3 ms·
I guess I should have specified, I was thinking of the benchmarks vs non-invokedynamic JRuby. No intention of starting a benchmark war. Ruby 1.9 is fast too.
by ZitchDog 14y ago
I guess I should have specified, I was thinking of the benchmarks vs non-invokedynamic JRuby. No intention of starting a benchmark war. Ruby 1.9 is fast too.
- igouy 14y agoOK. Back to August 29th, JRuby 1.7.0 preview2, invoke-dynamic=true on the right, the last number is elapsed time in seconds (so 451.046 vs 258.643 with invoke-dynamic=true) -- http://anonscm.debian.org/viewvc/shootout/shootout/website/websites/u32/data/data.csv?r1=1.1021&r2=1.1022 http://anonscm.debian.org/viewvc/shootout/shootout/website/w...
- igouy 14y agoI see that ViewVC link seems to have problems showing the diff at the moment, here are elapsed seconds (without invokedynamic and with invokedynamic) jruby 1.7.0 (1.9.3p203) 2012-10-22 ff1ebbe on Java HotSpot(TM) Server VM 1.7.0_09-b05 [linux-i386] invoke-dynamic= false true binarytrees #1 290.08 252.55 binarytrees #2 290.34 252.92 binarytreesredux #2 289.64 262.04 chameneosredux 31 39.85 32.73 fannkuchredux #2 1720.70 1481.23 fasta #6 372.59 418.07 fasta #5 294.54 270.02 fasta #1 278.98 260.14 knucleotide #2 428.75 377.28 knucleotide #1 470.40 461.46 mandelbrot #3 1404.58 1149.08 mandelbrot #1 2301.81 2167.50 meteor #1 17.59 17.42 meteor #2 13.19 13.37 nbody #1 678.62 568.72 pidigits #2 3.10 3.22 pidigits #3 25.89 25.73 pidigits #1 50.80 51.03 regexdna #1 78.76 78.60 regexdna #3 44.29 45.33 revcomp #2 24.05 24.49 spectralnorm #2 462.09 387.54 spectralnorm #1 496.94 438.55 YMMV