On Thu, Feb 06, 2003 at 11:53:06AM +0100, Michael Nordstr�m: > Now, I would prefer to fix the problem instead of the > symptom, but at least such an option would make it > possible for Visor users to run the viewer until the > real problem can be found.
I agree, but unfortunatly I wasn't able to fix (or even find) the source of the real problem. It could be a simple as the feature memeory on the Visor is buggy, but according to Palm's specifications, we're following it to the letter. I do know that this type of memory was added in OS 3.1, and was revised to support larger chunks in 3.5. Perhaps some additional bugfixes silently made it into 3.5? If that's the case then we could look at writing the cache directly do an actual database, but that may add in additional overhead at the cost of speed. I didn't entertain this possibility simply because of the effort involved, but it might be something to look at. -- Adam McDaniel Array.org Calgary, AB, Canada _______________________________________________ plucker-list mailing list [EMAIL PROTECTED] http://lists.rubberchicken.org/mailman/listinfo/plucker-list
