Attention grabbing headline? Check. It seems the ex-Valve engineer Rich Geldreich has noted a recent Phoronix benchmark
using apitest on how badly AMD's Catalyst driver performed.
RichThis excellent GL 4.x micro-benchmark that has been making waves recently is really interesting. Now that it's on Phoronix it's about as mainstream as it's going to get: NVIDIA Slaughters AMD Catalyst On Linux In OpenGL 4.x Micro-BenchmarksSource
At first glance the results sound great for NV: "The AMD Catalyst driver gets absolutely annihilated for these GL4 micro-benchmarks." But unfortunately it's bad news for everyone working in GL because it clearly demonstrates just how fractured and inconsistent the GL driver landscape actually is when the rubber hits the road.
The first thing to note is that the apitest utility may be Nvidia biased, and in addition to that there are AMD-related bugs
in apitest itself. So, looking over the test and peoples comments it does seem like there are bugs in apitest itself that could be causing such low performance.
I personally think AMD's future on Linux is with the open source driver which seems pretty decent and even comparable to AMD's Catalyst driver in some tests. It seems their Catalyst driver just constantly gets outclassed. The only problem with this is that the open source driver always plays catch-up with new hardware releases.
I hope that AMD won't just rely on the open source driver though, as a lot of people still look to Catalyst. With AMD's Mantle API how much effort are they really putting in? Who knows, but tests like this make it look bad.