Futuremark has made an official comment to Gainward his accusations:
The accusation is totally wrong because what it suggests is not even
feasible technically. 3DMark03 does not talk to graphics driver, it talks to
the DirectX API, which then talks to the driver. Thus, it is impossible for
the application to disable GPU compiler.
The only change in build 340 is the order of some instructions in the
shaders or the registers they use. This means that new shaders are
mathematically equivalent with previous shaders. A GPU compiler should
process the old and the new shader code basically with the same performance.
Of course, if there are application specific optimizations in the driver
that depend on identifying a shader or parts of it, then you might see
performance differences because these optimizations will not work if the
driver is not able to detect the shader.
Let's also repeat that 3DMark specific driver optimizations are forbidden in
our run rules because they invalidate the performance measurement and the
resulting score is not comparable to other hardware.
Thus, the right conclusion is that the new version of 3DMark03 is now very
suitable for objective performance measurement between different hardware.
Will the optimizations madness ever stop?
You will have problably noticed that there is a list on Futuremark his website, with approved drivers. You may be wondering why the ForceWare 52.16 is in that list having in mind that it still contains benchmark optimizations. Here is Futuremark his answer :
Reason why any given driver is listed there is that those drivers produce a
valid performance measurement result with 3DMark03 build 340. Looking from
our point of view, the most important thing for us to do is to enable our
customers get a comparable score.