RE: [xsl] [Ann] jd.xslt - a XSLT 1.1 processor

Subject: RE: [xsl] [Ann] jd.xslt - a XSLT 1.1 processor
From: "Kevin Jones" <kjouk@xxxxxxxxxxx>
Date: Mon, 5 Mar 2001 11:17:16 -0000
This really does look like a good performer, here are some direct
comparisons for the interested,

On XSLTMark 1.2.1
	jd.xslt 	439.39
	MSXML3	409.06
	Napa		404.13

The jd.xslt XSLTMark driver code does bend the rules slightly in opening the
output file outside of the timing loop. I don't think any of the other
drivers do this so I corrected this before running the times.  Timing was
done on a P550Mhz with 192Mb running Win2k.

jd.xslt performs particularly well on the highly recursive tests with string
manipulation, notably the queens, reverser and bottles tests. If you
interested in the details drop me a line and I can send you the results in a
excel spreadsheet.

If you remove some of the obviously unusual tests (very subjective, see
below) than you get the following figures.

	jd.xslt - 456.87
	MSXML3 - 474.99
	Napa - 531.94

Removed tests
	bottles - Bottles of beer verse generator
	encrypt - Simple encryption
	queens - Chess queen positioning solver
	reverser - Reverse words in text
	tower - Tower problem solver

Whichever way you look at it is a very good first release.

Disclaimer & shameless plug:

Before anybody complains Napa is my own work (see
http://www.tfi-technology.com/xml/napa.html) and it does not perform
exceptionally well on the tests I removed which is why I think they are not
typical ;-)

Regards,
Kev.




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


Current Thread