This project has moved. For the latest updates, please go here.
x64 Release ( 128 threads_per * 768 blocks = 98304 threads)
Cities  12;   Permutations:  479001600:
---------------------------------------
With disk cache empty and only a single class ...
           Total     Load      Run
              ms       ms       ms  distance
GpuTsp0     8736 =   2647 +   6089; 111.3318: - cold
GpuTsp0     7086 =   1167 +   5919; 111.3318: - warm

... now with separate classes
           Total     Load      Run
              ms       ms       ms  distance
GpuTsp1     7092 =   1170 +   5922; 111.3318: - 1_SeparateClass
GpuTsp2     4640 =   1190 +   3450; 111.3318: - 2_StructArray

GpuTsp3     5191 =   1868 +   3323; 111.3318: - 3_Architecture_x64_2_1
GpuTsp3a    4058 =   1841 +   2217; 111.3318: - 3_PathArrayStrided
GpuTsp3b    4068 =   1830 +   2238; 111.3318: - 3_DivisorsCachedGlobal

GpuTsp4    10704 =   1842 +   8862; 111.3318: - 4_Long
GpuTsp4a    9550 =   1804 +   7746; 111.3318: - 4_PathArrayStrided
GpuTsp4b    8337 =   1784 +   6553; 111.3318: - 4_DivisorsCachedGlobal

GpuTsp3c    3601 =   1764 +   1837; 111.3318: - 3_MultiplyInstead
GpuTsp4c    4323 =   1779 +   2544; 111.3318: - 4_MultiplyInstead

... and now with disk cache populated.
           Total     Load      Run
              ms       ms       ms  distance
GpuTsp1     6008 =     89 +   5919; 111.3318: - 1_SeparateClass
GpuTsp2     3537 =     95 +   3442; 111.3318: - 2_StructArray

GpuTsp3     3301 =     93 +   3208; 111.3318: - 3_Architecture_x64_2_1
GpuTsp3a    2308 =     98 +   2210; 111.3318: - 3_PathArrayStrided
GpuTsp3b    2328 =    101 +   2227; 111.3318: - 3_DivisorsCachedGlobal

GpuTsp4     8953 =    100 +   8853; 111.3318: - 4_Long
GpuTsp4a    7841 =    103 +   7738; 111.3318: - 4_PathArrayStrided
GpuTsp4b    6643 =    100 +   6543; 111.3318: - 4_DivisorsCachedGlobal

GpuTsp3c    1935 =    105 +   1830; 111.3318: - 3_MultiplyInstead
GpuTsp4c    2639 =    103 +   2536; 111.3318: - 4_MultiplyInstead

Done ... Press Enter to shutdown.


x64 Release ( 128 threads_per * 768 blocks = 98304 threads)
Cities  13;   Permutations: 6227020800:
---------------------------------------
           Total     Load      Run
              ms       ms       ms  distance
GpuTsp4c   38830 =   3228 +  35602; 111.9742: - 4_MultiplyInstead

Last edited Nov 4, 2012 at 11:27 PM by pgeerkens, version 1