" first, and have the next-gen architecture Sandy Bridge out next year. As well as Larrabee. :D 
November 20, 2009 8:37:31 PM

Either Intel is afraid to tip its hat, or, its aways off yet.
Usually, the hype machine starts aways ahead of release.
The gaing aspect wont be coming anytime soon. Its been said the gpgpu use would be shown and done first.
It fits here.
I think Intels pulling a Fermi, as yes, the first "showing" wasnt anything anyone could get their hands on, say, unlike the 5xxxseries showing.
But its loike a picture on facebook.
I remember all the crap Barcy got from a no hands on approach during first showings, and yes, the number werent typical then, whos to say different now?
If they had more, wed see more.
a b à CPUs
November 20, 2009 8:46:19 PM

JAYDEEJOHN said:
Either Intel is afraid to tip its hat, or, its aways off yet.
Usually, the hype machine starts aways ahead of release.
The gaing aspect wont be coming anytime soon. Its been said the gpgpu use would be shown and done first.
It fits here.
I think Intels pulling a Fermi, as yes, the first "showing" wasnt anything anyone could get their hands on, say, unlike the 5xxxseries showing.
But its loike a picture on facebook.
I remember all the crap Barcy got from a no hands on approach during first showings, and yes, the number werent typical then, whos to say different now?
If they had more, wed see more.

To pull a Fermi would require Otellini to hold up a badly assembled Lego model and exclaim.. "This is Larrabee". Intel seem to be making good progress on Larrabee in my opinion. To come out of the blue with a single product and try and tackle the big boys who have been at it for decades is no easy task. Last time around, Intel bought a GPU maker and released their product (i740 or something like that) now they're actually working on a product that is entirely theirs.

Even if it doesn't beat Fermi or RV870, if it get's somewhat close.. give it two to three generations and it will compete handsomely.
a b à CPUs
November 20, 2009 8:48:14 PM

We aren't even seeing numbers from intel, that's the thing. Maybe they realised like Nvidia, that making chips this big is '****ing hard' lol.

ATI must be laughing their backside off at the pair of them.
a b à CPUs
November 20, 2009 8:51:24 PM

ElMoIsEviL said:
Even if it doesn't beat Fermi or RV870, if it get's somewhat close.. give it two to three generations and it will compete handsomely.


Same old argument. :whistle: 

If larrabee comes in behind, there is no reason to believe it will ever get ahead. Do you see intel igp's catching up on the big two in graphics?
a b à CPUs
November 20, 2009 9:03:33 PM

jennyh said:
Same old argument. :whistle: 

If larrabee comes in behind, there is no reason to believe it will ever get ahead. Do you see intel igp's catching up on the big two in graphics?

Intel's IGP aren't getting the level of resources (investment in R&D) and aren't aiming to compete. They're aiming to simply offer basic functionality for a limited cost (cost to Intel and cost to customers/OEMs).

That it's not an old argument. It is how things have always played out.

Here's some history.

- 3Dfx releases the Voodoo Graphics card
- nVIDIA release the Riva 128

Riva 128 doesn't win but it offers up some level of competition. nVIDIA then release the Riva TNT which bests the Voodoo. 3Dfx reply with the Voodoo2 which bests the Riva TNT (and has capabilities for SLI).

nVIDIA respond with the TNT2 to which 3Dfx responds with the Voodoo3. nVIDIA is catching up. Their card may not be as fast but it can do 32-bit rendering whereas the Voodoo3 is stuck at 16-bit.

nVIDIA then releases the GeForce256, 3Dfx has nothing to offer yet...
nVIDIA releases the Geforce2 and 3Dfx responds with the Voodoo5 5500. GeForce 2 wins.

3Dfx goes belly up.

You claim what I am saying as being the "same old argument" but I am basing it on historical knowledge (precedence). You simply despise Intel with a passion and wish them failure in anything they do (blind hatred). And this blind faith you have is blinding you from what most of us see as rather obvious.
a b à CPUs
November 20, 2009 9:04:31 PM

Rather obvious being...Larrabee is imminent and going to be worth the $bn's spent?

Lol please...

If you are no good at doing something, trying hard isn't going to make enough of a difference vs somebody who is naturally good at it.
a b à CPUs
November 20, 2009 9:12:49 PM

jennyh said:
Rather obvious being...Larrabee is imminent and going to be worth the $bn's spent?

Lol please...

If you are no good at doing something, trying hard isn't going to make enough of a difference vs somebody who is naturally good at it.

