home *** CD-ROM | disk | FTP | other *** search
/ Executor 2.0 / executorv2.0.iso / pc / dos / extra / docs / maillist / text / archive.95 / text7109.txt < prev    next >
Encoding:
Internet Message Format  |  1996-03-31  |  2.9 KB
  1. Received: (from jeff@localhost) by Ika.umn.edu (8.6.12/8.6.12) id AAA00283; Fri, 24 Nov 1995 00:31:40 -0600
  2. Date: Fri, 24 Nov 1995 00:31:38 -0600 (CST)
  3. From: "Jeffrey A. Halverson" <jeff@Ika.umn.edu>
  4. Reply-To: halv0019@gold.tc.umn.edu
  5. To: executor@nacm.com
  6. Subject: Executor on Xaccel & XFree
  7. Message-ID: <Pine.LNX.3.91.951124001537.205A-100000@Ika.umn.edu>
  8. MIME-Version: 1.0
  9. Content-Type: TEXT/PLAIN; charset=US-ASCII
  10. Sender: owner-paper@nacm.com
  11. Precedence: bulk
  12.  
  13. Hello,
  14.  
  15. I thought Linux users might be interested in the following benchmarking 
  16. that I did on my system.  I ran speedometer under the standard 
  17. XFree3.1.2, then under the exact same configuration I ran it under 
  18. Xinside's Xaccel...  Here's what I found:
  19.  
  20.  
  21. ===========P.R. Results:
  22.                               Xaccel                          XFree
  23. CPU:                          8.606                           6.628
  24. Graphics:                     2.014                           1.682
  25. Disk:                         1.576                           1.462
  26. Math:                        22.401                          17.339
  27. Performance Rating (OLD PR):  6.602                           5.182
  28. Performance Rating (NEW PR):  3.060                           2.623
  29.  
  30.  
  31. ==========Benchmark Results:
  32.  
  33. KWhetstones:        384.615  52.687                 319.148  43.719 
  34. Dhrystones:        4942.339   5.072                4255.319   4.367
  35. Towers:                       8.432                           7.898
  36. Quicksort:                    7.253                           7.923
  37. BubbleSort:                   8.901                          11.250
  38. Queens:                      10.177                          11.170
  39. Puzzle:                      13.659                          14.095
  40. Permutations:                 8.191                           6.792
  41. FastFourier:                 17.952                          13.571
  42. F.P.MatrixMultiply:          19.192                          13.571
  43. IntegerMatrixMultiply:       11.445                           8.822
  44. Sieve:                       17.467                          13.069
  45.                              ======                          ====== 
  46. Benchmark Average:           15.036                          12.973
  47.         
  48.  
  49. ==========ColorTests
  50.  
  51. Black & White:                 .537                            .449
  52. 4 Colors:                      .730                            .574
  53. 16 Colors:                     .774                            .647
  54. 256 Colors:                    .719                            .624
  55.                                ====                            ====
  56. Color Average:                 .690                            .574
  57.  
  58.  
  59. What I thought was particularly interesting was that calculations seem to 
  60. actually speed up with AccelX also.  Why is this?  I was expecting those 
  61. to stay constant, but only have the video-related tests to improve...  
  62. Can anyone explain this odd phenomena?
  63.  
  64. Jeff
  65.  
  66.