Re: Cocoa framework for GPU utilization
Re: Cocoa framework for GPU utilization
- Subject: Re: Cocoa framework for GPU utilization
- From: Gabriel Jacoby-Cooper via Cocoa-dev <email@hidden>
- Date: Wed, 21 Sep 2022 19:47:06 -0400
You can try writing a GPU program in the Metal Shading Language
<https://developer.apple.com/metal/Metal-Shading-Language-Specification.pdf>
and then dispatch that program to the GPU using the Metal
<https://developer.apple.com/metal/> framework in your Cocoa app.
Thanks,
Gabriel Jacoby-Cooper
Rensselaer Polytechnic Institute
> On Sep 21, 2022, at 4:42 PM, Carl Hoefs via Cocoa-dev
> <email@hidden> wrote:
>
> I just got a new M1 Max Mac Studio (10-core CPU / 32-core GPU / 16-core
> neural engine). How can I take advantage of the GPUs and "neural" engine from
> Cocoa? Is there a Cocoa framework or other API to access these?
>
> In my case, I have a highly parallelizable Cocoa task that uses
> dispatch_apply() to run on all CPUs simultaneously. It averages about 0.2s
> per process thread, but doesn't utilize the GPUs, only the CPUs. I'm sure
> that offloading this task to the GPUs would speed this up tremendously, but I
> don't know where to start.
>
> Thx,
> -Carl
>
> _______________________________________________
>
> Cocoa-dev mailing list (email@hidden)
>
> Please do not post admin requests or moderator comments to the list.
> Contact the moderators at cocoa-dev-admins(at)lists.apple.com
>
> Help/Unsubscribe/Update your Subscription:
>
> This email sent to email@hidden
Attachment:
smime.p7s
Description: S/MIME cryptographic signature
_______________________________________________
Cocoa-dev mailing list (email@hidden)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden