Question Raptor Lake - Official Thread

[Hulk]

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.

 

[jpiniero]

The 12700 has a PL2 duration of 28 seconds and a PL2 of 180W. These details are available with an absolutely trivial amount of research.

Especially with a Z board who knows what the settings would be for PL2. Even the locked processors.

 

[Markfw]

The 12700 has a PL2 duration of 28 seconds and a PL2 of 180W. These details are available with an absolutely trivial amount of research.

Intel Core i7-12700F Specs

Alder Lake-S, 12 Cores, 20 Threads, 2.1 GHz, 65 W

www.techpowerup.com

Intel® Core™ i7-12700F Processor (25M Cache, up to 4.90 GHz) Product Specifications

Intel® Core™ i7-12700F Processor (25M Cache, up to 4.90 GHz) quick reference guide including specifications, features, pricing, compatibility, design documentation, ordering codes, spec codes and more.

ark.intel.com

And why should anyone believe your claims when you make ridiculous assertions like Ryzen being 2x the performance at half the power? Why don't you provide an independent source for once?

Lets see, first , all the reviewers say Alder lake takes a lot of power, verifying reviewers. I am not going to link them all. As far as performance, you don't care about the DC forum, or any points that they would generate or stats, so I say look it up yourself. But arguing with you is like talking to a brick wall. So be happy with your delusions.

Lets just wait until Raptor lake comes out. Even other posters have agreed with me. Sick of your BS

 

[Exist50]

Especially with a Z board who knows what the settings would be for PL2. Even the locked processors.

Of course, but then you're basically complaining that Intel doesn't lock down the PL2 or tau, which seems incredibly silly in an enthusiast thread.

Lets see, first , all the reviewers say Alder lake takes a lot of power, verifying reviewers.

Then why don't you go ahead and post a review proving that those numbers are wrong? I've provided hard sources, so why can't you do the same?

As far as performance, you don't care about the DC forum, or any points that they would generate or stats, so I say look it up yourself.

You say that because you know your claims are wildly inaccurate. There is no reviewer who claims a 5950x is 2x the performance of a 12900k, or that Ryzen is 4x as efficient. Do you honestly expect people to buy that BS?

Even other posters have agreed with me. Sick of your BS

Stop making false claims if you don't want anyone to respond to them.

 

[jpiniero]

Of course, but then you're basically complaining that Intel doesn't lock down the PL2 or tau, which seems incredibly silly in an enthusiast thread.

Not complaining. That's just the reality of it.

 

[Markfw]

Of course, but then you're basically complaining that Intel doesn't lock down the PL2 or tau, which seems incredibly silly in an enthusiast thread.


Then why don't you go ahead and post a review proving that those numbers are wrong? I've provided hard sources, so why can't you do the same?


You say that because you know your claims are wildly inaccurate. There is no reviewer who claims a 5950x is 2x the performance of a 12900k, or that Ryzen is 4x as efficient. Do you honestly expect people to buy that BS?


Stop making false claims if you don't want anyone to respond to them.

Do you even read ? 5950x in DC jobs is 2x performance of a 12700F, what I have. jeees....

 

[Exist50]

5950x in DC jobs is 2x performance of a 12700F

Ok, then why don't you provide a review or 3rd party benchmark to substantiate that claim? Or do you expect us to believe you without evidence, or actually, despite the evidence?

 

[eek2121]

The end goal is to have EVERY application specify how to be scheduled. Windows can't possibly do it all perfectly. That requires recoding. The code changes are minimal, but it does require time and effort to do so.

When a programmer creates a thread, there are several things that are assigned. You must give the thread a name, you must assign the thread to be background or foreground, you must assign a rough priority level (note a programmer can ignore setting these but they just are assigned to default values). The priority level wasn't very specific or granular as there were only 5 priority levels: https://docs.microsoft.com/en-us/dotnet/api/system.threading.threadpriority?view=net-6.0

But now there are a whole lot more options when you create a thread. For example, the programmer can specify what cores to strongly prefer (such as use the P-cores and not E-cores, use all cores, use only E-cores, use a mixture of P and E cores, etc): https://docs.microsoft.com/en-us/windows/win32/api/winbase/nf-winbase-setthreadaffinitymask Or, the programmer can weakly suggest what cores to use (such as use a P core if available otherwise use an E core): https://docs.microsoft.com/en-us/wi.../nf-processthreadsapi-setthreadidealprocessor The programmer must still specify priority, but now there are many more levels to choose from instead of just 5: https://docs.microsoft.com/en-us/wi...adsapi/nf-processthreadsapi-setthreadpriority

