Discussion Intel current and future Lakes & Rapids thread

Page 303 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

dmens

Platinum Member
Mar 18, 2005
2,271
917
136
So you're saying that AVX frequency penalty is a result of IR drop on the chip and voltage droop from the on-die VR? I always thought it was a power limit thing.

Yeah, this guard-banding is primarily the result from the response characteristics of the on-die VR, and some other issues which I will not go into.
 

mikk

Diamond Member
May 15, 2012
4,140
2,154
136
The cooling setup overall isn't great from XPS 9310, the bottom looks weak with no real ventilation grill areas. Lenovo Thinkpad 13s/14s Gen2 could have a really good cooling. According to the maintenance manual it's using an improved Yoga Slim 7 cooling system. The heat sink around the CPU die area looks a lot bigger than on the Yoga Slim 7. And unlike the XPS 9310 the Thinkbook 13s/14s features two big ventilation grill areas on the bottom.

From Thinkbook 13s/14s:


cooling92keg.png



Yoga Slim 7 (4800U): https://www.ultrabookreview.com/wp-content/uploads/2020/09/cooling-2.jpg
 

jpiniero

Lifer
Oct 1, 2010
14,599
5,218
136
Yeah, this guard-banding is primarily the result from the response characteristics of the on-die VR, and some other issues which I will not go into.

Pretty sure Skylake Client doesn't have an on die VR though, and it has the same issue.
 

dmens

Platinum Member
Mar 18, 2005
2,271
917
136
Pretty sure Skylake Client doesn't have an on die VR though, and it has the same issue.

Like I said, it is more than one issue. A poor VR response is very hard to overcome. But a better response does not mean the the rest of the power delivery chain won't have issues.
 

JoeRambo

Golden Member
Jun 13, 2013
1,814
2,105
136
@JoeRambo It's larger. Still the largest contributor to area is the doubled execution units. You can see from Skylake-SP. We'll see how much larger it is with Icelake-SP.

It was more in context of 2x256 vs 1x512 unit and ability to "convert" between the two. I think area wise 2x256 will be similar to 1x512, but that there is a huge jump in register file size and you need a proper 512bit unit to handle quite some AVX512 instruction.
 
  • Like
Reactions: Gideon

Gideon

Golden Member
Nov 27, 2007
1,641
3,678
136
It was more in context of 2x256 vs 1x512 unit and ability to "convert" between the two. I think area wise 2x256 will be similar to 1x512, but that there is a huge jump in register file size and you need a proper 512bit unit to handle quite some AVX512 instruction.
Yet you have the considerable advantage of not gimping your big-cores instruction set when running in "big-littte" mode. IMO that's a pretty important self-inflicted handicap, as the competitor doesn't have AVX-512 at all.
 

TheGiant

Senior member
Jun 12, 2017
748
353
106
Poor laptop just can't win for trying, lol.
:)
first setup can't clock higher than 3.0 and now second fu.d up thermal management- let it overheat so it can throttle to have lower overall throughput
my tip for the second setup is that it forces the CPU to go higher than the cooling system can manage
overall I am mightly dissapointed by the tiger lake lineup and launch overall
the best CPU mobile product in years for ultrabooks
Intel test laptop wasn't big, fat or what, the same as the renoir existing cooling solutions
just copy paste...no
the tiger lake cooling design even with major players looks like this
1603186988055.png
 

DrMrLordX

Lifer
Apr 27, 2000
21,632
10,845
136
@Zucker2k

Didn't expect the 11400 part. That may significantly undermine Comet Lake-S refresh products unless supplies on the 11400 are short.
 

mikk

Diamond Member
May 15, 2012
4,140
2,154
136
For Tiger Lake (and theoretically Rocket Lake as well), a new and significant optimization landed today in Mesa 20.3-devel. The optimization applies for Intel Gen12 graphics except for discrete/DG1 graphics. The patch is for enabling additional L1/HDC caching on Tiger Lake that results in "higher performance across the board, including 9-11% on a few titles."
 
  • Like
Reactions: lightmanek

DrMrLordX

Lifer
Apr 27, 2000
21,632
10,845
136
what is that? 14nm+++3i ?

I'll expand on what @Dayman1225 said.

Different rumours abound. Some suspect Rocket Lake-S will be cancelled following the cancellation of Rocket Lake-U, though I personally doubt this possibility. Others think Comet Lake-S will be refreshed for 4c parts. Still others suspect that Rocket Lake-S will ship in very low volumes (for whatever reason), and that some or all of the Comet Lake-S lineup will be refreshed with possible clockspeed bumps or other adjustments. There may be variants of Comet Lake-S released with PCIe 4.0 support, though I'll believe that when I see it.
 

coercitiv

Diamond Member
Jan 24, 2014
6,201
11,903
136
Some suspect Rocket Lake-S will be cancelled following the cancellation of Rocket Lake-U, though I personally doubt this possibility.
I think by now we can all agee RKL-S is not going to get canceled. They're publicly talking about it.
I’m also happy to confirm that the next generation 11th Gen Intel Core desktop processors (codenamed “Rocket Lake”) is coming in the first quarter of 2021 and will provide support for PCIe 4.0. It’ll be another fantastic processor for gaming, and we’re excited to disclose more details in the near future.
 
