On Tuesday 10 Apr 2007 07:08, Kalle Valo wrote: > "ext Michael Wetherell" <mike.wetherell at ntlworld.com> writes: > > I've found that applications that make fairly heavy use of the wlan > > for more than a few minutes make the whole machine unstable. I'm > > wondering if it is the same for everyone, or is there a fault with > > this particular machine? > > It might be a software problem or a hardware problem, difficult to > say. At least I'm not aware of such severe WLAN bugs currently. If > it's a software bug, it might related to your Access Point or radio > environment. There might be cases which could trigger some strange > bugs which others don't see. If possible, try running exactly the > same test cases in a different location with a differect AP. Ok I'll try that when I get a chance. I did try two different makes of AP already actually, though at the same location, so I'll try somewhere else too. > What AP are using and what are the settings? I'll send you the datails. > Also try to file a bug report about the issue. That way it would > easier to track. If the problem goes away at another location, then I'll open a bug report and send you the details of the APs here. If the problem is the same at another location then I think I'll return the device for repair or replacement - the problem is easy to reproduce with this device yet no one else seems see it, so I think it must have a defective wifi chip or something. > > I saw there was a thread about dynamic PSM a few weeks ago, so I've > > tried setting it to 0, i.e. full PSM. It reduces throughput which I > > don't mind, but I'm not sure how much it helps with stability, I > > think maybe a bit. > > Actually that might make the things worse. Try with 120000 instead, > with the newest software it will disable PSM altogether. It might > help, or not. But be aware that disabling PSM will drain your battery > really quickly, for example in 6 hours if the display is turned off. Full PSM didn't make things worse, though it didn't help much either. I've tried disabling PSM as you suggested, but same problem. Anyway, thanks for the help. Regards, Mike
