Register now to gain access to all of our features. Once registered and logged in, you will be able to create topics, post replies to existing threads, give reputation to your fellow members, get your own private messenger, post status updates, manage your profile and so much more. This message will be removed once you have signed in. Login to AccountCreate an Account
Graphics memory architecture and Apple Silicon (M1, etc)
The Apple M1 chip seems to have ridiculously good performance, especially for the power usage. My understanding is that a part of this is due to the unified graphics memory model, so GPGPU and graphics don’t require the same copying.
I’d assume classic integrated graphics, with some RAM devoted to graphics memory, is faster to copy than using PCI-e, while being held back by that copy and relatively slow system RAM.
Are any other hardware developers working on unified memory models? Do any Android phones have similar models? And how much of the split memory model is an assumption of most graphics APIs?
Sign In
Create Account
Back to top
Report