I think that, given the roll Intel is on and with Otellini in charge, Larrabee will be a success. I don't see it being a failure. I don't see what would lead you to think otherwise.

And as for claiming that Intel are "no good at doing something" I have to ask based on what? Their IGPs? Their IGPs weren't meant to be competitive with discrete solutions from other competitors (as I explained in my previous post).

I don't see any evidence to sustain your assertion that Larrabee is an imminent failure. Is it possible? Yes of course but the question you have to ask yourself is "Given Intel's current trend, is it plausible that Intel will release a product meant for failure?"... I think the answer is rather obvious... but I want to know what you think.
November 20, 2009 9:15:19 PM

But Intels response has been "Itll perform against highend gpus"
Its the bees knees for programming and can do anything, even power arrangement. etc
This is not a easy approach, but a "watch out, here comes LRB!!!" type of hype.
Lets see them do as well as the 5970, which should double those numbers you put up, which they cant.
Lets see them have working drivers for every 16-32-64 bit game ever made, and beating and competing with other gpus on the highend.
And, as far as FC2 goes, no one wins there, as the cpus of today are bottelnecking that game at 25x16, and no, its not an old game.
The use of x86 is being held high as a banner, yet they dont go on about the LRB language, and barely mention the c++ additions, and like these are to be easy as well.
It all sounds good, and from what weve actually seen?
You tell me, what have we actually seen?
Yes, its a Fermi
a b à CPUs
November 20, 2009 9:20:12 PM

JAYDEEJOHN said:
But Intels response has been "Itll perform against highend gpus"
Its the bees knees for programming and can do anything, even power arrangement. etc
This is not a easy approach, but a "watch out, here comes LRB!!!" type of hype.
Lets see them do as well as the 5970, which should double those numbers you put up, which they cant.
Lets see them have working drivers for every 16-32-64 bit game ever made, and beating and competing with other gpus on the highend.
And, as far as FC2 goes, no one wins there, as the cpus of today are bottelnecking that game at 25x16, and no, its not an old game.
The use of x86 is being held high as a banner, yet they dont go on about the LRB language, and barely mention the c++ additions, and like these are to be easy as well.
It all sounds good, and from what weve actually seen?
You tell me, what have we actually seen?
Yes, its a Fermi

The fact that it is based on the x86 architecture won't affect it's gaming performance but rather it's GPGPU performance. It can execute the entire C++ Library (where as Fermi and RV870 cannot). Fermi and RV870 both rely on a host processor to feed them data whereas Larrabee can fetch it's own data (MIMD). So you can program directly for Larrabee and not have to play around with workarounds for routines that are not compatible with RV870/Fermi. It is an advantage for Larrabee for sure.

Have we seen it? Umm no.. but I hadn't seen RV870 until a month or two before it was released either.

The difference with Fermi is that nVIDIAs CEO held up a fake card and claimed it was Fermi. So you can't call Larrabee a Fermi.
a b à CPUs
November 20, 2009 9:20:15 PM

Quote:
I think that, given the roll Intel is on and with Otellini in charge, Larrabee will be a success. I don't see it being a failure. I don't see what would lead you to think otherwise.


Interesting pov you have there elmo. Completely whacked in terms of reality, but interesting none the less.

I can say that given the roll intel is on...legally and with Otellini in charge, they won't have enough money left to pay for R&D if it continues.


Quote:
I don't see any evidence to sustain your assertion that Larrabee is an imminent failure. Is it possible? Yes of course but the question you have to ask yourself is "Given Intel's current trend, is it plausible that Intel will release a product meant for failure?"... I think the answer is rather obvious... but I want to know what you think.


You're right without realising it. Intel won't release a product meant for failure and that is why Larrabee is months behind schedule already.
November 20, 2009 9:36:57 PM

Ahh, but the similarities Im not ignoring either.
No ones seen it. Its been hyped.
Its been "shown" hint hint
And always close to another gpu release. hint hint
Yes, the only difference is, Otelini hasnt held up a fake yet, otherwise, its the same, like when will it be here?
Weve been hearing about it.
Weve been toldabout it.
Ive see DrWho bragging it up
Ive seen his poor arguments about the gpgpu usage and the x86 usage.
Ive also seen whats possible in Fermi too.
Its notthe norm with ECC tacked on.
L1 L2 cache, ECC, widening etc etc
But that gets no play? Or hasnt anyone taken the time?
David Kanter has stopped, as hes heard enough for him to sit and wait too.
New approaches, one familiar, the other saying the language is familiar.
One HW we know, works great. One HW, we havnt a clue but talk.
It all comes down to approaches, and since Fermi will fail bigtime if its only good at gpgpu, so too may LRB, if its no good at gaming
a b à CPUs
November 20, 2009 9:40:08 PM

I said it before and I believe it. Intel and Nvidia psyched each other out over this. Neither has a product worth bringing out for the simple reason that neither of them can compete with ATI.

