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: Personal Blind Listening Test of Bluetooth codecs (Read 59892 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Re: Personal Blind Listening Test of Bluetooth codecs

Reply #26 –
Thank you!
And turns out I'm not quite deaf yet. Trumpet sample indeed gets obliterated by SBC, when listening it through certain cheap earbuds with a smaller bitpool limit (37).
Meanwhile, whatever the latest Android uses for AAC, is no longer so broken, I don't really hear any problem with it on any samples, at least it's not worse than starved SBC.
a fan of AutoEq + Meier Crossfeed

Re: Personal Blind Listening Test of Bluetooth codecs

Reply #27 –
Pardon my ignorance but, is this test a valid representation of the real-world implementation of BT codec transmission? Since the actual devices need to encode the source stream and transmit the output simultaneously in real time.

Re: Personal Blind Listening Test of Bluetooth codecs

Reply #28 –
Quote from: Palladium on
Pardon my ignorance but, is this test a valid representation of the real-world implementation of BT codec transmission? Since the actual devices need to encode the source stream and transmit the output simultaneously in real time.
I use a Momentum 4 with a cheap QCC3056 chip blutooth transmitter and it connects with aptx HD that should even transmit 576 kbit/s so results may be different.
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

Re: Personal Blind Listening Test of Bluetooth codecs

Reply #29 –
Just forgive me about making this silly question, but I just can't be silent about it:

Why are LOSSY codecs used for Bluetooth audio?

(I know there are some explanations easily searchable in internet, but I think that here I will get direct and accurate answers  :D)

Re: Personal Blind Listening Test of Bluetooth codecs

Reply #31 –
A sidenote to my use of the Momentum 4. I never was really convinced by their sound and the cause was not blutooth. It was the same wired.
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

Re: Personal Blind Listening Test of Bluetooth codecs

Reply #33 –
I had an upgrade lately to blutooth aptx lossless. It transmits at ~1024kbps for easy to compress music and mixer set to 24/44.1. Loud Metal sometimes spikes above 1100kbps. White noise maximum is ~1120kbps.
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

Re: Personal Blind Listening Test of Bluetooth codecs

Reply #34 –
Quote from: Wombat on
I had an upgrade lately to blutooth aptx lossless. It transmits at ~1024kbps for easy to compress music and mixer set to 24/44.1. Loud Metal sometimes spikes above 1100kbps. White noise maximum is ~1120kbps.
Interesting finding. The newer version of aptx lossless supports 24/48 lossless according to qualcome website. May I know how to monitor real-time biterate of bluetooth in android.

Re: Personal Blind Listening Test of Bluetooth codecs

Reply #35 –
No idea if there is a possibility to check blutooth bitrate in Android. I have a win app for the blutooth dongle that shows the info in real time.
The description of aptx 'lossless' reads like the typical audio marketing fluff already used for the other aptx codecs. AFAIK there is no way to compress white noise.
In reality it may compress lossless until it doesn't.
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

Re: Personal Blind Listening Test of Bluetooth codecs

Reply #36 –
Quote from: Wombat on
AFAIK there is no way to compress white noise.
In reality it may compress lossless until it doesn't.
You mean there isn't an efficient way to compress noise? (Or are you implying that lossless codecs do not encode noise losslessly?)

Re: Personal Blind Listening Test of Bluetooth codecs

Reply #37 –
Quote from: a.ok.in on
Quote from: Wombat on
AFAIK there is no way to compress white noise.
In reality it may compress lossless until it doesn't.
You mean there isn't an efficient way to compress noise?
There way to compress (dropping some details about what that means) independent uniformly distributed signals (a.k.a. white noise) losslessly.
Here you got a challenge from back in the day: https://marknelson.us/posts/2012/10/09/the-random-compression-challenge-turns-ten.html
Obviously there cannot be an algorithm that will compress all files (because, if one input signal gets smaller then it "takes the place" of a smaller one among the set of all possible files - and so that other has to get bigger). The reason compression works, is that real-life files are not uniformly distributed over the set of all possible files of same size. So the ones that increase in size, are rare in real life although they are easy to generate if you want to.

Quote from: a.ok.in on
(Or are you implying that lossless codecs do not encode noise losslessly?)
"Every audio codec" has limitations to what signals it can carry. For example, none of the established ones support more than 32-bit signals. If you want to compress a 64-bit signal, you "have to" do some lossy pre-processing.
This particular codec has a bandwidth constraint, and it is supposed to deliver real-time playback. When it encounters an input signal where it cannot meet those constraints losslessly, it can do one out of two: stutter/reject/give up - or sacrifice losslessness.
Given the reason it exists in the first place, it should sacrifice losslessness. At https://www.qualcomm.com/news/releases/2021/09/qualcomm-adds-bluetooth-lossless-audio-technology-snapdragon-sound it is pointed out that yes it can transmit audio when the source is too high to do so losslessly over Bluetooth, and it is "Designed to scale-up to CD lossless audio based on Bluetooth link quality."

Re: Personal Blind Listening Test of Bluetooth codecs

Reply #38 –
Quote from: a.ok.in on
You mean there isn't an efficient way to compress noise? (Or are you implying that lossless codecs do not encode noise losslessly?)
I mean lossless codecs can't can't compress white noise well but of course without loss.
That means a 1411kbps 16/44.1 white noise lossless compressed file still has a bitrate above 1400kbps even at highest compression and monkeys audio.
When aptx lossless only uses 1120kbps for transmitting white noise i doubt it reinvented compression but simply leaves away data.
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

Re: Personal Blind Listening Test of Bluetooth codecs

Reply #39 –
Quote from: Wombat on
a 1411kbps 16/44.1 white noise lossless compressed file still has a bitrate above 1400kbps even at highest compression and monkeys audio.

This is an aside, but extremely noisy music isn't Monkey's fave banana. Merzbow's infamous "I lead you towards glorious times" (YouTube link) was enough to keep any .ape above 1423.
That's a track that the experimental SAC codec could - ultra-slowly - reduce by a third, but that's due to repetitive patterns. For "shorter" sections (6 seconds, frogs and apes might use longer) it didn't do nearly as good - but still it says something about the actual information content not being the maximum.

Results:
https://hydrogenaud.io/index.php/topic,120158.msg1001917.html#msg1001917 and the next reply for Monkey's (huh, I omitted "Fast"?)
https://hydrogenaud.io/index.php/topic,122040.msg1010086.html#msg1010086

Re: Personal Blind Listening Test of Bluetooth codecs

Reply #40 –
Why don't i wonder @Porcus already played with such edge cases in deep? :)
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

Re: Personal Blind Listening Test of Bluetooth codecs

Reply #41 –
> When aptx lossless only uses 1120kbps for transmitting white noise i doubt it reinvented compression but simply leaves away data.

Of course it does leave something away, zero doubt. These effective managers are just being "flexible" with their word usage.
a fan of AutoEq + Meier Crossfeed

Re: Personal Blind Listening Test of Bluetooth codecs

Reply #42 –
Quote from: magicgoose on
> When aptx lossless only uses 1120kbps for transmitting white noise i doubt it reinvented compression but simply leaves away data.

Of course it does leave something away, zero doubt. These effective managers are just being "flexible" with their word usage.

Reviewing the previous aptX profiles and other bluetooth codec families this wouldn't be too surprising. But actually there's almost no point in coding noise losslessly, at least just for hearing purpose. Output is still noise.
And they just want to use as less bandwidth with minimum effort.