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: What Kind Of Music For Testing? (Read 16195 times) previous topic - next topic
0 Members and 2 Guests are viewing this topic.

What Kind Of Music For Testing?

Reply #25
If people rejected every idea for which there was a pathological counter-example, we would probably still be back in dark ages.
Funny how the examples that disprove your theories are always "pathological", "straw men", "extreme" etc etc.


Newton's theory of gravity has been disproven yet it is widely used in engineering calculations. Somehow the buildings stay up and we made it out of the dark ages.

What Kind Of Music For Testing?

Reply #26
Oh, you want to change the basic nature of the examples?

I don't see what that would show other that that you need to make basic changes to your examples in order to demonstrate your point.
It's just an example of very simple lossy encoder preprocessor.

What about posting shorter but useful subsets of the 5 minute selections?
Ok, let's take this sample... Done!
[attachment=6001:LaOuJeSuisNee.7z]

What Kind Of Music For Testing?

Reply #27
The problem with this approach is that it would seem to presume that any given artifact is either inaudible to everybody or are audible for everybody. ...

By "practically inaudible" in my above reply I mean inaudible to the experienced testers who have took part in the development of the modern codecs. The early days of perceptual encoding when dramatical improvements were common are over. Although we have an established tradition of testing lossy codecs here at HA, interest in testing the high quality settings seems to have reduced during the last few years. This might indicate that the best encoders are already good enough for everyone.

A general misunderstanding is that lossy encoding produces a constant signature that can be detected in any encoded audio signal if the listener has golden ears and high quality equipment (I am not saying that you share this belief, surely you know better). Assuming a constant lowpass frequency is not too low (it isn't when a modern lossy encoder is used at a high quality setting) there is no such thing. Most if not all reported artifacts here at HA have been audible only during very brief moments with certain source signals.

I can pinpoint some momentary artifacts in my high bitrate LAME VBR encoded files in a proper blind test, but those are rare and even when they exist they are not really annoying, just a little bit different. The artifacts are very rarely audible in a casual listening situation.

For instance, a sudden and fierce cymbal crash may sound like having a slightly different pitch. Just a few days ago I uploaded a sample that demonstrates this effect here. This kind of artifact is audible only in a direct comparison. In general, the medium bitrate LAME -V5 encoded (133 kbps) version of the track is already very close to transparency.

What Kind Of Music For Testing?

Reply #28
Regarding to what kind of results we should expect when lossy is re-encoded to lossy Guruboolez published a practical test a bit over five years ago. Topic: Short re-encoding blind listening test, wavpack - mp3 - mpc - aac - vorbis. His findings are still interesting. These formats have not radically changed since then. (LossyWav works in a very different way. Its transcoding results cannot be compared to these formats.)

 

What Kind Of Music For Testing?

Reply #29
It's just an example of very simple lossy encoder preprocessor.

What about posting shorter but useful subsets of the 5 minute selections?
Ok, let's take this sample... Done!
[attachment=6001:LaOuJeSuisNee.7z]
That's a nice Lame/mp3 problem sample.

Even the 4th generation lossyWAV transcode doesn't sound bad at all. I can't ABX it (though others might be able to).

Cheers,
David.