The problem is that it takes years for new software to come out with these changes. Ultimately in the future when there are dozens of E cores, the choice is obvious. But right now when there could be just a few E cores the choice is quite difficult for Windows to make. @igor_kavinski was incorrect, Windows always can choose the priority--that functionality that igor wanted is already present. The thread director is simply there to give Windows hints for the really terrible processors with only a few E cores (especially the i7 series with just 4 E cores).

Respectfully, you are wrong. I have written code in everything from ASM and C/C++ to Python, Ruby, and Javascript. Unless you are using the Windows API or your code maps to it, you have no choice.

 

[Henry swagger]

13900k will hit 2500 points in cinebench r23 single core

 

[pakotlar]

13900k will hit 2500 points in cinebench r23 single core

That would be a 25% increase, which would be pretty surprising for single core uplift

 

[Thunder 57]

That would be a 25% increase, which would be pretty surprising for single core uplift

I wouldn't take anything they say seriously. Just a mouthpiece for Intel really.

 

[dullard]

Respectfully, you are wrong. I have written code in everything from ASM and C/C++ to Python, Ruby, and Javascript. Unless you are using the Windows API or your code maps to it, you have no choice.

Time to update your knowledge of Windows coding especially starting with Visual Studio 2022. Javascript isn't even multithreaded! Heck even if you don't want to touch Windows, Linux started the optimizations in version 5.18.

 

[DrMrLordX]

Undervolted 12900K will not be competitive with 5950X..

It will be in those benchmarks where it can win. Especially in a lot of single-threaded or sparsely-threaded workloads. It loses very little performance in those benchmarks/applications. At the very least Intel should have shown reviewers how to easily set the thing to ~150W mode (or whatever) for benching, so people could see for themselves. But all too many benches were carried out at 241W.

 

[igor_kavinski]

Javascript isn't even multithreaded!

Vanilla JS isn't. But, https://codersblock.com/blog/multi-threaded-javascript-with-web-workers/

 

[dullard]

But all too many benches were carried out at 241W.

I agree. They yank off the stock cooler (if it had one), put on a cooler for 250W+ cooling, let the motherboard be set to high power everything, and do their reviews. Honestly, that isn't relevant to anyone other than a niche group of enthusiasts. I think the reviewers really should also do many more tests at 65W or 125W. The performance is lower. But I think many will be shocked that the performance doesn't drop that much in most benchmarks, but the power used is much less.

 

[pakotlar]

Lets see, first , all the reviewers say Alder lake takes a lot of power, verifying reviewers. I am not going to link them all. As far as performance, you don't care about the DC forum, or any points that they would generate or stats, so I say look it up yourself. But arguing with you is like talking to a brick wall. So be happy with your delusions.

Lets just wait until Raptor lake comes out. Even other posters have agreed with me. Sick of your BS

This seems awfully rude for a moderator.




You cannot callout the mods, even though he was posting as a member.


esquared
Anandtech Forum Director

 

[pakotlar]

Vanilla JS isn't. But, https://codersblock.com/blog/multi-threaded-javascript-with-web-workers/

Javascript isn't multithreaded period. Web workers are kernel threads (processes) + message passing. Typically multithreading refers to user-space threads, since otherwise every language with the ability to run a bash/shell command is trivially multithreaded.

 

[Henry swagger]

That would be a 25% increase, which would be pretty surprising for single core uplift

It will hit 6ghz single core and 5.8 multi core

 

[Henry swagger]

I wouldn't take anything they say seriously. Just a mouthpiece for Intel really.

And you are a amd fanboy lol





you cannot use the term fanboy here. This is your one zero point warning.


esquared
Anandtech Forum Director

 

[Thunder 57]

It will hit 6ghz single core and 5.8 multi core

Oh, and you have some kind of proof of this?

