my crude benchmark results

James Damour jdamour@nycap.rr.com
Tue Jul 13 12:43:00 GMT 2004


On an Athlon 2600+ with 512Mb of RAM (with athcool v0.3.5 active), I got
the results listed below.  I've also scaled everything by the time taken
in C (8.5 seconds on Mr. Gabriel's machine, 3.229 on mine) and listed it
as "scaled".  Using the scaled numbers, I saw that Java interpreters
actually do worse on my system (relatively speaking), so I tried to
eliminate the "startup" penalty by moving to 263 iterations (the secons
set of results).  Obviously, the startup penalty is *not* that big a
factor; I must assume that my Java interpreters are inferior (but I gain
some satisfaction from having an Athlon 2600+ to make up for it:).

--- 100 iterations ---
GCC v3.3.4 (gcc -o c.exe -O2 -lm benchmark.c; time ./c.exe)
real    0m3.229s
user    0m3.178s
sys     0m0.002s
scaled    8.5  s

GCJ v3.3.4 (gcj -o java3.3.4.exe -O2 --main=Main Main.java; time
./java3.3.4.exe)
real    0m6.444s
user    0m6.367s
sys     0m0.010s
scaled    16.963s

Blackdown Java v1.3.1 (javac Main.java; time java Main)
real    0m10.217s
user    0m9.925s
sys     0m0.035s
scaled    26.895  s

GCJ v3.3.4 Interpreter (gcj -C Main.java; time gij Main)
real    0m12.255s
user    0m12.072s
sys     0m0.006s
scaled    32.260  s

--- 263 iterations ---
GCC v3.3.4 (gcc -o c.exe -O2 -lm benchmark.c; time ./c.exe)
real    0m8.423s
user    0m8.385s
sys     0m0.003s

GCJ v3.3.4 (gcj -o java3.3.4.exe -O2 --main=Main Main.java; time
./java3.3.4.exe)
real    0m16.651s
user    0m16.556s
sys     0m0.010s

Blackdown Java v1.3.1 (javac Main.java; time java Main)
real    0m25.892s
user    0m25.771s
sys     0m0.023s

GCJ v3.3.4 Interpreter (gcj -C Main.java; time gij Main)
real    0m31.734s
user    0m31.640s
sys     0m0.009s

P.S. Looking at the numbers, I think that Mr. Gabriel didn't use GCJ to
"ahead of time" compile the java code. 

P.P.S Mr. Gabriel, when you post your code where others can read it,
kindly include *some* sort of license.  Otherwise no one can legally
display your code due to copyright limitations.
-- 
James Damour (Suvarov454) <suvarov454@users.sourceforge.net>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
URL: <http://gcc.gnu.org/pipermail/java/attachments/20040713/4b1d2968/attachment.sig>


More information about the Java mailing list