Artemide Experimental Async
Changelog
-
Fully reworked the codebase, starting from scratch
-
Reapplied and cleaned up many legacy changes
-
Switched to async decoding by default, with automatic fallback to sync only on very old devices
-
FastGL (direct present) is now enforced
-
Major FSR fixes and performance improvements; FSR now supports presets:
Performance / Balanced / Quality (for both EASU and RCAS) -
Updated stats: in Advanced Lite, the new “Delay” metric now represents more than decode latency and includes upscaling latency
-
VSync toggle: uses Choreographer inside GlUpscaler to synchronize frame presentation with the display
-
FastVsync: uses a NanoPacer to align frame presentation timing with the display
-
Removed LFR, PreferBigCores, AdaptX, and related logic no longer needed
-
Max Smoothness now uses an adaptive dequeue timeout (500 → set value)
-
Exposed Decoder Output Dequeue Timeout (0–50000)
Stock Artemis uses 50000, LFR used 0 like the new toggle "Immediate Frame Delivery"
You can tune this value to find the best setting for your device
-
Exposed Decoder Output Drain Timeout
This should be left at 0, It will be removed in the next release (kept temporarily; do not change it)
-
Tuned decoder fetch / prefetch output buffer logic for this new branch
-
Many additional fixes and refinements see commit history for details
EDIT
Hotfix:
- fixed ULL black screen on Snapdragon 8gen2. Updated Apk's
Having issues?
If you experience problems with this release, please try one of the following versions:
6.5.1
6.2
As always,
⚠️ Heads up: this is experimental. I’ve only tested on my hardware, so things may break on yours. Use at your own risk.
Found a bug? Tell me and I’ll try to fix it.
Enjoy!