For comparison, I've now rerun my test on my Athlon at home, and have
obtained very similar results.  I expected a bit of variation in the
times, but what I didn't expect was a difference in the compressed
output FLAC files, since the input WAV file was the same in both cases.
This means that the same executable (FLAC 1.1.4) gives different
results depending on the processor that it runs on.  The obvious
question is, if I expand the files back up to WAV, do I get the same
original file in every case.  That's my next test.

0       00:00:16        288536
1       00:00:17        259350
2       00:00:19        258221
3       00:00:20        274497
4       00:00:24        244590
5       00:00:30        243261
6       00:00:31        243234
7       00:01:09        242093
8       00:01:32        239447

I've also rerun the test (again on my Athlon at home), but this time
using the lossless original of the CD, to see if the missing detail
from the OGG-orginated WAV file had an effect).

0       00:00:15        295080
1       00:00:16        269092
2       00:00:19        267969
3       00:00:20        283097
4       00:00:23        257177
5       00:00:29        256038
6       00:00:30        255979
7       00:01:11        255055
8       00:01:34        253869

The compression times are almost the same, but the resulting compressed
files are between 2% and 6% bigger, even though the difference in the
size of the two source WAV files was negligible.

It's also notable that compression level 3 gives consistently less
compression than level 1 - a bug in the compressor perhaps?  I suppose
I need to test this with some different CDs to be sure.


-- 
chill
------------------------------------------------------------------------
chill's Profile: http://forums.slimdevices.com/member.php?userid=10839
View this thread: http://forums.slimdevices.com/showthread.php?t=34156

_______________________________________________
audiophiles mailing list
[email protected]
http://lists.slimdevices.com/lists/listinfo/audiophiles

Reply via email to