Question Raptor Lake - Official Thread

Page 124 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Hulk

Diamond Member
Oct 9, 1999
4,525
2,519
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

exquisitechar

Senior member
Apr 18, 2017
684
942
136
Below is an AMD PowerPoint clearly shows Zen 4 5nm being released prior to 2022.
View attachment 69624
No. According to AMD, that means that anything left of 2022 should be released by the end of 2022, not before. They didn't fail to execute.
Had AMD released the 3D v-cache with the Zen 4 release. They would have basically won the war with Intel for a second straight CPU generation.
N5 stacking flat out isn't ready this year according to TSMC, V-Cache was impossible on launch for this generation.
 
  • Like
Reactions: lightmanek

Hans Gruber

Platinum Member
Dec 23, 2006
2,307
1,218
136
No. According to AMD, that means that anything left of 2022 should be released by the end of 2022, not before. They didn't fail to execute.

N5 stacking flat out isn't ready this year according to TSMC, V-Cache was impossible on launch for this generation.
Various sources have said AMD wanted 3d v-cache before the end of 2022 when they knew what raptor lake could do. It would have been nice to see on launch.

It would now make sense for AMD to make Zen 4 on AM4 as well. Just think how many Zen 4 CPU's they could sell as drop in upgrades.

Zen 3 was released on Nov 5th 2020. 23 months is a long time. They failed to execute on the timeline originally set out. It hurts sales for Zen 4 to be up against Raptor Lake almost from the start.
 

Henry swagger

Senior member
Feb 9, 2022
504
306
106

Computerbase only used a RTX 3090 Ti here but Raptor Lake is dominating pretty well at gaming in their tests.

Generally a 10-20% lead over the 5800X3D and 7950x with bigger leads in RT titles. If you look at 1% lows the differences are even bigger.

When considering this alongside the data from derbauer:

The 13900k is capable of being the top performer in gaming and mixed workloads while matching Zen 4 efficiency or beating it. You can pretty much set the 13900K to 88-90W and keep the top of the chart gaming experience.

Given the cheaper cost of the CPU and the cheaper platform costs, for gaming and mixed workloads it really is making little sense to buy a Zen 4 over RPL. This is especially true for the lower end Zen 4s.

Only the 7950x makes any sense and the only use case for where this applies is for people that run 100% load all the time and especially AVX workloads. Even in this case the 13900k isn't actually too bad but the 7950x is clearly more performant and more efficient here.


View attachment 69594
View attachment 69595


There is a lot of hyperbole about the 13th gen being horrifically inefficient, power hungry, hot and burning down your home but I think the objective data shows it has very reasonable efficiency at more sane power limits while keeping almost all of the performance.
Well said.. ✔️👍
 

TheELF

Diamond Member
Dec 22, 2012
4,027
753
126
Yeah, the one users are actually stuck with. Yet another generation of Intel products that can't conform to Intel's own power specs.
People that buy high end Z boards expect what from the board?


People that don't want or need overclocks are going for sensible boards and there 80% of boards conform to base power and below.
If you care about power and can't be bothered with settings you have to do research on what board to buy, it's nothing new, this has been the norm since forever.

Even for AMD you have to do research on the boards, especially if you plan on upgrading to a much larger CPU, you have to make sure that the mobo can handle it.
zfIQvko.jpg
 

pakotlar

Senior member
Aug 22, 2003
731
187
116
Prediction for next internet points, now that I won my 13900K 15% faster than 5800X 3D with non-potato ddr5 and no gpu bottlenecks (Mark and others, I expect my points by noon):

Intel counters 7950X 3D with 32 core raptor cove only HEDT Raptor Lake - S at $1,500, 24 core at 1,000 and 13900KS at 700. New mobo for the HEDT parts, makes Threadripper and 7950X 3D look silly.
 

Yosar

Member
Mar 28, 2019
28
136
106
Why would you need to test so many games unless you have that many variances in graphic engines to test from?
Its to me honestly wasted time.

