Question Raptor Lake - Official Thread

Page 196 - 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
4,367
2,232
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.
 
Reactions: vstar

uzzi38

Platinum Member
Oct 16, 2019
2,698
6,393
146
Computerbase has the 14600k using more power but TPU has it using less, so, *shrugs*.

View attachment 87335
There's some variance in silicon quality even within the same product tier already.

RPL-R isn't different enough that it really falls outside of that already existing range of quality. Which is why reviewer results seem all over the place - some worse, others better.
 

psolord

Platinum Member
Sep 16, 2009
2,015
1,224
136
Is that GEAR1 or GEAR2? GEAR1 would beat most "sane" DDR5 setups.
Just loading XMP and GEAR2 would not be that good.
I don't have the new RAM yet per se. I am thinking of getting these:


Right now I have a 3600kit on the 12400f, which will be moved to the other system. I only have enabled the XMP profile I believe. I don't know if that enables gear 1 or gear 2. Is it shown in cpuz or something? I am just looking for a rough estimation and maybe even not that.

I just need to skip the framedrops I am getting on the 12400f, while they are clearly not gpu related. I mean going from 12400f+3600 to 14600kf+4400, should suffice, right?

I see in TPU's test, that the 12400f scores 78.8% compared to the 100% of the 14600kf.


That means the 14600kf is 27% faster (1080p results for cpu limits only), which should be able to close the framedrops I have at 50s on the 12400f, right? Going from 50 to 60 stable is only 20% which....fits inside the 27%, in napkinian maths, right? xD Actually the 13600k aint far behind. If it gets a good discount, I might get that instead.
 

Gideon

Golden Member
Nov 27, 2007
1,703
3,912
136
The "not so great news" from the 14th gen in my view is the 14600K, no movement there in terms of value. But I'm not going to complain because 13600K was a big jump already (the champ of 13th gen), had they put 8 P cores in the 14600K it would have killed 14700+ sales completely. I certainly would have immediately bought a 14600K with 8+8 config though.
The biggest benefit 14600K can have, is to lower 13600K prices. Hopefully at least some fire-sales of old stock.
 

Gideon

Golden Member
Nov 27, 2007
1,703
3,912
136
I don't have the new RAM yet per se. I am thinking of getting these:


Right now I have a 3600kit on the 12400f, which will be moved to the other system. I only have enabled the XMP profile I believe. I don't know if that enables gear 1 or gear 2. Is it shown in cpuz or something? I am just looking for a rough estimation and maybe even not that.

I just need to skip the framedrops I am getting on the 12400f, while they are clearly not gpu related. I mean going from 12400f+3600 to 14600kf+4400, should suffice, right?

I see in TPU's test, that the 12400f scores 78.8% compared to the 100% of the 14600kf.


That means the 14600kf is 27% faster (1080p results for cpu limits only), which should be able to close the framedrops I have at 50s on the 12400f, right? Going from 50 to 60 stable is only 20% which....fits inside the 27%, in napkinian maths, right? xD Actually the 13600k aint far behind. If it gets a good discount, I might get that instead.

It really depends what you use it for and what is your budget for the upgrade.

As a guy who likes to squeeze the most out of a platfor, I do sympathize very much with this idea (I'm still running my B350 Tomahawk from 2017 with upgraded memory and CPU: 5800X3D), but considering the current mobo / memory prices I'm not 100% sure it makes sense, when you also need to upgrade the memory.

While F4-4400C19D-32GVK is one of the best DDR4 kits available, where I live there are plenty of DDR5 kits available which only cost slightly more:

And while you do lose in latency with DDR5 you do win quite massively in bandwidth,. 35% even when running at exactly the same MHz according to this. In AIDA64 Copy test it'll be closer to 40-60% (DDR5 7200 MT/s vs DDR4 4400MT/s):







Content creation definitely favors bandwidth and modern games tend to favor as well (though the mileage can vary).

Bear in mind, in games the differences only really come out when gaming at high refresh rates. Below 100-90 FPS It's all about the GPU and CPU..


Bottom line

If you need to upgrade the memory anyway, also take a look at some DDR5 boards. Some like B760 Pro RS or Z790 PG Lighting can be surprisingly cheap, as are some memory kits (links above).

