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.

 

[SmokSmog]

Cinebench r23
5950X 1640ST 25800MT (bone stock on average)
7950X 1890ST 34800MT (15% faster in ST and 35% faster in MT vs 5950X)
13900K 2200ST 37200MT
AMD is losing while having node advantage

 

[FangBLade]

Cinebench r23
5950X 1640ST 25800MT (bone stock on average)
7950X 1890ST 34800MT (15% faster in ST and 35% faster in MT vs 5950X)
13900K 2200ST 37200MT
AMD is losing while having node advantage

You are doing damage control for Intel in every forum lol. AMD dind''t lose anything, Zen 4 isn't out yet, and cinebench is irrelevant long time ago, cpu are much more complex than cinebench.

 

[pakotlar]

First 20%, now 15%, tomorrow will be below 8%.

You think I’ll edit my post tomorrow to 8%. Why?

 

[inf64]

Cinebench r23
5950X 1640ST 25800MT (bone stock on average)
7950X 1890ST 34800MT (15% faster in ST and 35% faster in MT vs 5950X)
13900K 2200ST 37200MT
AMD is losing while having node advantage

Where did you see 7950X benchmarks?

 

[SmokSmog]

You are doing damage control for Intel in every forum lol.

I just posted pure facts, AMD own slides are saying 15% faster single core and 35% faster multi core vs 5950X and we know the numbers for 5950X and now we even know them for 13900K.

RPL will draw more power but it will be faster and all this on inferior node, only Zen4 with v-cache can bring it down in games.

 

[inf64]

I just posted pure facts, AMD own slides are saying 15% faster single core and 35% faster multi core vs 5950X and we know the numbers for 5950X and now we even know them for 13900K.

RPL will draw more power but it will be faster and all this on inferior node, only Zen4 with v-cache can bring it down in games.

You haven't learned what a ">" sign means in math?

 

[coercitiv]

I’m expecting single digit against 12900KS in games, yeah, on average. 15% vs 12900K.

The 12900KS is ~2% faster than 12900K in games.

 

[FangBLade]

I just posted pure facts, AMD own slides are saying 15% faster single core and 35% faster multi core vs 5950X and we know the numbers for 5950X and now we even know them for 13900K.

RPL will draw more power but it will be faster and all this on inferior node, only Zen4 with v-cache can bring it down in games.

Fact will be when Zen 4 releases, they are still tuning Zen 4, and there is a reason why they said greater than , and not equal to 15%, in programming world there is big difference between those 2. Protecting Intel in every forum won't change anything for Intel or AMD, AMD isn't worried about RPL, at worst 10% less performance in synthetic benchmarks, and as i said, we will see about games, if original Zen 4 fails, Zen 4-3d will take care of everything easily, with cheap or expensive ram.

 

[FangBLade]

You haven't learned what a ">" sign means in math?

Yeah, nowdays kids.

 

[SmokSmog]

As per the rules for CPU forum.

Updated: 5-4-20
"your post must contain your own original written content and cannot consist solely of another's content, be it a quote, image, or link"


esquared
Anandtech Forum Director

 

[pakotlar]

Cinebench r23
5950X 1640ST 25800MT (bone stock on average)
7950X 1890ST 34800MT (15% faster in ST and 35% faster in MT vs 5950X)
13900K 2200ST 37200MT
AMD is losing while having node advantage

If they lose it will be by a very small amount, and will win in terms of efficiency.

 

[SmokSmog]

>15% will be 30%
>35% will be 50%

AMD is sandbagging !

Guys sitting in Intel articles on multiple forums damage controlling for AMD.
Please...

 

[inf64]

Well one thing is for sure : >15% is not the same as 15%. The math is awesome!

 

[FangBLade]

Cinebench alone doesn't tell us anything, not even for content creation workloads,even less for gaming. Facts are RPL will be small upgrade, Zen 4 will be very close in ST in cinebench like apps while consuming much less power (win for me), there will be 3dcache version for gaming that will destroy RPL if non3d fails to do so, there is AM5 longevity, while anyone with RPL will need to change platform for upgrade. I expected more from RPL.

 

[FangBLade]

>15% will be 30%
>35% will be 50%

AMD is sandbagging !

