5 ms·

Arm on windows may date to 2011, but it was mostly a side project with 1-2 maintainers. With sufficient investment, it shouldn’t take 13 years to build up RISC-

Arm on windows may date to 2011, but it was mostly a side project with 1-2 maintainers. With sufficient investment, it shouldn’t take 13 years to build up RISC-V support.

snvzz 2y ago
This. It is evident to anybody paying attention that Microsoft has RISC-V support well underway. But even if they had to start from scratch, it would be much easier, thanks to ARM having paved the way.
- dagmx 2y ago
  How do you imagine ARM having helped?
  - monocasa 2y ago
    The x86 emulator for one.
    - dagmx 2y ago
      Fair, though I don’t think translation is a good long term strategy. You need native apps otherwise you’re always dealing with a ~20-30% disadvantage. The competition isn’t sitting still either and QC already hit this with Intel stealing their thunder with Lunar Lake. They’re efficient enough that the difference in efficiency is far overshadowed by their compatibility story. Ecosystem support will always go to the incumbent and this would place RISC-V third behind x86 and ARM. macOS did this right by saying there’s only one true way forward. It forces adoption.
      - snvzz 2y ago
        >You need native apps For native apps, you need users. For users, you need emulation. It cannot be overstated how important successful x86 emulation is for the migration to anything else to be feasible.
        dagmx 2y ago
        I think you just ignored the rest of my comment though which specifically addresses why I don’t think just relying on translation is an effective strategy. Users aren’t going to switch to a platform that has lower compatibility when the incumbent has almost as good efficiency and performance.
        snvzz 2y ago
        >when the incumbent has almost as good efficiency and performance. The incumbent is the only two companies -Intel and AMD- that can make x86 hardware. The alternative is the rest of the industry. Thus having a migration path should be plenty on its own. Intel and AMD can both join by making RISC-V or ARM hardware themselves. My take is that they will too, eventually, come around. Or they'll just disappear from relevance.
        dagmx 2y ago
        The incumbent is not just x86 but now ARM as well. You have to think in network effects. You mention “the rest of the industry” yet ignore that it’s mostly arm , which would make arm the incumbent. x86 is the king for windows. But ARM has massive inroads with mobile, and now desktop with macOS, and servers with Amazon/Nvidia etc There’s a lot better incentive to support ARM than RISC-V for software developers. It isn’t one or the other , but it is a question of resources. Intel and AMD seem fine turning x86 around when threatened as can be seen by Lunar Lake and Stryx Point. Both have been good enough to steal QC’s thunder. You don’t think ARM manufacturers will do the same to RISC-V? TBH most of your arguments for RISC-V adoption seem to start from the position that it’s inevitable AND that competing platforms won’t also improve.
- MBCook 2y ago
  Like everything else, it doesn’t matter much. Windows ran on Itanium, Alpha, and as pointed out ARM for over a decade. Without the ISVs, it’s a flop for consumers. MS has had an abysmal time getting them to join in on ARM, only starting to have a little success now. Saying “Ha ha, just kidding, it’s RISC-V now” would be a disaster. That’s the kind of rug pull that helped kill Windows Mobile. Emulators aren’t good enough. They’re a stop gap. Unless the new chip is so much better than the old it’s faster with emulation then the old one was native no one will accept it long. Apple’s been there, but that’s not where MS sits today. And if your emulator is too good, what stops ISVs from saying “you did it for us, we don’t have to care”? So once again they don’t have to do it at all and you have no native software. MS can’t drop their ARM push unless they want to drop all non-x86 initiatives for a long time.
  - snvzz 2y ago
    >And if your emulator is too good, what stops ISVs from saying “you did it for us, we don’t have to care”? So once again they don’t have to do it at all and you have no native software. x86 emulation enables adoption. Adoption means having an user base. Having an user base means developers will consider making the platform a target. >Saying “Ha ha, just kidding, it’s RISC-V now” would be a disaster. Would it now? If anything, offering RISC-V support as well would further reinforce the idea that Windows is ISA-independent, and not tied to x86 anymore.
    - MBCook 2y ago
      > If anything, offering RISC-V support as well would further reinforce the idea that Windows is ISA-independent, and not tied to x86 anymore. Anymore? It’s been independent since the 90s. It’s only ISVs that have been an issue. And a rug pull is a fantastic way to scare all the ISVs far far away.
      - snvzz 2y ago
        Whose's rug would even be pulled?
      - 6SixTy 2y ago
dagmx 2y ago
So which of Microsoft’s false starts would you take as them taking ARM seriously? Why do you think they’d take RISC-V any more seriously than their previous attempts at ARM? There are two fallacies to overcome here.

