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.

 

[mikk]

That is actually a low MT score. the 7950X should reach 38,000 stock no problem and exceed 40,000 after OC.

Almost twice as fast as a 5950x 4.9Ghz.......keep on dreaming.

 

[nicalandia]

Almost twice as fast as a 5950x 4.9Ghz.......keep on dreaming.

I ninja edited and you didn't even notice?

By the way a PBO 5950X just posted above it reaches about 21,200 points a 7950X should be able to reach 25,000 points no problem.

Micro-Star International Co., Ltd. MS-7D51 - Geekbench

Benchmark results for a Micro-Star International Co., Ltd. MS-7D51 with an AMD Ryzen 9 5950X processor.

browser.geekbench.com

 

[nicalandia]

But a Zen 2 with 4 times the threads gets beaten ? I highly doubt that real, or geekbench is trash.

Geekbench is real finicky about memory subsystems. The 3990X Was held back by the 4 Channel and it was just 13% in MT than the 3970X Which had half the cores....

But.. the 3995WX fixed that(Stomping the 13900K)

ASUS System Product Name - Geekbench

Benchmark results for an ASUS System Product Name with an AMD Ryzen Threadripper PRO 3995WX processor.

browser.geekbench.com

That's Zen2... 5995WX Zen3 just murders the very capable but out it's league 13900K

WRX80 Creator - Geekbench

Benchmark results for a WRX80 Creator with an AMD Ryzen Threadripper PRO 5995WX s processor.

browser.geekbench.com

 

[pakotlar]

I ninja edited and you didn't even notice?

By the way a PBO 5950X just posted above it reaches about 21,200 points a 7950X should be able to reach 25,000 points no problem.

View attachment 65426

Micro-Star International Co., Ltd. MS-7D51 - Geekbench

Benchmark results for a Micro-Star International Co., Ltd. MS-7D51 with an AMD Ryzen 9 5950X processor.

browser.geekbench.com

Thats a very high score. I think my PBO2 5950X gets around 17,000. Will check later.

edit: Ok, in cold boot, in a 69 degree room, with most cores in CO at -25, a D15 and PPT at 190 (to keep temps below 80) I get 16,563 and 1600 single core. This is with 64GB of DDR 3600 ram, CL18, not best, not horrible. Cinebench r23 is around 28,000.

This seems pretty typical:

https://www.reddit.com/r/Amd/comments/mom9vf

This is a normal score. Average on geekbench 5 is right around 16,500.

The 21,200 5950X score is wildly optimistic.

 

[shady28]

Not sure if this was already posted, looks like someone got ahold of a retail 13900K.

Intel Core i9-13900K Retail Raptor Lake CPU Boosts Up To 5.8 GHz At Stock, Beats 64-Core Threadripper 3990X

Intel's Core i9-13900K 24-Core Raptor Lake CPU will offer a boost clock of up to 5.8 GHz and beats a 64-core AMD Threadripper CPU.

wccftech.com

 

[Carfax83]

Not sure if this was already posted, looks like someone got ahold of a retail 13900K.

Very impressive score no doubt. I wish we knew what the temps and power consumption was though.

Also, anyone find it strange that Zen 4 is supposedly launching next month but we don't have any good Geekbench scores yet?

 

[pakotlar]

Very impressive score no doubt. I wish we knew what the temps and power consumption was though.

Also, anyone find it strange that Zen 4 is supposedly launching next month but we don't have any good Geekbench scores yet?

My guess is its not all that great. I think we’re going to see perf/watt thats only slightly in favor of the zen 4, and perf overall going to raptor lake.

 

[tamz_msc]

Also, anyone find it strange that Zen 4 is supposedly launching next month but we don't have any good Geekbench scores yet?

It was the same with Zen 3 launch. AMD is very good at clamping down on potential leaks.

 

[pakotlar]

It was the same with Zen 3 launch. AMD is very good at clamping down on potential leaks.

There were CPU-Z benchiea a month out from release https://bit-tech.net/news/tech/cpus/possible-amd-ryzen-9-5900x-benchmarks-appear-via-twitter/1/

And Geekbench 2 weeks before release: https://bit-tech.net/news/tech/cpus/early-benchmarks-suggest-strong-figures-for-amd-ryzen-9-5950x/1/

 

[AtenRa]

An editor from hardwareluxx.de is working at Intel?

Crap I mistook him for another guy , sorry my bad.

 

[mikk]

Another Geekbench improvement.

https://twitter.com/x/status/1555836202619752448

 

[igor_kavinski]

Another Geekbench improvement.

https://twitter.com/x/status/1555836202619752448

OK now if 2300 ST can be had without too much trouble, Raptor Lake starts looking attractive.

 

[nicalandia]

Not sure if this was already posted, looks like someone got ahold of a retail 13900K.

Intel Core i9-13900K Retail Raptor Lake CPU Boosts Up To 5.8 GHz At Stock, Beats 64-Core Threadripper 3990X

Intel's Core i9-13900K 24-Core Raptor Lake CPU will offer a boost clock of up to 5.8 GHz and beats a 64-core AMD Threadripper CPU.

wccftech.com

View attachment 65432

View attachment 65433

It's click bait article. The 3995WX Stomps that 13900K in MT

 

[shady28]

Another Geekbench improvement.

https://twitter.com/x/status/1555836202619752448

This is getting close to twice the single thread performance of a 10850K, and is twice the single thread performance of the average 10400 or 3600X.

This is against an overclocked 10900K on a Maximus XII motherboard, running all core 5.4Ghz. This was the best you could get 2 years ago.

