MSI issues new Claw BIOS update, claims to outperform ROG Ally in top 100 most popular Steam games

MSI Claw
(Image credit: MSI)

MSI has been hard at work tuning its new Claw handheld gaming PC. The device maker has published another BIOS update for the Claw in under a month that improves performance by up to 30%. Performance has improved to the point where MSI states its Claw handheld is now capable of outperforming the Ryzen Z1 powered ROG Ally in most of the top 100 most popular Steam titles played right now, though we'll want to see that in some independent testing.

MSI showed off a performance graph of all the games it tested with the new BIOS update. Game selection features Hogwarts Legacy, Forza Horizon 5, Sreet Fight 6, Shadow of the Tomb Raider, NBA 2K24, FC24, Palworld, Sons of the Forest, PUBG, Dead Island 2, Manor Lord, 7 Days to Die, Helldivers 2, Fortnite, Stray, and Naraka: Bladepoint.

By far, Forza Horizon 5 saw the most noteworthy improvements from the new BIOS update, featuring a 30% performance lead over MSI's previous BIOS revision from May of this year. Other games that got huge improvements were Fortnite, Sons of the Forest, and Helldivers 2, which saw anywhere between a 28% and 20% performance boost, respectively. The remaining titles that saw double-digit percentage improvements were Hogwarts Legacy, FC24, NBA 2K24, and Manor Lord, which saw anywhere between an 11% to 15% performance improvement. The rest saw single-digit improvements anywhere between 5% and 9.6% respectively.

MSI also took a jab at Asus' ROG Ally with the Claw's new improvements. According to another slide from MSI, the MSI Claw outperforms the ROG Ally in nearly 90% of the top 100 most popular Steam games played right now. On average the MSI Claw is purportedly 26% faster on average in this Steam-based metric compared to the Asus ROG Ally. Frustratingly, MSI did not disclose which version of the ROG Ally it benchmarked (the vanilla ROG Ally or the Extreme version with the Ryzen Z1 Extreme chip).

The Claw's performance issues have been a problem since its debut. When the handheld came out several months ago, it was criticized by early reviewers for its underwhelming performance. This problem was attributed to optimization issues with Intel's Meteor Lake processors, which MSI chose to go with instead of AMD's handheld-optimized Ryzen Z1 chips. This decision did make the MSI Claw the world's first handheld gaming PC to utilize Meteor Lake processors but at the cost of immaturity issues with the new platform.

Not only was the performance terrible at launch, but power efficiency fared even worse, which is arguably the most important attribute in a handheld machine. An early review of the handheld reported that tuning the machine to 20 watts and 15 watts resulted in substantial performance losses, to the point where the Ryzen Z1 Extreme in Asus's competing ROG Ally was twice as fast as the Meteor Lake part at these lower power levels.

Thankfully, it appears the Claw's performance struggles are over, pending independent validation. The new BIOS is now available for download through the MSI website or the MSI Center M app for Claw owners to install.

Aaron Klotz
Freelance News Writer

Aaron Klotz is a freelance writer for Tom’s Hardware US, covering news topics related to computer hardware such as CPUs, and graphics cards.

  • Metal Messiah.
    Frustratingly, MSI did not disclose which version of the ROG Ally it benchmarked (the vanilla ROG Ally or the Extreme version with the Ryzen Z1 Extreme chip).

    I'm inclined that MSI has presumably compared the non-extreme vanilla Ryzen Z1 chip here.

    But if it is indeed the "Z1 Extreme" variant, then it would be a huge win for MSI's claw console, given how it has been struggling to compete with other gaming consoles, not only on the performance front, but also the current market pricing trend.

    But even the power consumption and battery time figures weren't shared though !
    Reply
  • thestryker
    While I didn't expect a whole lot from the Claw it was pretty clear on launch the firmware needed a ton of work as it wasn't even close to performing like laptops with the same chip did. I'd love to know what the reason was behind them shoving it out the door whether it was to hit a quarterly target, new hardware releases or some combination of both (though I know we'll never know). I'm very glad that they didn't just abandon it though as it should have been a decent alternative from the start hardware capability wise.
    Reply
  • SSGBryan
    I am sure it was to hit a quarterly target.
    Reply
  • cyrusfox
    Problem is you only get 1 chance to make a first impression, and after the disaster that was its outing... Good luck on changing public perception on Intel handhelds.
    JZG-WP8A_2cView: https://www.youtube.com/watch?v=JZG-WP8A_2c

    I was really looking forward to a competent handheld with Thunderbolt... Really see a good use case for it, but not sure this fits the bill. Would love to see a review with the new bios and how the battery and performance holds up. Sure wish the PC OEM's would take a note from Apple and not release it till fully polished.
    Reply
  • Amdlova
    First impression asus rma...
    Good look with your rma
    Reply
  • Notton
    are they finally up to the numbers that were teased at pre-launch?
    Reply
  • Pierce2623
    Metal Messiah. said:
    I'm inclined that MSI has presumably compared the non-extreme vanilla Ryzen Z1 chip here.

    But if it is indeed the "Z1 Extreme" variant, then it would be a huge win for MSI's claw console, given how it has been struggling to compete with other gaming consoles, not only on the performance front, but also the current market pricing trend.

    But even the power consumption and battery time figures weren't shared though !
    The percentage of the performance gains that are being reported are much too small to jump ahead of the z1 extreme. Most games would require 50-100% improvements to jump ahead of the z1 extreme.
    Reply
  • Pierce2623
    cyrusfox said:
    Problem is you only get 1 chance to make a first impression, and after the disaster that was its outing... Good luck on changing public perception on Intel handhelds.
    JZG-WP8A_2cView: https://www.youtube.com/watch?v=JZG-WP8A_2c

    I was really looking forward to a competent handheld with Thunderbolt... Really see a good use case for it, but not sure this fits the bill. Would love to see a review with the new bios and how the battery and performance holds up. Sure wish the PC OEM's would take a note from Apple and not release it till fully polished.
    Well the usb4 devices all supposedly will run a GPU dock without being TB certified if that’s why you wanted a TB port on a handheld. I don’t really see much other use for it on handheld unless you want to hook your handheld up with a TB external SSD, which seems kinda wasteful considering the price increase over USB external SSDs.
    Reply
  • Pierce2623
    thestryker said:
    While I didn't expect a whole lot from the Claw it was pretty clear on launch the firmware needed a ton of work as it wasn't even close to performing like laptops with the same chip did. I'd love to know what the reason was behind them shoving it out the door whether it was to hit a quarterly target, new hardware releases or some combination of both (though I know we'll never know). I'm very glad that they didn't just abandon it though as it should have been a decent alternative from the start hardware capability wise.
    From the testing I’ve seen on Meteor Lake, the TSMC GPU tile is fairly efficient and competitive but the compute tile made at Intel is still extremely inefficient compared to Zen4 and Zen4c.
    Reply
  • Quirkz
    Metal Messiah. said:


    But even the power consumption and battery time figures weren't shared though !
    Yeah, I'd like to see the power consumption figures - As the APU in the claw is optimised for a higher TDP.... which isn't great in a handheld.
    It's not much use if the Claw is running at 45W TDP vs the 25W Ally to get those performance improvements.
    Reply