Below is my original message with the test description and the results. I was using java 1.3.1 and batik 1.1.1.
Later, I have tried with the latest batik (got the source and built it) and also with java 1.4, but the results are so close to these that I won't look for them now.
Cheers
 
Milan
=========================================
Milan Trninic
Software Engineer
tel: 1 604 484-2764, 484-2750
[EMAIL PROTECTED]
Galdos Systems Inc. http://www.galdosinc.com
=========================================

===============
Privileged or confidential information may be contained in this message. If this message was not intended for you, destroy it and notify us immediately. Opinions, conclusions, recommendations, and other information presented in this message are not given or necessarily endorsed by my employer or firm.

----- Original Message -----
Sent: Tuesday, February 12, 2002 13:27
Subject: Rendering performance of Java

 
I made a simple test that loads an svg file of certain number of squares (paths) formed into a matrice. I was varying the number from 100 to 10, 000. This is what I've figured out:
 
1. I was using:
 
    _factory.createDocument
    to create the document out of url, and then
 
    _svgPane.setSVGDocument
 
to set the documtn to JSVGComponent.
 
2. I set the timer to capture ellapsed time after all processing done by my code is done.
 
3. Turns out the speed of Batik is very good. For example, with 10,000 squares it takes 3 seconds for batik to do it. So, basically for the varying number of <path> elements displayed (100 - 10, 000) the time spent was 0-3 sec.
4. But then goes the complicated part - After it was announced that the setting of the svg document is done, and basically my small test program has finished, java starts rendering that on the screen (so nothing is visible yet), and that part is trouble. Here are the times for it:
 
1000 items 4 sec
2000 items 8 sec
3000 items 15 sec
4000 items 30 sec
5000 items 65 sec
 
 

Reply via email to