 

[igor_kavinski]

This is getting close to twice the single thread performance of a 10850K, and is twice the single thread performance of the average 10400 or 3600X.

Yeah. 2300 would be more than double the ST score of my i7-5775C, so it would be a very nice, justifiable upgrade.

 

[nicalandia]

Yeah. 2300 would be more than double the ST score of my i7-5775C, so it would be a very nice, justifiable upgrade.

7800X should do 2,300 with PBO no problem.

 

[shady28]

Yeah. 2300 would be more than double the ST score of my i7-5775C, so it would be a very nice, justifiable upgrade.

What's crazy is it looks justifiable for anyone who doesn't have a high end Zen 3 or Alder Lake to me. It destroys everything else now, compounding alder lake with raptor lake the jumps over Zen 2 / Gen 10 are huge.

It kind a feels like 1995 again, when new CPUs did this crap and you had to buy a new system every couple of years.

 

[Asterox]

Cinebench/Geekbench/CPU-Z, or well they are not your everyday apps.

Hm, what do they have to do with real everyday applications or apps.One short example, "Zen 2 vs Alder Lake as comparison".

- Cinebench R23, Alder lake is 16% faster

- Autodesk Maya, Alder lake is only 4% faster

- Handbrake, Alder Lake is 7% faster

- Houdini, Alder Lake is 12% faster

- Blender Custom, Alder Lake is 9.3% faster

Page 33 - Zen 2 APUs/"Renoir" discussion thread

Page 33 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

forums.anandtech.com

What is conclusion, if someone somewhere on the internet throw some Zen 4 or Raptor Lake Cinebench test just for clicks.

 

[nicalandia]

Cinebench/Geekbench/CPU-Z, or well they are not your everyday apps.

Hm, what do they have to do with real everyday

What is conclusion, if someone somewhere on the internet throw some Zen 4 or Raptor Lake Cinebench test just for clicks.

The Neat thing is...... Zen4 will match or exceed Raptor Lake on CineBench R23 and in Geekbench5

 

[igor_kavinski]

7800X should do 23,000 with PBO no problem.

I'm not really interested in MT unless someone comes out with a game that pushes multiple threads as hard as Cinebench. If you meant 2300 ST for the 7800X, that would be awesome and would save me from Intel's space heater

 

[igor_kavinski]

It kind a feels like 1995 again, when new CPUs did this crap and you had to buy a new system every couple of years.

Thanks to Intel literally HEATING UP the IPC race

 

[nicalandia]

I'm not really interested in MT unless someone comes out with a game that pushes multiple threads as hard as Cinebench. If you meant 2300 ST for the 7800X, that would be awesome and would save me from Intel's space heater

I edited the original post. Yes, the 7800X with PBO will be able to reach 2,300 points in ST on GB5

 

[shady28]

Cinebench/Geekbench/CPU-Z, or well they are not your everyday apps.

Hm, what do they have to do with real everyday applications or apps.One short example, "Zen 2 vs Alder Lake as comparison".

- Cinebench R23, Alder lake is 16% faster

- Autodesk Maya, Alder lake is only 4% faster

- Handbrake, Alder Lake is 7% faster

- Houdini, Alder Lake is 12% faster

- Blender Custom, Alder Lake is 9.3% faster

Page 33 - Zen 2 APUs/"Renoir" discussion thread

Page 33 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

forums.anandtech.com

What is conclusion, if someone somewhere on the internet throw some Zen 4 or Raptor Lake Cinebench test just for clicks.

People on most on these sites long ago stopped looking at applications people actually use and benchmarks that were really relevant.

If people really looked at what was relevant to what they do, they'd probably just be looking at 4 areas. Web browser performance, MS Office performance, gaming performance, and overall system performance like PCMark - which also give one a good idea of how good a particular chipset / motherboard will work.

But no, cinebench and pov-ray are like the first thing up on every review. I see it as part of a cognitive dissonance that has infected the whole PC media over the last few years. Relative to those other categories, virtually no one uses cinebench or pov-ray.

I don't use any of those media apps listed, not one of them. But I do use a web browser...

 

[Zucker2k]

Cinebench/Geekbench/CPU-Z, or well they are not your everyday apps.

Hm, what do they have to do with real everyday applications or apps.One short example, "Zen 2 vs Alder Lake as comparison".

- Cinebench R23, Alder lake is 16% faster

- Autodesk Maya, Alder lake is only 4% faster

- Handbrake, Alder Lake is 7% faster

- Houdini, Alder Lake is 12% faster

- Blender Custom, Alder Lake is 9.3% faster

Page 33 - Zen 2 APUs/"Renoir" discussion thread

Page 33 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

forums.anandtech.com

What is conclusion, if someone somewhere on the internet throw some Zen 4 or Raptor Lake Cinebench test just for clicks.

Now, do Zen 2 vs Zen 3 and post the results let's see. With an average IPC of 18% by Zen 3 over Zen 2, maybe Zen 3 will beat ADL in those tests.
This is akin to ADL is only 3-5% faster than Zen 3 in games talk, and when the 5800X3D dropped with its 15% average gaming IPC over Zen 3, we found it's a virtual tie with ADL. That's even a given because the vast majority of games should run faster on ADL. It's only when you start selecting the few cache-sensitive games does the X3D flex its muscle.

 

[Carfax83]

Having less cache and using slow DDR5 memory in the low end CPUs hampers Alder Lake's performance as the cores are bandwidth starved. Alder Lake is an architecture that requires fast memory to achieve it's full potential.