Jannik Vogel

> I just applied all the suggested changes, ready to make more if needed No, you didn't. And it's very hard to keep track of what you did or how...

There's actually a bug in the implementation that was merged by #195. It assumes that the first argument in nvparse texture instructions is always a texture reference. That breaks `cull_fragment`...

I don't think we need an issue for this on this repository. I see it more as an ecosystem issue, rather than an issue with XQEMU itself. While it would...

I have [started work on improving my nv2a-step.py script](https://github.com/JayFoxRox/xbox-tools/pull/51). I'm mostly focused on dumping the traces at this point. I have not tested it in XQEMU yet and I can...

I'm not sure why it hasn't been mentioned yet, but this stepping script has moved to https://github.com/XboxDev/nv2a-trace and has become its own powerful tool. There's also some experiments with trace...

There's barely any HAL code of interest left I think. Also most of it is known to have been stolen elsewhere (so GPL is probably not even the right license)...

I've gone through these alphabetically: ## audio.c For what it's worth, the algorithms in https://github.com/XboxDev/nxdk/blame/231336bfba1cbc66ae7259f8a27a81448ef5421d/lib/hal/audio.c are largely by me (see #364). The remaining code is probably factual (constants, memory addresses,...

I have tried running XQEMU from visual studio now. some updated notes: ### Setup * Install MSYS2 (special rules apply: no spaces in path, no FAT filesystems) * MSVS doesn't...

Looks like it's related to circular buffers. It seems to be used for buffer 0xE (what we assume to be "circular" scratch-space). ~~I think this might modify the DMA block...

I have prototyped nkpatcher support myself, but haven't tested or debugged it yet: https://github.com/JayFoxRox/nxdk/pull/79.