Discussion Intel current and future Lakes & Rapids thread

Page 192 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Richie Rich

Senior member
Jul 28, 2019
470
229
76
2. Neither Sunny nor Willow is relevant in servers, so why does this matter? By the time they are, Zen 3/Zen4 will be in place (respectively), and AMD's server lineup will still have cores that are on-par at worst, as well as nearly twice as many in a lower TDP package than Ice Lake-SP. And if Ice Lake-U has shown us anything, it's that Sunny Cove on 10nm is a very, very minor improvement in power efficiency over heavily refined Skylake on 14nm. What Intel gain in IPC is mostly dropped in clocks whilst power draw remains almost identical.

Granite Rapids is where Intel has a chance again, so you're not wrong with Golden Cove. Just... Golden Cove in servers is 2022, it's competing with Zen 5. We're yet to see how that goes.
I agree, 2022 is the year where Intel could put everything together and strike back.

AMD has clear advantage in high core count server CPU, no doubt about that. However Intel has huge monolithic 28c cores with shared L3 cache (AMD is behind in development L3 ring bus cache, Zen3 should have that finaly) and this type of CPU is better in certain HPC load (CFD doesn't scale well). There are many applications where 28core at 5GHz is twice as fast than 64c Rome at 2.6 GHz. Nobody cares about 1000W consumption (SW license and engineer hour cost is 1000x higher).

Also 10nm production can be solved faster than everybody think. There is only one suplier for EUV machines: ASML. When Intel buys those machines then problem is solved because Intel struggles with last DUV. And they can move to 7nm EUV (equal to 5nm TSMC) almost in the same time as TSMC to 5nm. And this could be 10nm EUV (equal to 7nm+ TSMC) in 2021 Alder Lake (Golden Cove) and 7nm EUV in 2022. IMHO EUV will solve all Intel's production problems. This means AMD's advantage will be gone.

Anyway, Apple has +83% higher IPC using 6xALU wide core, so who cares about single digits advantage in some slow prehistoric 4xALU core. AMD and Intel should race for first 6xALU core in x86 world and bring those +50% IPC. They should challenge these ultimate Apple cores not each other.
 
  • Haha
Reactions: spursindonesia

D283W

Junior Member
Dec 11, 2017
7
14
81
One thing I have heard second hand from Fortune 100 companies is that they may not switch to AMD because of the lack of support for storage class memory for some workloads.
 

lobz

Platinum Member
Feb 10, 2017
2,057
2,856
136
I agree, 2022 is the year where Intel could put everything together and strike back.

AMD has clear advantage in high core count server CPU, no doubt about that. However Intel has huge monolithic 28c cores with shared L3 cache (AMD is behind in development L3 ring bus cache, Zen3 should have that finaly) and this type of CPU is better in certain HPC load (CFD doesn't scale well). There are many applications where 28core at 5GHz is twice as fast than 64c Rome at 2.6 GHz. Nobody cares about 1000W consumption (SW license and engineer hour cost is 1000x higher).

Also 10nm production can be solved faster than everybody think. There is only one suplier for EUV machines: ASML. When Intel buys those machines then problem is solved because Intel struggles with last DUV. And they can move to 7nm EUV (equal to 5nm TSMC) almost in the same time as TSMC to 5nm. And this could be 10nm EUV (equal to 7nm+ TSMC) in 2021 Alder Lake (Golden Cove) and 7nm EUV in 2022. IMHO EUV will solve all Intel's production problems. This means AMD's advantage will be gone.

Anyway, Apple has +83% higher IPC using 6xALU wide core, so who cares about single digits advantage in some slow prehistoric 4xALU core. AMD and Intel should race for first 6xALU core in x86 world and bring those +50% IPC. They should challenge these ultimate Apple cores not each other.
Almost everything you say is complete delusion.
 

lobz

Platinum Member
Feb 10, 2017
2,057
2,856
136
One thing I have heard second hand from Fortune 100 companies is that they may not switch to AMD because of the lack of support for storage class memory for some workloads.
That is just bogus information.
1: in the past 6 months, Papermaster and Norrod both said in interviews that they are monitoring the need for that and implement when needed - not very realistic that a second hand information is more reliable there.
2: 3rd gen. Optane will be DDR5 compliant.
 

uzzi38

Platinum Member
Oct 16, 2019
2,565
5,575
146
AMD has clear advantage in high core count server CPU, no doubt about that. However Intel has huge monolithic 28c cores with shared L3 cache (AMD is behind in development L3 ring bus cache, Zen3 should have that finaly) and this type of CPU is better in certain HPC load (CFD doesn't scale well). There are many applications where 28core at 5GHz is twice as fast than 64c Rome at 2.6 GHz. Nobody cares about 1000W consumption (SW license and engineer hour cost is 1000x higher).