Arm on windows may date to 2011, but it was mostly a side project with 1-2 maintainers. With sufficient investment, it shouldn’t take 13 years to build up RISC-V support.

snvzz 2y ago

This. It is evident to anybody paying attention that Microsoft has RISC-V support well underway. But even if they had to start from scratch, it would be much easier, thanks to ARM having paved the way.

dagmx 2y ago
How do you imagine ARM having helped?
- monocasa 2y ago
  The x86 emulator for one.
  - dagmx 2y ago
    Fair, though I don’t think translation is a good long term strategy. You need native apps otherwise you’re always dealing with a ~20-30% disadvantage. The competition isn’t sitting still either and QC already hit this with Intel stealing their thunder with Lunar Lake. They’re efficient enough that the difference in efficiency is far overshadowed by their compatibility story. Ecosystem support will always go to the incumbent and this would place RISC-V third behind x86 and ARM. macOS did this right by saying there’s only one true way forward. It forces adoption.
    - snvzz 2y ago
      >You need native apps For native apps, you need users. For users, you need emulation. It cannot be overstated how important successful x86 emulation is for the migration to anything else to be feasible.
      - dagmx 2y ago
        I think you just ignored the rest of my comment though which specifically addresses why I don’t think just relying on translation is an effective strategy. Users aren’t going to switch to a platform that has lower compatibility when the incumbent has almost as good efficiency and performance.
        snvzz 2y ago
        >when the incumbent has almost as good efficiency and performance. The incumbent is the only two companies -Intel and AMD- that can make x86 hardware. The alternative is the rest of the industry. Thus having a migration path should be plenty on its own. Intel and AMD can both join by making RISC-V or ARM hardware themselves. My take is that they will too, eventually, come around. Or they'll just disappear from relevance.
        dagmx 2y ago
        The incumbent is not just x86 but now ARM as well. You have to think in network effects. You mention “the rest of the industry” yet ignore that it’s mostly arm , which would make arm the incumbent. x86 is the king for windows. But ARM has massive inroads with mobile, and now desktop with macOS, and servers with Amazon/Nvidia etc There’s a lot better incentive to support ARM than RISC-V for software developers. It isn’t one or the other , but it is a question of resources. Intel and AMD seem fine turning x86 around when threatened as can be seen by Lunar Lake and Stryx Point. Both have been good enough to steal QC’s thunder. You don’t think ARM manufacturers will do the same to RISC-V? TBH most of your arguments for RISC-V adoption seem to start from the position that it’s inevitable AND that competing platforms won’t also improve.
MBCook 2y ago
Like everything else, it doesn’t matter much. Windows ran on Itanium, Alpha, and as pointed out ARM for over a decade. Without the ISVs, it’s a flop for consumers. MS has had an abysmal time getting them to join in on ARM, only starting to have a little success now. Saying “Ha ha, just kidding, it’s RISC-V now” would be a disaster. That’s the kind of rug pull that helped kill Windows Mobile. Emulators aren’t good enough. They’re a stop gap. Unless the new chip is so much better than the old it’s faster with emulation then the old one was native no one will accept it long. Apple’s been there, but that’s not where MS sits today. And if your emulator is too good, what stops ISVs from saying “you did it for us, we don’t have to care”? So once again they don’t have to do it at all and you have no native software. MS can’t drop their ARM push unless they want to drop all non-x86 initiatives for a long time.
- snvzz 2y ago
  >And if your emulator is too good, what stops ISVs from saying “you did it for us, we don’t have to care”? So once again they don’t have to do it at all and you have no native software. x86 emulation enables adoption. Adoption means having an user base. Having an user base means developers will consider making the platform a target. >Saying “Ha ha, just kidding, it’s RISC-V now” would be a disaster. Would it now? If anything, offering RISC-V support as well would further reinforce the idea that Windows is ISA-independent, and not tied to x86 anymore.
  - MBCook 2y ago
    > If anything, offering RISC-V support as well would further reinforce the idea that Windows is ISA-independent, and not tied to x86 anymore. Anymore? It’s been independent since the 90s. It’s only ISVs that have been an issue. And a rug pull is a fantastic way to scare all the ISVs far far away.
    - snvzz 2y ago
      Whose's rug would even be pulled?
    - 6SixTy 2y ago

dagmx 2y ago

So which of Microsoft’s false starts would you take as them taking ARM seriously? Why do you think they’d take RISC-V any more seriously than their previous attempts at ARM? There are two fallacies to overcome here.