Why would it? Apple hasn’t added anything hardware codec wise since M1 Pro/max except AV1 decoding - imho they should’ve added av1 encoding also - so speeds are faster but I don’t think quality would change.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
M3 Max GPU vs NVIDIA RTX 4090 Laptop
- Thread starter AnonFr13n45
- Start date
- Sort by reaction score
You might be correct but the GPU now has retracing. I’m not sure if that is used at all? I’m not sure that it purely uses the decoders.
AFAICT Apple doesn't use their GPU (directly) for encoding video when they are able to use their hardware media encoders for the task.
One thing I was surprised at (happy to be corrected if I'm wrong) is that there seems to be no performance improvements of the encode/decode engines from the M1 > M2 > M3. I know Apple said they now support 8k, but render performance seems identical. Surely these engines would scale with clock speed?
Encode/decode engines are embedded logic: data goes in one end and results come out the other. They do not process instructions but push the data through a configuration of logic gates. TBMK, there is not even any "microcode"-like action going on in there, just raw gate structures. As such, it should come as no surprise that their performance is consistent, as the units are far less dependent on clock speeds (which mostly affect dynamic code resolution logic), but should show some improvement as memory gets faster. This kind of dedicated processing is far more efficient than software, as long as it can handle contemporary standards.