Question Raptor Lake - Official Thread

Page 7 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Toggle sidebar Toggle sidebar
Hulk

Hulk

Diamond Member
Oct 9, 1999
4,225
2,015
136
Since we already have the first Raptor Lake leak I'm thinking it should have it's own thread.
What do we know so far?
From Anandtech's Intel Process Roadmap articles from July:

Built on Intel 7 with upgraded FinFET
10-15% PPW (performance-per-watt)
Last non-tiled consumer CPU as Meteor Lake will be tiled

I'm guessing this will be a minor update to ADL with just a few microarchitecture changes to the cores. The larger change will be the new process refinement allowing 8+16 at the top of the stack.

Will it work with current z690 motherboards? If yes then that could be a major selling point for people to move to ADL rather than wait.
 
  • Like
Reactions: vstar
Sort by date Sort by votes
N

nicalandia

Diamond Member
Jan 10, 2019
3,330
5,281
136
Exist50 said:
Bidirectional reduces the average distance, but doesn't change that some cores are closer to memory than others.
Click to expand...
In every SOC there has to be compromises, I am sure that the core nearest the memory controller really like his spot and the one furthest from is not as happy, when all things are taken into account the penalty is Negligible and if you could somehow measure it's impact using a benchmark it would be within the margin of error.
 
Upvote 0 Downvote
tamz_msc

tamz_msc

Diamond Member
Jan 5, 2017
3,821
3,641
136
Henry swagger said:
wccftech.com

Intel 13th Gen Raptor Lake Core i9-13900K With 24 Cores & 32 Threads Spotted Within The Ashes of The Singularity Benchmark

Intel's 13th Gen Raptor Lake flagship, the Core i9-13900K CPU, has been spotted within the Ashes of The Singularity benchmark database.
wccftech.com wccftech.com
Raptor lake looking good.. 13900k matching 12900k at just 3ghz 😀
Click to expand...
We don't know the clock speeds yet but given that the chip launches several months from now, it is definitely clocked around 3 GHz.
Click to expand...
So clock speed is unknown yet it's definitely 3 GHz? Par for the course coming from wccftech.
 
  • Haha
  • Like
Reactions: lightmanek, Saylick, Elfear and 2 others
Upvote 0 Downvote
IntelUser2000

IntelUser2000

Elite Member
Oct 14, 2003
8,686
3,785
136
coercitiv said:
Look at the P core data first, a ST scores of 1535 almost perfectly fits in a MT score of 12029 when HT is disabled. (1535x8= 12280)
Click to expand...

The near perfect scaling with cores makes me suspect R23 as a benchmark. It doesn't care about L3 caches, it scales perfectly with cores, and it doesn't care about memory.

It's a Dhrystone 2.0 benchmark.

That said, the overall performance figure lines up with basic theory:
-50% faster per clock
-20-30% with Hyperthreading

~1.8-2x.

While there are clock speeds to consider, the performance of the -H part pretty much lines up with 2x.

6+8 is roughly 10P. Assuming 20% gain it's equal to 11 Tigerlake-H cores, and we're seeing 20-40% gains.

Also don't count out how much refreshes can improve. Goldmont Plus Refresh increased peak frequency only by 5% or so but the sustained frequency improved by 10-15% over plain vanilla Goldmont Plus.
 
Last edited:
  • Like
Reactions: lightmanek
Upvote 0 Downvote
I

igor_kavinski

Lifer
Jul 27, 2020
16,328
10,340
106
The CPU is a very early ES chip but still managed to deliver similar performance as the Core i9-12900K which is very impressive. We don't know the clock speeds yet but given that the chip launches several months from now, it is definitely clocked around 3 GHz.
Click to expand...

So wccftech's logic is that if a CPU is several months from launch, it MUST be around 3GHz. Is this some silicon engineering rule regarding early engineering samples? Or they pulled that outta their voodoo crystal ball?
 
Upvote 0 Downvote
dullard

dullard

Elite Member
May 21, 2001
25,069
3,416
126
tamz_msc said:
So clock speed is unknown yet it's definitely 3 GHz? Par for the course coming from wccftech.
Click to expand...
I think you missed the "around" part of that quote.
igor_kavinski said:
So wccftech's logic is that if a CPU is several months from launch, it MUST be around 3GHz. Is this some silicon engineering rule regarding early engineering samples? Or they pulled that outta their voodoo crystal ball?
Click to expand...
They are just using common sense. The 12900K is 3.2 GHz base. The Raptor Lake 13900K does not have any significant efficiency gains over the 12900K, so it is physically impossible to be much faster than 3.2 GHz base unless P1 is increased or unless Intel was very conservative with the 12900K base clocks. P1 is not being increased. The added E cores will use some power, so there will be less available to the P cores. Thus, it is possible that the 13900K will be a touch lower clocked at base (hence, around 3 GHz).

