Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: converter in 0.9b is multithreaded! (Read 4133 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

converter in 0.9b is multithreaded!

Until now I had only used the new converter in foobar2000 in single cpu machines, without multi-threading. Tonight, I just upgraded foobar2000 to 0.9b10 on my fast workstation, with hyperthreading, and started a big encoding job.

Imagine my surprise seeing 2 files being converted at once! First I thought I made a mistake and somehow managed to launched 2 instances. Stopped the process and restarted it carefully, and same again :-)

Task manager confirms there are 2 instances of the encoder running, one on each (virtual) CPU. This is great, especially since it works without any configuration required. Thanks.

converter in 0.9b is multithreaded!

Reply #1 –
Quote
Until now I had only used the new converter in foobar2000 in single cpu machines, without multi-threading. Tonight, I just upgraded foobar2000 to 0.9b10 on my fast workstation, with hyperthreading, and started a big encoding job.

Imagine my surprise seeing 2 files being converted at once! First I thought I made a mistake and somehow managed to launched 2 instances. Stopped the process and restarted it carefully, and same again :-)

Task manager confirms there are 2 instances of the encoder running, one on each (virtual) CPU. This is great, especially since it works without any configuration required. Thanks.
[a href="index.php?act=findpost&pid=333790"][{POST_SNAPBACK}][/a]


cool 
on a pentium 4 D

converter in 0.9b is multithreaded!

Reply #2 –
Pity that HyperThreading isn't as hot as Intel would like everyone to think it is.

Yah, running two encoding processes on the same P4 HT results in about 10% performance boost, whereas the same thing on Athlon64 X2 results in near 100% performance boost. So I guess that would make a P4 D about 100% over a single process. Maybe. I don't know. The known P4 HT results were also more likely to be concurrent ReplayGain processes, rather than encoding.

Would anyone care to benchmark the concurrent processing, using various source/target formats? ( MP3, MusePack, 099 Verbis, FLAC, WavPack, whatever floats your boat )

converter in 0.9b is multithreaded!

Reply #3 –
Just a very quick test using an old P4 3 Ghz with HT, don't have time for more:

Converting mp3 to mp3 (yeah, I know, this is stupid...): 15 sec, so this makes 30 sec for two mp3s

Converting 2 identical mp3s at the same time using HT: 24 sec