Thursday, November 28th 2019

Intel "Rocket Lake-S" Desktop Processor Comes in Core Counts Up to 8, Gen12 iGPU Included

Intel's 11th generation Core "Rocket Lake-S" desktop processor will come in core-counts only up to 8, even as its predecessor, "Comet Lake-S," goes up to 10. Platform descriptors for Intel's next four microarchitectures surfaced on the web, detailing maximum values of their "S" (mainsteam desktop), "H" (mainstream notebook), "U" (ultrabook), and "Y" (low power portable) flavors. Both "Comet Lake-S" and "Rocket Lake-S" are 14 nm chips. "Comet Lake-S" comes with core counts of up to 10, a TDP of up to 125 Watts, Gen 9LP iGPU with 48 execution units, and native support for up to 128 GB of DDR4-2667.

The "Rocket Lake-S" silicon is interesting. Rumored to be yet another derivative of "Skylake," it features up to 8 CPU cores, the same 125 W maximum TDP, but swanky Gen12 iGPU with 32 execution units. The memory controller is also upgraded, which supports DDR4-2933 natively. There is no "Ice Lake-H" or "Ice Lake-S" in sight (no mainstream notebook or mainstream desktop implementations), ditto "Tiger Lake." For the foreseeable future, Intel will only make quad-core designs of the two 10 nm microarchitectures. "Rocket Lake-S" is slated for 2021 when, hopefully, we'll see Intel escape the 14 nm black hole.
Source: momomo_us (Twitter)
Add your own comment

41 Comments on Intel "Rocket Lake-S" Desktop Processor Comes in Core Counts Up to 8, Gen12 iGPU Included

#26
efikkan
ratirtYes, and we will have 14nm+++ till then. 10 gen was supposed to be on 10nm.
14nm+++? I suppose you mean 14nm++.
10nm was supposed to be ready in like 2016 or something.
ratirtI don't see 10nm for 10the desktop get but 14nm+++. The only thing that is 10nm is laptops.
Right now, yes, but that's beside the point.
ratirtI understand you want Intel to release 10nm for desktops and you want the desktop parts to be awesome, just don't be surprised if it won't happen.
What I want is irrelevant for what's going to happen, I don't believe I have the power to impact Intel's decisions.
I'm just citing the facts; Intel have said 10nm desktop is on the roadmap, and they have added driver support for 10nm Tiger Lake parts, clear evidence of them intending to release something, they don't do this just for fun. But it remains to see when they will ship, and in which quantities. And those are the facts, not my opinions.
ratirtBTW. Intel still has problems with 14nm supply and you expect 10nm desktops to fill the gap?
Not happening bro.
There is no problem with 14nm supply, just that demand is strong. Intel is making more than ever.

If you read my previous posts carefully you'd see that I'm not concerned whether Intel can make the desktop CPUs on 10nm, but the quantity of such chips. Both laptop chips and OEM desktop chips requires very high volumes compared to what AMD is shipping, which poses a challenge for Intel when moving to a new node. With 10nm+ arriving in 2020 we will see a larger selection of 10nm based products, but I don't expect a full lineup. 10nm++ in 2021 will expand the volume further at the same time as the first 7nm products arrive.
Posted on Reply
#27
ratirt
efikkan14nm+++? I suppose you mean 14nm++.
10nm was supposed to be ready in like 2016 or something.


Right now, yes, but that's beside the point.


What I want is irrelevant for what's going to happen, I don't believe I have the power to impact Intel's decisions.
I'm just citing the facts; Intel have said 10nm desktop is on the roadmap, and they have added driver support for 10nm Tiger Lake parts, clear evidence of them intending to release something, they don't do this just for fun. But it remains to see when they will ship, and in which quantities. And those are the facts, not my opinions.


There is no problem with 14nm supply, just that demand is strong. Intel is making more than ever.

If you read my previous posts carefully you'd see that I'm not concerned whether Intel can make the desktop CPUs on 10nm, but the quantity of such chips. Both laptop chips and OEM desktop chips requires very high volumes compared to what AMD is shipping, which poses a challenge for Intel when moving to a new node. With 10nm+ arriving in 2020 we will see a larger selection of 10nm based products, but I don't expect a full lineup. 10nm++ in 2021 will expand the volume further at the same time as the first 7nm products arrive.
Like I said 2020 is just around the corner. We will see if the 10nm desktops show up. My bet is there will be no 10nm desktops mass production.
14nm+++ is an obvious exaggeration just like 14nm ++++++ and so on. I can't see 10nm desktop Intel CPUs counter AMD's Ryzen but lets hope for the best.