  • Like
Reactions: lightmanek

AMDK11

Senior member
Jul 15, 2019
225
152
116
Maybe the x86 CypressCove cores are GoldenCove with an inbuilt inclusive cache subsystem with the same amount as SunnyCove? This is just my theory :)
 
Last edited:

jpiniero

Lifer
Oct 1, 2010
14,599
5,218
136
Rumours say that RKL covers i5/i7/i9 and that no RKL i3 exists and is intstead reusing CML for i3s.

If the split ends up being i3 and below, that would be better. They could have i5 be both Comet and Rocket. They could make the i3 be 6C12T Comet.
 

cortexa99

Senior member
Jul 2, 2018
319
505
136
Max 8 cores? No i3s? Max turbo not higher than CML? If this is RKL I think this release is a big disappointment or just correct me if anything is misinformed......
 

Ajay

Lifer
Jan 8, 2001
15,454
7,862
136
I think by now we can all agee RKL-S is not going to get canceled. They're publicly talking about it.
Uhh...
Though as you’d expect, we’re constantly looking ahead at what’s next and how we can make our desktop CPUs even better. With that said, I’m also happy to confirm that the next generation 11th Gen Intel Core desktop processors (codenamed “Rocket Lake”) is coming in the first quarter of 2021 and will provide support for PCIe 4.0. It’ll be another fantastic processor for gaming, and we’re excited to disclose more details in the near future. There’s a lot more to come, so stay tuned!
 

dullard

Elite Member
May 21, 2001
25,066
3,415
126
Max 8 cores? No i3s? Max turbo not higher than CML? If this is RKL I think this release is a big disappointment or just correct me if anything is misinformed......
Don't expect anything major from Rocket Lake. It is just a slight boost on a few stats and essentially Intel's way of doing a price cut without changing the prices of existing products. It will run well, but it will run hot and be power hungry. If you don't want a disappointment, then you'll need to wait a few more months after Rocket Lake launches for the really revolutionary new performance chips from Intel.
 

.vodka

Golden Member
Dec 5, 2014
1,203
1,537
136
Max 8 cores? No i3s? Max turbo not higher than CML? If this is RKL I think this release is a big disappointment or just correct me if anything is misinformed......

You can't expect much from a backport of a 10nm core to an inferior process.

14nm_(insert random number of plusses) Willow Cove will lose in all metrics to its 10nm full version. Area, power, performance, all worse. Features that were the icing on the cake on the full version most probably got cut in the process.

Is it better than Skylake and its five or six refinements? Yes. Is it good enough for a 2021 product considering what the competition has? I don't think so. 2018-19 would've been a very different situation for this product to see the light of day.

10 cores -> 8 is a requirement if you don't want these chips to fight nV's GA102 on power consumption, and the new Xe graphics, while much more capable and powerful, need a lot more area too. There are a lot of constraints here at play.

Rocket Lake was always going to be a stopgap solution for a dead end transition platform (LGA1200). Now that 10nm is viable in some shape or form, LGA1700/Alder Lake is where you should look for Intel's real show of whatever they have had collecting dust in the labs for the 80 times 10nm got delayed.

Problem for Intel is, Rocket Lake goes up against Zen3 closing the single threaded performance gap somewhat and only that, and Alder Lake will probably enjoy some time in peace if its launch is timed right until AM5 and Zen4's release.
 

Ajay

Lifer
Jan 8, 2001
15,454
7,862
136
Rocket Lake was always going to be a stopgap solution for a dead end transition platform (LGA1200). Now that 10nm is viable in some shape or form, LGA1700/Alder Lake is where you should look for Intel's real show of whatever they have had collecting dust in the labs for the 80 times 10nm got delayed.
AFAIK, Rocket Lake is there because Alder Lake will have lower performance (10nm) and be available in lower quantities. 10nm CPUs have, for the most part, become stopgap solution for what is becoming a terminally late 7nm products. Lucky for them that AMD is volume limited and hasn't yet overcome the credibility that Intel took decades to build with IT professionals.
 

jj109

Senior member
Dec 17, 2013
391
59
91
Don't expect anything major from Rocket Lake. It is just a slight boost on a few stats and essentially Intel's way of doing a price cut without changing the prices of existing products. It will run well, but it will run hot and be power hungry. If you don't want a disappointment, then you'll need to wait a few more months after Rocket Lake launches for the really revolutionary new performance chips from Intel.

I swear every time I visit this thread you guys substract a few 100Mhz and % IPC from Rocket Lake. Now it's just a refresh of Skylake?

Mebiuw still claims 18% IPC in Spec2017 over Comet Lake.
Which makes sense if the core is from Willow Cove bolted onto the Sunny Cove cache configuration.
 
  • Like
Reactions: ondma