On Wed, 13 Dec 2006 15:06:24 +0100, Bob Spelten <[EMAIL PROTECTED]> wrote:
> Op Tue, 12 Dec 2006 21:33:32 +0100 schreef Dilwyn Jones > <[EMAIL PROTECTED]>: > >> Could you and any MinisQL or Aurora users try running this short >> superbasic program for me and tell me if your results are as expected. >> > > I ran the test on my Auroras... > PAUSE 15000 should be 300 secs or 5 minutes but gave 294 secs and > according to my stopwatch 293 secs which suggest that the +1 sec in line > 160 is probably too much. > There is thus a 7 sec error every 5 mins, interesting, so don't rely on > it > for accurate timings. > Bob > IIRC the Aurora manual contains something about the differences between it and a standrad QL, and amongst those it says something like this: Because the Aurora can use many resolutions and refresh rates, the vertical synch (or frame) interrupt is not used any more, but is emulated instead. Because there is no compatible 50Hz timing source on teh Aurora, it uses a 51.2Hz signal instead,derived from it's on-board 40MHz osicllator. This means that all PAUSE statements will execute faster, i.e. pauses will be shorter by about 2.4%. A bit of calculation will reveal this is 7.03125 seconds error every 5 minutes :) So, the shorter time is normal for the Aurora. Bet you 2 things, though: 1) You thought I was dead or something :) 2) You never read that part of the manual :) Nasta _______________________________________________ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
