Article View: sci.image.processing
Article #305Re: image processing loops
From: Michael P. Jarreau
Date: Tue, 11 Aug 1992 20:00
Date: Tue, 11 Aug 1992 20:00
16 lines
1508 bytes
1508 bytes
I have been asked why this is so. First of all, I meant to write decrement rather than increment. The speed increase comes from the fact that there is a single machine instruction which is something to the effect of "decrement register and jump to address xxx IF NOT ZERO" thus costing one machine cycle. Other speed optimizations can be made by avoiding comparisons which contain expressions. These expressions are re-evaluated at each iteration of a loop without regard to whether any of the concerned variables have changed. I expect that this holds true with other processors, I just don't have any recent experience on them. Michael
Technical Details
Message-ID:
<1992Aug11.200012.24819@vuse.vanderbilt.edu>Path:
arc1-ca.pugleaf.net!sparky!uunet!convex!darwin.sura.net!news.Vanderbilt.Edu!vuse.vanderbilt.edu!jarreauReferences:
<1992Aug10.221247.8016@cfa.harvard.edu> <1992Aug11.151146.11849@vuse.vanderbilt.edu> <Bsu0zo.6qC@acsu.buffalo.edu>Show Full Headers
Complete Article Headers
Newsgroups: sci.image.processing Path: arc1-ca.pugleaf.net!sparky!uunet!convex!darwin.sura.net!news.Vanderbilt.Edu!vuse.vanderbilt.edu!jarreau From: jarreau@vuse.vanderbilt.edu (Michael P. Jarreau) Subject: Re: image processing loops Message-ID: <1992Aug11.200012.24819@vuse.vanderbilt.edu> Summary: A follow up Sender: news@vuse.vanderbilt.edu (News Manager) Nntp-Posting-Host: ip1 Organization: Vanderbilt University School of Engineering, Nashville, TN, USA ... on Sun Sparcstations you will probably find that loops which begin at the last pixel of an image and increment to 0 will perform significantly faster than otherwise. References: <1992Aug10.221247.8016@cfa.harvard.edu> <1992Aug11.151146.11849@vuse.vanderbilt.edu> <Bsu0zo.6qC@acsu.buffalo.edu> Distribution: sci.image.processing Date: Tue, 11 Aug 1992 20:00:12 GMT Lines: 16