Discussion Intel current and future Lakes & Rapids thread

[Abwx]

PPW gain for SF going from Intel 7 to Intel 3 would be 42%

The 42% is in comparison to the latest Intel 7 process used for RPL and wich is quite improved comparatively the one used for ADL, but still even using the latter as comparison that wouldnt be even close to 2.7x perf/watt improvement, more likely something like 60-65%.

 

[Khato]

Less than18 hours OS booting.?.

What does it mean exactly.?.

And they state excellent silicon health, there s some irony here...

It means that it took less than 18 hours from first power on to getting an OS to boot. Indicates no major issues with reset/initialization/basic execution.

 

[DrMrLordX]

Yes SRF maxes out at 144c, clear water forest goes higher

Wait what? Did Intel kill the 288c variant as well?

 

[SiliconFly]

1. Birch Stream-AP will support up to 500W.
2. Gracemont@3.9GHz roughly consumes 4W/core.
3. Hopping Gracemont to SRF is two node jump on perf/watt gain. Potentially speaking, two node jump will deliver 0.36x power consumption at same clock.

4 x 0.36 x 432 = 622 watt

Therefore, I guess that Intel will launch SRF-AP as 432cores/3~3.5GHz/500W, the phenomenon products.

We can expect 432 gracemont cores @ 3.5 GHz to draw in excess of 1000 to 1500 watts under full load. 500W doesn't sound likely.

 

[Abwx]

It means that it took less than 18 hours from first power on to getting an OS to boot. Indicates no major issues with reset/initialization/basic execution.

That s surely not a heavy OS but something basic enough to have the thing stable, but whatever, generaly firms do not market such technical things unless they feel that there s some lack of confidence in their doings, i see that as their current buying time strategy wich consist to say that they are not that lagging...

 

[RTX2080]

Yet no 32 E cores configuration being discovered on ARL-S

https://twitter.com/x/status/1692902684612059471

 

[SiliconFly]

I don't think there ever was. It was just a rumor. Still is.

 

[SiliconFly]

That s surely not a heavy OS but something basic enough to have the thing stable, but whatever, generaly firms do not market such technical things unless they feel that there s some lack of confidence in their doings, i see that as their current buying time strategy wich consist to say that they are not that lagging...

They typically use Linux & Windows.

 

[Ajay]

It means that it took less than 18 hours from first power on to getting an OS to boot. Indicates no major issues with reset/initialization/basic execution.

Probably a stripped down version of linux. Or, DOS (for the lolz!).

 

[Ajay]

Yet no 32 E cores configuration being discovered on ARL-S

https://twitter.com/x/status/1692902684612059471

Deleted.

 

[RTX2080]

Deleted.

Ouch, looks like the japanese guy leaf_hobby is too shy. However, despite that he also confirmed 3M L2 per core for ARL, and two die configuration(8+16, 6+8) existence, and dedicated L3 for GPU tile.

 

[H433x0n]

Ouch, looks like the japanese guy leaf_hobby is too shy. However, despite that he also confirmed 3M L2 per core for ARL, and two die configuration(8+16, 6+8) existence, and dedicated L3 for GPU tile.

He still has those posts, I just checked them out myself. The translation from Japanese -> English was a bit unclear when he was talking about ARL-S configurations.

ARL-S 6+8 only ES1 and ES2 for now?So i5-15600 (Ultra 5-160?) will be 8+0

^I've got no clue what he meant by that for example.

 

[Khato]

Probably a stripped down version of linux. Or, DOS (for the lolz!).

Really not sure why anyone is particularly surprised that Sierra Forest would be relatively functional right off the bat? Assuming that prior speculation/leaks of it using the same I/O die as Granite Rapids are correct, then it's just swapping compute die compared to Granite Rapids. So long as they already had Granite Rapids working it sure should be pretty quick to bring up Sierra Forest.

 

[A///]

what was the supposed makeup of a 32 core arl processor i9 again?