Just look at the 12700K with P cores at 3.6 GHz with 4 E cores and the 12900K with P cores at 3.2 GHz with 8 E cores. The 13900K will have 16 E cores, so the trend of slightly lower base P core clocks with more E cores will likely continue.
 
  • Like
Reactions: epsilon84 and igor_kavinski
Upvote 0 Downvote
tamz_msc

tamz_msc

Diamond Member
Jan 5, 2017
3,821
3,641
136
dullard said:
I think you missed the "around" part of that quote.
Click to expand...
"Clocked around 3 GHz" - for a leak presumably using an ES, wccf doesn't specify whether it is the clock speed while running the benchmark or base speed of the P-cores.

When using "definitely", it implies that the writer is certain that the clock speed is 3 GHz. Which is a direct contradiction of the previous part of the sentence "we don't know the clock speeds yet".
 
Upvote 0 Downvote
dullard

dullard

Elite Member
May 21, 2001
25,069
3,416
126
tamz_msc said:
"Clocked around 3 GHz" - for a leak presumably using an ES, wccf doesn't specify whether it is the clock speed while running the benchmark or base speed of the P-cores.

When using "definitely", it implies that the writer is certain that the clock speed is 3 GHz. Which is a direct contradiction of the previous part of the sentence "we don't know the clock speeds yet".
Click to expand...
You are correct that they could have been more specific calling it the P-core base clock. They could have been even more specific by saying it was the 13900K (instead of implying it since the whole leak was of a 13900K).

But, I still fail to see your point. "definitely around 3 GHz" does not mean exactly 3 GHz. It means near 3 GHz. The word "definitely" applies to the next word "around". It is "definitely around" 3 GHz. It doesn't skip words and mean "definitely 3 GHz". I think they can be pretty certain about that, given that the 12900K base P core is also around 3 GHz.
 
Upvote 0 Downvote
DrMrLordX

DrMrLordX

Lifer
Apr 27, 2000
21,634
10,849
136
dullard said:
They are just using common sense. The 12900K is 3.2 GHz base. The Raptor Lake 13900K does not have any significant efficiency gains over the 12900K, so it is physically impossible to be much faster than 3.2 GHz base unless P1 is increased or unless Intel was very conservative with the 12900K base clocks. P1 is not being increased. The added E cores will use some power, so there will be less available to the P cores. Thus, it is possible that the 13900K will be a touch lower clocked at base (hence, around 3 GHz).
Click to expand...

Neither Alder Lake nor Raptor Lake will stay at their base frequency running a game like AotS.
 
Upvote 0 Downvote
dullard

dullard

Elite Member
May 21, 2001
25,069
3,416
126
DrMrLordX said:
Neither Alder Lake nor Raptor Lake will stay at their base frequency running a game like AotS.
Click to expand...
While that statement is true in most situations (assuming good cooling, outside an enclosed area like many computer desks, without massive dust buildup), it has nothing to do with the base frequency that Intel will use to label the chip.
 
Upvote 0 Downvote
DrMrLordX

DrMrLordX

Lifer
Apr 27, 2000
21,634
10,849
136
dullard said:
While that statement is true in most situations (assuming good cooling, outside an enclosed area like many computer desks, without massive dust buildup), it has nothing to do with the base frequency that Intel will use to label the chip.
Click to expand...

It has everything to do with the assumption that the leaked Raptor Lake AotS score was running at only 3 GHz (this assumption is most likely incorrect).
 
Upvote 0 Downvote
dullard

dullard

Elite Member
May 21, 2001
25,069
3,416
126
DrMrLordX said:
It has everything to do with the assumption that the leaked Raptor Lake AotS score was running at only 3 GHz (this assumption is most likely incorrect).
Click to expand...
No where did is say the leaked AotS score was 3 GHz. It said the final chip when launched will be about 3 GHz.
 