If you need to pay $320 for the CPU and $100+ for the memory anyway, perhaps spending extra $150 on a new MoBo and $50 extra on a DDR5 kit can be worth it? (~ $620 vs ~ $420)

The difference isn't drastic by any means, but it's there
 
Reactions: Tlh97 and psolord

DrMrLordX

Lifer
Apr 27, 2000
21,790
11,131
136
I think Meteor Lake as HTPC or SFF will do quite well.

Given the poor availability and high prices in Q4 of this year, I think it'll be awhile before any OEM takes Meteor Lake seriously as an SFF SoC.

You make it sound like a terrible thing if Intel will fix the heat and power issues.

If their performance effectively stands still while fixing those heat and power issues while their competitor doesn't necessarily suffer those problems while still advancing performance generation-over-generation, then it's not going to be good for Intel.
 

Heartbreaker

Diamond Member
Apr 3, 2006
4,260
5,257
136
Would the disappointed people have been happier if it was called e.g. 13950K?
Or is it a product that shouldn't exist which such a low performance difference?

It's basically irrelevant, and 13900K and 14900K (or 13950K) are all kind of terrible power hogs, that I would never buy.

I kept seeing posts on reddit asking if it was worth waiting for 14th gen. Which baffled me, because it was obvious this was going to be an irrelevant product with tiny clock speed boost on the same silicon.

Hopefully Intel has something better up their sleeve for next gen. They need to have a cache boost for gamers, to compete with AMD X3D CPUs.
 

inf64

Diamond Member
Mar 11, 2011
3,759
4,212
136
It's so weird to see the same Intel that brought us Sandy Bridge and Skylake, the Intel with it's boot on AMD's neck through all the Bulldozer years, struggling so badly now.

Granted they're not getting their stuff pushed in like AMD was, but having to basically double the power consumption of your competitor to still lose or at best draw even is... not a great look.
It's a horrible look : https://www.pcgameshardware.de/Core...14600K-UVP-Release-Benchmark-Specs-1431342/3/

Losing the gaming performance comparison while drawing 2.5x+ more power in the process vs 7950X3D (and 3.6x more power vs 7800X3D).



Overall performance index (combines apps and gaming):

 

Hitman928

Diamond Member
Apr 15, 2012
5,567
8,717
136
There's some variance in silicon quality even within the same product tier already.

RPL-R isn't different enough that it really falls outside of that already existing range of quality. Which is why reviewer results seem all over the place - some worse, others better.

Yes, I am aware of variance in sample quality. RPL-R should have an advantage in performance/efficiency due to better thermals but even with that advantage, there seems to be no clear improvement over RPL which indicates that there isn't any difference in silicon process used between RPL and RPL-R, hence the shrug.
 

evident

Lifer
Apr 5, 2005
11,936
538
126
can anyone confirm if there's AV1 Encode/Decode support? This might be the kicker for me to get the 14 series vs a 12 or 13 series if it does. would be for an over kill nas /home server build
 
Last edited:

Heartbreaker

Diamond Member
Apr 3, 2006
4,260
5,257
136
can anyone confirm if there's AV1 Encode/Decode support? This might be the kicker for me to get the 14 series vs a 12 or 13 series if it does. would be for an over kill nas /home server build

No change. It's the same chip, just clocked higher.

14th Gen, Meteor Lake, Laptop parts will get AV1 support.
 

coercitiv

Diamond Member
Jan 24, 2014
6,371
12,748
136
can anyone confirm if there's AV1 Encode/Decode support? This might be the kicker for me to get the 14 series vs a 12 or 13 series if it does. would be for an over kill nas /home server build
I'm a bit confused, AFAIK there are no differences between 13th and 14th except maybe firmware tweaks. I would expect the same AV1 decode support and no AV1 encode support.
 

evident

Lifer
Apr 5, 2005
11,936
538
126
I'm a bit confused, AFAIK there are no differences between 13th and 14th except maybe firmware tweaks. I would expect the same AV1 decode support and no AV1 encode support.
early rumors were that it would have had it. quick google search was very unclear and just had speculation that it WOULD be included. guess its a 13th gen microcenter combo deal for me then!
 

