Andre Weissflog

Yeah somehow my brain skipped this part: > ...because they will have full access to these features via the all-powerful build system... ...tbh though, if a self-contained cross-platform Clang toolchain...

> Hopefully not, otherwise your code is horribly broken beyond repair There are quite a few reasons to pin the toolchain to a specific version, especially in bigger projects and...

Hmm, probably some change in behaviour either in the sokol_app.h dpi handling code or in Windows since when I implemented dpi-changed event handler. The function which contains that assert shouldn't...

Lol, looks like I bricked my Windows laptop (it suddenly wants a Bitlocker recovery key but I had setup Windows with a local account and I wasn't even aware that...

Hmm... interestingly enough I cannot reproduce the problem so far... I'm testing with the clear-sapp sample which doesn't set `sapp_desc.high_dpi`. My config is a laptop with the builtin display scaling...

Great! I'll keep this ticket open for me as reminder to find a less harsh solution (e.g. no asserts triggering when the dpi-awareness mode has been set outside of sokol_app.h.

> I'm afraid there is no support for cpu reading/copying a gpu buffer That's correct, currently there's no way to get the data back to the CPU side since that's...

> Is there may be some very non-performant/ugly way around it? Only by calling directly into backend APIs like GL, D3D11 or Metal (which sokol-gfx allows but with many caveats)....

Not sure if it's useful for the discussion, but in Oryol I had a mapping layer over the gamepad input because axis and button indices are not standardized in some...

No plans to merge this PR as is, sorry. And getting consistent gamepad support across all platforms and gamepad types (e.g. anything that's not an Xbox controller on Windows, or...