Also 10nm production can be solved faster than everybody think. There is only one suplier for EUV machines: ASML. When Intel buys those machines then problem is solved because Intel struggles with last DUV. And they can move to 7nm EUV (equal to 5nm TSMC) almost in the same time as TSMC to 5nm. And this could be 10nm EUV (equal to 7nm+ TSMC) in 2021 Alder Lake (Golden Cove) and 7nm EUV in 2022. IMHO EUV will solve all Intel's production problems. This means AMD's advantage will be gone.

Anyway, Apple has +83% higher IPC using 6xALU wide core, so who cares about single digits advantage in some slow prehistoric 4xALU core. AMD and Intel should race for first 6xALU core in x86 world and bring those +50% IPC. They should challenge these ultimate Apple cores not each other.

5GHz 28 cores? Those... don't exist. What you're thinking of is probably the 9990XE, which basically doesn't exist either - it can only be bought in limited quantities at auction. Oh, and has half the cores.

Also, you couldn't be more wrong about TDP. Nobody wants a server CPU that pulls more than 250W really, because past 250W you need watercooling to keep temps under control. And that's a pain.

On 7nm, it's not a matter of it being done faster than we think, Intel have literally said that by EoY 2021 they'll be producing GPUs on 7nm (cough cough Ponte Vecchio) and within a year will have filled out the CPU stack. By default, that means 7nm server CPU in 2021... I'm not sure how it's a case of everything else can come faster.

Also, to say that it's soon after TSMC's 5nm isn't accurate either. N5 goes into HVM early this year (March?). AMD will just be adopting it later than Apple/Huawei.

There is no 10nm EUV coming from Intel. There will be PDK updates, but EUV is reserved for 7nm and onwards.

EUV isn't a silver bullet. See: Samsung.

Anyway, Apple has +83% higher IPC using 6xALU wide core, so who cares about single digits advantage in some slow prehistoric 4xALU core. AMD and Intel should race for first 6xALU core in x86 world and bring those +50% IPC. They should challenge these ultimate Apple cores not each other.

Not particularily on topic, are we?

Also, imagine calling a core prehistoric purely because of it's ALU count.
 

beginner99

Diamond Member
Jun 2, 2009
5,208
1,580
136
However Intel has huge monolithic 28c cores with shared L3 cache (AMD is behind in development L3 ring bus cache, Zen3 should have that finaly)

OK, now you confirmed that your mostly spewing buzzwords with mediocre understanding at best. The 28c intel monolithic dies do not have a ring bus, they have a mesh (since Skylake-SP). AMD could go ring bus within one 8-core chiplet but I don't really see the need. Chiplet is already small and hence allows to solve the problem by having a lot of L3.
 
  • Like
Reactions: lobz and uzzi38

Markfw

Moderator Emeritus, Elite Member
May 16, 2002
25,478
14,434
136
I agree, 2022 is the year where Intel could put everything together and strike back.

AMD has clear advantage in high core count server CPU, no doubt about that. However Intel has huge monolithic 28c cores with shared L3 cache (AMD is behind in development L3 ring bus cache, Zen3 should have that finaly) and this type of CPU is better in certain HPC load (CFD doesn't scale well). There are many applications where 28core at 5GHz is twice as fast than 64c Rome at 2.6 GHz. Nobody cares about 1000W consumption (SW license and engineer hour cost is 1000x higher).

Also 10nm production can be solved faster than everybody think. There is only one suplier for EUV machines: ASML. When Intel buys those machines then problem is solved because Intel struggles with last DUV. And they can move to 7nm EUV (equal to 5nm TSMC) almost in the same time as TSMC to 5nm. And this could be 10nm EUV (equal to 7nm+ TSMC) in 2021 Alder Lake (Golden Cove) and 7nm EUV in 2022. IMHO EUV will solve all Intel's production problems. This means AMD's advantage will be gone.

Anyway, Apple has +83% higher IPC using 6xALU wide core, so who cares about single digits advantage in some slow prehistoric 4xALU core. AMD and Intel should race for first 6xALU core in x86 world and bring those +50% IPC. They should challenge these ultimate Apple cores not each other.
Your statement on the 28c @5.0 ghz is something NOBODY will or has done, except Intel on the one demo they got called out on.. Why do you even bring this crap up ?
 