You would need to select the most commonly used engines and test performance that way, as each game can change by slight texture or optimizations the developers put into it, which would again scale if the same applied on the same engine.

And there are not 25 different engines.

It's not about engines only. 70% of games are on Unreal engine. And you don't want to sum up 70% of games to one example that is as much important as engine with one game on it.
Moreover implementation of engine can vary in a big way. There are games on Unreal engine where intel fares better, there are games on Unreal engine where AMD fares better.
That's why it is important to have in gaming test as many games as possible.
And that's why tests with 10-12 games can be dismissed, you can easily manipulate them to get picture you want.

And when you see a test of Doom Eternal when there is barely any difference between processors, you can be sure something went wrong with test and you can dismiss it (yes I look at you TPU review of 13600K in 1080p -> I didn't bother farther even, it was useless).
It means reviewer is incompetent, or simply biased. Either way it's a waste of my time, and money if I take their tests seriously.
 

Heartbreaker

Diamond Member
Apr 3, 2006
4,340
5,464
136
Any review of 13600K with DDR4 3200/3600 ??

HWUB tests the 13600K, and it includes DDR4 3600 for the gaming tests. In short AMD 7600x leads gaming by a tiny bit compared to the 13600K with DDR5 (He considers it essentially identical performance), but the gap increases when the 13600K uses DDR4.
In productivity workloads the 13600K is in a different class, it's often neck and neck with the 5950X. But he sticks with DDR5 here.

 
Last edited:

Hulk

Diamond Member
Oct 9, 1999
4,525
2,519
136
Ehh, its 230W for the 7950x and 253W for the 13900k, it's not far enough apart to even care.
This is stock as in still inside warranty for the cpus, what every mobo does is a different kind of stock.

I stand corrected. I didn't know "stock-for-stock" power levels were so close.

I don't think AMD or Intel "failed" in any sense of the word with Zen 4 and Raptor. In fact, both are brilliant and considering the world wide pandemic were released in a pretty timely fashion. It's also how they each followed completely different paths but ended up at roughly the same destination from a performance point of view. Ain't competition a beautiful thing?

I've been following this market since the early 1990's and except for a quick blip with the Athlon this is the first time Intel has had their feet held to the fire for an extended period of time. Not only are we seeing great products but you KNOW they are both scrambling to make sure their next combatant is prepared to fight in the CPU wars. Many of you probably don't remember the steady flow of $999 dollar top of the stack Intel products. It sucked. Now if Intel could just get those ARC drivers together...
 

Kaluan

Senior member
Jan 4, 2022
504
1,074
106
Ehh, its 230W for the 7950x and 253W for the 13900k, it's not far enough apart to even care.
This is stock as in still inside warranty for the cpus, what every mobo does is a different kind of stock.
Except it's not really 253W is it?
Particularly in the reviews that tested on Z790s, the socket load power is more like ~300W. As many outlets have mentioned, the default is basically unlocked usually. Intel managed to destroy both their own meaning of Tau boosting and PL2 values within 1 year. After a long history of both Intel and it's partners already playing a very loose defintions game of what stock is, with stuff like MCE and infinite boost periods.

Unless someone wants to tell me both 13700K/F and 13900K/F consume exactly the same amount of power at "stock" load cause of 253W PL2 with a straight face lol
 
  • Like
Reactions: Thor86 and Tlh97

Kocicak

Golden Member
Jan 17, 2019
1,090
1,141
136
I created a video showing that you can cool the 13900K at 300W easily with a small 240 AIO cooler, even after 10 minutes the temps were mostly below 90°C.


BTW I do not like idea of this power draw at all, my CPU is at the moment limited to 160W, I am getting ready to put my modest hyper 212 type air cooler back on.
 
Last edited:
  • Like
Reactions: clemsyn

DrMrLordX

Lifer
Apr 27, 2000
22,065
11,693
136
People that buy high end Z boards expect what from the board?

I don't think anyone seriously expected 350W unlimited mode to be ubiquitous, except for skeptics.

We are in the Raptor Lake thread. So of course the topic is on client CPUs.

