Discussion Intel current and future Lakes & Rapids thread

[Ajay]

That really baffles me. I know they are expensive, but the finance guys love to depreciate stuff. And if they are on the super, duper critical path for future growth that seems hugely irresponsible to everyone, even the shareholders.

Wouldn't you buy them just so the competition didn't get them at some point? I am thinking about the (really old now) Only the Paranoid Survive and I feel like that Intel might have been busy burying knives into backs to keep their dominance.

Shareholder value something, something...

 

[jpiniero]

That really baffles me. I know they are expensive, but the finance guys love to depreciate stuff. And if they are on the super, duper critical path for future growth that seems hugely irresponsible to everyone, even the shareholders.

It seemed they were on the fence about continuing with 7 nm.

 

[Hulk]

I think much of Intel's process woes will be mitigated if they can move desktop parts to 10ESF in quantities ASAP. From a power and transistor density it seems as though 10SF is competitive with TMSC 7nm so I'm thinking 10ESF will be even more competitive. Assuming AMD hasn't moved on to 5nm by then. But even if they have it seems likely the density/power gains will be getting smaller as we approach the end of this node shrinking game that started 50 years ago.

 

[DrMrLordX]

We will know when we see 7nm ramp.

Maybe they will release exact wpm numbers in investor calls, but given the lack of transparency as of late, I doubt it. Intel's 10nm+, 10SF, and 10SFE capacities still aren't public knowledge, and we still don't know how well 10SF/10SFE yield. The availability of TigerLake-H 8c will answer some questions there, assuming we ever get good availability data on those parts.

 

[mikk]

AlderLake-S ES with DDR5-4800 spotted on CapFrameX database. The more interesting part for me is the base of 2.2 Ghz, this quite a bit higher than the Intel Core-1800 ES from last week.

 

[JoeRambo]

What we really need is GB5 leak, that would give us decent estimate for performance improvements, even more so if done at some fixed base clock.

 

[Dayman1225]

Maybe they will release exact wpm numbers in investor calls, but given the lack of transparency as of late, I doubt it. Intel's 10nm+, 10SF, and 10SFE capacities still aren't public knowledge, and we still don't know how well 10SF/10SFE yield. The availability of TigerLake-H 8c will answer some questions there, assuming we ever get good availability data on those parts.

I don’t think Intel would release their WPM numbers publicly any time soon. Maybe to their close partners now that IFS is a thing.

 

[beginner99]

Do we have any leaks or release date of Icelake XEON-d products? didn't keep track of this thread so excuse me if this was already mentioned.

 

[IntelUser2000]

So Tigerlake-H is launched.

Intel Launches 11th Generation Core Tiger Lake-H: Eight Core 10nm Mobile Processors

www.anandtech.com

Base frequency
@ 65W: 3.3GHz
@ 45W: 2.6GHz

For 11900K
@125W: 3.5GHz
@95W: 3GHz

11900T
@35W: 1.5GHz
@25W: 1GHz

11900
@65W: 2.5GHz

Backport = fail

 

[Dayman1225]

So Tigerlake-H is launched.

Intel Launches 11th Generation Core Tiger Lake-H: Eight Core 10nm Mobile Processors

www.anandtech.com

Base frequency
@ 65W: 3.3GHz
@ 45W: 2.6GHz

For 11900K
@125W: 3.5GHz
@95W: 3GHz

11900T
@35W: 1.5GHz
@25W: 1GHz

11900
@65W: 2.5GHz

Backport = fail

PL2 on 11980HK is 107w too, lower than I expected.

https://twitter.com/x/status/1392076095567106049

 

[jpiniero]

Die size is ~190 mm2 which is very large. They did put the Thunderbolt 4 on the die which I am sure takes up a bit of space.

 

[Gideon]

Die size is ~190 mm2 which is very large. They did put the Thunderbolt 4 on the die which I am sure takes up a bit of space.

They seem to have a considerable amount of dark silicon, where the GPU used to be. I wonder why couldn't they just add a couple of EUs instead:

https://twitter.com/x/status/1392088812600365057

 

[eek2121]

So Tigerlake-H is launched.

Intel Launches 11th Generation Core Tiger Lake-H: Eight Core 10nm Mobile Processors

www.anandtech.com

Base frequency
@ 65W: 3.3GHz
@ 45W: 2.6GHz

For 11900K
@125W: 3.5GHz
@95W: 3GHz

11900T
@35W: 1.5GHz
@25W: 1GHz

11900
@65W: 2.5GHz

Backport = fail

Without a doubt. Can't wait for the reviews.

 

[DrMrLordX]

They seem to have a considerable amount of dark silicon, where the GPU used to be. I wonder why couldn't they just add a couple of EUs instead:

Kind of makes you wonder, doesn't it? Really though, let's see how many of these things actually show up on the market. Time for Intel to prove themselves.

