Subj : Re: ansi escape detection
To   : Tracker1
From : Digital Man
Date : Sun Aug 28 2005 10:20 pm

  Re: Re: ansi escape detection
  By: Tracker1 to Digital Man on Sun Aug 28 2005 09:58 pm

 > Digital Man wrote:
 > >> Mainly in some of my scrolling list type menus, where you would use up/do
 > >> arrow keys, but have escape as a back option.. why would it be a bad idea
 > >> reduce the poll/wait time for an escape followed by the ansi sequence.
 > >
 > > Because arrow keys are sent as ANSI escape sequences (e.g. ESC[A is the up
 > > arrow). If there is > 500ms latency between the ESC and the '[', then the 
 > > is exited from the menu rather than changing the selection up, like they
 > > intended.
 > 
 > I understand that, but what is the probability of having a > 500ms
 > ping/latency, or having it be that long between one packet and the next (if
 > the sequence was in and of itself split)?  I mean even dialup is usually und
 > a 400ms ping time... 700ms may be a much safer number, but still under a ful
 > second.

"ping times" don't really have relevancy to TCP latency. And the potential
latency we're talking about is transient, not consistent (or even average).
Even at a 1 second delay, there are still users that occasionlly don't have
arrow keys interpretted 100% of the time. <shrug>

In any case, goto http://bugs.synchro.net and submit the feature request to
make this value configurable, and I'll gladly do that in a future version.

                                                  digital man

Snapple "Real Fact" #107:
You burn more calories sleeping than watching TV.

.