Whoa. Big miscalculation.
this post was submitted on 20 Mar 2025
25 points (100.0% liked)
Linux
6600 readers
293 users here now
A community for everything relating to the GNU/Linux operating system
Also check out:
Original icon base courtesy of lewing@isc.tamu.edu and The GIMP
founded 2 years ago
MODERATORS
Maybe a big miscalculation to adopt it. The lead developer lists many important reasons for dropping it, including:
- None of their users actually care for it
- Lead dev does not have the time for it
- Chimera's innovation is not hardware side
- Risc hardware is very slow
You're probably right. But, point by point
- RISC-V is still in early adopter phase, and probably in, or just starting to come out of, the chasm. Organizations that get in at the ground floor have an advantage in the technology later on, but
- Early adoption requires resources to survive the early dearth of users. It's a bet on the future, and you only place bets if you have the capital. RISC-V is a sure bet, but it could be 100% guaranteed and you still may not be able to afford to buy in, which it sounds like is happening here.
- This is a non-point.
- For now. RISC-V is still extremely young technology.
It was probably the right decision for Chimera. We all have to prioritize. I love RISC architecture, ever since college and CPU architecture classes, and back when I was still writing Assembly on occasion. And, yet, I still don't own a RISC-V computer. I'll probably be in the "early majority" group somewhere.
I don't believe it's a miscalculation to adopt it if you can afford to. Early adopters have later advantages, in experience and mindshare. This may come back to bite them; maybe not.
I will say this, though: although I'm not the target and I really don't understand the appeal of mobile gaming, it's a space where RISC-V really could be an overnight game-changer. Its extensibility and configurability, and low cost of entry for CPU makers, makes it an ideal target for a submarine hardware developer who's focusing on gaming, and releases a core that outperforms ARM in graphics and AI tasks. It makes it a great platform form one of Apple's notorious architecture shifts - they have enough money to develop an in-house powerhouse chip, and they've built bespoke CPUs before. If that happens, this decision will, in retrospect, look like a terrible mistake; Chimera will be a late majority adopter scrambling to add support. No-one who's buying The Thing (whatever it is) will be thinking "Chimera!" They'll be just yet another distribution in a sea of late majority adopters.
I mean, yeah probably makes sense at this point. Nobody is seriously running desktop Linux on RISC-V.
On the other hand nobody is seriously running desktop Linux on PowerPC either. Why do they support that?
Probably going to be 5 years or so until RISC-V gets to the point that you might actually want to use it for "normal" stuff.
In other news: a sack of rice fell over in China.