133794m3r
May 5th, 2010, 03:06 AM
Firefox is the build that comes with ubuntu(might be why it's faster than swiftfox), and swiftfox latest aka 3.6.3 I've also cleared all of the cache before doing either test to try to make it so that no results were skewed.
Here is swiftfox amd64's results.
Swiftfox results (http://www2.webkit.org/perf/sunspider-0.9/sunspider-results.html?%7B%223d-cube%22:%5B58,62,63,60,63%5D,%223d-morph%22:%5B31,31,30,31,31%5D,%223d-raytrace%22:%5B77,79,85,78,85%5D,%22access-binary-trees%22:%5B54,49,48,50,49%5D,%22access-fannkuch%22:%5B81,79,80,83,91%5D,%22access-nbody%22:%5B41,41,36,40,35%5D,%22access-nsieve%22:%5B15,15,15,24,14%5D,%22bitops-3bit-bits-in-byte%22:%5B2,2,2,2,2%5D,%22bitops-bits-in-byte%22:%5B12,11,11,11,11%5D,%22bitops-bitwise-and%22:%5B3,2,2,3,2%5D,%22bitops-nsieve-bits%22:%5B36,32,37,31,32%5D,%22controlflow-recursive%22:%5B41,41,49,45,41%5D,%22crypto-aes%22:%5B38,39,35,43,39%5D,%22crypto-md5%22:%5B17,16,17,16,17%5D,%22crypto-sha1%22:%5B9,9,9,9,10%5D,%22date-format-tofte%22:%5B125,131,128,124,124%5D,%22date-format-xparb%22:%5B79,83,79,84,78%5D,%22math-cordic%22:%5B16,12,20,20,16%5D,%22math-partial-sums%22:%5B29,28,28,28,29%5D,%22math-spectral-norm%22:%5B8,8,8,8,8%5D,%22regexp-dna%22:%5B59,61,61,65,71%5D,%22string-base64%22:%5B13,13,14,13,14%5D,%22string-fasta%22:%5B87,90,89,88,94%5D,%22string-tagcloud%22:%5B107,116,113,111,114%5D,%22string-unpack-code%22:%5B118,112,113,113,112%5D,%22string-validate-input%22:%5B42,41,36,35,44%5D%7D)
Firefox Results (http://www2.webkit.org/perf/sunspider-0.9/sunspider-results.html?{%223d-cube%22:[102,113,102,104,105],%223d-morph%22:[96,102,97,96,101],%223d-raytrace%22:[101,98,98,99,103],%22access-binary-trees%22:[42,42,49,43,42],%22access-fannkuch%22:[167,171,168,175,167],%22access-nbody%22:[114,117,107,109,119],%22access-nsieve%22:[47,47,47,51,46],%22bitops-3bit-bits-in-byte%22:[46,47,46,50,48],%22bitops-bits-in-byte%22:[73,73,73,72,72],%22bitops-bitwise-and%22:[65,62,68,62,61],%22bitops-nsieve-bits%22:[84,89,90,83,89],%22controlflow-recursive%22:[37,36,38,38,38],%22crypto-aes%22:[64,62,59,66,60],%22crypto-md5%22:[46,46,48,47,49],%22crypto-sha1%22:[49,48,49,49,51],%22date-format-tofte%22:[95,92,93,93,92],%22date-format-xparb%22:[99,102,99,99,98],%22math-cordic%22:[118,118,116,124,127],%22math-partial-sums%22:[102,100,99,119,106],%22math-spectral-norm%22:[51,50,61,57,50],%22regexp-dna%22:[229,225,226,229,214],%22string-base64%22:[46,43,48,44,43],%22string-fasta%22:[119,116,117,112,115],%22string-tagcloud%22:[127,124,140,142,149],%22string-unpack-code%22:[173,156,209,212,243],%22string-validate-input%22:[62,67,70,78,79]})
Firefox the base one on 10.04 is ~1.2x faster at its slowest and ~5x faster at its fastest. This leads me to believe that the swiftfox build has some serious javascript engine changes to it or something similar to warrant this extreme speed decrease. Also, can anyone else out there do a similar test on your system so that we can get some more benchmarks/comparisons going for a more legitimate results thread so that the findings will be more likely to be correct.
