JProfiler Get rid of your performance problems and memory leaks! Modern Garbage Collection Tuning Java Performance Training Courses COURSES AVAILABLE NOW. We can provide training courses to handle all your Java performance needs Java Performance Tuning, 2nd ed The classic and most comprehensive book on tuning Java Java Performance Tuning Newsletter Your source of Java performance news. Subscribe now! Enter email: Training online Threading Essentials course JProfiler Get rid of your performance problems and memory leaks!

Back to newsletter 057 contents

James Gosling comments on sin() and cos() performance. The summary is that they are slow because of the implementation and he goes on to say "This error has tragically become un-fixable because of the compatibility requirements".

But I don't understand why that's the attitude, we already have java.lang.Math and java.lang.StrictMath, why can't we have java.lang.FastMath? There are no issues in doing so, it could clearly state that the impementation would provide a fast but slightly inaccurate implementation. Then, instead of James' tragic last comment "This is one of those area where no matter what we do, we're screwed.", he could say "This is one of those areas where we can please everyone at the same time".

A note from this newsletter's sponsor

Oh, and if anyone reading this has a hotline to James, do ask him to to please get rid of that background. Interesting it may be, but it makes the text much more difficult to read.

As usual in the newsletter we list our usual raft of articles, performance news, items and more. And, naturally, we have many new performance tips extracted in concise form.

A note from this newsletter's sponsor

News

Java performance tuning related news.

• clone is much slower than new - use use new() with System.arraycopy() to copy arrays if speed matters
• Windows pages your application even when there is lots of free RAM, can hit performance especcialy on full GCs, and a workaround.
• Input for the next version of servlet report. Useful discussion thread is you need to consider factors when measuring application performance
• performance improvements in Mustang OGL
• Java To EXE - Why, When, When Not And How
• Sun submits SPECjAppServer2004 score on free stack

A note from this newsletter's sponsor

Tools

• toolkit to thread JSP tags
• JMS Performance testing harness

Articles

• Scalable icons with Java2D (Kirill Grouchnikov/java.net)
• Grizzly: An HTTP Listener Using Java Technology NIO (Jean-Francois Arcand/java.net)
• Java Technology Performance Myths (Janice J. Heiss, Clifford Click/java.net)
• High Performance Testing (Scott Barber/Better Software Magazine)
• What is faster - LinkedList of ArrayList? (Heinz Kabutz/JavaSpecialists)
• J2EE Clustering (Wang Yu/TheServerSide)
• The Collections Framework in Tiger (John Zukowski/DeveloperWorks)
• XML Performance Checklists (Atsushi Eno/monogatari)
• JDBC 4.0 (John Goodson/JavaDevelopersJournal)

Jack Shirazi

Back to newsletter 057 contents