hi John,

 BS>> It _seriously_ annoys me  that IBM chose to make syssleep() with a
 BS>> resolution of one second, given that the DosSleep() api can resolve
 BS>> done to a millisecond-you can recompile the syssleep() example code
 BS>> in the OS2 toolkit to get a syssleep() with a finer granularity(
 BS>> tenths and hundreths are more useful for me).

 JP> My solution to the keypress problem would be to use an inkey()

What problem?  CHARS() and SysGetKey() do the job.

 JP> function as provided with the Quercus Systems REXX library functions,
 JP> which incidentally, also offers a delay() function - accurate to one
 JP> tenth of a second.

This one is freeware:


/* SLEEP.CMD : Rexx/2 with RexxIPC.DLL */
parse arg delay
if 0=DataType(delay,"W") then do
  say "Usage: SLEEP "
  say "where 'delay' is the number of milliseconds you want to sleep"
  say "SLEEP returns the number of milliseconds slept."
  return -1
end
call RxFuncAdd "SemLoadFuncs", "RexxIPC",
"SemLoadFuncs"
call SemLoadFuncs
call SemEventCreate "sem"
rval = Time("E")
call SemEventWait sem, delay
rval = 1000*(Time("E")-rval)
call SemEventClose sem
return rval


I could have used YdbaUtil instead with narry a change...

regards,
         hugo