Intel "Cooper Lake" Latest 14nm Stopgap Between "Cascade Lake" and "Ice Lake"

[btarunr]

With no end to its 10 nm transition woes in sight (at least not until late-2019), Intel is left with refinement of its existing CPU micro-architectures on the 14 nanometer node. The client-desktop segment sees the introduction of the "Whiskey Lake" (aka Coffee Lake Refresh) later this year; while the enterprise segment gets the 14 nm "Cascade Lake." To its credit, Cascade Lake introduces a few major platform innovations, such as support for Optane Persistent Memory, silicon-level hardening against recent security vulnerabilities, and Deep Learning Boost, which is hardware-accelerated neural net building/training, and the introduction of VNNI (Variable Length Neural Network Instructions). "Cascade Lake" makes its debut towards the end of 2018. It will be succeeded in 2019 by Ice Lake the new "Cooper Lake" architecture.

"Cooper Lake" is a refresh of "Cascade Lake," and a stopgap in Intel's saga of getting 10 nm right, so it could build "Ice Lake" on it. It will be built on the final (hopefully) iteration of the 14 nm node. It will share its platform with "Cascade Lake," and so Optane Persistent Memory support carriers over. What's changed is the Deep Learning Boost feature-set, which will be augmented with a few new instructions, including BFLOAT16 (a possible half-precision floating point instruction). Intel could also be presented with the opportunity to crank up clock speeds across the board.



View at TechPowerUp Main Site

 

[Caring1]

Sounds more like BSFLOAT.

 

[stimpy88]

Ahhh, I smell another 1-2% IPC gain on its way... Just what Intel need to knock out AMD...

I wonder if Intel will break out the soldering machine for this one...

 

[bug]

Looks like Intel could use automakers' labeling. 14nm 2016, 14nm 2017, 14nm 2018... It scales as long as needed

 

[TheinsanegamerN]

So is this just another admission they cant get 10nm off of the ground? How many times can they re-release haswell?

 

[ssdpro]

14nm has been serving Intel quite well for consumer CPUs. Over the last 12 months Intel's stock has gone from 36.59 in August 17 to $49.96 August 2018. Revenue in Q2 was 16.96 billion up 15%. If the product does what it is supposed to and consumers prefer it 4-1 why rush?

 

[TheinsanegamerN]

14nm has been serving Intel quite well for consumer CPUs. Over the last 12 months Intel's stock has gone from 36.59 in August 17 to $49.96 August 2018. Revenue in Q2 was 16.96 billion up 15%. If the product does what it is supposed to and consumers prefer it 4-1 why rush?

Because intel, one the leader of new fab processes with some of the best manufacturing facilities, is falling behind TSMC. Intel has watched their lead slip as AMD gets ryzen going, and seems to have nothing in the pipeline.

Tesla has rising share prices and revenues too, doesnt mean the company isnt on massively shaky ground going forward. Same applies here. If intel cant get a new arch and new process off of the ground, AMD could end up catching or even passing them in a few years time. The MCM approach to CPUs is proving to scale better with multi core then intels monolith, and AMD's continued efforts to combat inter chip latency is proving fruitful.

You really cant call it a "rush" when 10nm is now 3 years, looking to be 4 years late. At that point we call it "get your act together".

 

[$ReaPeR$]

pathetic.

 

[eidairaman1]

14nm has been serving Intel quite well for consumer CPUs. Over the last 12 months Intel's stock has gone from 36.59 in August 17 to $49.96 August 2018. Revenue in Q2 was 16.96 billion up 15%. If the product does what it is supposed to and consumers prefer it 4-1 why rush?

Thank AMD for bringing competition via Ryzen.

 

[bug]

Because intel, one the leader of new fab processes with some of the best manufacturing facilities, is falling behind TSMC.

This is where you're wrong. TSMC isn't ahead of Intel, because they're all using the same equipment built by ASML.
It just happens that Intel wants to add an extra layer of patterning to draw smaller features. They feel the pain right now, but going forward everybody else will need that extra layer at some point. And then the tables will be turned.

 

[TheinsanegamerN]

This is where you're wrong. TSMC isn't ahead of Intel, because they're all using the same equipment built by ASML.
It just happens that Intel wants to add an extra layer of patterning to draw smaller features. They feel the pain right now, but going forward everybody else will need that extra layer at some point. And then the tables will be turned.

True, but intel should get the main pattern down first before bothering with smaller features. It's costing them a ton of money and wont pay off if it is just another marginal improvement.