I remember not long ago, when Ryzen CPUs came out and people were saying supply problem, bad management, oh what other stuff there was. When I mentioned market demands I was quickly shot down, it is not the demands but AMD's production problem. Yet with Intel is always demand right although Intel is losing shares so how does the demand look for Intel now? It was an extensive conversation and I'm sure you participated in that as well. I guess winds have changed again.
Posted on Reply
#28
ppn
10nm for desktop will show up in the form of 50% IPC. 20+30 for sunny and willow cove total. might be worth the wait. pcie lanes will move to the pch. I hope it doesn't require yet another socket, other than ddr5.
Posted on Reply
#29
ratirt
ppn10nm for desktop will show up in the form of 50% IPC. 20+30 for sunny and willow cove total. might be worth the wait. pcie lanes will move to the pch. I hope it doesn't require yet another socket, other than ddr5.
We are discussing 10nm never making to the desktop market and you say it will show up with 50% IPC? On top of that you say it is worth the wait? 50% means the IPC is going to be halved in comparison to what Intel has now so how is this worth waiting? Unless you meant 150% IPC up which I don't see coming. Intel needs new architecture to achieve this since the "lake" architecture is basically pushed to the max. What else Intel can do with it? 10nm for desktop is not coming since if it really was, it would have been here now.
Posted on Reply
#30
R0H1T
ppn10nm for desktop will show up in the form of 50% IPC. 20+30 for sunny and willow cove total. might be worth the wait. pcie lanes will move to the pch. I hope it doesn't require yet another socket, other than ddr5.
Yes & I have a bridge to sell you, by any chance would you be interested :rolleyes:
Posted on Reply
#31
ppn
average +18% + the result upto 30%. this is a big core. skylake will look like pentium4 next to it.

"On average 18% increase in IPC in comparison to 2015 Skylake "
"willow Cove is forecasted to be up to 30% increase clock for clock compare to Sunny Cove."
Posted on Reply
#32
hat
Enthusiast
Weren't there supposed to be >8 core mainstream desktop CPUs anyway? Unless they scrapped those ideas... /shrug

Either way, any -lake CPU remains uninteresting to me. If I'm not mistaken, Sunny Cove is supposed to be the next big thing? New architecture, hopefully with improved performance and none of these nasty security bugs. We had a good run, but the lake is dried up.
Posted on Reply
#33
ratirt
ppnaverage +18% + the result upto 30%. this is a big core. skylake will look like pentium4 next to it.

"On average 18% increase in IPC in comparison to 2015 Skylake "
"willow Cove is forecasted to be up to 30% increase clock for clock compare to Sunny Cove."
What is this forecast based on? 18% later on 30% for Intel CPUs year after year when I remember pretty well how things were with Intel CPU's IPC back in the days. 1% 3% up not more. Why do you think now it will be any different? Because AMD is a competition now? Well true but Intel still doesnt have any answer for Ryzen nowadays. Today Intel got nothing and now it is really needed. I remember a year or 2 ago people here stating that intel already has CPU way more efficient than Kabylake but there is no competition and there's no point of releasing them. The competition is now and still nothing. Now you say 18% 30% willow cove, sunny cove, whatever cove. Maybe this AMD situation will mobilize Intel for some work now but I seriously doubt we will see 10nm launch for desktops with 18% or even 30% IPC uplift when there was absolutely nothing for over 2 years while AMD was flooding the market with Ryzens. Don't get me wrong I'd love to see it but saying 18% IPC increase from Intel is crazy?
Posted on Reply
#34
efikkan
ppnaverage +18% + the result upto 30%. this is a big core. skylake will look like pentium4 next to it.

