AMD Expands Copilot+ Capable Ryzen AI 300 Series, Debuts Ryzen 200 Series Mainstream Mobile Processors

[btarunr]

AMD today vastly fleshed out its mobile processor lineup with the introduction of two new processor lines besides the Ryzen AI Max 300 series. This includes the introduction of more processor models in the Ryzen AI 300 series that are powered by the "Strix Point" silicon, and the introduction of the Ryzen 200 series mobile processors, which are based on the older "Hawk Point" silicon. In 2024, AMD had debuted the Ryzen AI 300 series "Strix Point," but with just the top-end Ryzen AI 9 370 and 365, which came with maxed out 12-core/24-thread (4x Zen 5 + 8x Zen 5c) core configuration, and a maxed out iGPU with 16 CU. Today the company is introducing the Ryzen AI 7 350, the Ryzen AI 5 340, and their AMD PRO variants for commercial notebooks. Both the consumer and commercial parts have identical specs, except for the latter featuring the AMD PRO feature-set.

The Ryzen AI 7 350 comes with a CPU configuration of 8-core/16-thread (4x Zen 5 + 4x Zen 5c). All cores have a base frequency of 2.00 GHz, the Zen 5 cores boost up to 5.00 GHz. The iGPU on offer is the Radeon 860M, with 12 CU and an engine clock of up to 3.00 GHz. TDP is configurable between 15 W to 55 W. The Ryzen AI 5 340 comes with a 6-core/12-thread configuration (3x Zen 5 + 3x Zen 5c), and CPU clock speeds of 2.00 GHz base with 4.80 GHz boost achievable on the Zen 5 cores. The iGPU is heavily cut down, with just 4 CU available, and an iGPU engine clock of 2.90 GHz. Notebook designers can configure this chip with a wide power range from 15 W to 55 W. All four processor models mentioned above come with a Ryzen AI XDNA 2 NPU that's capable of 50 AI TOPS, which means they're all Microsoft Copilot+ AI PC logo eligible.



Next up, AMD launched at least 11 Ryzen 200 series processor models. All these chips are based on the older-generation "Hawk Point" silicon that features "Zen 4" CPU cores, and a Radeon 700M series iGPU based on the older RDNA 3 graphics architecture, with up to 12 CU. Some models come with the NPU enabled, but this is the older Ryzen AI XDNA NPU with 16 TOPS of performance, which is enough for lightweight AI acceleration, but doesn't meet Copilot+ specs.



View at TechPowerUp Main Site

 

[R0H1T]

3x Zen 5 + 3x Zen 5c

Don't remember the last time I saw tri core (clusters?) in an x86 chip, Phenom being the last one of course

 

[csendesmark]

Don't remember the last time I saw tri core (clusters?) in an x86 chip, Phenom being the last one of course

Where you had a good chance to unlock it to a full 4 core CPU

 

[SL2]

The Ryzen AI 7 350 comes with a CPU configuration of 8-core/16-thread (4x Zen 5 + 4x Zen 5c). All cores have a base frequency of 2.00 GHz, the Zen 5 cores boost up to 5.00 GHz. The iGPU on offer is the Radeon 860M, with 12 CU and an engine clock of up to 3.00 GHz.

No, the 350 has fewer CU's than Phoenix and even Rembrandt, which in turn makes me doubt that anything in the 200 series have 12CU..

https://www.amd.com/en/products/processors/laptop/ryzen.html#tabs-1181ea0b44-item-6ccfea5f65-tab

The only APU's presented today that are faster than the two year old Phoenix is the Strix halo. The rest is just resegmentation and rebadging IMO.

 

[R0H1T]

Where you had a good chance to unlock it to a full 4 core CPU

And 4->6 cores with Zosma. Also didn't dual cores have the ability to be unlocked to full quads?

 

[csendesmark]

And 4->6 with Zosma. Also didn't dual cores also have the ability to be unlocked to full quads?

Well actually I never did it, I remember selling X3-s to customers where the guy bough a dozen of those just to check which are 4-core able,
During that time I use the good old Core 2 Duo E6300, then the original i7

 

[sudothelinuxwizard]

I think this will be more of the same as Strix Point. Small CPU performance uplift, if at all, and a blatant downgrade on iGPU (R7 has been downgraded to x60M tier and R5 to x40M tier) but much improved battery life and thermals.


It is sad to see them waste die space on a NPU to bootlick MS, but overall it should be a good product for the target audience IMO.

 

[Carillon]

Don't remember the last time I saw tri core (clusters?) in an x86 chip, Phenom being the last one of course

zen to zen2 had the 6 core variants with cores in clusters of 3

sad this is the fourth rebrand of phoenix

 

[theouto]

I am still baffled at them comparing "AI" to the industrial revolution, or electricity.

 

[R0H1T]

zen to zen2 had the 6 core variants with cores in clusters of 3

sad this is the fourth rebrand of phoenix

Pretty sure that's wrong, they were 4+2 core cofigs. I'm assuming you're talking about xx60 (desktop) chips?

 

[SL2]

I think this will be more of the same as Strix Point. Small CPU performance uplift, if at all, and a blatant downgrade on iGPU

Uplift?

I have a really hard time understanding how 4 x 5 GHz Zen5 + 4 x 3.5 GHz Zen5C could be faster or even equal to 8 x 5.1 GHz Zen4.
4.25 Ghz average against 5.1 GHz. I dunno if the insane Zen5 improvements can help that much.
Well unless I'm missing something here.

As for graphics we already know the answer, just cut the Strix point benchmarks in half (only 100 MHz difference). Yeah that's way less than Phoenix..

sad this is the fourth rebrand of phoenix

That's way too generous lol

More like Hawk point 2 part 2, given the clow (pronounced "slow") cores.

 

[Carillon]

Pretty sure that's wrong, they were 4+2 core cofigs. I'm assuming you're talking about xx60 (desktop) chips?

yes the 600 desktop parts, but i believe the mobile hexacores like renoir to be the same.

Well unless I'm missing something here.

the power limits should keep the performance on similar levels

 

[sudothelinuxwizard]

Uplift?

I have a really hard time understanding how 4 x 5 GHz Zen5 + 4 x 3.5 GHz Zen5C could be faster or even equal to 8 x 5.1 GHz Zen4.
4.25 Ghz average against 5.1 GHz. I dunno if the insane Zen5 improvements can help that

According to some quick napkin math, before accounting for the node improvement + other factors, it is 2.5 percent slower then 8840U if we believe AMDs claim of 16% IPC uplift. So its probably a wash in my opinion. I do predict this will struggle against higher TDP Hawk Point. And of course the ST performance will be much improved.

Regarding the GPU, it will be a downgrade, I never said anything to the contrary.

 

[Visible Noise]

Their new brand: AMD Rebadgathon

 

[kondamin]

It’s been like a year or something of x86 npu blocks has anything remotely interesting been done with them?