News AMD's upcoming Ryzen AI 9 HX 370 beats the company's current best mobile chip – Strix Point ES Geekbench results show big improvements

D

Deleted member 2731765

Guest
Here is another GB6 entry, ASUS TUF Gaming A14 laptop.

AMD-Ryzen-AI-9-HX-370-Zen-5-Strix-Point-APU-Benchmark-Leak-_2-964x1456.png



Today's test scores let the processor fully loose, with two chips hitting 4.5 and 4.8 GHz.

But if you check the log file, the chip reached a maximum core clock close to 5.0 GHz, on one of the benchmarks.
 
  • Like
Reactions: prtskg
D

Deleted member 2731765

Guest
Geekbench GPU results also popped up today for the integrated graphics within the Ryzen AI 9 HX 370 APU, though the Radeon 880M is named rather than the 890M which is confirmed to be in the HX 370.

We would assume that the integrated graphics were listed incorrectly in the Geekbench tests, as the HX 370 has the Radeon 890M inside, but engineering sample weirdness or some other problem could be to blame. The RDNA 3.5-based Radeon 890M GPU has 16 compute units, which allows it to soundly beat its predecessor, the Radeon 780M.

Please check the entry again, the SKU name is shown with the older nomenclature i.e., "AMD Ryzen AI 9 HX 170", before AMD made the last minute change.

There is no weirdness here, LOL.

AMD also made a last minute change to the iGPU as well, that's why benchmark reads it as the 880M instead. And, the 880M igpu was also earlier dubbed as "870M" instead.

The benchmark does NOT read it as the new Ryzen AI 9 HX 370, hence the 880M shows up there.

A last minute change was made by AMD, and also at Computex most of the companies were showcasing prototype Laptops running early samples of this APU. Even the Ryzen AI 9 365 was previously codenamed as "Ryzen AI 9 165" sporting the 870M igpu.

gSTX01f.jpeg
 
Last edited by a moderator:

TechyIT223

Prominent
BANNED
Jun 30, 2023
277
66
760
It's kind of odd to also change the igpu number as well, cause this makes no sense.

For the APU side it obviously makes sense to go for the 300 series nomenclature to stay ahead of Intel's Arrow Lake series. But integrated graphics model number should have remained the same IMO.
 

KnightShadey

Reputable
Sep 16, 2020
147
88
4,670
It's kind of odd to also change the igpu number as well, cause this makes no sense.

For the APU side it obviously makes sense to go for the 300 series nomenclature to stay ahead of Intel's Arrow Lake series. But integrated graphics model number should have remained the same IMO.

Except for intel's iGPU naming scheme is also similar so AMD R780M > UHD770 , but the only naming I've seen for Arc 8 core is a longer 3 grouping Identifier so not sure if they were aiming for R880/890M being ahead of UHD880 (as the past would indicate) or just boost8ng the name +1+1+0? 🤷🏻‍♂️

It does make me wonder if they change the naming strategy for StrixHalo which looks like an enormous leap, does it become R899M or R9??M or do they start another category for a mid-desktop level performance ; something ala Radeon 9700 pro refresh naming to R x800, do the same to mobile to something like Radeon X???M or does that confuse it with older mobile parts like the X700M etc ? 🤔
 
  • Like
Reactions: TechyIT223

KnightShadey

Reputable
Sep 16, 2020
147
88
4,670
With Strix Halo they'll have to use a different nomenclature for the igpu to avoid any confusion.
I would hope so, but as we're seen from AMD, intel, and Micro$oft's naming teams... there's lotsa people there that aren't very good at their jobs, so they could easily mess this up too, just like the Pro confusion above.
There seems to be obvious options, but all the better to bungle it. 🤔🤣
 
It's kind of odd to also change the igpu number as well, cause this makes no sense.

For the APU side it obviously makes sense to go for the 300 series nomenclature to stay ahead of Intel's Arrow Lake series. But integrated graphics model number should have remained the same IMO.
I think the logic (if there is any) is that the 680M/780M were both 12CU so now the 880M is also 12CU and we now have an 890M at 16CU. The naming honestly is just ridiculous and all a game as there was really no reason to shift any of it APU or IGP.
 

TechyIT223

Prominent
BANNED
Jun 30, 2023
277
66
760
I would hope so, but as we're seen from AMD, intel, and Micro$oft's naming teams... there's lotsa people there that aren't very good at their jobs, so they could easily mess this up too, just like the Pro confusion above.
There seems to be obvious options, but all the better to bungle it. 🤔🤣

LoL. Fire them from their jobs 😁
 
  • Like
Reactions: KnightShadey
D

Deleted member 2731765

Guest
And even 870M igpu was sporting 12 CUs, so there is some logic to rename it as the "880M" as well.


An early engineering sample of Ryzen AI 9 365/165 was spotted running within the Acer Swift SF14-61 laptop. Chip scored 2730 points in single-core and 13,032 points in the multi-core tests.

It peaked out at 4414 MHz which is much lower than its official 5.0 GHz boost clock.

AMD-Ryzen-AI-7-365-Strix-Point-APU-Benchmark-Leak.png
 
Last edited by a moderator:

darcotech

Distinguished
Oct 26, 2004
34
1
18,535
Regarding iGPU Vulkan score, if my calculations are right, expected score should be somewhere 48K+, as it loses its score for basically failed particle physics test (it is presented as the last one).
See comparison versus older AMD 780m:

https://browser.geekbench.com/v6/compute/compare/2374984?baseline=2381138

If that one was the same as 780m, so not better, result would be 48840 from average of other tests improvements. That puts it close to nVidia 3050 laptop and nvidia 1070. Not bad for an iGPU.
 
  • Like
Reactions: KnightShadey

KnightShadey

Reputable
Sep 16, 2020
147
88
4,670
Regarding iGPU Vulkan score, if my calculations are right, expected score should be somewhere 48K+...
...That puts it close to nVidia 3050 laptop and nvidia 1070. Not bad for an iGPU.
Nice. 🤙

Also nice to see a 64GB configuration innthe mix making them not seem like true unicorns😎