On 11/19/01 12:33 PM, "Grant [Irradiated] Hutchinson" <grant_at_splorp.com>
wrote:
> In a previous message, Laurent Daudelin typed vigorously:
>
>> Just wondering a bit. What is the use of a stopwatch that goes down to 1/100
>> of a second? I mean, unless your Newt is plugged to some detection device,
>> you'll have to tap the screen with your pen to start/stop the timer. By the
>> time your pen tip touches the screen, there might have been 200 or 300
>> 100ths of a second that passed. I guess my real question is what's the point
>> to get such a precise timer when the mode in which you start and stop it is
>> so imprecise?
>
> Hmmm... I'll have to ask the fellow who was looking for this in the first
> place. Even if there was a slight delay created by the time it takes for
> your stylus to touch the screen and for any UI widget to highlight,
> wouldn't the delay be the same when starting the timing as when ending
> the timing? The two delays caused by these particular user events would
> cancel each other out, and you would end up with a reasonably accurate
> timing. And how would this be any different than the delay involved with
> clicking a mechanical button on a handheld stopwatch?
Yes, if they were the very same delays. But here, with the time that the tip
of the pen touches the screen, that the touchscreen reacts, sends the
information to the OS, which then would search for the widget that actually
received the tap, and then the OS transmitting the message, I mean, a lot
here is very imprecise. Even when the event has been acknowledged by the OS,
based on the OS scheduling, I wouldn't be surprised that if you were to
timed 2 identical events, you might get different delays. Now, add the pen's
trip to the touch screen, the touch screen reaction, the OS taking over, you
have already a few sources for variable delays...
-Laurent.
-- ===================================================================== Laurent Daudelin Developer, Multifamily, ESO, Fannie Mae mailto:Laurent_Daudelin_at_fanniemae.com Washington, DC, USA ********************** Usual disclaimers apply ********************** evil and rude adj.: Both evil and rude, but with the additional connotation that the rudeness was due to malice rather than incompetence. Thus, for example: Microsoft's Windows NT is evil because it's a competent implementation of a bad design; it's rude because it's gratuitously incompatible with Unix in places where compatibility would have been as easy and effective to do; but it's evil and rude because the incompatibilities are apparently there not to fix design bugs in Unix but rather to lock hapless customers and developers into the Microsoft way. Hackish evil and rude is close to the mainstream sense of `evil'.-- This is the Newtontalk mailinglist - http://www.newtontalk.net To unsubscribe or manage: visit the above link or mailto:newtontalk-request_at_newtontalk.net?Subject=unsubscribe
This archive was generated by hypermail 2.1.2 : Sat Dec 01 2001 - 20:03:14 EST