Upvote 0 Downvote
E

eek2121

Platinum Member
Aug 2, 2005
2,930
4,026
136
nicalandia said:
Okay, so perhaps the TPU MT core numbers were off(most likely they did not know they had an active P core )

So lets use ST to test Performance Per Area at stock speed, also would like to compare it with Apple M1 Firestorm Performance Core and AMD Zen 3 for reference

View attachment 57150


Intel Golden Cove core with L2$ as measured by Locuza is 7.04 mm2 and it gets 1,937 points in GB5
Apple Firestorm core with L2$ measured by Semianalysis is 3.83 mm2 and gets 1,745 points in GB5
AMD Zen 3 core with L2$ as measured by Locuza is 4.27 mm2 and it gets 1,506 points in GB5
Intel Gracemont core with L2$ as measured by Locuza is 2.19 mm2 and gets 1,168 in GB5


Performance/mm2

1st place is Intel Gracemont core with 532 Geekbench5 points per mm2
2nd place is Apple Firestorm core with 455.6 Geekbench5 points per mm2
3rd place is AMD Zen3 core with 352.7 Geekbench5 points per mm2
4th place is Intel Golden Cove with 275.1 Geekbench5 points per mm2

I left out L3$ because Apple Firestorm lack L3$ and the L3$ along with the Ring Bus are Huge in Intel recent CPU uArchs skewing the numbers.
Click to expand...

  1. You are comparing a flagship desktop chip to 2 mobile chips. You should use the 6+8 mobile variant or the test is meaningless.
  2. You are ignoring hyper-threading.
 
Upvote 0 Downvote
DrMrLordX

DrMrLordX

Lifer
Apr 27, 2000
21,634
10,849
136
dullard said:
No where did is say the leaked AotS score was 3 GHz. It said the final chip when launched will be about 3 GHz.
Click to expand...

Ahem

Page 7 - Question - Raptor Lake - Official Thread

Page 7 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
forums.anandtech.com forums.anandtech.com

This one poster said it. You entered a conversation aimed at debunking his post.
 
Upvote 0 Downvote
E

Exist50

Platinum Member
Aug 18, 2016
2,445
3,043
136
Upvote 0 Downvote
dullard

dullard

Elite Member
May 21, 2001
25,069
3,416
126
Henry swagger said:
wccftech.com

Intel 13th Gen Raptor Lake CPUs get further support in latest Linux kernel

Intel Raptor Lake to offer further software integrations into Linux 5.18, this time adding audio support for the upcoming new processor line.
wccftech.com wccftech.com
Launch is not that far off
Click to expand...
Like Exist50 said, you are jumping the gun. Even that article says "last quarter of this year", although I have seen some claiming sooner than that. Intel is going to have a rapid transition from Alder Lake to Raptor Lake, but many Alder Lake chips aren't even available for sale in the US yet (outside of "used" ones on Ebay).
 
Upvote 0 Downvote
IntelUser2000

IntelUser2000

Elite Member
Oct 14, 2003
8,686
3,785
136
I am seeing variety of Alderlake chips on Newegg US, but the low end chips are way over MSRP. $150 for 12100F is $50 more. Celeron G6900 is 60-70% higher. Also not seeing the iGPU versions of 12100.

Ok, I guess we're missing lots of the non-K versions. They might be reserved for the retailers first as the non-K and the low power -T models aren't as critical for the DIY market.

Henry swagger said:
wccftech.com

Intel 13th Gen Raptor Lake CPUs get further support in latest Linux kernel

Intel Raptor Lake to offer further software integrations into Linux 5.18, this time adding audio support for the upcoming new processor line.
wccftech.com wccftech.com
Launch is not that far off
Click to expand...

A simple Rule of Thumb. They are a business. So they need to make money. Introducing Raptorlake with a significant performance enhancement doesn't make them lots of money.

Generally, that's about a year. They 9 months they are trying are pushing it. They said 12 months is an optimal balance between introducing it too fast(which lowers the revenue of the previous generation) and too slow(which means the revenue tapers off).
 
  • Like
Reactions: coercitiv
Upvote 0 Downvote
dullard

dullard

Elite Member
May 21, 2001
25,069
3,416
126
IntelUser2000 said:
Ok, I guess we're missing lots of the non-K versions. They might be reserved for the retailers first as the non-K and the low power -T models aren't as critical for the DIY market.
Click to expand...
Yes, that is what I was talking about. I realize that the T models are not commonly purchased, but they are for sale outside the US just not in the US. Retailers have announced plenty of products using them, but they are always listed as "coming soon" or similar. I've been anxiously waiting since I need a low power HTPC.
I just can't see Raptor Lake being launched while Alder Lake hasn't fully rolled out yet (which ties into your maximizing profit comments).
 
Upvote 0 Downvote
You must log in or register to reply here.

TRENDING THREADS

COMPANY
RESOURCES
FOLLOW
Top Bottom