home *** CD-ROM | disk | FTP | other *** search
/ Executor 2.0 / executorv2.0.iso / pc / dos / extra / docs / maillist / text / archive.95 / text7112.txt < prev    next >
Encoding:
Text File  |  1996-03-31  |  2.7 KB  |  63 lines
  1. Reply-To: halv0019@gold.tc.umn.edu
  2. To: executor@nacm.com
  3. Subject: Executor on Xaccel & XFree
  4. Message-ID: <Pine.LNX.3.91.951124001537.205A-100000@Ika.umn.edu>
  5. MIME-Version: 1.0
  6. Content-Type: TEXT/PLAIN; charset=US-ASCII
  7. Sender: owner-paper@nacm.com
  8. Precedence: bulk
  9.  
  10. Hello,
  11.  
  12. I thought Linux users might be interested in the following benchmarking 
  13. that I did on my system.  I ran speedometer under the standard 
  14. XFree3.1.2, then under the exact same configuration I ran it under 
  15. Xinside's Xaccel...  Here's what I found:
  16.  
  17.  
  18. ===========P.R. Results:
  19.                               Xaccel                          XFree
  20. CPU:                          8.606                           6.628
  21. Graphics:                     2.014                           1.682
  22. Disk:                         1.576                           1.462
  23. Math:                        22.401                          17.339
  24. Performance Rating (OLD PR):  6.602                           5.182
  25. Performance Rating (NEW PR):  3.060                           2.623
  26.  
  27.  
  28. ==========Benchmark Results:
  29.  
  30. KWhetstones:        384.615  52.687                 319.148  43.719 
  31. Dhrystones:        4942.339   5.072                4255.319   4.367
  32. Towers:                       8.432                           7.898
  33. Quicksort:                    7.253                           7.923
  34. BubbleSort:                   8.901                          11.250
  35. Queens:                      10.177                          11.170
  36. Puzzle:                      13.659                          14.095
  37. Permutations:                 8.191                           6.792
  38. FastFourier:                 17.952                          13.571
  39. F.P.MatrixMultiply:          19.192                          13.571
  40. IntegerMatrixMultiply:       11.445                           8.822
  41. Sieve:                       17.467                          13.069
  42.                              ======                          ====== 
  43. Benchmark Average:           15.036                          12.973
  44.         
  45.  
  46. ==========ColorTests
  47.  
  48. Black & White:                 .537                            .449
  49. 4 Colors:                      .730                            .574
  50. 16 Colors:                     .774                            .647
  51. 256 Colors:                    .719                            .624
  52.                                ====                            ====
  53. Color Average:                 .690                            .574
  54.  
  55.  
  56. What I thought was particularly interesting was that calculations seem to 
  57. actually speed up with AccelX also.  Why is this?  I was expecting those 
  58. to stay constant, but only have the video-related tests to improve...  
  59. Can anyone explain this odd phenomena?
  60.  
  61. Jeff
  62.  
  63.