Guys sitting in Intel articles on multiple forums damage controlling for AMD.
Please...

I'm glad you showed your true face, i wasn't sure who you are, but now i am.

 

[SmokSmog]

It can be 15,3% and 35,4%

 

[inf64]

Or you can apply logic : IPC (~10%) x higher ST clocks (at least 5.5Ghz but I bet it's more) => 1.1 x (5.5/4.9)= 1.23 or 23%. See how hard you are coping with 15% number. Please go away.

 

[FangBLade]

It can be 15,3% and 35,4%

It can yes, they used that because every workload benefits differently, cpu-s are much more complex than simple cinebench IPC measurement.

 

[pakotlar]

The 12900KS is ~2% faster than 12900K in games.

Looks like you’re right. Tom’s saw 2.7% average, though they do note 6.9% 99th percentile framerates, which they suggest is because of GPU limitation on max framerates https://www.tomshardware.com/news/intel-core-i9-12900ks-cpu-review

Anyway, I think expecting ~15-20% bump over 12900K is reasonable in games due to cache boost + clockspeed, but we’ll see. In any case neither 13900K nor 7950X is particularly exciting. I’ll probably get 7950X or wait for some higher core AM5 part. My 5950X is fine and neither a 35% nor a 40% MT bump is all that interesting given that both architectures are increasing power draw (with 7950X being a much larger % power draw increase over 5950X, but 13900K drawing substantially more power overall).

 

[FangBLade]

And here is a link for our dear friend SmokSmog: https://en.wikipedia.org/wiki/Greater-than_sign

 

[SmokSmog]

Until AMD will prove more I will stick with 15%~ST CB and 35%~MT CB over stock 5950X. Lucky Zen4 is rumored to release in mid september so in two months we should know everything.

@FangBLade
I know what > < sign means

 

[FangBLade]

Until AMD will prove more I will stick with 15%~ST CB and 35%~MT CB over stock 5950X. Lucky Zen4 is rumored to release in mid september so in two months we should know everything.

Whatever makes you happy, everyone deserve happiness, even you, have a nice day.

 

[SmokSmog]

Whatever makes you happy, everyone deserve happiness, even you, have a nice day.

Funny that we know more about Raptor Lake than about Zen4.

 

[FangBLade]

Looks like you’re right. Tom’s saw 2.7% average, though they do note 6.9% 99th percentile framerates, which they suggest is because of GPU limitation on max framerates https://www.tomshardware.com/news/intel-core-i9-12900ks-cpu-review

Anyway, I think expecting ~15-20% bump over 12900K is reasonable in games due to cache boost + clockspeed, but we’ll see. In any case neither 13900K nor 7950X is particularly exciting. I’ll probably get 7950X or wait for some higher core AM5 part. My 5950X is fine and neither a 35% nor a 40% MT bump is all that interesting given that both architectures are increasing power draw (with 7950X being a much larger % power draw increase over 5950X, but 13900K drawing substantially more power overall).

RPL have more cache yes, but Zen 4 too, and every architecture have different sweet spot, 2mb l2 of RPL doesn't mean it is better than 1mb l2 of Zen 4, they are 2 very different architectures, but what is important that they both have increased cache, + Zen 4 will have 3d versions with more l3 combined with l2 so i'm curious how will that translate to gaming + also ddr5.

 

[pakotlar]

Cinebench alone doesn't tell us anything, not even for content creation workloads,even less for gaming. Facts are RPL will be small upgrade, Zen 4 will be very close in ST in cinebench like apps while consuming much less power (win for me), there will be 3dcache version for gaming that will destroy RPL if non3d fails to do so, there is AM5 longevity, while anyone with RPL will need to change platform for upgrade. I expected more from RPL.

Both are small upgrades. The noteworthy thing here is probably that Raptor Lake fixes some low hanging issues with the hybrid architecture (hopefully ring bus not being down-clocked when E cores enabled, better voltage regulation, more L2 cache to offset the poor L3 performance relative to Zen 3, more E cores hopefully allowing lower E core clocks and better PPW), and AM5 brings to AMD cpus DDR5 and PCIE5. Performance wise both are boring.