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: AAC 160 vs 256 (Read 11585 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

AAC 160 vs 256

All my songs are encoded AAC 256 VBR and I have so much stuff that I can't fit all my songs into my micro sd 8gb. Is a good choice rip everything at AAC 160 VBR? (from a lossless source of course) I listen music with earphones and AAC 160, 256 and ALAC sound the same to me '-' I dont know what to do and google doesnt help

AAC 160 vs 256

Reply #1 –
Is that really an issue given the price of a larger memory card?

AAC 160 vs 256

Reply #2 –
Quote from: pelmazo on
Is that really an issue given the price of a larger memory card?

Yes it is, memory cards are not hard drives.

I use AAC ~96 kbps for my wife's car. Convert few at ~96/~128/~160/etc. and ABX them with the original, you decide.
 

AAC 160 vs 256

Reply #3 –
ABX? what is this?

AAC 160 vs 256

Reply #4 –
You can get twice the capacity you currently have for under 10 bucks! I wouldn't spend much time for doing all the recoding for this kind of money.

AAC 160 vs 256

Reply #5 –
There are 2 answers. Pelmazo is correct that you could simply buy a larger SD card for little expense and not bother re-encoding your library at all. On the other hand, there isn't much point in wasting tons of space on your micro SD card to attain quality levels well beyond your ability to distinguish.

If you really want to stick with your current card then encoding to AAC at 128kbps will yield results that I highly doubt you would ever distinguish from the lossless sources while doubling the amount of music you can store on the card. You can probably push those bitrates right down into the 96kbps range without noticing. So Eahm is right as well.

For info about ABX testing:
http://wiki.hydrogenaud.io/index.php?title=ABX

AAC 160 vs 256

Reply #6 –
If AAC 256 is so high why itunes store uses this setting? I use cuz I thought it was the standard lossy '-'

128kbps seems so low quality :/

AAC 160 vs 256

Reply #7 –
Just because it seems like a low quality setting doesn't mean you can hear it. That's what the ABX test is for.

AAC 160 vs 256

Reply #8 –
they all sound the same to me but i dont know maybe i'll keep everything the way it is (aac 256) just in case... and buy a new micro sd '-'

AAC 160 vs 256

Reply #9 –
Quote from: thatguyraphael on
i dont know

People have tried to help you understand.  It's really not that difficult.

Here's another link about ABX testing, but I think the wiki link provided earlier should be sufficient:
http://www.hydrogenaud.io/forums/index.php?showtopic=16295

You have the lossless originals and can double your capacity just by encoding a new set of files.  It costs you nothing.

If you realize later through ABX testing that 128k is insufficient you can always encode a new set.

AAC 160 vs 256

Reply #10 –
Quote from: thatguyraphael on
they all sound the same to me but...


That's the point. Modern lossy codecs (AAC, Vorbis, Opus, etc) and their encoders are now good enough that in the vast majority of situations they are effectively transparent at relatively low "quality" settings. This means that sitting down and trying to hear a difference (via an ABX test) between the lossy version and the original is exceptionally difficult and often impossible.

If YOU can't hear a difference between 128kbps and 256kbps encodes then anything above 128kbps is just wasted space. There isn't any magic stuff going on that you're missing. If you can't hear a difference then your perceived quality between 128kbps, 256kbps, or the lossless original is the SAME.

I personally encode to OGG Vorbis at around 96kbps (quality level 2). For the most part it's transparent to me. Any rare problem spots where it isn't the difference is so obscure I'll probably never notice it.

AAC 160 vs 256

Reply #11 –
Several years ago I ran an extensive series of ABX sessions using 128k mp3 vs. AIFF and found that with a lot of painstaking mental effort I could just barely distinguish them.  I'm not convinced it was even the encoding, because I was unable to put what I heard into words; it just seemed like there was an extremely subtle difference which was audible on direct comparison.  In fact it was so much effort to make each comparison that I didn't even finish the full 20 tests with each sample; after getting 10/10 on several samples I decided that it wasn't worth the stress to complete 20 and that I would go ahead and use 128k for the portable and listening in the car.  Except for direct comparison, I can't detect any differences.  I normally rip everything at 256k AAC and use those versions for sessions where I'm in my office, more likely to be relaxed and just enjoying the music. I have never felt that there is any lack of perceptible detail in either.

