Intel Rocket Lake 11th gen desktop CPUs confirmed for Q1 2021

[porina]

Quote

Quote

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!

 

- John Bonini, Intel VP and GM of Client Computing Group Desktop, Workstations and Gaming

 

Sources

https://medium.com/intel-tech/intels-commitment-to-gaming-and-a-sneak-peek-at-intel-technology-to-come-83677833be7f

 

Summary

This is kind of an official pre-announcement but does acknowledge the existence of the next desktop CPU, and also a timeframe. Support for PCIe 4.0 is mentioned as a feature, but no other specs for now.

 

My thoughts

Rocket Lake has been simmering away at a low level in the rumours for a while but there's been practically no news from Intel until now. Absent further information, the rumours suggest it will be made on 14nm, but will have a back-ported Willow Cove cores similar to that used on Tiger Lake, although there may be differences required due to the older process used. This should offer an improvement in IPC over the Skylake architecture we've had for the last 5 years or so while likely to maintain clocks. PCIe 4.0 support being confirmed brings it back to parity with AMD offerings. We'll likely get a new chipset to go with it, but it is expected to share the same socket as current Comet Lake. Zen 3 announcement is happening tomorrow, which will be its competitor. Rocket Lake's successor Alder Lake has already been mentioned by Intel as expected 2H 2021 and will be on 10nm process.

 

[aisle9]

So releasing 14nm++++++++++++² in Q1, then a brand new architecture on 10nm in Q3 or Q4? It's like the Pointless Lake 7000 series all over again. Why the hell would you be an early adopter for a brand new CPU knowing it'll be a generation behind in 6-9 months?

[Levent]

Sounds like they are just pumping up stocks. IIRC they never did this kind of announcement before.

[Unixsystem]

3 minutes ago, aisle9 said:

So releasing 14nm++++++++++++² in Q1, then a brand new architecture on 10nm in Q3 or Q4? It's like the Pointless Lake 7000 series all over again. Why the hell would you be an early adopter for a brand new CPU knowing it'll be a generation behind in 6-9 months?

If Intel could be trusted on new-node-based products then we would have been 10nm parts 4 years ago. At this point I would just take 2H 2021 as "we'll wait until then to announce the next delay". 

[porina]

6 minutes ago, aisle9 said:

So releasing 14nm++++++++++++² in Q1, then a brand new architecture on 10nm in Q3 or Q4? It's like the Pointless Lake 7000 series all over again. Why the hell would you be an early adopter for a brand new CPU knowing it'll be a generation behind in 6-9 months?

Intel know they are behind, so they need to get newer CPUs out ASAP. Also generally when there is a release window, assume it will be towards the end of that release window. People are setting their expectations for March availability. Alder Lake could be at the end of the year, with rumours that it'll be delayed.

 

Expectations are for Zen 3 to build upon and improve over Zen 2. Intel are unlikely to push clock any further but we finally get the architecture update. How much improvement remains to be seen. Ice Lake was already Zen 2 IPC competitive and Tiger Lake builds on top of that. Might not be enough to pass Zen 3 but it'll keep them closer than if they did nothing at all.

 

1 minute ago, Levent said:

Sounds like they are just pumping up stocks. IIRC they never did this kind of announcement before.

Intel have released information via posts on Medium before, as a less formal setting than the main website.

[.Apex.]

2 minutes ago, aisle9 said:

So releasing 14nm++++++++++++² in Q1, then a brand new architecture on 10nm in Q3 or Q4? It's like the Pointless Lake 7000 series all over again. Why the hell would you be an early adopter for a brand new CPU knowing it'll be a generation behind in 6-9 months?

That's a good point, though it is likely done to soften the blow of the amount of stock that is lost in that first half of 2021 rather if they had not released anything at all because they know that won't compete with Zen 3 considering the rumors that we've seen recently, and an IPC improvement wouldn't matter much if the CPU still guzzles power

[Fasauceome]

Intel:

releases another 14nm CPU

 

Shareholders:

[S w a t s o n]

YAY 8 core 14nm++++++++++++++++++++++++++ backport CPUs, what a great deal. Totally sure they wont cost more than Zen3 and suck 2x power per core

 

Also this is clearly Intel trying to get ahead of the Zen 3 announcement tomorrow

[ouroesa]

Quote

This is kinda a official

Ouchies my brain. In future please write this as "This is kind of an official"

[Fatih19]

LMAO, it's Kaby Lake all over again. Replaced in less than a year.

