Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads

Page 613 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Tigerick

Senior member
Apr 1, 2022
790
757
106
PPT1.jpg
PPT2.jpg
PPT3.jpg



As Hot Chips 34 starting this week, Intel will unveil technical information of upcoming Meteor Lake (MTL) and Arrow Lake (ARL), new generation platform after Raptor Lake. Both MTL and ARL represent new direction which Intel will move to multiple chiplets and combine as one SoC platform.

MTL also represents new compute tile that based on Intel 4 process which is based on EUV lithography, a first from Intel. Intel expects to ship MTL mobile SoC in 2023.

ARL will come after MTL so Intel should be shipping it in 2024, that is what Intel roadmap is telling us. ARL compute tile will be manufactured by Intel 20A process, a first from Intel to use GAA transistors called RibbonFET.



LNL-MX.png

Intel Core Ultra 100 - Meteor Lake

INTEL-CORE-100-ULTRA-METEOR-LAKE-OFFCIAL-SLIDE-2.jpg

As mentioned by Tomshardware, TSMC will manufacture the I/O, SoC, and GPU tiles. That means Intel will manufacture only the CPU and Foveros tiles. (Notably, Intel calls the I/O tile an 'I/O Expander,' hence the IOE moniker.)



Clockspeed.png
 

Attachments

  • PantherLake.png
    PantherLake.png
    283.5 KB · Views: 24,025
  • LNL.png
    LNL.png
    881.8 KB · Views: 25,517
Last edited:

DZero

Golden Member
Jun 20, 2024
1,397
509
96
Intel pulled a Conroe Jump it seems and with their next in house node, seems that they are about to retake the crown. Still, since E core might be discarded, they should give the advancements to a minor maker like DM&P (taiwanese one)
 

Hitman928

Diamond Member
Apr 15, 2012
6,667
12,314
136

Results are better than the last two, but still behind overall some of those results seem strangely GPU bound at 1080p with a 4090. Hopefully the picture gets clearer with higher quality reviews.
 

tsamolotoff

Senior member
May 19, 2019
253
510
136
Gaming perf is weird
CS2 probably bugged / GPU-bound benchmark, as x3d is typically 20-30% faster than non-x3d in this game. Also, 7600x somehow faster than 7950x which shouldn't happen with substantial frequency gap (I guess gamebar is broken / disabled).
 
Last edited:

vanplayer

Member
May 9, 2024
64
113
66
They shut down E cores and gain 4% fps in gaming. WOW surprise! And still slower than 9700x (default 285k is 100% while 9700x is 107%).

Don't ask why there's no 9950x in gaming test.

For this reason, we specifically turned off E-Core in BIOS and retested the game results using 8 P-Cores. The average frame rate of 14 games can be increased by as much as 4%!

While Chadmont is so hyped then people are still forced to shut it down and only 8C8T P cores left. What a design, what a day.
 

MS_AT

Senior member
Jul 15, 2024
791
1,601
96
Lol, in CB 2024 optimised for Intel, barely better in Povray that enable AVX2 only for Intel and of course CPU Z who definitively crippled AMD back in 2017, quite some "wins".
I will be grateful if you could point out where Pov-ray enables AVX2 for Intel only. Here is the source code https://github.com/POV-Ray/povray. According to windows readme the official way to build is using Visual Studio 2015 that supports either SSE2 or AVX.
 

GTracing

Senior member
Aug 6, 2021
478
1,114
106
They shut down E cores and gain 4% fps in gaming. WOW surprise! And still slower than 9700x (default 285k is 100% while 9700x is 107%).

Don't ask why there's no 9950x in gaming test.



While Chadmont is so hyped then people are still forced to shut it down and only 8C8T P cores left. What a design, what a day.
Honestly, a 4% performance drop from enabling the e cores doesn't sound bad at all to me. Hyperthreading has a similar performance hit, and Intel choosing e cores over hyperthreading gives consumers much more multithreaded performance for the dollar.
 

SolidQ

Golden Member
Jul 13, 2023
1,468
2,398
106
CS2 probably bugged / GPU-bound benchmark, as x3d is typically 20-30% faster than non-x3d in this game. Also, 7600x somehow faster than 7950x which shouldn't happen with substantial frequency gap (I guess gamebar is broken / disabled).
That strange test, because in every test i see X3D beat easily 14900k and every other CPU.

3months ago.
2c263271426ada1e1ee9ab0dd9425bd6.jpg
 

Abwx

Lifer
Apr 2, 2011
11,856
4,834
136
I will be grateful if you could point out where Pov-ray enables AVX2 for Intel only. Here is the source code https://github.com/POV-Ray/povray. According to windows readme the official way to build is using Visual Studio 2015 that supports either SSE2 or AVX.

It was demonstrated by forum member MarkPost, he showed the difference before and after recompiling Povray to enable AVX2 for AMD, his numbers can be found somewhere in this forum.
How about wait before you declare the results rigged? Who knows, Zen 5 might win and it won’t be rigged.

It s documented that Povray use AVX2 only for Intel, and it s also documented that CPU Z crippled AMD since 2017, their excuse was that Zen was taking advantage of their bench because the 4 ALUs did execute very well some parts of the code, too well for their liking apparently, but i wont abund on this debate in this thread.
 
Last edited:

tsamolotoff

Senior member
May 19, 2019
253
510
136
3months ago.
In their case gamebar is also broken as 7950x3d should be on top if this list. I've recently tried to 'fix' it on my ages-old W10 setup and it's actually not that evident that it is broken (when it is actually broken), even reinstalling W10 on top of existing iinstallation did not fix it for me (but copying some files from the clean installation did)
 

MS_AT

Senior member
Jul 15, 2024
791
1,601
96
It was demonstrated by forum member MarkPost, he showed the difference before and after recompiling Povray to enable AVX2 for AMD, his numbers can be found somewhere in this forum.
I don't know what he compiled, but the problem with Visual Studio compiler is it doesn't have a notion of CPU architecture or vendor. Just instruction sets. I hope Mark will see this post and will join the discussion explaining what he did exactly.

Oh and for Skymont running in SSE mode is more beneficial than AVX if the SSE only binary is used to compare all CPUs. Just a detail.
it s also documented that CPU Z crippled AMD since 2017, their excuse was that Zen was taking advantage of their bench because the 4 ALUs did execute very well some parts of the code, to well for their liking apparently, but i wont abund on this debate in this thread
Well, CPU-Z does a poor job as a benchmark, so it's results should be generally ignored either for Intel or AMD.