 

[Ajay]

Really not sure why anyone is particularly surprised that Sierra Forest would be relatively functional right off the bat? Assuming that prior speculation/leaks of it using the same I/O die as Granite Rapids are correct, then it's just swapping compute die compared to Granite Rapids. So long as they already had Granite Rapids working it sure should be pretty quick to bring up Sierra Forest.

I'm not surprised. We used to bring up hardware on a minimal firmware 'kernel'. The DOS part was for the lolz.

 

[A///]

not speaking for ajay but I think most people have lost enough faith in intel over the last 7 years or whatever that they assume the engineers at intel are behaving like a pack of monkeys and flinging bananas at each other, which would excite igor. it's surprises people, even those who've got some faith in them still such as myself that they're hitting milestones,.

 

[SiliconFly]

not speaking for ajay but I think most people have lost enough faith in intel over the last 7 years or whatever that they assume the engineers at intel are behaving like a pack of monkeys and flinging bananas at each other, which would excite igor. it's surprises people, even those who've got some faith in them still such as myself that they're hitting milestones,.

Actually, what they're doing is a bit annoying. Hitting milestones shouldn't even make it to the news. After all, thats what they're supposed to do!

But Intel is trying to make a splash everytime they hit a milestone and it looks a bit pathetic for such a large semiconductor company thats been around for many decades.

 

[igor_kavinski]

Actually, what they're doing is a bit annoying. Hitting milestones shouldn't even make it to the news. After all, thats what they're supposed to do!

But Intel is trying to make a splash everytime they hit a milestone and it looks a bit pathetic for such a large semiconductor company thats been around for many decades.

You have a point!

They can certainly do one better and put a live tracker of their progress online.

Silicon bug found!
Silicon bug squashed!
Frequency increased by 100 MHz!
144 hours stable at 4 GHz!

 

[A///]

But Intel is trying to make a splash everytime they hit a milestone and it looks a bit pathetic for such a large semiconductor company thats been around for many decades.

you must've not been around when they were making fun of amd with the glue comments.

 

[Thunder 57]

you must've not been around when they were making fun of amd with the glue comments.

Intel Says AMD EPYC Processors "Glued-together" in Official Slide Deck

So, yes, Intel, I think the AMD engineers who have developed the Zen architecture from the ground-up would take issue with that. Especially when AMD's "Glued-together" dies actually wipe the proverbial floor with the blue company's chips in power-performance ratios, and deliver much better...

www.techpowerup.com

My how times have changed.

 

[H433x0n]

Don't forget Pat Gelsinger once said that AMD / Ryzen was in the rear view mirror.

"AMD in the rearview mirror in clients [consumer market],"

I will say that the people who favor AMD are extremely sensitive about that comment. It gets mentioned all of the time in numerous discords and subreddits I frequent.

 

[Abwx]

Don't forget Pat Gelsinger once said that AMD / Ryzen was in the rear view mirror.



I will say that the people who favor AMD are extremely sensitive about that comment. It gets mentioned all of the time in numerous discords and subreddits I frequent.

Because in its whole history AMD never made such derogatory statements, Intel on the other hand routinely practice such baseless statements, among others they once made a slide to spread FUD and saying that AMD was using defective dies for the Phenom X3...

 

[igor_kavinski]

AMD should always feel flattered that they are enough of a threat to get mentioned in Intel slides

 

[A///]

Intel Says AMD EPYC Processors "Glued-together" in Official Slide Deck

So, yes, Intel, I think the AMD engineers who have developed the Zen architecture from the ground-up would take issue with that. Especially when AMD's "Glued-together" dies actually wipe the proverbial floor with the blue company's chips in power-performance ratios, and deliver much better...

www.techpowerup.com

My how times have changed.

Some time after this intel began a purge of its marketing dept.

 

[igor_kavinski]

Some time after this intel began a purge of its marketing dept.

Is Gregory Bryant's leaving Intel related to that? I remember he infamously called AMD a copycat during the Tiger Lake launch presentation without mentioning AMD. He kept his job after that and but left soon after when the Alder Lake live overclocking failed during its launch presentation.