• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

AMD Ryzen AI 9 300 Posts a 20% Performance Upgrade with Both Graphics and CPU Over Previous Gen

btarunr

Editor & Senior Moderator
Staff member
Joined
Oct 9, 2007
Messages
47,668 (7.43/day)
Location
Dublin, Ireland
System Name RBMK-1000
Processor AMD Ryzen 7 5700G
Motherboard Gigabyte B550 AORUS Elite V2
Cooling DeepCool Gammax L240 V2
Memory 2x 16GB DDR4-3200
Video Card(s) Galax RTX 4070 Ti EX
Storage Samsung 990 1TB
Display(s) BenQ 1440p 60 Hz 27-inch
Case Corsair Carbide 100R
Audio Device(s) ASUS SupremeFX S1220A
Power Supply Cooler Master MWE Gold 650W
Mouse ASUS ROG Strix Impact
Keyboard Gamdias Hermes E2
Software Windows 11 Pro
The top-spec AMD Ryzen AI 9 300 series "Strix Point" processor, the Ryzen AI 9 HX 370, is expected to post a 20% performance improvement over both the CPU and integrated graphics fronts, over its predecessor, the Ryzen 9 8945HS "Hawk Point," according to leak by Golden Pig Upgrade. On the CPU front, the HX 370 packs a 12-core/24-thread CPU based on a combination of four "Zen 5" and eight "Zen 5c" cores. The single-thread performance gains on the basis of the "Zen 5" microarchitecture's generational IPC increase, besides higher clock speeds; while the multithreaded performance increases on account on more cores. This performance increase isn't linearly scaling with the 50% increase in core-count.

On "Hawk Point," all eight cores are "Zen 4," capable of boosting to high frequencies, with two of them being marked as CPPC preferred cores, capable of boosting the highest. On "Strix Point," however, only four cores are based on the "Zen 5" architecture and capable of boosting to high frequency bands; while the other eight are "Zen 5c," which don't boost as high. While the IPC of "Zen 5c" is identical to "Zen 5," the fact that it doesn't boost as high, means that the generational multithreaded performance gain from the core-count increase is expected to be closer to 20%, with Golden Pig Upgrade talking about a Cinebench R23 nT score of over 20000 points, with "Hawk Point" scoring around 16000 points.



Things get interesting with graphics. The new RDNA 3.5 iGPU on "Strix Point" packs 16 compute units (CU), compared to 12 CU on the "Hawk Point." These 16 CU work out to 1,024 stream processors, a 33% increase over the 768 stream processors of "Hawk Point," and yet there are many other factors that decide graphics performance besides CU count, and so Golden Pig Upgrade expects a 20% graphics performance improvement, which should make the new iGPU beat the Intel Arc Xe-LPG graphics of Core Ultra "Meteor Lake" processors by at least 20%. AMD, in its product announcement slide, claimed a 36% graphics performance lead over the Arc Graphics iGPU of the Core Ultra 9 185H processor.

As for the NPU, AMD has already claimed an AI inferencing performance of 50 TOPS, which goes a fair bit above the 40 TOPS required to meet Microsoft's Copilot+ AI PC program. This figure of NPU performance is needed for Windows to run local sessions of Copilot, minimizing back-and-forth from the Cloud, and improving privacy.

View at TechPowerUp Main Site | Source
 
I dont care if the cpu increases performance by 90000% and makes my breakfast. The naming scheme now has arrived at such a high level of BS where I simply refuse to buy until they are back at a comprehensive level.
 
Well if you extrapolate those iGPU numbers, in raw numbers at least, the new Arc Xe2 iGPU in the Lunar Lake chips, should be faster than this. Intel claimed a 50% uplift. It will be interesting to see if its the case.
 
I dont care if the cpu increases performance by 90000% and makes my breakfast. The naming scheme now has arrived at such a high level of BS where I simply refuse to buy until they are back at a comprehensive level.
I'm sure they'll fix it when an AMD rep sees this post so they don't lose out on the sale of one processer from you
 
20% more at the same TDP is quiet respectable, no doubt about that.
 
I dont care if the cpu increases performance by 90000% and makes my breakfast. The naming scheme now has arrived at such a high level of BS where I simply refuse to buy until they are back at a comprehensive level.
There'a processor for everyone. Ryzen 9X AI-TXT-XTX moreX-moreAI 19950X-3D+3D for me, it will trade AMD stock, forge my tax returns and indirectly make my lunch all by itself. Intel® Processor 300 for you.
 
With this stupid naming, you don't even know what the previous gen is anymore.
 
I dont care if the cpu increases performance by 90000% and makes my breakfast. The naming scheme now has arrived at such a high level of BS where I simply refuse to buy until they are back at a comprehensive level.

Tell people with weird names that they could only become your friends if they change their names to John and Mary.

Well if you extrapolate those iGPU numbers, in raw numbers at least, the new Arc Xe2 iGPU in the Lunar Lake chips, should be faster than this. Intel claimed a 50% uplift. It will be interesting to see if its the case.
It's heavily optimized TimeSpy where Intel iGPU scores proportionally higher than competition, in comparison to real life workloads.
 
