• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Larrabee - Update

Lol at the 5870/GTX280

Yes the GTX480 is out now. Yes it is more expensive but 5870/GTX470 would be the least you could say.
 
thought intel abandoned it? or took it back to the drawing board so to speck
 
imo, larrabee was, and probably always will be simply an intel plaything. Methinks some derivative of the technology will surface if we ever see GPU and CPU merged into one general chip (ie, more fused than fusion - something will be able to process either CPU or GPU instructions, instead of diffierent parts of the chip doing defferent things ... if that makes sense)
 
He said that if they released it NOW (Q2 this year) that it would face the 5870/GTX280.

The GTX480 is out NOW.

"Larrabee 1, aka the part you have heard of, was beset by delays. By the time it was deprecated, it was well over a year and a half late. The first official release date to come our way was mid to late 2008."

"For a chip that was targeting the HD4870/GTX285 generation of GPUs, a year's delay meant that it would be facing the HD5870/GTX285, one of them moving the bar up substantially."

These are the 2 quotes i have taken from the article about when it was going to be released. It says it was aimed at launching mid to late 2008 which would have put it up against HD4870/GTX285. Then there are 2 bits saying it was 1 to 1 1/2 years late which would be puting it up against gtx285/5870. If you go by these time scales then the gtx480 was not released.

I can't remember him saying anything about releasing the original larabee now.
 
Larrabee is dead, in the water if not the actual project. Intels entire mentality and approach will NEVER triumph in the gaming GPU market even if they make a superb piece of hardware (which knowing intel they will in the end as the core 2 duo line, etc. bares testament).
 
I bet Intel are kicking themselves for not releasing. They could have cleaned up by default over the last 8 months.

Intel could make a serious bid for second spot over the next 12-18 months, and give ATI a hard time on the folding front. It would be great for the industry if Intel could get something out the door.
 
Last edited:
I bet Intel are kicking themselves for not releasing. They could have cleaned up by default over the last 8 months.

Intel could make a serious bid for second spot over the next 12-18 months, and give ATI a hard time on the folding front. It would be great for the industry if Intel could get something out the door.

The 5 series would have destroyed whatever intel released.

The chip was hot and only competative perhaps with the 4 series.

Intel wants to come out and WIN.

They are not going to release a graphics cards for folders.
 
Larrabee is dead, in the water if not the actual project. Intels entire mentality and approach will NEVER triumph in the gaming GPU market even if they make a superb piece of hardware (which knowing intel they will in the end as the core 2 duo line, etc. bares testament).

You could not be more wrong, or hilarious.

The reason Intel failed, is the SAME reason Nvidia had such problems with Fermi, to big a die with too many "large" cores.

Larabee is without question not dead, you'd infact have to be rather silly to suggest it.

Lets see, AMD is taking GPU's on die, and eventually sucking in the low end FOR CERTAIN and probably the low mid end and possibly the mid end also in years to come. Intel are doing this, with completely uncompetitive, ill featured and massively slower GPU's.

Intel have more money than Nvidia could ever hope to have, they have deeper pockets and higher revenue than most other companies in the world.

Worst comes to worst and they simply buy their way through all the problems and absorb multiple companies and they have the pockets for that, it won't come to that.

As has been suggested, lots of people have rumoured the latest Larabee is pretty damn competitive to a 4870, yes its late, its very late. But AMD/Nvidia have been in the game for MUCH longer and take years to get something from first thought to final vision.

Unlike when AMD/Nvidia started you can't just make any old 1 pipeline card and sell a few, and unlike when AMD/Nvidia started when it cost very little and turn around was uber quick and you could launch a crap product(as both did, more than once) and just bring out something new very quickly, thats just not the case now.

Larabee was always going to go through a few internal revisions and lets be honest, it hoped to have something to show by now but AMD and Nvidia would have slipped a few generations, the way the industry is now, before releasing something useful.

The difference is, Intel have more than enough cash to through away on this pet project till it gets on top of production.

You're also mistaking what they are trying to do with Larabee, they are trying to compete with what will eventually be on core massive GPU acceleration rather than make a brilliant GPU, but as the industry moves towards gpu's on die, the gaming industry WILL adapt to the way Intel and AMD are going.

Larabee's got some growing up to do and some expanding in numbers of cores. Larabee has far more of a future than Nvidia does in gaming GPU's.
 
Last edited:
You missed the point of what I was saying... even if they make an absolutely superb piece of hardware - and being intel I'm pretty sure the will in the end - they absolutely won't make it in the gaming GPU market... intel simply lack the mentality to make it work and no amount of R&D, money or time will fix that. They should stick to what they are good at like CPUs.
 
You missed the point of what I was saying... even if they make an absolutely superb piece of hardware - and being intel I'm pretty sure the will in the end - they absolutely won't make it in the gaming GPU market... intel simply lack the mentality to make it work and no amount of R&D, money or time will fix that. They should stick to what they are good at like CPUs.

I don't think intel can stay away from the gpu market with amd's plans for cpu+gpu.
 
I don't think intel can stay away from the gpu market with amd's plans for cpu+gpu.

Thats as may be, but intel doesn't have the flexibility, for want of a better way to describe it, to make it far in the gaming GPU arena - their whole approach is wrong and until that adjusts its not going to happen.

Would I be mad if I though that nvidia and intel joining forces would make sense?

Never going to happen... if anything intel would sabotage nVidia covertly and then buy them up while they were on their knees.
 
If Nvidia lose enough money, then I suppose its a possibility that Nvidia could end up as Intels GPU wing.

I think that'd be quite a likely outcome if it turned out that APUs (AMD's name for chips that have the GPU integrated on the CPU) were both very lucrative and that Intel felt it would begin to struggle to compete with AMD in the long term because of them. Now, if it's true that AMD's Llano chips will have around 400-480 shaders, there's little doubt of Intel potentially struggling to compete against such a chip in terms of graphical capabilities (short of getting Larrabee in a working state that could be incorporated in a chip that isn't the size of your average crop field); I think the only other question is if that chip turns out to be popular enough for Intel to start worrying about. I think if both of those things are true, Intel will be taking some very long and hard looks at Nvidia to say the least.
 
I don't think Intel have any interest in fighting it out in the ultra high end gaming card market. Something they can integrate into their high volume OEM CPU's (which is better than the current laughable attempts, and give a decent-ish speedup to certain tasks), and be competitive with AMD(fusion) integrated setups is what they want. The high end stuff will be highly parallel compute chips that can sit in a QPI socket, eating the market NV are trying to build with the Tesla.
 
from what i remember reading on a atrical I don't remember were intel have only killed that deign so to speck I got the impression they may be throwing it away and starting again
 
Back
Top Bottom