For your information, I send you a performance test result of Embedded CaffeineMark on my mobile phones. P905i + JVM(JBlend) is the fastest in three tests and the Dalvik VM Interpreter on Java is very slow.
P905i (500MHz - UniPhier 4M + JBlend) Sieve: 1569 Loop: 2421 Logic: 2251 String: 5109 Float: 973 Method: 830 Overall: 1811 Spec sheet: http://ja.wikipedia.org/wiki/P905i http://panasonic.co.jp/corp/news/official.data/data.dir/jn080716-2/jn080716-2.html Google Dev Phone 1 (528MHz - MSM7201A) Sieve: 464 Loop: 551 Logic: 394 String: 491 Float: 336 Method: 397 Overall: 433 Spec sheet: http://en.wikipedia.org/wiki/Google_phone P905i (Java Dalvik VM on JBlend) Sieve: 6 Loop: 3 Logic: 26 String: 85 Float: 3 Method: 6 Overall: 9 --- Koji Hisano On Apr 2, 9:42 pm, blindfold <[email protected]> wrote: > Could be fun to see the performance of a CPU intensive Android > function, after compilation to Dalvik bytecodes, and executed through > your Java implementation of it, running on a Java ME mobile phone with > JIT compiler for the Java VM, and compare that to the performance of > running the same Dalvik bytecodes in the usual way with the Dalvik > interpreter on the G1. In theory the JIT compiler might compensate for > inefficiencies added by the Dalvik_bytecode-to-Java-to-JVM_bytecode > translations, as long as those added inefficiencies do not cause a > speed penalty larger than about a factor 5. :-) > > On Apr 1, 4:56 pm, hisano <[email protected]> wrote: > > > I developed a Dalvik VM implementation on the Java > > platform.http://code.google.com/p/android-dalvik-vm-on-java/ > > > --- > > Koji Hisano --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Android Discuss" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/android-discuss?hl=en -~----------~----~----~----~------~----~------~--~---