eek2121

Diamond Member
Aug 2, 2005
3,042
4,258
136
Gahhh dayum! Those power figures. Is this thing a CPU or is it a GPU? I know perf/W doesn't matter too much in the desktop space but everything has a limit.
Oh it matters to some of us…
By the way, anyone wishing to pull the trigger on the 14900K, I'll pay you $400 for it when you upgrade next year. Buy with confidence!
I may actually grab one at some point. Working on something interesting. Having a bunch of e-cores will make it even more interesting.
In another thread I mentioned that RPL-R was rumored to have higher power consumption but was quickly told they have the same TDP. Well, according to Intel, that is true. But, anyone with a brain can see it is not. It's a damn joke honestly. Give us Arrow Lake Intel. Although rumors on that look pretty bleak as well.

Gamers Nexus did better testing than AT. In their testing there was a 5W difference between 13th gen and 14th gen. It looks like many are testing with power limits disabled (GN tested this as well)
Depends on your definition of desktop. I think Meteor Lake as HTPC or SFF will do quite well. The company that I work for only buys SFF desktop computers now (not that I have any say in it). My HTPC is less than 2 years old, so I'm not really in the market for one now. But what the rumors are saying would make Meteor lake into a great HTPC.


And if you look at all the posts the last two days they are almost exclusively talking about poor efficiency. You make it sound like a terrible thing if Intel will fix the heat and power issues.
Hoping to see Geekom or another company give us something with Meteor Lake soon!
It's so weird to see the same Intel that brought us Sandy Bridge and Skylake, the Intel with it's boot on AMD's neck through all the Bulldozer years, struggling so badly now.

Granted they're not getting their stuff pushed in like AMD was, but having to basically double the power consumption of your competitor to still lose or at best draw even is... not a great look.
There is nothing wrong with the chip being able to scale with power limits. Intel does a far better job than AMD in that regard.

However, perf/watt leaves a lot to be desired.
 
Jul 27, 2020
17,714
11,499
106

14900K iGPU seems to be slightly starved. They probably diverted some of its power to the CPU cores.
 

Hulk

Diamond Member
Oct 9, 1999
4,367
2,232
136
What i hate about tests like these is reviewers that have their cranial cavity empty. What does that test have to do with CPU power consumption?
Did it occur to those people that 14900K will have HIGHER fps than 14600K (due to more L3) and to generate those FPS, 4090 needs to work harder. The power increase for 4090 is not linear either.
They even have a graph there where 14900K has like 10% more FPS, and it is safe to say when talking about 400W system consumption that 10% increase will have large impact.

For all we know, 14900K might have consumed least and got swamped by 4090.
Correct. Also the Intel 7 node can handle insane amounts of power/heat and reach really high clock speeds. If Intel has capped the parts at 5.2 GHz or something like then they power usage would be much lower and benches just a bit lower.

But instead they gave us full control over the parts. To clock them sanely or insanely. Then people clock them insanely and complain about insane power consumption.

We "fought" with Intel for years for them to offer unlocked parts and these complaints are basically telling them we can't handle it, please lock them back up.

All of this does not refute the fact that Zen 4 is more efficient than Raptor Lake. I'm only posting to illustrate that max clocks only exacerbate the efficiency discrepancy. If you want to show Raptor in the worst light for efficiency then cool it with a custom loop and volt it to the moon. But as we all know that 100% power increase for 10% performance increase.
 

Hulk

Diamond Member
Oct 9, 1999
4,367
2,232
136
Not on desktop, it isn't.



That's a year from now, if not (slightly) longer (Dec 2024 anyone?). Also that's probably going to be more of an efficiency jump than anything else over Raptor Lake. Lower PL2 values, but not much more performance. +5% ST, +15-20% MT? Maybe? That would be fine today, but next year? It will have new competition.
Intel is going to have a heck of a time surpassing the 14900K performance-wise with the top of the stack ARL part due to the astronomical clocks of the 14900K. MT they can do if they put like 20 E's on it. P's are gonna be tougher with ST due to those 6GHz clocks we are seeing in non special edition parts.
 
