AMD’s Ryzen 7040U chips have proven very popular with handheld gaming PC makers this year, thanks to their blend of AMD Zen 4 CPU cores and RDNA 3 integrated graphics. But now that AMD has introduced its next-gen chips, we’re starting to see handheld gaming PC makers announce plans to launch models with Ryzen 8040U “Hawk Point” processors.

GPD plans to use the new chips for refreshed versions of its Win Mini, Win 4, and Win Max 2 handhelds, while AYA says the upcoming AYA Neo Flip handheld will be available with Ryzen 7 7840U and 8840U options. The only catch? Users probably won’t see any significant performance improvements.

That’s because the new AMD Ryzen 7 8840U and Ryzen 5 8640U chips are virtually identical to the previous-gen Ryzen 7 7840U and Ryzen 5 7640U processors in all the ways that really matter for handheld gaming.

The new 8040 series chips have the same CPU and GPU architecture, CPU and graphics frequencies, and GPU core counts as their 7040 series counterparts. And they’re still 15-30 watt chips with a default TDP of 28 watts.

So why did AMD bother launching new chips at all? First, the new chips feature an upgraded neural processing unit for up to a 60% boost in AI performance… which probably has little to no impact on gaming. And second? Bigger numbers sound better. AMD has a habit of reusing older technologies for new chips, while updating the names to reflect the year they were launched.

Honestly, it would probably make more sense for handheld makers to continue using Ryzen 7040U chips while supplies last, as they’re probably easier to obtain at the moment, while offering virtually identical performance. But moving to “next-gen” chips might make more sense from a marketing perspective. And it’s unclear whether using older chips will actually save any money, as AMD is most likely trying to convince most PC makers to migrate from Ryzen 7000 to 8000 series processors anyway.

Support Liliputing

Liliputing's primary sources of revenue are advertising and affiliate links (if you click the "Shop" button at the top of the page and buy something on Amazon, for example, we'll get a small commission).

But there are several ways you can support the site directly even if you're using an ad blocker* and hate online shopping.

Contribute to our Patreon campaign

or...

Contribute via PayPal

* If you are using an ad blocker like uBlock Origin and seeing a pop-up message at the bottom of the screen, we have a guide that may help you disable it.

Subscribe to Liliputing via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 9,543 other subscribers

Join the Conversation

5 Comments

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

  1. “AMD has a habit of reusing older technologies for new chips, while updating the names to reflect the year they were launched.”

    Common practice in the industry which presumably the author knows and surely remembers just a few weeks ago when Intel tried to make AMD look bad for doing so when they themselves are far more guilty. At least amd calls out the core generation in the name and clearly indicates it on the part pages on their website, something Intel would neglect.

    Also funny that Phoenix cores which hawk point uses beat Intel’s spanking new uarch in real world performance in more cases than not which is why Intel started their 6 hour long prior to pulling it down smear campaign.

  2. I wonder if they’ll be leveraging the improved AI for improved upscaling tech a la DLSS. If they can use it to improve graphics/res it could be out benefit, especially when connecting to an external monitor

    1. That isn’t part of rdna3, it won’t be introduced until rdna4 launches. PS5 pro os using rdna3 with the GPU AI block from rdna4 tacked on but it’s a different kind of ai. The integration of AI in hawk point is for the CPU which in theory could be used for frame gen but isn’t necessary. It wouldn’t help with ai upscaling the image. It’s really for general ai language model acceleration aka stuff like chatgpt but running locally instead of on the cloud. Eventually windows copilot will likely use that part of the CPU and technically apps can use it if coded to do so specifically but no consumer apps do as of yet that I know of.

  3. The addition of the NPU for AI acceleration is actually pretty interesting. I’ll be waiting for something like the Win Max 2 with the 8000 series APU due to the expected Oculink and the new NPU. I’ve been getting into locally run LLMs and while my ASUS G15 can run larger models very quickly, my current 11 inch AMD laptop (netbook?) struggles to run a 7b model at a snails pace and even a 1.1b model is slow (not to mention useless). I was sad to hear that the GPU component of the new 8000 series APUs are basically the same as last year now that Intel’s competitors are allegedly receiving a large graphics boost. Because of this I may wait for the 9000 series to arrive.

    1. This is kind of off topic, but what 11 inch AMD laptop do you have??? The only one I know of is the Win Max 2 which I have but it’s closer to 10 inches as well as the upcoming OneXPlayer X1 which is a tablet. I’d love to know if there are more because 11 inch laptops are my favorite size laptops. There are low end ones like the Ideapad with the lower end non-Ryzen AMD chips.