home *** CD-ROM | disk | FTP | other *** search
/ Executor 2.0 / executorv2.0.iso / pc / dos / extra / docs / maillist / text / archive.95 / text7111.txt < prev    next >
Encoding:
Internet Message Format  |  1996-03-31  |  2.8 KB
  1. From: "Jeffrey A. Halverson" <jeff@Ika.umn.edu>
  2. Reply-To: halv0019@gold.tc.umn.edu
  3. To: executor@nacm.com
  4. Subject: Executor on Xaccel & XFree
  5. Message-ID: <Pine.LNX.3.91.951124001537.205A-100000@Ika.umn.edu>
  6. MIME-Version: 1.0
  7. Content-Type: TEXT/PLAIN; charset=US-ASCII
  8. Sender: owner-paper@nacm.com
  9. Precedence: bulk
  10.  
  11. Hello,
  12.  
  13. I thought Linux users might be interested in the following benchmarking 
  14. that I did on my system.  I ran speedometer under the standard 
  15. XFree3.1.2, then under the exact same configuration I ran it under 
  16. Xinside's Xaccel...  Here's what I found:
  17.  
  18.  
  19. ===========P.R. Results:
  20.                               Xaccel                          XFree
  21. CPU:                          8.606                           6.628
  22. Graphics:                     2.014                           1.682
  23. Disk:                         1.576                           1.462
  24. Math:                        22.401                          17.339
  25. Performance Rating (OLD PR):  6.602                           5.182
  26. Performance Rating (NEW PR):  3.060                           2.623
  27.  
  28.  
  29. ==========Benchmark Results:
  30.  
  31. KWhetstones:        384.615  52.687                 319.148  43.719 
  32. Dhrystones:        4942.339   5.072                4255.319   4.367
  33. Towers:                       8.432                           7.898
  34. Quicksort:                    7.253                           7.923
  35. BubbleSort:                   8.901                          11.250
  36. Queens:                      10.177                          11.170
  37. Puzzle:                      13.659                          14.095
  38. Permutations:                 8.191                           6.792
  39. FastFourier:                 17.952                          13.571
  40. F.P.MatrixMultiply:          19.192                          13.571
  41. IntegerMatrixMultiply:       11.445                           8.822
  42. Sieve:                       17.467                          13.069
  43.                              ======                          ====== 
  44. Benchmark Average:           15.036                          12.973
  45.         
  46.  
  47. ==========ColorTests
  48.  
  49. Black & White:                 .537                            .449
  50. 4 Colors:                      .730                            .574
  51. 16 Colors:                     .774                            .647
  52. 256 Colors:                    .719                            .624
  53.                                ====                            ====
  54. Color Average:                 .690                            .574
  55.  
  56.  
  57. What I thought was particularly interesting was that calculations seem to 
  58. actually speed up with AccelX also.  Why is this?  I was expecting those 
  59. to stay constant, but only have the video-related tests to improve...  
  60. Can anyone explain this odd phenomena?
  61.  
  62. Jeff
  63.  
  64.