[Techstorm970]

Intel is about to get clobbered and they know it.

[KnoT]

I rly wish that ppl would stop with this 14nm vs 7nm nonsense, those are JUST names they have no correlation to anything for quite some time now... apart from 7nm TSMC being 'better' than 14nm TSMC.

 

Great explanation here: 

 

[NotTheFirstDaniel]

Can anyone explain why Intel has stuck with "X Lake" for the past 5 years?

 

Is it because they've been stuck on the same node or have they just stopped caring? Can't wait for "Moon Lake" in 2028...

[Bombastinator]

26 minutes ago, KnoT said:

I rly wish that ppl would stop with this 14nm vs 7nm nonsense, those are JUST names they have no correlation to anything for quite some time now... apart from 7nm TSMC being 'better' than 14nm TSMC.

 

Great explanation here: 

 

Ok.  So there’s Samsung8 (omitting the nm because it’s misleading) TSMC7, intel14, and intel10.  My understanding is TSMC7 is marginally better(?) than Samsung8, and more better than intel14, whereas intel10 is supposed to be more or less on par with TSMC7, and marginally better(?) than Samsung8. Does this sound right?

[KnoT]

9 minutes ago, Bombastinator said:

Ok.  So there’s Samsung8 (omitting the nm because it’s misleading) TSMC7, intel14, and intel10.  My understanding is TSMC7 is marginally better(?) than Samsung8, and more better than intel14, whereas intel10 is supposed to be more or less on par with TSMC7, and marginally better(?) than Samsung8. Does this sound right?

Nope, you simply cant compare intel, TSMC or Samsung node as those Xnm values there are just names and names alone they could calll them "The Best Node" and it would be the same as calling it Xnm node, but if had to guess current 14nm+++++++++++ intel node is comparable with 7nm+ of TSMC but you cant rly tell as they are simply different. 

 

In the end all you should care about is Perf/watt/price interesting example here is Nvidia Turing 12nm TSMC vs AMD RDNA1 7nm TSMC

[schwellmo92]

At the moment it looks like Rocket Lake will use Willow Cove cores. Willow Cove is essentially a higher clocking Sunny Cove which had a 15-20% IPC increase over Skylake. If Intel can manage the same clocks as their current Comet Lake lineup we could expect 15-20% better performance per core. I'm not confident they will manage to attain 5.3GHz and Rocket Lake is also dropping down to 8 cores at the top end from 10, as far as we currently know at least.

 

Based off all of the above it doesn't sound like Rocket Lake would be a very competitive product so there must be something we are missing. Maybe they will have a new core architecture, or maybe 8 cores is not the top end.

 

I'll be looking forward to the release and finding out more information on Rocket Lake, they've been poking Skylake for a long time now so it will be refreshing to have a new core architecture in desktop.

[Bombastinator]

4 minutes ago, KnoT said:

Nope, you simply cant compare intel, TSMC or Samsung node as those Xnm values there are just names and names alone they could calll them "The Best Node" and it would be the same as calling it Xnm node, but if had to guess current 14nm+++++++++++ intel node is comparable with 7nm+ of TSMC but you cant rly tell as they are simply different. 

 

In the end all you should care about is Perf/watt/price interesting example here is Nvidia Turing 12nm TSMC vs AMD RDNA1 7nm TSMC

Guessing is exactly what I am trying to minimize.  You apparently said more or less “no you can’t do that but doing so anyway the intel process produces stuff just as good as the other stuff even though it clearly doesn’t when measured” 

 

that’s silly. “They’re all not really comparable” would make some sense, but that wasn’t what you said.  You said “they’re all effectively the same” which is not only not what the poster said but not what the video said.

[Fatih19]

21 minutes ago, schwellmo92 said:

At the moment it looks like Rocket Lake will use Willow Cove cores. Willow Cove is essentially a higher clocking Sunny Cove which had a 15-20% IPC increase over Skylake. If Intel can manage the same clocks as their current Comet Lake lineup we could expect 15-20% better performance per core. I'm not confident they will manage to attain 5.3GHz and Rocket Lake is also dropping down to 8 cores at the top end from 10, as far as we currently know at least.

