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: any1 tried OggEncLC ? (Read 3772 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

any1 tried OggEncLC ?

Hi guys,
actually i've been in this forum once it is got started but never register. but i find some very interesting fact & findings u guys have made. 

ok. recently i downloaded the Intel Compiled Oggenc which came with OggEncLC.    the lossy coupling enabled up to 6.99. 

For me, i can hear stereo image more than the other artifacts.
I think it is a good idea to implement a switch to manipulate the coupling channel.  If not, may be i would like another compile which have lower -q value with loseless coupling. I would like to see how Ogg perform with loseless coupling at lower -q value , lets say....start from -q1 ??  Any comment?

Btw .. i really start to like Ogg. especially in low bitrate.

Tq

any1 tried OggEncLC ?

Reply #1 –
I can email you a version of oggenc with lossless coupling at all -q values. I wouldnt recommend it at all, but if you really want Ill email it.
I guess if youre hearing is very sensitive to stereo diffrences maybe there would be some benefit for you, but it will cause higher bitrates at lower -q settings possibly about 80 kbit/s at -q 0.

If any one else is wanting this then maybe Roberto might want to host this.

any1 tried OggEncLC ?

Reply #2 –
Quote
Originally posted by lithoc
For me, i can hear stereo image more than the other artifacts. 
I think it is a good idea to implement a switch to manipulate the coupling channel.  If not, may be i would like another compile which have lower -q value with loseless coupling. I would like to see how Ogg perform with loseless coupling at lower -q value , lets say....start from -q1 ??   Any comment?


That's cool. Can you do a comparison of -q 5 between the lossy and lossless stereo mode and describe (hopefully with ABX) any differences in quality you can hear?

That would be interesting result for us to see.
Thanks

any1 tried OggEncLC ?

Reply #3 –
Quote
Originally posted by Alex
If any one else is wanting this then maybe Roberto might want to host this.


No problem with me.

Actually, I'm not caring that much about my quota any more. It will be increased soon. What scares me is, the more programs I add, the worst my page seems.

Gotta fix it soon...

Regards;

Roberto.

any1 tried OggEncLC ?

Reply #4 –
Ill have to post quickly for now, here are some EAQUAL results the first is with lossless channel coupling and the other two are standard RC3.

Bitrate                      ODG
135   -q 3   -0.5714
132   -q 4.99   -0.5257
118   -q 3   -0.6901

at the same bitrate(132/135) the quality is very close.
On the sample i choose, which was quite a hard sample, lossless coupling didnt have any especially strong atrifacts but there was more of a constant slight hiss/noise. with lossy coupling there were a few artifacts at certain points and lower hiss/noise.

Its possible to adjust the lossyness of lossy coupling, only in the source code at the moment. I believe Monty is going to add some control of coupling in RC4 to what level I do not know.

Ill update this post later when I have more time.

any1 tried OggEncLC ?

Reply #5 –
Oh it's really a interesting result which shows that lossless coupling don't always do good to quality. This result really is the same as Dibrom's with LAME stuff. Dibrom eventually use a dynamical coupling strength to achieve best quality/size ratio with "standard" preset. Hope Ogg will do similar thing to achieve better quality -- especially on low bitrate encoding.

any1 tried OggEncLC ?

Reply #6 –
hey... the files is available in rjamorim page. Just downloaded it and encoded 2 albums with it. Now listening to it actually. For starting I use -q1 and slowly raise the q value only if i've time to test it out.

As expected the bitrate will raise higher than nominal bitrate(80~100).
The quality is not much different, but i'm happy some of the songs sound more dynamic as it use loseless stereo.

Overall i can still hear some artifacts(at least i know it is not from lossy stereo) and frequency cut-off at -q1. If only Oggenc have higher cut-off. 

Personally I used to encode using -q5, i think its a bit overkill for normal pop music.

Keep up the good works guys