Tell people with weird names that they could only become your friends if they change their names to John and Mary.
Thats exactly the reason why nicknames exist. Too many unstreamlined Names. And its sad we need them for Hardware now. Otherwise you need a codewheel to decipher what a product actually resembles.
 
Things get interesting with graphics.
It'd be good to clarify to readers that the posted score in TimeSpy is desktop iGPU 780M on 8700G, the chip that consumers more power. Mobile iGPU 780M graphics scores are on average lower, around ~3060.

Thats exactly the reason why nicknames exist. Too many unstreamlined Names. And its sad we need them for Hardware now. Otherwise you need a codewheel to decipher what a product actually resembles.
Yes, it's called Strix. That's all you need to know.
If you want the top model, look for the number 370, or look into spec that reads 12 cores.
So, you have several hints: Ryzen 9, Strix, 370 or 12 cores.
What else do you need to be happy?
 
They've already explained that people hated the older naming, so they changed it.

Right now it looks like revenge, tho.
People hated the naming because instead of product generations, the numbers meant release years. So if they released a Zen 3 CPU a year after a Zen 4 one, it would bear a higher number. And now, the new name means absolutely nothing, but has "AI" in it to make investors happy, I guess. They effectively turned bad to worse.
 
CPU names haven't made sense for at least a decade, but as long as you can search whatever name they give them and get the right result, then whatever. Honestly, it was about time for renaming anyway since they were about to clear the 10k mark. I didn't care for AMD's old mobile approach, which didn't follow the desktop SKUs at all, but they did that in order to sell various generations at the same time but make them all look current.
 
With this stupid naming, you don't even know what the previous gen is anymore.
For a company that just loves to mix several generations (CPUs or GPUs) under the same naming, that would be "mission accomplished", wouldn't it? ;)
 
The naming is becoming as confusing as Intel's with their Core Ultras.
 
The naming is becoming as confusing as Intel's with their Core Ultras.
It's not doing us any favor, but let's be honest: those the could see past the name on the box will continue to do so, the others will keep buying the higher number, as usual.
 
It's not doing us any favor, but let's be honest: those the could see past the name on the box will continue to do so, the others will keep buying the higher number, as usual.
Yeah, true. The same goes for GPUs when there are so many SKUs with similar naming etc.
 
  • Like
Reactions: bug
They should just name all their CPUs...Bob.

Slow Bob
Medium Bob
Fast Bob
Faster Bob
That's how I read naming schemes anyway.
Ever since we've gone multicore, any number on a box became useless. You really need to understand how the CPU is built in order to determine if it fits your needs.
Do I need more cores? Do I need faster cores? Do I need more cache? Do I need more memory channels? No naming scheme will help you to answer that.
 
So Time Spy CPU, 13.5k which is about 15% better than the current 8945HS which is around 11.8K, which my current 2020 laptop (4900HS) is around 9.4k which is about a 44% increase... so yea, it might be worth upgrading at some point...
 
People hated the naming because instead of product generations, the numbers meant release years. So if they released a Zen 3 CPU a year after a Zen 4 one, it would bear a higher number. And now, the new name means absolutely nothing, but has "AI" in it to make investors happy, I guess. They effectively turned bad to worse.
I don't even know what CPU this is supposed to be. Clearly I'm not their market, but I have no idea what CPU this is and what CPU was it's predecessor.
 
I don't even know what CPU this is supposed to be. Clearly I'm not their market, but I have no idea what CPU this is and what CPU was it's predecessor.
Someone said that it starts with 3 because it's the third (don't debate me, I'm the messenger) generation with an NPU. (And yeah, AMD want's to be ahead of Intel's upcoming 200)

It's the first APU with 12 cores, so it's quite a step up from 4000/5000/6000/7040(8000) which has 8. Graphics is now 16 CU instead of 12.

There are only two models, 365 and 370, with 10 or 12 cores, 4 regular Zen5 cores and the rest Zen5C cores.
 
It's not doing us any favor, but let's be honest: those the could see past the name on the box will continue to do so, the others will keep buying the higher number, as usual.
If that's the case I can't wait for the many lost sales after they fell back to 100's for these CPUs...

I mean, I'll take one that's over 9000 directly! But this?

They should just name all their CPUs...Bob.

Slow Bob
Medium Bob
Fast Bob
Faster Bob
Next gen: Hank.

Oh yeah I want this. If you think of it, Nvidia's already kinda doing it. Ampere, Ada...
 
Well if you extrapolate those iGPU numbers, in raw numbers at least, the new Arc Xe2 iGPU in the Lunar Lake chips, should be faster than this. Intel claimed a 50% uplift. It will be interesting to see if its the case.
They claimed it was 50% faster than a "ULV" meteor lake chip with 4CU/XEs iGPU (Half of the 155/185H model).
 
Back
Top