The idea of using the date/time stamp to generate a unique id/filename is
great but if you are concerned about duplicates (files created w/in the same
 min/second) then it might not ne kosher...

My idea is to use a "unique" filename server/command that uses a static
variable/counter to generate ... along the lines of this cmd file...

IF BOUND(CNTR) THEN
   SETVAR CNTR, (CNTR + 1)
ELSE
   SETVAR CNTR, 1
ENDIF
SETVAR X1, CHR(HPYEAR + 65)
SETVAR NEWFNAME, "!X1!HPDOY" + RHT("00"+"!CNTR",3)
ECHO !NEWFNAME
DELETEVAR X1

****
Converting the year to letter, (2000 as "A") ensures an alpha start till
2026...You can still figure out the create date easily (pseudo-julian).
Above will create 1-999 unique filenames a day - (expandable to 1-9999)

I hope this helps...