Fix winmm's check_position() tests

[Andreas Mohr]

Hi,

On Tue, Jul 20, 2004 at 02:19:22AM +0200, Francois Gouget wrote:
> Sometimes in the VMware virtual machines I do get 96000 instead of 96001 
> for the 96kHz tests. But that does not seem systematic. Maybe that's the 
> pcm conversion bug you mentioned?
> 
> So do you have any idea why we are getting different results? Maybe the 
> result depends on the individual sound driver and the winmm test should 
> accept both?
What if it's simply the sound driver?
One is very correct, while the other one is simply WRONG?
Accepting both sounds like a bad (since wrong?) semi-solution to me
(without knowing further specifics, that is).

We might want to do extensive testing on further Windows installations...
(to find the most commonly used offset calculation method, which is not
necessarily the correct one -- just as the most widely used PC OS is not
necessarily the correct one ;-)

Greetings,

Andreas