"On average 18% increase in IPC in comparison to 2015 Skylake "
"willow Cove is forecasted to be up to 30% increase clock for clock compare to Sunny Cove."
Where does this "forecast" come from?
Willow Cove will feature a cache redesign, but I was under the impression that it was a "smaller" upgrade over Sunny Cove. Golden Cove is expected to be the next jump in single threaded performance.
ratirt<snip>
I remember a year or 2 ago people here stating that intel already has CPU way more efficient than Kabylake but there is no competition and there's no point of releasing them. The competition is now and still nothing. Now you say 18% 30% willow cove, sunny cove, whatever cove. Maybe this AMD situation will mobilize Intel for some work now but I seriously doubt we will see 10nm launch for desktops with 18% or even 30% IPC uplift when there was absolutely nothing for over 2 years while AMD was flooding the market with Ryzens. Don't get me wrong I'd love to see it but saying 18% IPC increase from Intel is crazy?
Not sure if I'm reading you correctly, but Ice Lake(Sunny Cove) were originally targeted for 2017(!) and Sapphire Rapids(Golden Cove) for 2019/2020, and both of these have been in development for over five years. Also, Cannon Lake, the failed 10nm port of Skylake, did have some smaller improvements to the design, improvements which are carried over to Sunny Cove of course. While some may claim that Intel have been holding back, but there is no substance to that claim. That's not to say that more competition wouldn't have motivated Intel further.
I think the big difference is that Zen has spawned attention around the term "IPC" (though often incorrectly applied), which is now practically in almost every CPU discussion, but were largely absent five years ago.
Posted on Reply
#35
ratirt
efikkanNot sure if I'm reading you correctly, but Ice Lake(Sunny Cove) were originally targeted for 2017(!) and Sapphire Rapids(Golden Cove) for 2019/2020, and both of these have been in development for over five years. Also, Cannon Lake, the failed 10nm port of Skylake, did have some smaller improvements to the design, improvements which are carried over to Sunny Cove of course. While some may claim that Intel have been holding back, but there is no substance to that claim. That's not to say that more competition wouldn't have motivated Intel further.
I think the big difference is that Zen has spawned attention around the term "IPC" (though often incorrectly applied), which is now practically in almost every CPU discussion, but were largely absent five years ago.
Honestly there's so many of these names from Intel that people are getting confused. These coves and lakes? It is starting to get harder to count all of these. Problem is that these is all just Intel theories about releasing something new and fresh.
10nm is not happening and the improvements that you are mentioning maybe are there but these are marginal. There should be 3 gen Intel not 10th. Some people here say 2 gen considering performance uplift. I wish Intel the best but I can see and start to realize it is not happening. I will believe it when I see it and as of now all you can be sure off is delay.
Posted on Reply
#36
hat
Enthusiast
ratirtHonestly there's so many of these names from Intel that people are getting confused. These coves and lakes? It is starting to get harder to count all of these. Problem is that these is all just Intel theories about releasing something new and fresh.
10nm is not happening and the improvements that you are mentioning maybe are there but these are marginal. There should be 3 gen Intel not 10th. Some people here say 2 gen considering performance uplift. I wish Intel the best but I can see and start to realize it is not happening. I will believe it when I see it and as of now all you can be sure off is delay.
You're not kidding. Remember when Ice Lake was supposed to be on 10nm, super awesome and not have security vulnerabilities? :laugh:
Posted on Reply
#37
ratirt
hatYou're not kidding. Remember when Ice Lake was supposed to be on 10nm, super awesome and not have security vulnerabilities?
I never kid about serious tech stuff. :) I remember that each year we get info about Intel's new upcoming processors and how great they will be and the IPC gains will be enormous per core (sarcasm) and then it turns out it is not enormous per core but basically it is a refresh of a refreshed refresh processor and vulnerabilities are swarming. I don't care much about the vulnerabilities actually but since Intel claims the premium and best of all these should not be there ever and yet here they are.
Maybe Intel should call it quits and just focus on its 14nm++++ stopping this embarrassment and promises Intel can't keep, brainwashing young people with its attitude and bogus fairy-tales about one of a kind processor that is just around the corner.
Posted on Reply
#38
hat
Enthusiast
I do care about the vulnerabilities. Not so much that i'm worried about being hacked through them, but the other significant downside is all the fixes that come out for said vulnerabilities that break more than they fix.
Posted on Reply
#39
ratirt
hatI do care about the vulnerabilities. Not so much that i'm worried about being hacked through them, but the other significant downside is all the fixes that come out for said vulnerabilities that break more than they fix.
Well I said I don't care thinking it is self explanatory because who would want to hack my PC and what would they get. Hesitation from Intel's side with fixing the vulnerabilities is the performance hit the CPUs get. 10th gen is a good example and yet not all fixes are applied. On the other hand, OEMs are pissed with this and the value of the product is going down, OEMs are displeased because Intel cannot deliver and the constant delays of new CPUs and supply problems of older. Well this doesn't look good cause Intel doesn't have anything not even an excuse to justify all the delays.
I remember hype about AMD CPUs at least these were justified. this 10th gen from Intel is a joke simply put. A joke that some people say here Intel's been nurturing, developing to get AMD? If that's Intel's 5-4 year work then maybe it would have been better for Intel to focus on something else.
Posted on Reply
#40
Midland Dog
lemonadesodaIntel's 11th generation Core "Rocket Lake-S" with Gen12 GPU, what a mess
deadhorse lake, like heck they used to be bridges and wells, like why not performancewall and nodewall
Posted on Reply
#41
R0H1T
Huh? I think it's obvious that someone at Intel is fond of swimming or phfishing hence all these lakes :shadedshu:
Posted on Reply
Add your own comment
Nov 21st, 2024 12:26 EST change timezone

New Forum Posts

Popular Reviews

Controversial News Posts