PowerPC had support from Apple, IBM and Motorola, plus it saw use in every video game console for a time period, it got a Windows port... it's still around and has its uses, but I don't know that you'd say it "succeeded." RISC-V may have support, but it's a hard market to break into.
It would be very interesting to see and Intel or AMD designed RISC-V chip. I'm curious just how the ISA could do given similar resources brought to it.
AMD should be in a better position to try because they were already planning to put ARM cores in their SoCs a few years ago. While that effort was dropped, I assume they still have a bunch of people around that know what the issues are and are ready to tackle them. Problem is ARM doesn't really have enough of those people to divert to alternatives.
AMD or Intel could beat others to the punch and make a mobile SoC with RISC-V applications cores. They'd be the first to face the pain of porting Dalvik and NDK apps, but in the end they'd be way out ahead of the competition.
I've just been assuming Google would bring Dalvik to RISC-V. They're already porting Go and some other things. If they were to release Android for RISC-V it could be devastating to ARM. Not that I think Google cares about ARM, they've been enjoying competition among SoC designers due to ARMs reasonable licensing (that's a relative term of course).
I meant the more abstract pain of introducing the new ISA and probably taking blame for problems with functionality and performance. Yes, Google would probably be the ones executing this task.
Re:Intel and the Danger of Integration
by imtringued on Tuesday June 26, 2018 @05:25AM
(#17399058)
Does Google force developers to upload an IR or bytecode for the NDK so that they can just recompile for a new architecture?
Oops. I meant AMD doesn't have enough resources...
RISC-V is a much bigger problem for ARM than it is for Intel.
Re:Intel and the Danger of Integration
by PedroBatista on Monday June 25, 2018 @10:10AM
(#17392534)
Therefore Intel getting into the RISC-V bandwagon and making it a freight train.
But I suspect they won't do that.
Nokia/Symbian vs the Open Handset Alliance/Android, all over again?
Beyond that, I think it will end up a legit heavy-duty apps processor for Android phones. It might start out on the mid/low-end phones, but manufacturers will absolutely love the idea of cutting out ARM from their BOM.