This is the mail archive of the xsl-list@mulberrytech.com mailing list .


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

RE: Latest XSLTMark benchmark



>
>> for some of the tests. To give Java processors a chance to warm
>up I think
>> you should really do some un-timed runs followed by some timed ones where
>> runs is significantly >10.
>
>Good idea... So good that the current version already does that!
>Here is what XSLTMark 2.0 prints out when running with XT.
>
>dummy initialization run:
>xslbench3:          xslbench3.xsl       xslbenchdream.xml   ... done in
>644ms.

This is clearly better than none at all but I doubt its enough for
technologies like Hotspot to really get to grips with a particular
stylesheet. The advise I got was to run *each* test at least 100 times
before starting timing to get maximum Java speed.

There also appears to be a difference depending on if/how the processor
implements output indenting. I just got some better scores by making
indenting the default as apposed to no indenting. I guess the stylesheets
could just be set to turn this off.

Regards,
Kev.

p.s. I had a problem with dgnorm on my Win2k box. The distributed one just
kept crashing but if built manually it was ok.



 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]