Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!apple!uokmax!rrmorris
From: rrmorris@uokmax.ecn.uoknor.edu (Rodney Raym Morrison)
Newsgroups: comp.sys.amiga.programmer
Subject: Re: Combsort algorithm
Keywords: sorting, Combsort
Message-ID: <1991May3.165222.31097@uokmax.ecn.uoknor.edu>
Date: 3 May 91 16:52:22 GMT
References: <1193@cbmger.UUCP>
Organization: Engineering Computer Network, University of Oklahoma, Norman, OK
Lines: 22

In article <1193@cbmger.UUCP> peterk@cbmger.UUCP (Peter Kittel GERMANY) writes:
>Anybody read the April Byte and the article about the new Combsort
>algorithm? Recommended to everyone. Not only that it's a pure miracle
>how one can achieve such a performance with such a little beastie
>(you take bubble sort and *add two lines* and *change one other line*).
>ABSOLUTELY INCREDIBLE.
>But, for the best, if you look closely into one of the explanation
>boxes, you find all this research (they obviously did an awful lot)
>was done on an A2000.
>
>I also thought at first, be careful it's the April issue, but I
>immediately tested it and it seems to work, flawlessly and blindingly
>fast. It looks as if it will replace my old loved heapsort that I used
>the last years (I hate recursive algorithms like quicksort), because
>Combsort beats Heapsort by a factor of 33 % on my machine (e.g. 12 s
>against 16 s).
>
>-- 
>Best regards, Dr. Peter Kittel  // E-Mail to  \\  Only my personal opinions... 

Could you post your combsort code or email it to me?
 rrmorris@uokmax.ecn.uoknor.edu