> Mmmm redid the 1.8.7 tests. Now the results are a little different. Very similar to 1.8.6. Rescanned the original 1.8.7 binaries and these turned out to be of exactly the same size. And indeed produce almost equal results. Looks like something was not done properly while building these binaries... Well at least it happend to put me in the right track ;-) Ok here the results of 1.8.7. Same optimization level. All non-shared. Option Real Sys --disabled-pthread 23.70 0.02 --enabled-pthread 41.39 10.14 --enabled-pthread and config.h edited 23.22 0.02 BTW this is the output of running rubybench: Running benchmark BinaryTrees Argument is 13 stretch tree of depth 14 check: -1 16384 trees of depth 4 check: -16384 4096 trees of depth 6 check: -4096 1024 trees of depth 8 check: -1024 256 trees of depth 10 check: -256 64 trees of depth 12 check: -64 16 trees of depth 14 check: -16 long lived tree of depth 13 check: -1 Name Param Real Sys Usr BinaryTrees 13 23.22 23.20 0.02 Please note the Sys and Usr times are swapped. This run only consumed USR time from an OS perspective. It is a --disable-pthread run. This, btw, is not 1.8.7 specific. It showed in 1.8.6 too. happened Regards Paul van den Bogaard --------------------------------------------------------------------------------------------- Paul van den Bogaard Paul.vandenBogaard / sun.com ISV-E -- ISV Engineering, Opensource Engineering group Sun Microsystems, Inc phone: +31 334 515 918 Saturnus 1 extentsion: x (70)15918 3824 ME Amersfoort mobile: +31 651 913 354 The Netherlands fax: +31 334 515 001