More strong reason for AndCooper

It is stronger for passing the Google ADT plugins to a build system full this small Benchmark of the Dalvik VM on Java project:

P905i (500 MHz - UniPhier 4 M + JBlend)
@Sieve: 1569
@Loop: 2421
@Logic: 2251
@String: 5109
@Float: 973
@Method: 830
@Overall: 1811
Fact 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 (528 MHz - MSM7201A)
@Sieve: 464
@Loop: 551
@Logic: 394
@String: 491
@Float: 336
@Method: 397
@Overall: 433
Fact sheet:
http://en.Wikipedia.org/wiki/Google_phone

P905i (Dalvik VM on JBlend Java)
@Sieve: 6
@Loop: 3
@Logic: 26
@String: 85
@Float: 3
@Method: 6
@Overall: 9

For those who have not read before compass sales faster high speed. Not only slow speed but also notice that there are no have in the Dalvik on benchmark Android data. This means that Dalvik VM optimized double conversion battery based transfer instructions to register to the base. Thus, a large part of the strategy of optimization of old J2ME which were the basis of dupon eliminating transfer instructions by modifying the structure of the code will not work.

So we need ways to collect some data on the code, so that we can determine some new optimization strategies because the platform Dalvik JIT is still far away at the moment. Well, unless there is today an announcement that I do not know about. Thus, it is one of the main drives for the AndCooper project.

View the original article here