Then nobody's "wiping the floor with AMD" now are they? Not that Raptor Lake is doing that either . . .
 
  • Like
Reactions: Kaluan

Doug S

Platinum Member
Feb 8, 2020
2,785
4,749
136
Sigh... You cannot release something that is not ready. Why don't they just go ahead and release Zen 5 at CES?

Yep no one wants to see AMD & Intel compete against each other with paper launches, announcing stuff that trickles out in tiny volumes with little QA that no one can buy. AMD announces Zen 5 at CES, Intel announces Arrow Lake at IDF, a few months later maybe we see a handful of risk production chips sent to a couple benchmarking sites and no one wins.
 
  • Like
Reactions: Kaluan

nicalandia

Diamond Member
Jan 10, 2019
3,331
5,282
136
Has anyone with a 13900K being able to disable the e cores and see if the AVX-512 Was somehow not laser off by Intel? Or that ship has sailed way long long ago?
 

Harry_Wild

Senior member
Dec 14, 2012
841
152
106
Some CPU comparison YouTuber should do a CPU low wattage comparisons on performance. I find zero right now, Only PC World seems interested on this subject. Waiting for their 13900K and 13600K low wattage video(s).
 

DAPUNISHER

Super Moderator CPU Forum Mod and Elite Member
Super Moderator
Aug 22, 2001
29,622
24,615
146
I stand corrected. I didn't know "stock-for-stock" power levels were so close.

I don't think AMD or Intel "failed" in any sense of the word with Zen 4 and Raptor. In fact, both are brilliant and considering the world wide pandemic were released in a pretty timely fashion. It's also how they each followed completely different paths but ended up at roughly the same destination from a performance point of view. Ain't competition a beautiful thing?

I've been following this market since the early 1990's and except for a quick blip with the Athlon this is the first time Intel has had their feet held to the fire for an extended period of time. Not only are we seeing great products but you KNOW they are both scrambling to make sure their next combatant is prepared to fight in the CPU wars. Many of you probably don't remember the steady flow of $999 dollar top of the stack Intel products. It sucked. Now if Intel could just get those ARC drivers together...
Quoted for emphasis. This is the thinking of any reasonable DIY hobbyist. We are in a golden age of PC hardware right now. AMD became competitive again, forcing Intel to stop sitting on their hands. Seems like Intel is already eating it on margins to win back lost market share. Having fired the latest shots in a price war, in which we the consumer will be the winners. Good times!
 

biostud

Lifer
Feb 27, 2003
18,700
5,434
136
To a gamer, overclocking, and pushing that envelope is more important then how efficient the cpu is.
To me as a gamer, it made sense when overclocking gave +15% more fps, but now overclocking is not something I would ever consider for my gaming rig, as the benefits are minimal and the electricity prices are so high here in Europe.
Also all modern CPUs are in 99% not going to be limiting games, compared to the video card.
 

DAPUNISHER

Super Moderator CPU Forum Mod and Elite Member
Super Moderator
Aug 22, 2001
29,622
24,615
146
To me as a gamer, it made sense when overclocking gave +15% more fps, but now overclocking is not something I would ever consider for my gaming rig, as the benefits are minimal and the electricity prices are so high here in Europe.
Also all modern CPUs are in 99% not going to be limiting games, compared to the video card.
Well stated. I have all of my Ryzens in ECO with +200MHz. I get an overclock with lower heat and power use. There isn't a single game where I could tell you the difference though, since it doesn't make any perceivable difference to gameplay.