Here is swiftfox amd64's results.
Swiftfox results (http://www2.webkit.org/perf/sunspider-0.9/sunspider-results.html?%7B%223d-cube%22:%5B58,62,63,60,63%5D,%223d-morph%22:%5B31,31,30,31,31%5D,%223d-raytrace%22:%5B77,79,85,78,85%5D,%22access-binary-trees%22:%5B54,49,48,50,49%5D,%22access-fannkuch%22:%5B81,79,80,83,91%5D,%22access-nbody%22:%5B41,41,36,40,35%5D,%22access-nsieve%22:%5B15,15,15,24,14%5D,%22bitops-3bit-bits-in-byte%22:%5B2,2,2,2,2%5D,%22bitops-bits-in-byte%22:%5B12,11,11,11,11%5D,%22bitops-bitwise-and%22:%5B3,2,2,3,2%5D,%22bitops-nsieve-bits%22:%5B36,32,37,31,32%5D,%22controlflow-recursive%22:%5B41,41,49,45,41%5D,%22crypto-aes%22:%5B38,39,35,43,39%5D,%22crypto-md5%22:%5B17,16,17,16,17%5D,%22crypto-sha1%22:%5B9,9,9,9,10%5D,%22date-format-tofte%22:%5B125,131,128,124,124%5D,%22date-format-xparb%22:%5B79,83,79,84,78%5D,%22math-cordic%22:%5B16,12,20,20,16%5D,%22math-partial-sums%22:%5B29,28,28,28,29%5D,%22math-spectral-norm%22:%5B8,8,8,8,8%5D,%22regexp-dna%22:%5B59,61,61,65,71%5D,%22string-base64%22:%5B13,13,14,13,14%5D,%22string-fasta%22:%5B87,90,89,88,94%5D,%22string-tagcloud%22:%5B107,116,113,111,114%5D,%22string-unpack-code%22:%5B118,112,113,113,112%5D,%22string-validate-input%22:%5B42,41,36,35,44%5D%7D)
Firefox Results (http://www2.webkit.org/perf/sunspider-0.9/sunspider-results.html?{%223d-cube%22:[102,113,102,104,105],%223d-morph%22:[96,102,97,96,101],%223d-raytrace%22:[101,98,98,99,103],%22access-binary-trees%22:[42,42,49,43,42],%22access-fannkuch%22:[167,171,168,175,167],%22access-nbody%22:[114,117,107,109,119],%22access-nsieve%22:[47,47,47,51,46],%22bitops-3bit-bits-in-byte%22:[46,47,46,50,48],%22bitops-bits-in-byte%22:[73,73,73,72,72],%22bitops-bitwise-and%22:[65,62,68,62,61],%22bitops-nsieve-bits%22:[84,89,90,83,89],%22controlflow-recursive%22:[37,36,38,38,38],%22crypto-aes%22:[64,62,59,66,60],%22crypto-md5%22:[46,46,48,47,49],%22crypto-sha1%22:[49,48,49,49,51],%22date-format-tofte%22:[95,92,93,93,92],%22date-format-xparb%22:[99,102,99,99,98],%22math-cordic%22:[118,118,116,124,127],%22math-partial-sums%22:[102,100,99,119,106],%22math-spectral-norm%22:[51,50,61,57,50],%22regexp-dna%22:[229,225,226,229,214],%22string-base64%22:[46,43,48,44,43],%22string-fasta%22:[119,116,117,112,115],%22string-tagcloud%22:[127,124,140,142,149],%22string-unpack-code%22:[173,156,209,212,243],%22string-validate-input%22:[62,67,70,78,79]})
Firefox the base one on 10.04 is ~1.2x faster at its slowest and ~5x faster at its fastest. This leads me to believe that the swiftfox build has some serious javascript engine changes to it or something similar to warrant this extreme speed decrease. Also, can anyone else out there do a similar test on your system so that we can get some more benchmarks/comparisons going for a more legitimate results thread so that the findings will be more likely to be correct.