Not sure if I would bet on Raspberry Pi, because to me it seems they mostly ride on inertia and their okay-ish drivers. Looking only at hardware, the Raspberry Pis have gotten quite expensive, not that far off from the cheap and faster x86 boxes with intel’s e cores when including accessories.
There are many competitors in the SBCs space offering products that generally exceed the offerings of the Raspberry Pi. The only issue is software, often you get something like an image based on an old Debian with kernel patches. (Or you must use a 3rd party project like Armbian). But I can't imagine it's impossible that one of the usual SBC-SoC manufacturers would finally start doing proper upstream Linux kernel work.
Raspberry Pi seems really unpopular in some circles, the IPO probably doesn't help.
I'm wondering how Broadcom sees the bigger focus on commercial customers rather than education and hobbyists.
So like half the thing then? It is kinda important bit if you actually want to use the thing.
Plus in my experience many of "the hardware is ready we just need software to catch up" issues turn quickly into "uhm, oh, we have several hardware issues but hopefully the software can work around them eventually".
> But I can't imagine it's impossible that one of the usual SBC-SoC manufacturers would finally start doing proper upstream Linux kernel work.
One can always hope. After a while one might think that the reasons they are unable to grow up to the task is structural.
Either way the reasoning "somebody one day might start doing what Raspberry Pi is already doing" is not as damning as you make it sound.
Not sure if it's half the work, since they did do the work at one point. And projects like armbian do the work to provide recent versions of debian/ubuntu that work on the popular SBCs.
Raspberry Pis also need some work to be supported, Raspberry Pi OS is just maintained by Rasperry Pi and not some third party.
> Either way the reasoning "somebody one day might start doing what Raspberry Pi is already doing" is not as damning as you make it sound.
4 comments
[ 1.2 ms ] story [ 18.5 ms ] threadThere are many competitors in the SBCs space offering products that generally exceed the offerings of the Raspberry Pi. The only issue is software, often you get something like an image based on an old Debian with kernel patches. (Or you must use a 3rd party project like Armbian). But I can't imagine it's impossible that one of the usual SBC-SoC manufacturers would finally start doing proper upstream Linux kernel work.
Raspberry Pi seems really unpopular in some circles, the IPO probably doesn't help. I'm wondering how Broadcom sees the bigger focus on commercial customers rather than education and hobbyists.
So like half the thing then? It is kinda important bit if you actually want to use the thing.
Plus in my experience many of "the hardware is ready we just need software to catch up" issues turn quickly into "uhm, oh, we have several hardware issues but hopefully the software can work around them eventually".
> But I can't imagine it's impossible that one of the usual SBC-SoC manufacturers would finally start doing proper upstream Linux kernel work.
One can always hope. After a while one might think that the reasons they are unable to grow up to the task is structural.
Either way the reasoning "somebody one day might start doing what Raspberry Pi is already doing" is not as damning as you make it sound.
> Either way the reasoning "somebody one day might start doing what Raspberry Pi is already doing" is not as damning as you make it sound.
I'd say "Plenty of companies do similar things".