4 ms·

Only speaking for myself but I suspect others might feel the same way: I can't get excited about another essentially closed platform. I'm not interested in wha

by blihp 10y ago

Only speaking for myself but I suspect others might feel the same way: I can't get excited about another essentially closed platform. I'm not interested in what are overpriced and underpowered devices that pretty much behave at best as slightly more open, yet still closed, Android-like devices. I've been living this for the last several years both with Android and ARM devel/hobbyist boards and it isn't much fun. It would be nice these quasi-business entities (i.e. Mozilla, Canonical) would forget about their brand-exclusive strategies and try to rally around an open ARM-based platform (you know, with support for things like, I don't know, GPUs and such) they can all run on. THAT I could get excited about and we could all win. Short of that, I don't get what I want and they continue to lose relevance in the mobile era. Granted, I'm part of the 1% that runs Linux. But here's a newsflash: right now we're the only ones who care about Canonical, Mozilla, etc. and ~1% beats 0%.

slgeorge 10y ago
What examples are there of an "open ARM-based platform" with support for GPU's etc? I don't know of any situation where the ARM ecosystem works like that, the companies compete for differentiation, they have a very limited interest in standardisation. Maybe I'm missing what you're referring to. It also wouldn't get you a phone, it would get you a base board but all the other bits like radios compete on the different manufacturers platforms.
- blihp 10y ago
  They're aren't, and that was my point. I think that is one of the reasons (there are others) why something like the Raspberry Pi has become so dominant. It's disastrously bad from the standpoint of openness as a platform, but they've gotten the needed technical info and binary blobs to make it work better than just about everything else out there right now on ARM. It seems like there ought to be a couple of 2nd tier GPU design houses (i.e. not nVidia or AMD) that would be motivated to cut a deal to ship on the order of at least millions of units in the relatively near term. And the larger supposed supporters of open source have funding on the order of over a billion dollars (Mozilla's last deal with Google was a billion dollars alone) that it seems like at least a fraction of could be directed to help make this happen. They don't need to build anything that competes with this or even last year's flagship devices, iterate nearly as quickly, or even play the 'we made it .00001mm thinner' industrial design game. They would just need to get something serviceable out there and the design costs could be amortized over several years with only minor revisions needed (i.e occasional bumps to CPU and memory specs etc.) I'm not suggesting that this be done purely from the standpoint of altruism: for both the IP providers and the various foundations, this would provide them with a possible path to participate in mobile that is closed to them currently so it seems like it's in their interest from an existential standpoint. That said, maybe it really is impossible to motivate any of the players necessary to go down this path. In which case it seems worthwhile for the open source crowd to go all-in on something like RISC-V/OpenGPU and jettison the proprietary IP as completely as possible, as soon as possible.

by blihp 10y ago

slgeorge 10y ago

What examples are there of an "open ARM-based platform" with support for GPU's etc? I don't know of any situation where the ARM ecosystem works like that, the companies compete for differentiation, they have a very limited interest in standardisation. Maybe I'm missing what you're referring to. It also wouldn't get you a phone, it would get you a base board but all the other bits like radios compete on the different manufacturers platforms.

blihp 10y ago
They're aren't, and that was my point. I think that is one of the reasons (there are others) why something like the Raspberry Pi has become so dominant. It's disastrously bad from the standpoint of openness as a platform, but they've gotten the needed technical info and binary blobs to make it work better than just about everything else out there right now on ARM. It seems like there ought to be a couple of 2nd tier GPU design houses (i.e. not nVidia or AMD) that would be motivated to cut a deal to ship on the order of at least millions of units in the relatively near term. And the larger supposed supporters of open source have funding on the order of over a billion dollars (Mozilla's last deal with Google was a billion dollars alone) that it seems like at least a fraction of could be directed to help make this happen. They don't need to build anything that competes with this or even last year's flagship devices, iterate nearly as quickly, or even play the 'we made it .00001mm thinner' industrial design game. They would just need to get something serviceable out there and the design costs could be amortized over several years with only minor revisions needed (i.e occasional bumps to CPU and memory specs etc.) I'm not suggesting that this be done purely from the standpoint of altruism: for both the IP providers and the various foundations, this would provide them with a possible path to participate in mobile that is closed to them currently so it seems like it's in their interest from an existential standpoint. That said, maybe it really is impossible to motivate any of the players necessary to go down this path. In which case it seems worthwhile for the open source crowd to go all-in on something like RISC-V/OpenGPU and jettison the proprietary IP as completely as possible, as soon as possible.