Intel and giving up on HPC GPUs, name a more iconic duo
Intel and giving up on HPC GPUs, name a more iconic duoOoooh crikey, they surrendered the GPGPU race. That's bad. Like real bad.Intel’s Falcon Shores Future Looks Bleak as It Concedes AI Training to GPU Rivals
Intel’s Falcon Shores future looks bleak as it concedes AI training to GPU rivals On Monday, Intel sent a letter to employees detailing its comeback plan after an abysmal second-quarter […]www.hpcwire.com
This is a bit extra hilarious, sinceIntel and giving up on HPC GPUs, name a more iconic duo
They've been trying to enter and corner ML training market since 2016 and their acquisition of Nervana.I don't think can compete with NVIDIA and AMD in the space anyway
WhoAI enthuast at home
They do not have the money for vanity s/w efforts anymore.And try to build a community that may help you when you try to move to higher level
And that's skipping the decade ofThis is a bit extra hilarious, since
They've been trying to enter and corner ML training market since 2016 and their acquisition of Nervana.
A bunch of Nervana products (very dead), then GPGPU stuff (also very dead!), then Habana, then back to GPGPU stuff and now again, all dead.
Delightfully schizophrenic ability of not being able to focus and commit to a single roadmap (granted, PVC being an utter and absolute trainwreck didn't help either).
It sounds like Pat is still obsessed with Larrabee and wants to see the idea return. I don't see this turning out well for Intel...And that's skipping the decade ofLarrabeeXeon Phi failure
It's where all the growth is. If only the CEOs before had been as obsessed maybe they could have done something other than PVC. Xeon Phi had added neural net instructions a decade ago and had a large memory capacity, it was genuinely on the right track.It sounds like Pat is still obsessed with Larrabee and wants to see the idea return. I don't see this turning out well for Intel...
It's where all the growth is. If only the CEOs before had been as obsessed maybe they could have done something other than PVC.
Nobody cared at the time. But had they continued developing Xeon Phi they would have had a product, like MI, that could have been iterated into a ML accelerator by now. And it would have had some pre-existing soft support.That was why they bought Nervana and then Habana. Nobody cared.
TBF, it's not entirely clear that people buying AMD are even using them for anything beyond selling AI hype to "Investors".
Oddly enough, products like Sierra Forest etc. seem like semi-successors to Phi. All they're missing are AVX512 and quad SMT.Nobody cared at the time. But had they continued developing Xeon Phi they would have had a product, like MI, that could have been iterated into a ML accelerator by now. And it would have had some pre-existing soft support.
Ooooh crikey, they surrendered the GPGPU race. That's bad. Like real bad.Intel’s Falcon Shores Future Looks Bleak as It Concedes AI Training to GPU Rivals
Intel’s Falcon Shores future looks bleak as it concedes AI training to GPU rivals On Monday, Intel sent a letter to employees detailing its comeback plan after an abysmal second-quarter […]www.hpcwire.com
It's where all the growth is. If only the CEOs before had been as obsessed maybe they could have done something other than PVC. Xeon Phi had added neural net instructions a decade ago and had a large memory capacity, it was genuinely on the right track.
For awhile there it looked like they would sell off Altera.You want to see some other hard decisions? They need to divest Altera and Mobileye.
Well that doesn't really apply to their GPGPU efforts.Intel's GPU efforts have been plagued by software/driver issues since the early days
No, it's definitely surrender, they've been trying to enter ML training market since before AMD reanimation. It's been a long ass trainwreck, in case you've missed it.Giving up in a race where the leaders were so far ahead you can't even see them is not "surrender", its admitting reality.
They just brought the wrong leadership for the project. However, even when they brought in people with good credentials like Jim Keller, instead of listening to them they just made the leave. Well, look where they are now.Yeah Intel "surrendered" the GPGPU race in the same way I've surrendered in the NYC marathon.
Intel's GPU efforts have been plagued by software/driver issues since the early days. They have a (not great but) credible iGPU, and every time they've tried to move beyond that into big boy territory it has been a joke. Giving up in a race where the leaders were so far ahead you can't even see them is not "surrender", its admitting reality.
They should have made the hard decisions way earlier, they surely knew where they're headed. The writing was on the wall years ago.Intel no longer has the luxury of pouring billions into projects with zero certainty of commercial success. They have to make the hard decisions, and as hard decisions go this is an easy one.
Yes, I'm not sure why they're keeping Mobileye, however they've already cut some stuff like Lidar development.You want to see some other hard decisions? They need to divest Altera and Mobileye. Intel needs to focus on what's important, not the forever niche idea of FPGAs in servers and a long term play in autonomous vehicles. They need the capital those would free up more than they need to own either.
It's where all the growth is. If only the CEOs before had been as obsessed maybe they could have done something other than PVC. Xeon Phi had added neural net instructions a decade ago and had a large memory capacity, it was genuinely on the right track.
Did Aurora ever complete hardware install? Is it running production code, or still in system development mode?
Nothing in that link convinces me that it is in production.Aurora (supercomputer) - Wikipedia
en.wikipedia.org
So yes, currently in production use.
Nothing in that link convinces me that it is in production.
What about the fact that Falcon Shores was recently said to be on TSMC N3?Intel and giving up on HPC GPUs, name a more iconic duo
Most people want a thin and light laptop (that they can also plug into a monitor and keyboard if needed), efficient enough to last an entire day of work on battery power, with enough performance to handle anything they can throw at it. I'd say we're close, but not quite at that point yet- certainly not at affordable price points. Give it another five years or so and we'll be there.Here is a question, have we reached a point where what we have is good enough for 95% of the market? If so, there is little incentive for general public to buy the new shiny thing even the relative performance is high. Majority of general purpose software is not limited my modern hardware, and many use cloud based software. It would take few years for software to catch up to the current hardware. What I am trying to say is, Intel (and AMD) have to find different market/target for new CPUs. Otherwise they won't able cover the R&D and manufacturing cost.
Having said that, personally I want more than 16 cores(p-cores) in desktop .
The thing you speak of exists in the Apple ecosystem.Most people want a thin and light laptop (that they can also plug into a monitor and keyboard if needed), efficient enough to last an entire day of work on battery power, with enough performance to handle anything they can throw at it. I'd say we're close, but not quite at that point yet- certainly not at affordable price points. Give it another five years or so and we'll be there.
affordability doesn’tThe thing you speak of exists in the Apple ecosystem.