- 2 years
heads up the GPU drivers are proprietary as of posting:
The SpacemiT K1 also doesn’t have any upstream open-source graphics driver as another disappointing aspect.
- 2 years
That’s a huge deal breaker. If that’s not resolved by launch they should be ashamed to put their name behind it.
- 2 years
Don’t let perfect be the enemy of the good.
If the product doesn’t fit your needs, don’t buy it. But we’re not going to get a completely open source laptop that competes with mass market options at the same price over night.
- 2 years
That’s not really a relevant argument here. One of the massive benefits of RISC-V is the lack of proprietary instruction sets.
- 2 years
But we’re talking about the supply chain for a GPU that is compatible with this new RISC-V main board that is also good enough to compete with another laptop at the same price point (looks like it’s an IMG BXE-2-32).
That’s what I’m saying, we’re on the right path, but we’re not going to get there over night. If you want a working viable daily driver today, there are some compromises that have to be made still.
- 2 years
Are there any companies making discrete laptop graphics that don’t have proprietary drivers? I don’t think I’ve ever seen an AMD powered laptop unless it used an APU. I shudder to think of what proprietary Linux drivers from a company less resourced than Nvidia are like.
- SteveTech@programming.devEnglish2 years
I don’t think I’ve ever seen an AMD powered laptop unless it used an APU.
There’s at least 4 on AMD’s website, so they do exist but they don’t seem very common.
Also Intel has laptop chips, but I’m not sure if it’s actually discrete or just another die on the CPU.
- 2 years
Doesn’t look like it’s ready for prime time. Probably only an option for RISC-V developers.
- 2 years
Yeah I think RISC-V is probably still about 10 years away from being a sensible choice for a laptop. There’s a load of platform stuff around things like ACPI and Device Tree that’s still being decided. Also some ISA extensions that are standard on x86/ARM are either unratified or very recently ratified (e.g. Vector).
For microcontrollers it’s ready now, and for server applications it’s probably doable now and will be solid in a few years. Laptops & phones will be last though.
- 2 years
Yeah, last time I tried this it was on POWERPC. And almost every version of JRE was interpreter mode only.
This their chance to use coreboot, but knowing canonical, they probably won’t.
- 2 years
Fuck yeah, I’m really hyped about getting some solid ARM support. I want that battery life.
