Earlier quoted context omitted.
Let's be honest here, the something is 99% drivers. It was printers, then it was wifi, now it's gpus. DNNs and LLMs at least made the compute part of the gpu driver Linux-native (from the vendors' perspective), one would hope the graphics part of the graphics processor would come next, but alas making big iron do work is done from macos and windows laptops.
Weird I haven't had a graphics issue since like the 8000 series of cards for like, the normal driving graphics right off the card use case. Optimus is definitely a little fucky with multiple GPUs and sometimes new CUDA versions break libs if you're keeping it bleeding edge updated, but most ML workflows use environment management to get around the clusterfuck that is the python ML ANN ecosystem anyway. Nvidia does a…
- gaming: Windows with a sprinkle of Steam Deck (yay!)
- compute: Linux
- video decode: Windows (DRM on Linux... yeah)
- video encode: mixed (Windows for consumers and smbs, Linux for cloud transcode)
Of these, the Steam Deck is the closest to a laptop that we got to. This is actually great, but it didn't result in a noticeable shift in anything, because LLMs took it all.