Rezultati pretraživanja
  1. 1. velj
    Odgovor korisniku/ci

    I'm not sure I have the time, still finishing up v2, and managing ... But I can assist where documentation is lacking if you choose to take it on.

  2. . Find all prime numbers less than 10,000,000 in just 0.012 seconds using plain old in your web browser:

  3. 21. sij

    v2.6.3 out just now fixes modulo performance, accuracy on , and CPU type inference. Shout out to for helping us!

  4. 1. stu 2019.

    A shout out to for letting us use their massive servers to test . Right now we're using a 8 (yes, that is right 8!) x 2080 Ti w/ 384 GB RAM on a Xeon Gold 3ghz Way overkill, but it is what they could spare ;)

  5. 30. lis 2019.

    v2.2.0 released yesterday! With it come the ability to use code coverage with GPU code. Also a bit better random ty to help from

    Prikaži ovu nit
  6. 5. lis 2019.

    GPU.js is a JavaScript Acceleration library for GPGPU in JavaScript for Web and Node.

  7. 20. lip 2019.

    Writing test case to compare script performance when using gpu.js with the JavaScript only. Hopefully the overhead of using gpu.js is not bigger than its benefits.

  8. 24. tra 2019.

    RT : RT : v2.0.0 rc8 out now - ! Strongly typed end to end inference, highly tuned index lookups, getting very stable, fresh docs, and support! Making be blistering fast for ,

  9. This JavaScript GPU acceleration library seems really nifty. Should be very useful for on-device tasks that could benefit greatly from GPU acceleration. Has anyone played around with it yet?

  10. 18. ruj 2018.

    Nice! I'd love to see more examples on how to simply render features on a leaflet map using

  11. 3. ruj 2018.

    v1.7.0 "Immortal" released! Some nice performance fixes and added support for constants of various types.

  12. 12. lip 2018.
  13. 3. lip 2018.
  14. 9. ožu 2018.

    RT : Today gpu.js 1.2.0 "Irukandji" is released with WebGL 2 support HUzzah!

  15. 23. pro 2017.

    RT : On GPU.js v1 here is what you can expect from texture mode in firefox on an i7 OSX MacBook Pro: ..."266.45 times faster!" than cpu. We've gone to plaid.

  16. 26. srp 2017.

    GPU.JS - GPU Accelerated

  17. 24. srp 2017.
  18. 22. srp 2017.

    machine learning is overcomplicated. It's all just simple math. Let's change this.

  19. 20. srp 2017.

    is an interesting concept:

  20. 12. srp 2017.
    Odgovor korisnicima

    Your right. There is some overhead for the CPU & GPU side, texture mode / multi-kernel. 😂 My job to optimize our messy code haha.

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.