Jul 27, 2020
17,714
11,499
106
This guy got lucky with what appeared to be an inferior 14900KF sample: https://www.overclock.net/threads/o...lts-bins-and-discussion.1807439/post-29248564

Quote from the next post:

We have a local PC repair shop called "Northridge Fix" who handles all of the ASUS North America repair claims/jobs in the US. The shop owner has all but confirmed the 13900KS and 14900K were all likely pulled from the same bin. Physically speaking, you could probably grab 10x 14900K's and 10x 13900KS's and if you were manually OC them, it would be a complete coinflip on which out performs which.
 
Jul 27, 2020
17,714
11,499
106

dullard

Elite Member
May 21, 2001
25,184
3,608
126
This guy got lucky with what appeared to be an inferior 14900KF sample: https://www.overclock.net/threads/o...lts-bins-and-discussion.1807439/post-29248564

Quote from the next post:
We have a local PC repair shop called "Northridge Fix" who handles all of the ASUS North America repair claims/jobs in the US. The shop owner has all but confirmed the 13900KS and 14900K were all likely pulled from the same bin. Physically speaking, you could probably grab 10x 14900K's and 10x 13900KS's and if you were manually OC them, it would be a complete coinflip on which out performs which.
Intel has been doing that for decades. It is a way to give customers a price cut without actually cutting prices on any specific chip. Instead of a $699 13900KS, you can now buy a $589 14900K. Essentially the same CPU but 15.7% cheaper and people complain.
 

DrMrLordX

Lifer
Apr 27, 2000
21,790
11,131
136
We "fought" with Intel for years for them to offer unlocked parts and these complaints are basically telling them we can't handle it, please lock them back up.

Sorry but that's bunk, on multiple levels. Two main ones:

1). Nobody asked Intel "for years" to release the same part over and over again, just at a higher TDP. At least Raptor Cove has some improvements over Golden Cove, but Raptor Lake Refresh doesn't even give you that much. Once all the dust is settled and people are no longer emotionally or financially-invested in Intel 10nm/Intel 7 releases, it would be interesting to see just how well a fully-tuned 12900k on Z690 fares against a 14900k on Z790 in a suite of benchmarks. As it stands the 14900k offers nothing much over a tuned 13900k or a 13900ks so mischaracterizing this part as an "unlocked part" that Intel fans have been clamoring for years is disingenuous. People want new parts with new designs on a new process AND unlocked halo parts as part of the stack.

2). These parts aren't so much unlocked as they are overclocked-from-the-factory. These things are nearly pushed to their limits out-of-the-box, and many users are being forced to learn to lower their power consumption by hand rather than being forced to learn to overclock them.

Intel is going to have a heck of a time surpassing the 14900K performance-wise with the top of the stack ARL part due to the astronomical clocks of the 14900K.

Their competition seems to be doing just that, so . . .

Intel has been doing that for decades. It is a way to give customers a price cut without actually cutting prices on any specific chip. Instead of a $699 13900KS, you can now buy a $589 14900K. Essentially the same CPU but 15.7% cheaper and people complain.

If that were really true, shouldn't they just cut the price on the 13900ks and be done with it? Don't lie to people and pretend like it's a fundamentally new part.
 

CropDuster

Senior member
Jan 2, 2014
369
50
91
Kinda miffed that I bought into LGA1700 partially on the hope of future platform updates, but it looks like they're pretty meaningless. Not sure if even going from 12700k to 14900k would make any meaningful difference for mostly gaming.

Am I missing anything?
 

Thunder 57

Platinum Member
Aug 19, 2007
2,794
4,075
136
Kinda miffed that I bought into LGA1700 partially on the hope of future platform updates, but it looks like they're pretty meaningless. Not sure if even going from 12700k to 14900k would make any meaningful difference for mostly gaming.

Am I missing anything?

Yes, don't buy Intel if you plan on scoket longevity. This is the first time in a long time (if ever) that Intel has blessed us with three "generations" on the same socket/chipset. If meteor Lake was where they wanted it to be I'd bet we would've gotten that on a new socet instead of RPL-R.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |