AMD launched the first chips based on its new Zen architecture this year, and so far we’ve seen a whole bunch of options, ranging from entry-level Ryzen 3 processors to the company’s high-end Ryzen Threadripper chips with support for up to 16 CPU cores and 32 threads.

But all of the Ryzen chips launched to date have a few things in common: they’re desktop processors and they’re designed to be used with a separate graphics card, since they don’t have integrated graphics.

AMD plans to launch a line of Ryzen Mobile chips with integrated graphics aimed at laptops by the end of the year.

We got an early look at some details earlier this year, thanks to a series of leaks, and now some benchmark results give us an even better idea of what to expect.

Formerly codenamed “Raven Ridge,” Ryzen Mobile chips will be part of AMD’s APU or “Accelerated Processing Unit” family and the Ryzen 5 2500U is expected to be a 4-core/8-thread processor with AMD Radeon Vega graphics.

A listing at Geekbench seems to confirm those specs, and an image posted to reddit seems to suggest that the chip is showing up in 3DMark 11 benchmark results as well.

I’d take the actual benchmark scores with a grain of salt, since we’re talking about unreleased products. But at this point it looks like the Ryzen 5 mobile chip outperforms a Core i7-7660U Kaby Lake processor with Intel Iris Plus 640 graphics in 3D Mark 11.

It’s not easy to crown a clear winner in a Geekbench comparison: AMD’s chip seems to outperform Intel’s Core i5-7200U and Core i7-7500U chips based on 7th-gen Core technology when it comes to multi-core performance. But that’s hardly surprising, since the quad-core Ryzen 5 2500U has twice as many CPU cores as those Intel chips.

Single-core performance scores put AMD’s Raven Ridge chip on par with Intel’s Core i5-7200U.

Those aren’t Intel’s newest chips though. The new 8th-gen Kaby Lake Refresh chips including the Intel Core i5-8250U and Core i7-8550U are quad-core chips that seem to best the Raven Ridge chip in both single-core and multi-core performance, according to Geekbench.

Again, there’s no telling how accurate these test results are at this point, since it’s an unreleased chip. And the name suggests it may not be the most powerful processors in the Ryzen Mobile lineup.

A Geekbench listing from August shows that there may also be a Ryzen 7 2700U processor on the way, but there’s something a bit fishy about that page, since the chip allegedly has a higher clock speed than the Ryzen 5 2500U, but scores lower in both single and multi-CPU tests.

It’s unclear how many different models AMD plans to launch, but I suspect the Ryzen 5 2500U is just one of several: if the company wants to compete with Intel in the laptop space, it’d make sense for there to be alternatives to Intel’s Core i3, Core i5, and Core i7 chips.

Oh, and one more fun fact from the Geekbench results: It looks like someone is running the test on at least two different systems feature Ryzen 5 2500U chips:

  • A Windows 64-bit machine with a motherboard called AMD Mandolin
  • A 32-bit Android system with a board called AMD Tambourine

via Guru3D



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

9 replies on “AMD Ryzen 5 2500U benchmarks hint that Ryzen is coming to laptops soon”

  1. I would be really excited to see a model that would cater to Mini gaming PCs, and deliver performance AND cost around the level of PS4 and Xbox one.

    And when I say cost, I mean that it would fit a budget where the whole system would match that cost

    1. You’re gonna have trouble with the BluRay player.
      The PS4 console (and Xbox One) is heavily subsidised, remember that.

  2. Based on what we know about Intel’s desktop chips, and how they scale down into mobile chips…. we can actually infer how the Raven Ridge chips will behave since we can make some educated guesstimates based on Ryzen, Vega, and previous APU’s.

    We do have current offerings from Intel in the shape of:
    35W TDP, 3.4GHz 4c/8t, chips. Which draws around 8W power at Idle, around 24W mixed-load, and tops out at 34W when full-load. The performance in CineBench R15 single-core score roughly of 160cb and multicore of 705cb. And the Intel Iris HD iGPU can produce a FireStrike graphics score roughly of 1,200. Temperatures between the two should be similar, but ultimately depend immensely on the specific laptop’s design.

    AMD would do something similar in the shape of:
    35W TDP, 2.9GHz 4c/8t chips. Which draws around 10W power at Idle, around 20W mixed-load, and tops out at 30W when full-load. The performance could be around CineBench R15 single-core score roughly of 130cb and multicore of 650cb. And the refined-polished mobileVega GPU could perhaps produce a FireStrike graphics score roughly of around 3,100. Temperatures between the two should be similar, but ultimately depend immensely on the specific laptop’s design.

    We can’t be too sure about how AMD’s Vega will behave in a mobile setup.
    Will it be better than their desktop counterparts?
    Will it be more efficient than Intel’s great Iris GPUs?
    Will it be more efficient than Nvidia’s new Pascal GPUs?

    I have a hunch it will be between Intel and Nvidia in-terms of efficiency, and a lot lot better than how their desktop Vega cards have been behaving. Which is why I rated it the way I did. I used laptop graphics cards like the GTX 960M, MX 150, and GTX 1050 to get a better gauge.
    So far AMD’s Vega GPU’s are slightly above Polaris, but definitely below Pascal.
    And their AMD’s IPC is about equal to Sandy Bridge-Ivy Bridge-Haswell.
    Hence, why I put the figure where they are.
    Keep in mind that Pascal on a 25W TDP power draw gets around 3,800 points FireStrike graphics score (GT 1030). So it would be ridiculous to ask for more performance for Vega and a lower power drain. We know Pascal is the more efficient microarchitecture.

    So if you (postulates) had a laptop with an Intel Core i7-7700HQ and Nvidia MX150… I believe it should perform a little faster at both CPU and GPU functions, and use slightly less battery life….. in contrast, to a comparable laptop with AMD Ryzen and AMD Vega graphics.
    Is that a bad thing? Not necessarily.
    Look what Sony could do with the PS4, and that things a pathetic APU. Think how much better a modern APU would be with these vastly improved microarchitecture in the CPU and the GPU.

    1. And on what do you base AMD’s much lower single threaded score in CB to?
      If AMD clocks Raven Ridge at base to 3 GhZ (which they probably can and will), single threaded performance will be likely on par if not go in AMD favor.
      IPC-wise, Intel doesn’t really have any edge over AMD which isn’t accounted by architectural optimizations the industry gave them. Right now, the difference is basically 5%, which would produce a score of 152 in single threaded cinebench at same clocks… but if Raven Ridge is clocked as I said to 3.0 ghZ, or maybe above at stock (again, better manuf. process and Ryzens design will likely allow for it) will either equalize the single threaded score or be better.

      As for Vega… given how current AMD’s igp performs quite well over Intel’s igps, Vega will likely behave better… and of course there’s further optimizations that AMD will likely do, followed by dev support, etc.

      1. Generally, Ryzen has lower IPC than Intel’s Sky/Kaby Lake.
        Ryzen is more efficient at the higher clocks, but Intel is more efficient at the lower clocks.

        So I’m generalising, but I think a laptop chipset based on AMD APU would be slower and not as efficient. I would love to be proven wrong though. But so far all the leaks are actually NOT agreeing with my assessment, they’re showing Raven Ridge with LESS performance AND efficiency than I gave them credit.

        But these are early leak-days.
        It’s possible AMD hits the high-bar I set them. Its even possible they exceed it.
        I find it unlikely, but I’m always happy to be wrong. They’re likely to fall short, not that I blame them much for it… its hard to innovate when your company is constantly battling Bankruptcy and you have very little R&D margins.

  3. A small typo (missed the v out of Raven):

    “that seem to best the Raen Ridge chip in both”

Comments are closed.