Re:Apple Silicon M1 – What to Expect, Based on A14
by coldtea on Wednesday November 11, 2020 @01:16PM
(#25061024)
>Will they build a discrete GPU, or continue their AMD partnership, or try to convince everyone that their integrated graphics are good enough?
Well, what if they are? There's no law/reason integrated graphics can't be, especially with significant pipeline changes and faster/nocopy access to memory...
Re:Apple Silicon M1 – What to Expect, Based on A14
by modeless on Wednesday November 11, 2020 @04:11PM
(#25063027)
There are several reasons actually. One is that GPUs have different memory tradeoffs than CPUs and prefer higher bandwidth at the cost of latency; unified memory is always a compromise, usually in favor of the CPU. Another is that top discrete GPUs are reticle limited, which means it's not possible to make the chip any bigger, so adding other SoC parts to the same chip would necessarily reduce the space dedicated to graphics. Yet another is that separating the GPU and CPU makes it easier to effectively cool both of them.
Will they build a discrete GPU, or continue their AMD partnership, or try to convince everyone that their integrated graphics are good enough? Will they build a massive multicore chip to challenge Threadripper? Will they go to higher TDP and push single core performance even more, or will they relax and just use medium TDP parts that still outperform Intel's best?
I'm excited to learn the answers to these questions in the next year or two.