Path: utzoo!utgpu!jarvis.csri.toronto.edu!clyde.concordia.ca!uunet!willett!GEnie
From: GEnie@willett.UUCP (ForthNet articles from GEnie)
Newsgroups: comp.lang.forth
Subject: OPTIMIZATION
Message-ID: <50.UUL1.3#5129@willett.UUCP>
Date: 21 Dec 89 03:19:48 GMT
Organization: Latest Link in ForthNet Chain
Lines: 43


 Date: 12-20-89 (05:23)              Number: 1681 (Echo)
   To: MIKE SPERL                    Refer#: NONE
 From: MARK SMILEY                     Read: NO
 Subj: OPTIMIZATION                  Status: PUBLIC MESSAGE

 Mike,
    Here's another timing question for you from GRAFKRNL.SEQ.
 \ YOURS:
 code v320.dot  ( x y color -- )
     pop ax
     mov di, ax                   \ color in di
     ...
 \ PROPOSED:
 code v320.dot  ( x y color -- )
     POP DI
     ...
 \ Here are my times, with similar code to that of my previous note.
 \ Here, T2 is the POP DI method.
 100 SS !
 TIMER T1
 Elapsed time   =  00:00:28.39
 TIMER T2
 Elapsed time   =  00:00:27.63
 1000 SS !
 TIMER T1
 Elapsed time   =  00:04:43.80
 TIMER T2
 Elapsed time   =  00:04:36.22
 \ Once again, it seems that the smaller code is (slightly) faster.  Any
 \ comments?
                               Mark
 P.S. I haven't been able to bring myself to try your 22 minute timer
 test, I'm not sure I can part with my computer for 44+ minutes!!  (My
 addiction to my computer must be worse than I thought.)  Besides, all
 the tests agree that the smaller code is faster, so I see no real need
 to run the test.

 ---
  ~ EZ-Reader 1.13 ~ 
-----
This message came from GEnie via willett through a semi-automated program.
Report problems to: 'uunet!willett!dwp' or 'willett!dwp@gateway.sei.cmu.edu'