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.

 

[PJVol]

What is the typical latency for the ADL and ddr5 6000 cl40 kit?

 

[itsmydamnation]

We are talking about pcs with 12900ks and high end gpus? My guess is closer to 0.

the first result from google

Dell XPS Desktop with up to 12th Gen Intel Processor

Take creativity up a notch, with the expandability, elite performance, fast speeds, improved thermals and liquid cooling options of the XPS Desktop.

www.dell.com

 

[Just Benching]

the first result from google

Dell XPS Desktop with up to 12th Gen Intel Processor

Take creativity up a notch, with the expandability, elite performance, fast speeds, improved thermals and liquid cooling options of the XPS Desktop.

www.dell.com

And what does that supposed to mean? Is the person buying a dell prebuild interested in reviews and benchmarks?

I mean what are you actually saying, that amds zen 4 testing was useless since they used 6000c30 ram?

 

[Det0x]

What is the typical latency for the ADL and ddr5 6000 cl40 kit?

Found this:



pcgameshardware.de

 

[Just Benching]

What is the typical latency for the ADL and ddr5 6000 cl40 kit?

Should be around 60ns with just xmp enabled. You can drop it down to the 50ns range with just a little bit of tunning, just the primaries and trefi + trfc

 

[Det0x]

Should be around 60ns with just xmp enabled. You can drop it down to the 50ns range with just a little bit of tunning, just the primaries and trefi + trfc

You forgot to mention you need to disable e-cores, clock ringbus higher for that to be possible.. As proof above show that you will get around ~65ns @ 6000 MT/s 40-40-40 XMP with all cores enabled and stock clocks otherwise.

 

[Just Benching]

You forgot to mention you need to disable e-cores, clock ringbus higher for that to be possible.. As proof above show that you will get around ~65ns @ 6000 MT/s 40-40-40 XMP with all cores enabled and stock clocks otherwise.

No, not really, i get 51ns with ecores on and stock cache. Only tuned primaries and trefi + trfc. Stock ram frequency as well

 

[Just Benching]

Found this:
View attachment 67233


pcgameshardware.de
View attachment 67234

Thats exactly the kit im using, but at 6000c30 34 34 58 1t,got 51ns

 

[Carfax83]

That's actually factually incorrect.

The IMC (UCLK) on Zen 4 is capable of 3000MHz+ - it runs 1:1 with the actual memory frequency (MEMCLK). It's the FCLK that craps out past 2000MHz. We actually don't know how far Raphael's memory controller can go.

As an example, AMD's test rig for their launch day used DDR5-6000, which means:

- 2000MHz FCLK (IF frequency)

- 3000MHz UCLK (memory controller frequency)

- 3000MHz MEMCLK (memory frequency)

Intel's memory controller can only do 2:1 ratio and not 1:1 with DDR5, but comparing latency with Zen 4 Alder Lake still comes out on top, let alone Raptor Lake which has a much better IMC.

I'm guessing this is because AMD's chiplet design comes with a big latency penalty vs the monolithic design in Alder/Raptor Lake.

Will be interesting to see how Zen 4 performs in 2:1 mode.

 

[uzzi38]

Intel's memory controller can only do 2:1 ratio and not 1:1 with DDR5, but comparing latency with Zen 4 Alder Lake still comes out on top, let alone Raptor Lake which has a much better IMC.

I'm guessing this is because AMD's chiplet design comes with a big latency penalty vs the monolithic design in Alder/Raptor Lake.

Will be interesting to see how Zen 4 performs in 2:1 mode.

I think if you try running higher frequency memory you'd just drop from the 3:2:2 ratio to 2:1:1 (or 4:2:2 - same thing) I don't think you'd see a whole lot of difference to latency. I don't want to say too much because we've been doing some testing, but actually it seems like our understanding of how Zen reacts to memory OCs even on Zen 3 is quite off tune. Obviously 1:1:1 is ideal, but other neat ratios seem to perform reasonably well.

 

[Det0x]

I think if you try running higher frequency memory you'd just drop from the 3:2:2 ratio to 2:1:1 (or 4:2:2 - same thing) I don't think you'd see a whole lot of difference to latency. I don't want to say too much because we've been doing some testing, but actually it seems like our understanding of how Zen reacts to memory OCs even on Zen 3 is quite off tune. Obviously 1:1:1 is ideal, but other neat ratios seem to perform reasonably well.

With Zen3 i dont actually think it was the asynced fclk:mclk that was the problem contrary to people beliefs.. It was rather that the uclk (memory controller) dropped to half the memory speed when doing so..
On Zen4 we can run mclk and uclk synced (1:1) even if are unsycned with fclk, hence the auto:1:1* setting is recommended
* = fclk:mclk:uclk

 

[Rigg]

Yes but the 12900k can actually support 7000+ memory speeds, probably even 8k with the new hynix A dies.