FriedMoose

Member
Dec 14, 2019
48
28
51
I love how everyone is talking about how AVX 512 is so niche, but Intel sold $3.5 billion worth of chips in 2019 solely for use in AI. AMD had $6.5 billion in revenue in 2018, and will probably be at less than $10 billion for 2019. Intel's AVX 512 DL library is one of the main drivers of Xeon scalable growth.
 

lobz

Platinum Member
Feb 10, 2017
2,057
2,856
136
I love how everyone is talking about how AVX 512 is so niche, but Intel sold $3.5 billion worth of chips in 2019 solely for use in AI. AMD had $6.5 billion in revenue in 2018, and will probably be at less than $10 billion for 2019. Intel's AVX 512 DL library is one of the main drivers of Xeon scalable growth.
Please tell me you're kidding
 

FriedMoose

Member
Dec 14, 2019
48
28
51
Please tell me you're kidding
"Several years ago, we began a transformation to reposition the company to take advantage of the data revolution that is reshaping computing," CEO Bob Swan said in the earnings call. "We are accelerating growth by expanding the capabilities of our workload-optimized platforms and playing a larger role in our customers' success. Demand for our Intel Xeon Scalable processors is very strong as customers continue to make Xeon the foundation for their AI-infused data center workloads."



"In 2019, Intel expects to generate over $3.5 billion in AI-driven revenue, up more than 20 percent year-over-year. "



"Intel’s collection of data-centric businesses achieved record revenue in the fourth quarter, led by record Data Center Group (DCG) revenue. DCG revenue grew 19 percent YoY in the fourth quarter, driven by robust demand from cloud service provider customers and a continued strong mix of high-performance 2nd-Generation Intel Xeon Scalable processors. Intel acquired Habana Labs in the fourth quarter, strengthening its artificial intelligence portfolio for the data center. "



Seems pretty clear to me.
 
  • Like
Reactions: pcp7 and liahos1

Adonisds

Member
Oct 27, 2019
98
33
51
I think that when/if Intel solves its problems and finally releases a significantly better desktop processor than what AMD has, they will drop the Core brand and use something new
 

lobz

Platinum Member
Feb 10, 2017
2,057
2,856
136
lets not get to carried away with what VNNI is :

VPDPBUSDMultiply and add unsigned and signed bytes
VPDPBUSDSMultiply and add unsigned and signed bytes with saturation
VPDPWSSDMultiply and add signed word integers
VPDPWSSDSMultiply and add word integers with saturation
It's the driving force behind Intel's enterprise revenue. I've learned that here today.
 

FriedMoose

Member
Dec 14, 2019
48
28
51
It's the driving force behind Intel's enterprise revenue. I've learned that here today.
Ah yes, please tell us of your vast expertise in Intel's financial details. I'm sure you can refute Bob Swan with your decades of knowledge. Please enlighten us on how AI isnt driving the growth of Xeon Scalable Processors.
 
  • Like
Reactions: liahos1

liahos1

Senior member
Aug 28, 2013
573
45
91
It looks like intel is set to gain pc share in q1 per amds guidance. I think once icelake server is out in 2h things get ugly. gg thanks for playing!
 

lobz

Platinum Member
Feb 10, 2017
2,057
2,856
136
Ah yes, please tell us of your vast expertise in Intel's financial details. I'm sure you can refute Bob Swan with your decades of knowledge. Please enlighten us on how AI isnt driving the growth of Xeon Scalable Processors.
The fact that liahos liked your comment, negates every need to even try to argue with you, so like I said, you're right and intel's AVX512 AI instructions are the main driving force there. Why even bother talking to me when I have no expertise?
 

CHADBOGA

Platinum Member
Mar 31, 2009
2,135
832
136
I think that when/if Intel solves its problems and finally releases a significantly better desktop processor than what AMD has, they will drop the Core brand and use something new
They should, "Core" is the most ridiculous marketing term I have ever seen a semi-conductor company use.

And even more ridiculous, someone being paid $100,000+ , likely came up with it and it has been allowed to run for years now.
 
  • Like
Reactions: spursindonesia

BigDaveX

Senior member
Jun 12, 2014
440
216
116
I think that when/if Intel solves its problems and finally releases a significantly better desktop processor than what AMD has, they will drop the Core brand and use something new
Depends how long that takes, and how far AMD is able to pull ahead in the interim. For the time being at least, the Core name hasn't been rendered as toxic as the Pentium (4) brand had become by 2006.
 

Adonisds

Member
Oct 27, 2019
98
33
51
Depends how long that takes, and how far AMD is able to pull ahead in the interim. For the time being at least, the Core name hasn't been rendered as toxic as the Pentium (4) brand had become by 2006.
Yeah but the 10900K vs Zen 3 should make it ugly