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: Can lossy codecs be better optimised at some bitrates than others? (Read 1792 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Can lossy codecs be better optimised at some bitrates than others?

So, for example, 128kbps sounds better than 160kbps despite 160kbps being higher due to more optimisation work?

If so, please provide examples.

Re: Can lossy codecs be better optimised at some bitrates than others?

Reply #1 –
You can go through the Listening Tests forum to see the results and discussions to help answer your question.

For example, in this test Personal test - MP3 vs AAC (high rates, VBR), the results show that Apple's AAC at ~192kbps performs better than MP3 LAME ~192kbps, ~260kbps & 320kbps.

Re: Can lossy codecs be better optimised at some bitrates than others?

Reply #2 –
Yes.   AAC is supposed to be better than MP3 and from what I've read OGG may be better too.   That was the goal of AAC.    By "better" I mean the same sound quality at a lower bitrate or better sound at the same bitrate.

But there is the POSSIBILITY that song-A sounds better with MP3 and song-B sounds better with AAC, etc. 

Of course ,if your Mp3 is transparent at 160kbps (sounds identical to the uncompressed original in a blind listening test) you can't get better sound quality.  ;)

...Some codecs are designed for the best sound quality and some are optimized for the lowest bitrate while achieving speech intelligibility.

Re: Can lossy codecs be better optimised at some bitrates than others?

Reply #3 –
Perhaps I'm mistaken but I understood the question to mean within a codec. I thought it is well known that good modern codecs can easily beat old bad ones even at lower bitrates.
And if my understanding is correct, giving a codec more bits to use should always increase quality. Though a bug in the encoder could cause an exception to this rule, see for example https://hydrogenaud.io/index.php/topic,85135.msg1056191.html#msg1056191. Apple's AAC encoder has a bug where 80 kbps CBR is worse than 64 kbps CBR as 80 kbps causes very audible glitches.

Re: Can lossy codecs be better optimised at some bitrates than others?

Reply #4 –
Quote
And if my understanding is correct, giving a codec more bits to use should always increase quality.
Right.  At lower bitrates more data is being thrown-away so it's "more lossy".

But, if you can achieve transparency higher bitrates can't be "better",  Or, you may run into limitations/artifacts that are a characteristic of the particular coded and don't improve with higher bitrates.'

Re: Can lossy codecs be better optimised at some bitrates than others?

Reply #5 –
Original question has several interpretations and it is not obvious that the OP has one particular in mind (it might be that they would rather have answers they haven't thought of, and that is fine).

In addition to outright bugs like pointed out above, there have been some discussions for MP3 encoders.
* https://hydrogenaud.io/index.php/topic,126120.0.html
* LAME V0 vs CBR320, at least one sound clip appears to have more artifacts on CBR320 to trained ears:  https://hydrogenaud.io/index.php/topic,121579.msg1003462.html#msg1003462

But note, it is possible to "repack" a low-bitrate MP3 into higher-bitrate MP3, losslessly. Proving that the best possible 192kbit/s MP3 (of a particular signal) cannot be worse than the best 96 kbit/s - just repack the latter as CBR 192, and waste the space, but it sounds the same. The process is even reversible.

Re: Can lossy codecs be better optimised at some bitrates than others?

Reply #6 –
Quote from: Case on
Perhaps I'm mistaken but I understood the question to mean within a codec. I thought it is well known that good modern codecs can easily beat old bad ones even at lower bitrates.
And if my understanding is correct, giving a codec more bits to use should always increase quality. Though a bug in the encoder could cause an exception to this rule, see for example https://hydrogenaud.io/index.php/topic,85135.msg1056191.html#msg1056191. Apple's AAC encoder has a bug where 80 kbps CBR is worse than 64 kbps CBR as 80 kbps causes very audible glitches.

Things like that 80kbps bug are exactly what I'm talking about. My question is about different bitrates with the same encoder.

Re: Can lossy codecs be better optimised at some bitrates than others?

Reply #7 –
Why are you asking this? What's your motivation? Many useless questions recently, but possibly I'm just not seeing it.
 

Re: Can lossy codecs be better optimised at some bitrates than others?

Reply #8 –
Quote from: DVDdoug on
Yes.   AAC is supposed to be better than MP3 and from what I've read OGG may be better too.   That was the goal of AAC.    By "better" I mean the same sound quality at a lower bitrate or better sound at the same bitrate.

But there is the POSSIBILITY that song-A sounds better with MP3 and song-B sounds better with AAC, etc. 

Of course ,if your Mp3 is transparent at 160kbps (sounds identical to the uncompressed original in a blind listening test) you can't get better sound quality.  ;)

...Some codecs are designed for the best sound quality and some are optimized for the lowest bitrate while achieving speech intelligibility.

Well LAME MP3 & Apple AAC don't have the same perceptual model. So It very likely someone could get 4.9 ~ 5 at V3 ~ V0 but they get 4.25 ~ 4.65 with Q82 ~ Q109.
LAME MP3 at V2 ~ V1 | FLAC for archiving |