The respondents here, mostly represent a tiny niche, inside of our niche. Any effort to extrapolate that to the larger group falls flat for me. I did way too much client facing I.T. in my time to have delusions about the common PC gamers level of knowledge or needs. Heck, you can read PCMR and see how many have no clue beyond the meme level knowledge dispensed. Which is why I think the big tech tubers fail them by making these hyperbolic videos about how hot and power hungry Raptor is. They meme it, the others read the memes, and then it has a bad rep is doesn't deserve.
 

dr1337

Senior member
May 25, 2020
417
691
136
In what world are reviewers immune from the game/software/driver bugs that MS, Nvidia, AMD, game developers, etc cannot even get right? As I type this there is a current thread on Ryzen 7000 issues with games and Windows 11: (2) Question - Windows 11 hobbling Ryzen 7000 performance | AnandTech Forums: Technology, Hardware, Software, and Deals

No review is ever more than a snapshot in time, and we all know performance can change for better or worse.
At some point you need to ask yourself why all of these 'journalists' aren't focusing in on these intel and nvidia bugs and instead only talk about the performance? Arent things like the XTU bug and nvidia driver overhead really big deals for the average diy/enthusiast consumer? Why wouldn't the headline be about the bugs? Because they know their testing is bad and don't care about the quality of their reporting, only the clicks and the money.
 
Last edited:

ondma

Diamond Member
Mar 18, 2018
3,005
1,528
136
AMD failed to execute. Intel has failed to execute many times. Every time after their failures they wipe the floor with AMD. This should not be acceptable over at AMD. I hope AMD has saved their profits. They are going to need all that capital moving forward for R & D.

I just want to point out that Raptor Lake wasn't even on the Intel timeline. They added Raptor Lake in the last 18 months because their fabs were struggling with their process nodes. It's a stop gap product. Meteor Lake is going to be a problem for AMD because of the (intel 4) 7nm process. The core advantage will be gone for AMD when Meteor Lake arrives. I still think AMD will have an efficiency lead because 3nm is the next step.

Had AMD released the 3D v-cache with the Zen 4 release. They would have basically won the war with Intel for a second straight CPU generation.
View attachment 69624
[/QUOTE]
I would not say intel has "wiped the floor" with AMD since the Bulldozer days. They have come back to be compete strongly in the consumer space while still drawing more power. I also think they will be in trouble with Zen 5 until Arrow Lake comes out. Meteor Lake seems like a non-competitive chip for the desktop. Now if they would put together two 6P+8E tiles for 12P plus 16E........ I think the criticism of the big/little architecture may be excessive by some, but eventually I think Intel will have to increase the P cores instead of just spamming E cores.
 
Last edited:

aigomorla

CPU, Cases&Cooling Mod PC Gaming Mod Elite Member
Super Moderator
Sep 28, 2005
20,894
3,247
126
well which ever side amd or intel comes out with a 18-24core HEDT that gives me at least 64 pci-e 5.0 lanes, and wont end up costing me enterprise parts prices, will probably take my money this gen.

Im not looking hopeful at ThreadRipper Pro, as it seems they want to inflate the prices on them to keep them competitive with EYPC.
And im not hopeful in Intel giving us anything but expensive W CPU's which i believe started at 3000 dollars, hence keeping on line with Enterprise prices.

But man.. i realy want a new HEDT platform that can hold me off at least a good 4-5 yrs, with ample PCI-E, and one that wont trip my breaker each time it loads up to 100%, or require me to install a dedicated 240V outlet so i can install a 2200W PSU.
 

ondma

Diamond Member
Mar 18, 2018
3,005
1,528
136
HWUB tests the 13600K, and it includes DDR4 3600 for the gaming tests. In short AMD 7600x leads gaming by a tiny bit compared to the 13600K with DDR5 (He considers it essentially identical performance), but the gap increases when the 13600K uses DDR4.
In productivity workloads the 13600K is in a different class, it's often neck and neck with the 5950X. But he sticks with DDR5 here.

I dont really understand this video. The *cpu* price of the 7600x is fine. It is slightly faster in gaming (at least in this video, results seem all over the place with different sites) while being equal to or cheaper than the 13600k. Yes it gets destroyed in productivity, but for that use, I would move up the stack anyway. What AMD really needs is cheaper motherboards. I guess saying the cpu should be cheaper is the same thing in a way, but it is actually focusing on the wrong component, IMO.

Edit: one other factor, is that the 12600k probably has some overclocking headroom, while the 7600x is closer to maxed out, so there is that.