Posts by toader@fosstodon.org
 (DIR) Post #B5i283GrTdy0jJWnRY by toader@fosstodon.org
       0 likes, 0 repeats
       
       @azonenberg Definitely try UTM first. Last time I tried macOS guests, tjey looked GPU accelerated, but didn’t dig in the features. Linux guests have the usual QEMU backends, some of which are GPU enabled, but might be beta-ish. Bonus points for linux guests, if you don’t require QEMU for them: you can use rosetta, so you can run x86 apps/libs also in Linux. There is also libkrun which advertises gpu accelerated containers on macOS via venus. Haven’t tried this tho.
       
 (DIR) Post #B5i4kdmcNEjEsnuUOe by toader@fosstodon.org
       0 likes, 0 repeats
       
       @azonenberg Definitely give it a try, never not a good reason to also have ARM64 support. Regarding performance, my M4 Mini is giving a lot of Ryzen boxes a run for their money in Vivado synthesis (in a Linux VM, ‘emulating’ x86 via Rosetta). The single core perf is that good and virtualization is seamless.  UTM is stable, and also has a server for managing VMs via an API.
       
 (DIR) Post #B5vSX60BaFcYIpoG7k by toader@fosstodon.org
       0 likes, 0 repeats
       
       @azonenberg @jenesuispersonne How were the shaders written before? e.g was it opencl forAmd/apple and cuda for nvidia? Or all were Vulkan and the comparison applies only to the 1/4 thread optimization?Also, is it correct that having a shader for this is “more” worth it when later stages are also on gpu, cutting memory transfers?
       
 (DIR) Post #B6WwYNSWLnSACx6jJo by toader@fosstodon.org
       0 likes, 0 repeats
       
       @azonenberg @dtl how do you keep track of all your ongoing projects? Any special tool or simple todolists/text notes?