Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!jarthur!nntp-server.caltech.edu!piglet!madler
From: madler@piglet.caltech.edu (Mark Adler)
Newsgroups: comp.sys.handhelds
Subject: Re: hp48 graphics, timings, terminal emulator and more!
Message-ID: <1990Apr11.043121.5506@laguna.ccsf.caltech.edu>
Date: 11 Apr 90 04:31:21 GMT
References: <5783@jarthur.Claremont.EDU> <TJB.90Apr10211822@tadtec.uucp>
Sender: news@laguna.ccsf.caltech.edu
Organization: California Institute of Technology, Pasadena
Lines: 26


I use the following benchmark program:

%%HP: T(3)A(R)F(.);
@ TIMR crc #D3C0h length 74
\\<<
  MEM DROP              @ do garbage collection
  TICKS \-> $$$ \<<     @ save current tick count
    EVAL                @ run program with stack contents
    TICKS               @ get tick count when done
    $$$ - B\->R         @ compute ticks elapsed
    86 -                @ adjust for time to assign $$$ and do TICKS
    8192 /              @ convert to seconds
  \>>
\>>

I believe storing the ticks in a local is more appropriate, since the
execution time of the program will depend on how much stuff is on the
stack, so the timing routine should not change that.  Also, the
garbage collection is necessary to get consistent results.  Even with
the MEM DROP, the timing is still dependant on the amount of free
memory, since that determines how often garbage collection is done.
I'll leave how to figure out the "86 -" as an exercise for the reader.

Mark Adler
madler@tybalt.caltech.edu