Well the status is much better actually. Actually I didn't expect to be 100% clean.
Part 6 is working on my side. Where Part 11 is working on your side. The 2 drivers didn't interpret the spec the same way. Part 12 fail in both drivers, maybe my interpretation of the spec is incorrect. That said the spec doesn't follow any logic.
By the way, can you test transform feedback test too now that GL3.0 is working.
Code:
for feedback_test in useprogstage_noactive useprogstage_active bind_pipeline
do
echo "******** RUN $feedback_test"
./ext_transform_feedback-api-errors $feedback_test -auto
done
So is it going to be that mesa will be the gold standard soon? Is this the way we will get HW mode eventually (perhaps the last reason why some games not yet playable due to demanding cutscenes, etc)?
OS: Linux Mint 17.2 64 bit (occasional Antergos/Arch user)
(I am no longer a Windows user)
CPU: Intel Pentium G3258
GPU: Nvidia GTX 650 Ti
Mesa won't be really gold mostly due to the performance. Beside I need to discuss my patch with upstream and eventually they will accept them. That said from a developper point of view:
1/ knowing how the driver behave help a lots to make the code more clever.
2/ it would be 100x times easier to debug crash. On FGLRX the only info is "there is a segmentation fault in the driver!!!".
3/ Intel GPU will be able to handle GSdx, and so I would be able to test GSdx on Intel hardware too.
4/ Waiting 1 years (actually I'm still waiting) a basic fix in the driver when I could patch it in a couple of days....