Release v0.8.1

Some tweaks to nvcuda to minimize overhead in multithread situations.
Added support for OptiX ABI 93 (SDK 8.1.0)
Upstream updates of dxvk-nvapi and wine-nvml.

Remove cufft64_11.dll helper library for FFT usage. This may however
come with some missing tweaks for untested FFT functions.

Various cleanups and removal of unused headers.

Release v0.8.0

This release have quite a few highlights.

Redo folder structure to use x32/x64 for 32 and 64 bit
Rename all winelib dll.so files to .dll for easier integration
Drop generating "fake" .dll files, as these are not really useful
with newer wine versions.

Add NVENC libraries - nvcuvid and nvencodeapi. These can be used
with hardware level encoding/decoding using NVIDIA GPU.

Some update to upstream dxvk-nvapi
Some minor updates and tweaks to nvcuda

Updated installscripts and added script for easier installation
when using "Bottles". This will install the libraries into the
bottle of your choosing.

Release v0.7.20

Quite some updates to dxvk-nvapi.

Add new library nvofapi (64-bit only).
No significant changes to other support libraries

Release v0.7.19

Some upstream changes to dxvk-nvapi

Added some tweaks to be able to run the "Elevator" Demo (Star Wars)

No change to CUDA or OptiX

Release v0.7.18

Mostly bugfixes and minor updates to nvcuda.
Some updates to dxvk-nvapi and nvml library.
Removed CUDA SDK 3.0 helper library, as it does not seem needed.

Added note about DAZ Studio and the usage of nvml (wine-nvml).

Release v0.7.17

Add support for CUDA SDK 12.5

Add helper library for FFT (cufft64_11.dll) when running FFT functions from
CUDA SDK 12.3 or newer.

Add 32-bit helper library for cuda RunTime 3.0 (cudart32_30_14.dll) for running
old CUDA SDK 3.0 samples and NVIDIA demo's. This however requires some
handheld fiddling. See README.md for info.

Various bugfixes and tweaks.

Release v0.7.16

Quick fix for stupid mistake, and added some missing cuda function stubs.

Release v0.7.15

Some updates to DXVK-NVAPI.

Added "Encryption scheme" from ZLUDA for the "Unknown" internal function. This replaces the "stack-return-hack" that was used to bypass this runtime check.

CUFFT (Cuda FFT) functionality for CUDA 12.3+ is still not working.

Release v0.7.14

Added support for most CUDA 12.4 functions although the CUFFT demos does not run when using CUDA 12.3 and 12.4.

Updated proton installscript to make it slightly more readable.

Added upstream DXVK-NVAPI and WINE-NVML changes.

Release v0.7.13

This should fix issues when running CUDA SDK 11.6 and newer. Also DAZ Studio 4.22 seems to work now.
Added some stubs from SDK 12.3, but some functions may still be missing so this is a WIP.