 

[bug]

True, but intel should get the main pattern down first before bothering with smaller features. It's costing them a ton of money and wont pay off if it is just another marginal improvement.

Those are details only known to Intel, I won't speculate.
They're in a tough spot, no doubt about that. But the manufacturing process could turn out to be a boon, once they get past their troubles.

 

[ensabrenoir]

....seems I won't be building a new intel system for a while.

 

[TheGuruStud]

This is where you're wrong. TSMC isn't ahead of Intel, because they're all using the same equipment built by ASML.
It just happens that Intel wants to add an extra layer of patterning to draw smaller features. They feel the pain right now, but going forward everybody else will need that extra layer at some point. And then the tables will be turned.

Intel admitted their 10nm is worse performing than their 14. It's over. Everyone needs to deal with it. Even their 10nm+ will be inconsequentially better than 14++++++++++++++++++ lol.

 

[zelnep]

14nm has been serving Intel quite well for consumer CPUs. Over the last 12 months Intel's stock has gone from 36.59 in August 17 to $49.96 August 2018. Revenue in Q2 was 16.96 billion up 15%. If the product does what it is supposed to and consumers prefer it 4-1 why rush?

you are talking like as if they could rush anything... only thing intel could and did rush is x299 (with that super weak price/performance) and also they did rush that 28-core presentation that was cooled with refrigerator and about that strong stock price - dont get me started... from that period SP500 went +15% (yes even with that january-feruary hick-up -we are in the full-mode Bull market right now) and most tech stocks do +30% or more in bull markets like these (not to mention AMD, nvidia that did even more)

 

[bug]

Intel admitted their 10nm is worse performing than their 14. It's over. Everyone needs to deal with it. Even their 10nm+ will be inconsequentially better than 14++++++++++++++++++ lol.

Do you have a link for that? I must have missed that part.

 

[TheGuruStud]

Do you have a link for that? I must have missed that part.

Oops, I mean 10++ is slightly better lol. And we know how they exaggerate slides. 10 is broken and can't be fixed. By the time lower power usage is achieved at high clocks, they'll be years behind. Just look at how far behind they are compared to this slide. There's no real 10nm part coming until 2020 and on this slide they expected refined EUV lolololol

 

[bug]

I get none of your claims from that chart.
As for "slightly better", you do know what a log scale is, don't you?

 

[TheGuruStud]

I get none of your claims from that chart.
As for "slightly better", you do know what a log scale is, don't you?

This is a marketing baloney scale and they're years behind it. It's time to move on. It's so bad the CEO was fired lol.

 

[bug]

This is a marketing baloney scale and they're years behind it. It's time to move on. It's so bad the CEO was fired lol.

Yes, it's behind schedule and way over budget. But 10nm performing worse than 14nm is just something you pulled out of thin air.
That graph only shows the _projected 2018 iteration_ of 10nm would be performing below the third 14nm iteration. That's already outdated, but if you compare similar iterations of 14 and 10nm, 10nm is still supposed to be better.

"broken and can't be fixed" - even if true, you can't know that now.

 

[ppn]

meltdown; spectre hardware fix? Whatever instructions intel introduces if they just kept selling sandybridge 4 and and now 6 core for mainstream, and refine 32nm to +++++++, it wouldn't have mattered. Now fp16 should remain in very specialised asics. Why cpu.

 

[Hood]

You really cant call it a "rush" when 10nm is now 3 years, looking to be 4 years late. At that point we call it "get your act together".

I think it's because Intel doesn't want to transition to 10nm until it's perfect. I mean, why release some half-baked product that's not ready - that's AMD's M.O. - (Ryzen took 3 years to release and another year or more to work out the quirks - some would say it's still not ready for prime time). Intel has a reputation to uphold, it would be unseemly to spit out sub-standard products like their competition.

 

[ToxicTaZ]

10nm+ Icelake Dual Channel coming out October 2019.

10nm+ Icelake Quad channels coming out Q3 2020.

14nm+++ CopperLake 9th Gen Q3 2019

14nm+++ CascadeLake 8 Gen Q4 2018 (28 Cores 5GHz) upcoming ASUS ROG Rampage VII board will be interesting.

 

[Prima.Vera]

Why do I have an almost evil desire that Ryzen 2 or/and 3 to smoke all those 14nm+++++++++ CPUs that most likely will only bring higher freqs and IPC increase in 2 or 3% again.

 

[Caring1]

Intel ran the "well" dry, now they are draining the "lake"