Bottom line: I think I have a pretty good appreciation for detail and subtlety in musical recordings/performances.  If mp3/AAC sounds effectively transparent to me, then either my ears truly suck or the coders had a darn good idea what they were doing when they designed the codecs.  Either way, it's nice not to have to worry about it

AAC 160 vs 256

Reply #12 –
Quote from: Nichttaub on
Bottom line: I think I have a pretty good appreciation for detail and subtlety in musical recordings/performances.  If mp3/AAC sounds effectively transparent to me, then either my ears truly suck or the coders had a darn good idea what they were doing when they designed the codecs.  Either way, it's nice not to have to worry about it


AAC is superior to mp3 (as is Vorbis and Opus). You should retry that test now with AAC, Vorbis, or Opus at 128kbps. If you thought it was hard distinguishing mp3, you're in for a shock.

AAC 160 vs 256

Reply #13 –
I thought the thrust of Opus was to give acceptable results at low bitrates, providing near-transparency. Was it also designed to provide transparency with typical samples at higher bitrates to trained critical listeners?  From what little I've read I've gotten the impression that the answer was no; that it was perhaps along the lines of something like HE-ACC.

AAC 160 vs 256

Reply #16 –
Quote from: yourlord on
Quote from: Nichttaub on
Bottom line: I think I have a pretty good appreciation for detail and subtlety in musical recordings/performances.  If mp3/AAC sounds effectively transparent to me, then either my ears truly suck or the coders had a darn good idea what they were doing when they designed the codecs.  Either way, it's nice not to have to worry about it


AAC is superior to mp3 (as is Vorbis and Opus). You should retry that test now with AAC, Vorbis, or Opus at 128kbps. If you thought it was hard distinguishing mp3, you're in for a shock.


The ABX software I used at the time (ABC-HR?) didn't support AAC decoding for the ABX tests, so I used the HA recommended LAME MP3 encoder.  I will be happy (well, OK, not happy but willing) to re-do the tests with AAC if I can find a tool that works with them.  I don't doubt that AAC will provide excellent results.

AAC 160 vs 256

Reply #17 –
Try the ABX component for foobar2000. You can ABX pretty much any format that way. You can also try python-abx for Linux (via Python) or Lacinato ABX on OSX (via Java).

AAC 160 vs 256

Reply #19 –
The problem with the listening tests (ABC/HR, more specifically) as they relate to this thread is they are preference tests for subjects that are not specifically intended to give transparent results, whereas this thread is primarily concerned with transparency.  Still, they point to a region that perhaps begins to provide transparent results.

ABX is the proper tool to be using to determine if a lossy encoding can be distinguished from its lossless source.

AAC 160 vs 256

Reply #20 –
Quote from: Aleron Ives on
Try the ABX component for foobar2000. You can ABX pretty much any format that way. You can also try python-abx for Linux (via Python) or Lacinato ABX on OSX (via Java).


Thanks!  Lacinato is exactly what I've been looking for, since I use OS X.  It seems to be able to do what I need, so next time I get an opportunity I'll run myself through some AAC 128k ABX tests.

AAC 160 vs 256

Reply #22 –
Quote from: Nichttaub on
I'll run myself through some AAC 128k ABX tests.


Don't limit yourself to 128kbps. If you fail to detect a difference at 128kbps then keep pushing that rate lower until you do. When you are able to start reliably hearing a difference you can encode at a rate just above that level with reasonable confidence.

AAC 160 vs 256

Reply #23 –
Quote from: yourlord on
AAC is superior to mp3 (as is Vorbis and Opus). You should retry that test now with AAC, Vorbis, or Opus at 128kbps. If you thought it was hard distinguishing mp3, you're in for a shock.
Vorbis, Opus, yes. AAC? Not for me. I might have found a killer sample for that, but I had the same impression about AAC with other music as well. If forced into compatibility, I very much prefer MP3.