And you are a amd fanboy lol

Says the one citing rumors as facts. And now your opinion as fact. Others will see right through that if they haven't already. Only a chump (or shill) would believe they are going to hit 6GHz or better yet 5.8GHz multicore when they are doubling the E core total and need to stay at the same TDP on the same node.

 

[TheELF]

Lets see, first , all the reviewers say Alder lake takes a lot of power, verifying reviewers. I am not going to link them all. As far as performance, you don't care about the DC forum, or any points that they would generate or stats, so I say look it up yourself. But arguing with you is like talking to a brick wall. So be happy with your delusions.

Lets just wait until Raptor lake comes out. Even other posters have agreed with me. Sick of your BS

Yes, but you are using "does" and "has to" interchangeably and they aren't.
To post my picture again, you can see 300W usage there but the difference from 160W to 300W is less than 10% ,between the stock 241W and the "unlimited" 300W the difference is basically zeroand it gets the same performance as the 5950x at 160W even though the 5950x has more than 30% more threads.
Have you bothered using IXTU on your DC workload for your 12700k to figure out the most efficient settings to run it to see how much difference there really is?

Well, maybe for you. What about the millions that will buy an OEM system, and have no idea about any of this. If the product is not "out of the box great", then it has a problem. Tweaking on AMD can achieve nice results, but out of the box they are great. For Intel to suceed, they need an "out of the box" success.

For 99% of those people if the system doesn't freeze or shut down, like ryzen does if it gets too hot, and is fast enough for them not to notice any issues then it's a great product. Users of OEM systems don't have any idea about the things you talk about.
Also they don't use their systems for DC like you do.

 

[DrMrLordX]

For 99% of those people if the system doesn't freeze or shut down, like ryzen does if it gets too hot

lol wut

 

[Timmah!]

Yes, intel should have enforced a max TDP of 160W, maybe up to 190W and punish any reviewer that used unlimited TDP MCE or anything other that would increase power draw.

Messungen: Leistungsaufnahme, Temperatur und Energieeffizienz - Seite 9 - Hardwareluxx

Core i9-12900K und Core i5-12600K: Hybrid-Desktop-CPUs Alder Lake im Test.

www.hardwareluxx.de

I wonder what the sustained top clocks are at those lower wattages. Since there is little difference between 300W and 200W score, should i assume so is the difference between clocks? Or the performance does not scale with clocks linearly?

 

[Accord99]

I wonder what the sustained top clocks are at those lower wattages. Since there is little difference between 300W and 200W score, should i assume so is the difference between clocks? Or the performance does not scale with clocks linearly?

The used benchmark, Cinebench R23 scales almost perfectly with clock speeds. The low difference between 300W and 200W is probably from the 300W runs hitting the overheat limit and frequently throttling the P-cores while at 200W, the P-cores can run continuously at 4.8-4.9 GHz.

 

[nicalandia]

It will be in those benchmarks where it can win. Especially in a lot of single-threaded or sparsely-threaded workloads. It loses very little performance in those

That is Alder Lake in general, but if you are going to under volt a 12900K/KS and compare it against a 5950X on Single Threaded Apps/Games then might as well get a 12700K/F.

They Pushed the 12900K/S So much to match or get a win for a short burst even in MT apps(like Cinebench R23) but they throttle down once the work load is sustained for 5-10 minutes so for content creators that put their computer to work Amd makes more sense because at sustained workloads AMD is not nearly affected as Intel CPUs

 

[dullard]

Well, maybe for you. What about the millions that will buy an OEM system, and have no idea about any of this. If the product is not "out of the box great", then it has a problem. Tweaking on AMD can achieve nice results, but out of the box they are great. For Intel to suceed, they need an "out of the box" success.

Psst: most OEMs don't throw away the stock cooler, replace it with a high-end cooler, put it in a motherboard that continuously supplies 250W+, and pair it with a power supply that can handle it. Example: The typical Dell Alder Lake computer comes with either a 180 W or a 300 W power supply total. There is no way the 180 W power supply is providing 241 W to the CPU long term. https://www.dell.com/en-us/member/s...desktop/spd/inspiron-3910-desktop/nd3910fjwhs


Instead most OEMs are running much closer to the 65W/125W limits that the stock coolers supply. I do realize the K chips don't come with stock coolers, but your run of the mill OEM computer still doesn't dedicate that much power to the CPU.