 

[Dayman1225]

Kind of makes you wonder, doesn't it? Really though, let's see how many of these things actually show up on the market. Time for Intel to prove themselves.

They say 80+ designs this year and have already shipped 1million+ units of TGL H

 

[coercitiv]

I don't have the energy to look into this, but something is off if the info bellow is true:

MT scores as per Intel PR:
11980HK vs 10980HK = +19%
11950H vs 10885H = +16%

 

[DrMrLordX]

They say 80+ designs this year and have already shipped 1million+ units of TGL H

Which units of TigerLake-H though? The new 8c SKUs or the not-as-new 4c high TDP SoCs?

 

[Hulk]

So Tigerlake-H is launched.

Intel Launches 11th Generation Core Tiger Lake-H: Eight Core 10nm Mobile Processors

www.anandtech.com

Base frequency
@ 65W: 3.3GHz
@ 45W: 2.6GHz

For 11900K
@125W: 3.5GHz
@95W: 3GHz

11900T
@35W: 1.5GHz
@25W: 1GHz

11900
@65W: 2.5GHz

Backport = fail

So these frequency/power numbers look to be very close to Zen 3, right?

 

[eek2121]

Which units of TigerLake-H though? The new 8c SKUs or the not-as-new 4c high TDP SoCs?

So far it looks like the 11800H may have been a significant part of that. So there are definitely some 8-core parts.

So these frequency/power numbers look to be very close to Zen 3, right?

Boost clocks are significantly higher, base clocks are lower. AVX will drag clocks down like always. wait for the reviews…

 

[Hitman928]

So these frequency/power numbers look to be very close to Zen 3, right?

Someone can check me on this but here's what I came up with.

CPU​	TDP	Cores/Threads	Base	Max/All Boost
Intel 11980HK	45 W	8/16	2.6 GHz	5.0/4.5 GHz
Intel 11980HK	65 W	8/16	3.3 GHz	5.0/? GHz
AMD 5980HS	35 W	8/16	3 GHz	4.8/? GHz
AMD 5980HX	45+ W	8/16	3.3 GHz	4.8/? GHz
Intel 11900H	35 W	8/16	2.1 GHz	4.9/? GHz
Intel 11900H	45 W	8/16	2.5 GHz	4.9/4.4 GHz
AMD 5900HS	35 W	8/16	3 GHz	4.6/? GHz
AMD 5900HX	45 W	8/16	3.3 GHz	4.6/? GHz
Intel 11400H	35 W	6/12	2.2 GHz	4.5/? GHz
Intel 11400H	45 W	6/12	2.7 GHz	4.5/4.1 GHz
AMD 5600HS	35 W	6/12	3 GHz	4.2/? GHz
AMD 5600H	45 W	6/12	3.3 GHz	4.2/? GHz

 

[jpiniero]

Not really comparable since you are talking about AVX-512 versus AVX2 presumably. Also pretty sure AMD doesn't technically have any kind of multiplier limits.

 

[blckgrffn]

I know it's fantasy, but wouldn't it be great to see a desktop 11980HK as a true LGA1200 halo CPU?

I mean, way way way back in the day I bought 1.6ghz mobile Celerons and doubled the FSB to have 3.2 ghz desktop CPUs. There is extremely dubious precedent here

(I am pretty sure those were Northwood P4 based Celerons....)

 

[coercitiv]

PL2 on 11980HK is 107w too, lower than I expected.

https://twitter.com/x/status/1392076095567106049

After a double check both 107 and 135W are possible for PL2.

https://twitter.com/x/status/1392153307158827012

I double checked with Intel: Tiger Lake-H45 (Core i9-11980HK) has a PL2 of 107/135 watts for the base/performance profile.

 

[Geranium]

I think much of Intel's process woes will be mitigated if they can move desktop parts to 10ESF in quantities ASAP. From a power and transistor density it seems as though 10SF is competitive with TMSC 7nm so I'm thinking 10ESF will be even more competitive. Assuming AMD hasn't moved on to 5nm by then. But even if they have it seems likely the density/power gains will be getting smaller as we approach the end of this node shrinking game that started 50 years ago.

Intel's 10nm is similar to TSMCs 7nm?? So what is the real chip density of Intel's 10nm? And I am not talking about paper density, real usable & production quality chip density. And Tiger Lake already shows us how efficient Intel's 10nm compared to TSMC's 7nm.

AFAIK Intel's 14nm & 10nm is so dense that Intel's don't publish real chip density number nowdays.
TSMC's 7nm density range from 40M/mm2 - 85-90 M/mm2 and chip power range from 2W SoC to 400W GPU.

 

[RTX]

PL2 on 11980HK is 107w too, lower than I expected.

https://twitter.com/x/status/1392076095567106049

His other twitter post is within the same hour as that Intel AMA
Base PL2 is 107w
There's also a PL2 135w performance

https://www.reddit.com/r/intel/comments/na00cv