home *** CD-ROM | disk | FTP | other *** search
/ Executor 2.0 / executorv2.0.iso / pc / dos / extra / docs / maillist / text / archive.95 / text7108.txt < prev    next >
Encoding:
Internet Message Format  |  1996-03-31  |  3.1 KB
  1. Received: from Ika.umn.edu (dialup-2-105.gw.umn.edu [134.84.101.105]) by nacm.com (8.6.10/8.6.9) with ESMTP id WAA10649 for <executor@nacm.com>; Thu, 23 Nov 1995 22:32:50 -0800
  2. Received: (from jeff@localhost) by Ika.umn.edu (8.6.12/8.6.12) id AAA00283; Fri, 24 Nov 1995 00:31:40 -0600
  3. Date: Fri, 24 Nov 1995 00:31:38 -0600 (CST)
  4. From: "Jeffrey A. Halverson" <jeff@Ika.umn.edu>
  5. Reply-To: halv0019@gold.tc.umn.edu
  6. To: executor@nacm.com
  7. Subject: Executor on Xaccel & XFree
  8. Message-ID: <Pine.LNX.3.91.951124001537.205A-100000@Ika.umn.edu>
  9. MIME-Version: 1.0
  10. Content-Type: TEXT/PLAIN; charset=US-ASCII
  11. Sender: owner-paper@nacm.com
  12. Precedence: bulk
  13.  
  14. Hello,
  15.  
  16. I thought Linux users might be interested in the following benchmarking 
  17. that I did on my system.  I ran speedometer under the standard 
  18. XFree3.1.2, then under the exact same configuration I ran it under 
  19. Xinside's Xaccel...  Here's what I found:
  20.  
  21.  
  22. ===========P.R. Results:
  23.                               Xaccel                          XFree
  24. CPU:                          8.606                           6.628
  25. Graphics:                     2.014                           1.682
  26. Disk:                         1.576                           1.462
  27. Math:                        22.401                          17.339
  28. Performance Rating (OLD PR):  6.602                           5.182
  29. Performance Rating (NEW PR):  3.060                           2.623
  30.  
  31.  
  32. ==========Benchmark Results:
  33.  
  34. KWhetstones:        384.615  52.687                 319.148  43.719 
  35. Dhrystones:        4942.339   5.072                4255.319   4.367
  36. Towers:                       8.432                           7.898
  37. Quicksort:                    7.253                           7.923
  38. BubbleSort:                   8.901                          11.250
  39. Queens:                      10.177                          11.170
  40. Puzzle:                      13.659                          14.095
  41. Permutations:                 8.191                           6.792
  42. FastFourier:                 17.952                          13.571
  43. F.P.MatrixMultiply:          19.192                          13.571
  44. IntegerMatrixMultiply:       11.445                           8.822
  45. Sieve:                       17.467                          13.069
  46.                              ======                          ====== 
  47. Benchmark Average:           15.036                          12.973
  48.         
  49.  
  50. ==========ColorTests
  51.  
  52. Black & White:                 .537                            .449
  53. 4 Colors:                      .730                            .574
  54. 16 Colors:                     .774                            .647
  55. 256 Colors:                    .719                            .624
  56.                                ====                            ====
  57. Color Average:                 .690                            .574
  58.  
  59.  
  60. What I thought was particularly interesting was that calculations seem to 
  61. actually speed up with AccelX also.  Why is this?  I was expecting those 
  62. to stay constant, but only have the video-related tests to improve...  
  63. Can anyone explain this odd phenomena?
  64.  
  65. Jeff
  66.  
  67.