 

Based off all of the above it doesn't sound like Rocket Lake would be a very competitive product so there must be something we are missing. Maybe they will have a new core architecture, or maybe 8 cores is not the top end.

 

I'll be looking forward to the release and finding out more information on Rocket Lake, they've been poking Skylake for a long time now so it will be refreshing to have a new core architecture in desktop.

Or maybe it's just that. Not a competitive product. I highly doubt it'll be a new architecture since they got 10nm delayed.

[KnoT]

14 minutes ago, Bombastinator said:

Guessing is exactly what I am trying to minimize.  You apparently said more or less “no you can’t do that but doing so anyway the intel process produces stuff just as good as the other stuff even though it clearly doesn’t when measured” 

 

that’s silly. “They’re all not really comparable” would make some sense, but that wasn’t what you said.  You said “they’re all effectively the same” which is not only not what the poster said but not what the video said.

But you dont have to guess at all and or care about in what process node name your CPU is made just check you benchmarks by LTT for example and then chose your CPU not by in what process node it its made...

 

https://youtu.be/1kQUXpZpLXI?t=765

[Bombastinator]

8 minutes ago, KnoT said:

But you dont have to guess at all and or care about in what process node name your CPU is made just check you benchmarks by LTT for example and then chose your CPU not by in what process node it its made...

 

https://youtu.be/1kQUXpZpLXI?t=765

Yes that is the posted video I referred to earlier. We seem to have drawn very different conclusions from it.  You seem to have drawn the conclusion that there is no functional difference between the intel and TSMC processes, whereas the impression I got was that there WAS, and that the TSMC process produced sharper definitions which allowed for smaller structures. But it was by no means 2 to 1 as is implied by the number of nm, and terming the processes by nm was something that hasn’t been accurate for a very long time. As to which one is functionally more useful from a user perspective is a completely separate topic, as the structures used by each processor maker are totally different. 
 

One example of that would be this superfin thing which is different yet again from finfet.  I don’t know where that comes in though iirc superfin will not be in rocket lake so it’s irrelevant. Unless of course it is.

[KnoT]

Nope my conclusion is that you should not care about the node or fab CPU is made. And yes you are right TSMC's gates seams to look sharper (could be just to the picture) and are thinner by 1nm but we have no idea if that matter anything at all. And yes you cant compare those 2 by just those number alone.

 

My comparative guess is made solely on their performance as in my opinion current Intel CPU's are competitive with their AMD counter parts Intel still wins for gaming while AMD trumps them with multicore workloads.

 

So to make my point clear you or anyone else should not compare CPUs/GPUs by on what node they are made on and should NOT care. W8 for Benchmarks and then draw your conclusion which is the best product for you.

[Bombastinator]

21 minutes ago, KnoT said:

Nope my conclusion is that you should not care about the node or fab CPU is made. And yes you are right TSMC's gates seams to look sharper (could be just to the picture) and are thinner by 1nm but we have no idea if that matter anything at all. And yes you cant compare those 2 by just those number alone.

 

My comparative guess is made solely on their performance as in my opinion current Intel CPU's are competitive with their AMD counter parts Intel still wins for gaming while AMD trumps them with multicore workloads.

 

So to make my point clear you or anyone else should not compare CPUs/GPUs by on what node they are made on and should NOT care. W8 for Benchmarks and then draw your conclusion which is the best product for you.

So not applying to processes at all.

[KnoT]

Just now, Bombastinator said:

So not applying to processes at all.

Yes, just ignore the process as it does NOT matter to the end customer.

[Bombastinator]

Just now, KnoT said:

Yes, just ignore the process as it does NOT matter to the end customer.

Yeah.  The question was about processes though.  Why would you think I was asking about anything else? 

[KnoT]

5 minutes ago, Bombastinator said:

Yeah.  The question was about processes though.  Why would you think I was asking about anything else? 

I might have miss read or misunderstood something, sorry as English in not my 1st language.

 

PS. Lets stop of topic here