It looks impressive in AIDA64, GB3 multi-core memory, and Timespy. Not much else.

Spoiler

Next generation DDR5 ICs Review: SK hynix 16 Gbit A-Die - 2x 16 GB with overclocking and gaming | igor´sLAB

Today I have a glimpse into the near future of RAM technology for you, and as is so often the case in the computer hardware industry, it comes from China, where SK Hynix’s new DDR5 memory ICs are…

www.igorslab.de

 

[Carfax83]

It looks impressive in AIDA64, GB3 multi-core memory, and Timespy. Not much else.

Quite the conclusion for such a small battery of tests. There are many other games that show much larger benefits from using high speed DDR5.

 

[Rigg]

Quite the conclusion for such a small battery of tests. There are many other games that show much larger benefits from using high speed DDR5.

Great news! Can you post links to some charts? I agree its hard to draw any concrete conclusions from such a small sample size. I can't find anything else to work with though. The data I linked seems to indicate that if latency is the same or similar the bandwidth doesn't help much over 6000 MT/s. If the game even benefits from more bandwidth to begin with.

 

[nicalandia]

Another 13700K Geekbench5

Z690 Phantom Gaming 4/D5 - Geekbench

Benchmark results for a Z690 Phantom Gaming 4/D5 with an Intel Core i7-13700K processor.

browser.geekbench.com

 

[mrtastee]

..let alone Raptor Lake which has a much better IMC.

Is this just speculation, or do we know this for certain?

 

[tamz_msc]

Is this just speculation, or do we know this for certain?

Educated guess that has a fairly high chance of being true, given that official memory support goes up from 4800 MT/s to 5600 MT/s.

 

[tamz_msc]

Intel's "official" slides for Raptor Lake, courtesy Igor's Lab.

Intel Raptor Lake - Argumentation Aids from the blue Foil King for the Sale of their new CPUs | Page 10 | igor´sLAB

How does Intel actually want to sell the new Raptor Lake CPUs? Today’s sales foils, which I show you as a picture gallery, suggest that the blue foil king sees enough strong arguments to give the…

www.igorslab.de

 

[Det0x]

Intel's "official" slides for Raptor Lake, courtesy Igor's Lab.

Intel Raptor Lake - Argumentation Aids from the blue Foil King for the Sale of their new CPUs | Page 10 | igor´sLAB

How does Intel actually want to sell the new Raptor Lake CPUs? Today’s sales foils, which I show you as a picture gallery, suggest that the blue foil king sees enough strong arguments to give the…

www.igorslab.de

In case anyone is wondering if that’s all for today: No! At 2 p.m. sharp, there will be a product launch with two test samples, one of which is a hot candidate for an absolute top spot in its crawl group and the other of which at least doesn’t send cold shivers down your spine. But you’ll have to come back again, but you like to do that, I know you do.

Not quite sure i understand what he is saying, but i think igors will preview 7950x vs 13900k later today (?)

 

[uzzi38]

With Zen3 i dont actually think it was the asynced fclk:mclk that was the problem contrary to people beliefs.. It was rather that the uclk (memory controller) dropped to half the memory speed when doing so..
On Zen4 we can run mclk and uclk synced (1:1) even if are unsycned with fclk, hence the auto:1:1* setting is recommended
* = fclk:mclk:uclk
View attachment 67244

There's another issue in that usually people tested if higher FCLKs were better by just adding 33-133MHz to FCLK and testing.

Like a I said earlier, it sounds like the optimal memory configuration is always going to be whenever you can have a neat ratio like 1:1:1, 2:1:1, 3:2:2 etc.

 

[coercitiv]

Not quite sure i understand what he is saying, but i think igors will preview 7950x vs 13900k later today (?)

Based on the language we can conclude they have a 13900K, but the second CPU may simply be a lower SKU, probably the 13600K.

 

[coercitiv]

Intel's "official" slides for Raptor Lake, courtesy Igor's Lab.

So, the 13900K won't be the efficiency black swan of the group, now the spot is probably reserved for the 13700K.

 

[deathBOB]

I count 5 different boosts. Too many boosts!

 

[Zucker2k]

This probably means either he has a Zen4 chip on hand, or he's testing performance at DDR5 6000 Mhz CL30 with optimal settings on Intel chips because Zen4 is going to be reviewed with those speeds.

 

[jpiniero]

You know, the 13600K does have a lower stock boost than the 7600X. And the 13700K and 13900K only have the same boost as the 7700X and 7950X when using Turbo Boost Max, which Intel doesn't guarantee that all cores can actually hit that at stock.