That's why ATI graphics are running the #5 supercomputer and not nvidia or intel graphics. Both of them are chasing a rainbow.
a b à CPUs
November 20, 2009 9:46:11 PM

jennyh said:
I said it before and I believe it. Intel and Nvidia psyched each other out over this. Neither has a product worth bringing out for the simple reason that neither of them can compete with ATI.

That's why ATI graphics are running the #5 supercomputer and not nvidia or intel graphics. Both of them are chasing a rainbow.

That's an AMDZone argument if I ever saw one (in fact we were making fun of such arguments earlier at X C P U S).

Bringing a Super Computer into this argument is irrelevant. Larrabee is an unreleased product so thinking that something that is unreleased should already be running a super computer is ridiculous.

The level of bias you're under is very strong. You are asking more of Intel then you are of AMD or nVIDIA.

It's like Intel must jump through hoops to convince you when you feel comfortable taking AMD at their word.

This argument is over. I'll let the THG readers decide who won.
a b à CPUs
November 20, 2009 9:53:41 PM

You and JDJ are just grasping at straws. Neither of you have one shred of evidence on the failure of LRB. AMD should pay you two for all the propaganda.
a b à CPUs
November 20, 2009 9:55:50 PM

Quote:
That's an AMDZone argument if I ever saw one (in fact we were making fun of such arguments earlier at X C P U S).

Bringing a Super Computer into this argument is irrelevant. Larrabee is an unreleased product so thinking that something that is unreleased should already be running a super computer is ridiculous.


Yes Larrabee is unreleased...and not imminent. That is the point of the thread lol? :D 

Quote:
The level of bias you're under is very strong. You are asking more of Intel then you are of AMD or nVIDIA.


That's interesting. You are the one suggesting that intel have so much money that they can hardly fail. I didn't put any expectation on intel, you did.

Quote:
It's like Intel must jump through hoops to convince you when you feel comfortable taking AMD at their word.

This argument is over. I'll let the THG readers decide who won.


Won? Argument? Lol, this was never an argument elmo so I don't see how anyone can win.
a b à CPUs
November 20, 2009 10:00:02 PM

BadTrip said:
You and JDJ are just grasping at straws. Neither of you have one shred of evidence on the failure of LRB. AMD should pay you two for all the propaganda.


That is true, we don't have any evidence. But that is also the point. There is no evidence proving otherwise either.

Larrabee Q4 2009 'gauranteed'. Ottellini said so, nearly.

Where is it? We've been seeing Clarkdale benches for *months* now but we haven't seen a single thing about Larrabee. Where is it?
November 20, 2009 10:01:55 PM

Quote:
You're right without realising it. Intel won't release a product meant for failure and that is why Larrabee is months behind schedule already

And you have proof of this from where?...
a b à CPUs
November 20, 2009 10:11:35 PM

You should look at what Otellini is saying.

Quote:
"Larabee first silicon should be late this year in terms of samples and we'll start playing with it and sampling it to developers," said Otellini. "I still think we are on track for a product in late 2009, 2010 timeframe."


If Larrabee exists in the wild? It must be under the best kept nda yet because nobody has the slightest inkling that it is doing the rounds in samples.

Nothing. Nothing at all is what the Larrabee news equates to. You can bury your head in the sand if you like, but in the tech sector, no news = no product.
November 20, 2009 10:19:20 PM

Hey ninja, LTNS.
If the ROI isnt there, it just isnt there, Intel has lasting power, but so does nVidia in this scenario,
If neither does well in gaming, the ROI simply wont be there.
They could both fail, if this market never takes off.
That being said, its better that both are arriving to this, so the market to be will have options, and possibly shake enough interest for growth within it.
AMD on the other hand can well afford to wait, and just profit from the market, if they too can eventually create something, tho, as jennyh pointed out, AMD does have a few things already in the overall market.
But, to sustain a slow market, gfx usage is primary for ROI, until the market takes hold
    • 1 / 7
    • 2
    • 3
    • 4
    • 5
    • More pages
    • Next
    • Newest
!