4 ms·

There is a reason for that: Windows offers super-stable APIs that allow WinUAE to be supremely good at fidelity and accuracy. When it comes to graphics and audi

There is a reason for that: Windows offers super-stable APIs that allow WinUAE to be supremely good at fidelity and accuracy. When it comes to graphics and audio, the Linux experience is pretty much a disaster across the board if one moves away from a tightly controlled environment. If you want proof, look at the best Unix port of WinUAE (FS-UAE). It works, but compared to the original, the experience is so bad in many different ways (graphical glitches, crashes, audio stuttering, latency, bugs and more bugs) that you're better off running WinUAE through Wine. I'm a hardcore Amiga user and enthusiast, and own multiple original units in working condition. Regarding preservation, I feel so much better knowing that WinUAE runs on Windows. Linux is simply not a stable/quality graphical platform and with the Wayland/X.org fiasco taken into account, should inspire even less confidence for the future.

troad 3y ago
That's interesting, thanks for sharing your point of view.
- jychang 3y ago
  Random thought: If you want code that you know can be run in 100 years, you should target WINE. This is assuming that the code you’re writing is a medium sized project- a small project would be too trivial, a large one (think MySQL or Java) would probably get ported to a future architecture. Native Windows isn’t great, a lot of Windows 95 binaries have issues running on Windows 11. Linux binaries aren’t great either, I don’t trust Linux binaries from 1999 to work perfectly even now. But WINE? There are millions of pieces of Win32 software from 1995-2030+ (at the very least) that define this period in history. This means that even if we all use some weird alien OS or architecture in 100 years, at least it’s probable that there’s a fork of WINE that future-people use to run obsolete Windows software. So yeah, if I was hypothetically writing an emulator which was for (an obscure enough platform that no other emulators would exist in 100 years), and assuming that the emulator would not be getting patched/ported after 100 years, then WINE-Win32 would probably be the platform I’d aim to support for maximum compatibility in the far future.

troad 3y ago

That's interesting, thanks for sharing your point of view.

jychang 3y ago
Random thought: If you want code that you know can be run in 100 years, you should target WINE. This is assuming that the code you’re writing is a medium sized project- a small project would be too trivial, a large one (think MySQL or Java) would probably get ported to a future architecture. Native Windows isn’t great, a lot of Windows 95 binaries have issues running on Windows 11. Linux binaries aren’t great either, I don’t trust Linux binaries from 1999 to work perfectly even now. But WINE? There are millions of pieces of Win32 software from 1995-2030+ (at the very least) that define this period in history. This means that even if we all use some weird alien OS or architecture in 100 years, at least it’s probable that there’s a fork of WINE that future-people use to run obsolete Windows software. So yeah, if I was hypothetically writing an emulator which was for (an obscure enough platform that no other emulators would exist in 100 years), and assuming that the emulator would not be getting patched/ported after 100 years, then WINE-Win32 would probably be the platform I’d aim to support for maximum compatibility in the far future.