Stories
Slash Boxes
Comments

SoylentNews is people

posted by hubie on Monday May 08, @02:25AM   Printer-friendly
from the why?-because-you-can,-I-suppose dept.

Run your whole system from your GPU:

WTF?! Now you can bypass your hard drive and store your whole operating system in your VRAM (should you want to). Well-known Windows modder NTDEV has demonstrated how, and it's surprisingly painless.

Step one is to create a RAM drive in your GPU's memory. A VRAM drive, if you will. There's an open-source tool that can do it for you called GpuRamDrive. It only takes a couple clicks but the tool was abandoned before it reached stability, so you might need to try it a few times.

Step two is to use your pick of tools to create a virtual machine. NTDEV used Windows' baked-in Hyper-V manager, which is a simple yet powerful tool for spawning virtual machines available to Windows 10 and 11 Pro, Education, and Enterprise users. You'll need to change just a couple of defaults in Hyper-V, and you can pick those out in NTDEV's video.

If you have an RTX 4090 or RX 7900 XTX or a GPU with more than 20 GB of VRAM then you should be able to squeeze a vanilla Windows 11 installation onto the VRAM drive you created. If you didn't drop a grand on a new GPU this past year, you'll have to use an alternate operating system with less demanding storage requirements. NTDEV uses Tiny11, a stripped-down version of Windows 11 that he created.

[...] I can't think of a single reason why anyone would want to run Windows 11 on their VRAM. And yet, it's astonishingly practical: quick to set up and seemingly as fast and stable as a regular virtual machine. Why not, then, I suppose?

A YouTube video showing it running


Original Submission

This discussion was created by hubie (1068) for logged-in users only. Log in and try again!
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
(1)
  • (Score: 5, Funny) by jb on Monday May 08, @03:44AM (2 children)

    by jb (338) on Monday May 08, @03:44AM (#1305233)

    I can't think of a single reason why anyone would want to run Windows 11 on their VRAM.

    There, FTFY.

    • (Score: 3, Insightful) by takyon on Monday May 08, @04:49AM

      by takyon (881) <takyonNO@SPAMsoylentnews.org> on Monday May 08, @04:49AM (#1305236) Journal

      This is at least better than the first article I read about this.

      I assume you can't use the dGPU while this is happening, so better have an iGPU as well.

      --
      [SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
    • (Score: 2) by corey on Tuesday May 09, @12:03AM

      by corey (2202) on Tuesday May 09, @12:03AM (#1305399)

      Agree, but unfortunately I need it for work as Altium don’t make a Linux edition. I’m considering buying a MacBook M2 and running it in a virtual machine though as I need to get a new laptop.

  • (Score: 3, Funny) by Rosco P. Coltrane on Monday May 08, @05:21AM (1 child)

    by Rosco P. Coltrane (4757) on Monday May 08, @05:21AM (#1305238)

    Wow... All the excitement of virtual machines and ramdisks...

    • (Score: 2) by inertnet on Monday May 08, @08:31AM

      by inertnet (4071) Subscriber Badge on Monday May 08, @08:31AM (#1305256) Journal

      I guess the idea is to have a faster Windows experience. But, as tradition will have it, Microsoft will soon find ways to nullify that advantage.

  • (Score: 3, Funny) by Username on Monday May 08, @09:00AM (5 children)

    by Username (4557) on Monday May 08, @09:00AM (#1305260)

    I need win 11 installed, then run another win11 install through a VM? How is it faster than running off the main install?

    • (Score: 2) by kazzie on Monday May 08, @02:04PM (4 children)

      by kazzie (5309) Subscriber Badge on Monday May 08, @02:04PM (#1305288)

      Instead of just using your GPU for accelerated graphics, you can use it for accelerated EVERYTHING!!!

      • (Score: 2) by takyon on Tuesday May 09, @03:30PM (3 children)

        by takyon (881) <takyonNO@SPAMsoylentnews.org> on Tuesday May 09, @03:30PM (#1305537) Journal

        What we need are SSGs so we can run everything on the GPU with separate RAM and storage.

        https://www.anandtech.com/show/11968/now-shipping-radeon-pro-wx-9100-and-ssg-with-new-drivers [anandtech.com]

        --
        [SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
        • (Score: 2) by kazzie on Tuesday May 09, @06:33PM (2 children)

          by kazzie (5309) Subscriber Badge on Tuesday May 09, @06:33PM (#1305574)

          What I needed was an explanation of the term SSG, but you helpfully provided one. SSDs on a graphics card, whatever next?

          • (Score: 3, Informative) by takyon on Tuesday May 09, @07:08PM (1 child)

            by takyon (881) <takyonNO@SPAMsoylentnews.org> on Tuesday May 09, @07:08PM (#1305578) Journal

            3D Mega APU/SoC with CPU cores, GPU cores, accelerators, and a large amount of universal memory [wikipedia.org]. So close, yet so far.

            I don't know how well the AMD SSG caught on with businesses. Probably not so good given Nvidia's dominance. If an SSG ever came to consumers, the selling point would be that you can duplicate an entire 200+ GB game or the necessary parts of one into the SSG's storage, and there would be less latency and better performance when loading textures or other assets from the SSG's storage instead of a connected PCIe drive. Similarly, big L3/L4 cache in CPUs can increase performance.

            Getting logic and memory as close as physically possible is the endgame for computing. Or running Windows 11 on your IoT lightbulb.

            --
            [SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
            • (Score: 3, Insightful) by kazzie on Wednesday May 10, @06:20AM

              by kazzie (5309) Subscriber Badge on Wednesday May 10, @06:20AM (#1305661)

              Getting logic and memory as close as physically possible is the endgame for computing

              And preferably without a high latency link to the display (i.e. Google Stadia).

(1)