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: Equivalent Bitrates - mp3 and OGG (Read 13073 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Equivalent Bitrates - mp3 and OGG

Ok guys, I'm new here, so please don't kill me if I've done something wrong by posting this... 

Can anyone give me some sort of table to show me what mp3 bitrates are equivalent to certain ogg bitrates? Something like 128kbps ogg = 192kbps mp3, for example.

Thanks,

nazgulord.

Equivalent Bitrates - mp3 and OGG

Reply #1 –
The only way to tell equivalence without a doubt is to ABX the track in question at the formats and bitrates in question, as far as my knowledge affords me.
EAC>1)fb2k>LAME3.99 -V 0 --vbr-new>WMP12 2)MAC-Extra High

Equivalent Bitrates - mp3 and OGG

Reply #2 –
Quote
Ok guys, I'm new here, so please don't kill me if I've done something wrong by posting this... 

Can anyone give me some sort of table to show me what mp3 bitrates are equivalent to certain ogg bitrates? Something like 128kbps ogg = 192kbps mp3, for example.[{POST_SNAPBACK}][/a]

I don't think anyone can. That kind of statement would violate the forum terms unless backed up with conclusive listening test results of every bitrate pair. The test should be something like the guruboolez's newest test: [a href="http://www.hydrogenaudio.org/forums/index.php?showtopic=38792]Personal evaluation at ~130..135 kbps, 200 samples[/url]. Even that test is valid only for the exact encoder versions and settings used.

Equivalent Bitrates - mp3 and OGG

Reply #3 –
Thanks for the info and prompt replies. 

LANjackal, what is ABX? 

Thanks,

nazgulord.

Equivalent Bitrates - mp3 and OGG

Reply #5 –
Quote
Ok guys, I'm new here, so please don't kill me if I've done something wrong by posting this... 

Can anyone give me some sort of table to show me what mp3 bitrates are equivalent to certain ogg bitrates? Something like 128kbps ogg = 192kbps mp3, for example.

Thanks,

nazgulord.
[a href="index.php?act=findpost&pid=342105"][{POST_SNAPBACK}][/a]


First of all, Vorbis and MP3 are only formats defined by a specification describing how to interpret a bitstream. These specifications don't force encoders to use a specific bit allocation strategy -- so, if you're going to implement a lossy encoder for one of those formats you have to build something we all call "psychoacoustic model". Fact is, the current models (used by LAME (mp3 encoder) and libVorbis (Vorbis encoder)) differ greatly. (For example: all Vorbis encoders I tested tend to use higher SNRs in the lower frequency region and lower SNRs in the higher frequency region compared to LAME). To answer your question correctly you first need to define "equality".

What definition of "equality" makes sense ? Perhaps this one: Two encoders behave equally if they quantize the signal with similar signal-to-noise ratios on all frequency/time regions. That'd be a simple definition. Others would be possible as well -- maybe incorporating stereo image stability and stuff.

Examining your example: Ogg Vorbis 128 kbps -- produced by any current Ogg Vorbis encoder -- versus LAME at 192 kbps (ABR):
As mentioned earlier the Vorbis encoder tends to quantize the highs more heavily whereas LAME tends to quantize the lower frequencies more heavily. Also, I'd like to mention the usage of intensity stereo (or "point stereo" -- exact same idea, other name *) within the Vorbis encoder at above 8 kHz whereas LAME preserves interchannel phase correlations and only adds quantization noise in the data reduction step. So, using the above definition of equality: They aren't. And they can't be unless at least one of the parties (LAME devs or Vorbis-Enc devs) change their psy model.

If by equality you mean the score of an ABCHR test for a bunch of test samples is equal -- this is another story. However this can only give you a hint for what the formats are capable of because format and encoder are two different things (I cannot emphasize this enough). Perhaps you remember the early days of AAC. There was hardly a reason to prefer AAC (due to early, untuned encoders) even though the format is technically superior to MP3. The situation now changed quite a bit.

(* Monty seems to have great fun to devise different names for the same stuff. He still says that Vorbis (in general) works totally differently compared to MP3 everytime someone asks about the differences -- wow! -- I guess, we first need to define an abstraction level we work on, if we want to answer whether Vorbis works totally differently or not)

bye,
Sebi

edit: fixed a typo. Consider all remaining ones a feature.

Equivalent Bitrates - mp3 and OGG

Reply #6 –
maybe i can offer a slightly more practical response....

transparency is the point at which the quality of the lossy encoded file is high enough, that the listner can not longer tell it from the source.  for me lame mp3's are transparent at -V3 and above (i can abx -V4 and below).  vorbis files on the other hand, are transparent for me -q 4 and above.  therefore, for me, vorbis -q4 (about 128 kbps) is more or less "equivalent" to lame -V3 (about 175 kbps).

keep in mind though, that this will be different for differnt people.  i strongly recommend that you try some abx tests for yourself; you would probably learn alot and then you should share your results with all of us.    good luck

edit: typo
a windows-free, linux user since 1/31/06.

Equivalent Bitrates - mp3 and OGG

Reply #7 –
From Wikipedia:
Quote
Many users feel that Vorbis reaches transparency (sound quality that is indistinguishable from the original source recording) at a quality setting of -q5, approximately 160 kbit/s. For comparison, it is commonly felt that MP3 reaches transparency at around 192 kbit/s (except for the frequency range, which only occasionally exceeds 16kHz), resulting in larger file sizes for the same sound quality.


So transparency is surelly the point of comparsion, as VCSkier said.
But there are two problems:
1. transparency is often a personal perception.
2. Different bitrates should mean different efficiency (for example Vorbis is more efficent on low bitrates than MP3).

Anyway if you'll be able to compile a table, i'll surelly use it in my thesis. Good luck.