We should use Telemetry to better understand real-world usage of dual-GPUs, and the extent to which we can rely on particular approaches for detecting the active GPU. For example: -what percentage of our users have dual-GPU systems -how many of these systems have exactly one GPU that is blacklisted (in which case, correctly detecting the active GPU in our blacklisting logic is crucial for correct behaviour) -do D3D9 and D3D10 always get the same GPU
Is there anything else needed beyond the gfxinfo that is now included in telemetry data?
(In reply to John Daggett (:jtd) from comment #1) > Is there anything else needed beyond the gfxinfo that is now included in > telemetry data? It'd be nice to also know which GPU is being used for D3D9 and D3D10 (and how frequently this is GPU #1 in gfxinfo), since this would tell us how frequently our current blacklisting logic is making a decision based on the wrong GPU.
You need to log in before you can comment on or make changes to this bug.