mirrors/Ryujinx
1
0
Fork 0
mirror of https://github.com/Ryubing/Ryujinx.git synced 2025-03-17 07:01:48 +00:00
Code Issues Projects Releases Packages Wiki Activity

Clarify that this branch should not really be used yet

Browse source
This commit is contained in:
Evan Husted 2025-03-02 02:26:37 -06:00
parent 39ced98cb6
commit 6c159372f5
1 changed files with 9 additions and 0 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

9
README.md
View file

@ -1,3 +1,12 @@
# You have stumbled on a gimped version of Ryujinx.
This is the NativeAOT branch. It should run much faster than the normal version, however it's ruined by one thing: the Macro JIT.
This single part of Ryujinx, which handles common Maxwell GPU Macros by using .NET Reflection Emit, is entirely unusable via NativeAOT and it kills the performance.
I was getting 20FPS in BOTW with 98% FIFO usage.
This is compared to over 100FPS with about 70% FIFO usage in standard, managed C#, Ryujinx.
If there are any JIT wizards out there; I am calling to you. I believe getting this to work would be a game changer.
[This is the offending JIT implementation.](https://github.com/Ryubing/Ryujinx/blob/feature/native_aot/src/Ryujinx.Graphics.Gpu/Engine/MME/MacroJitCompiler.cs)
<table align="center">
<tr>
<td align="center" width="25%">