It doesn't seem to me that ABX testing is flawed, just its asking the wrong question for those of us who are looking to reproduce more enjoyable music from our equipment. When enjoyment is the goal who cares if there's a difference between A and B? That's an academic question - what I want to know is does B satisfy me more than A? And that's a question ABX has never set out to answer.
Put it this way, AD8610 and LME49720 sound different, they are "little instruments" with their own sound, even if it's little.
Most people will say ES9018 is better sounding than ES9023.
The evidence / ABX enthusiasts tend to think the above all sound identical, but maybe they will change their mind in 2025 when someone can prove it, until then they can listen to their car radio.
Most people will say ES9018 is better sounding than ES9023.
The evidence / ABX enthusiasts tend to think the above all sound identical, but maybe they will change their mind in 2025 when someone can prove it, until then they can listen to their car radio.
It's easier with the components you and Lampizator like, Abraxalito, since the consensus is that vintage components actually do sound quite different.
There must be a huge error in how low THD we can hear, perhaps.
Just like that "huge error" about sonic memory.
There must be a huge error in how low THD we can hear, perhaps.
Just like that "huge error" about sonic memory.
Last edited:
People don't change their mind on evidence they just claim that they do, two entirely different things. For a contemporary example, go over to the thread on WBF where Amir has aced a DB test that Arny Krueger put up - but Arny's looking distinctly unmoved.
Ah I like TDA1541 a lot too, just I don't use it in designs as availability (and authenticity) isn't that great being so long out of production.
Main problem with the ABX tests that are done by audio enthusiasts, in forums like this, is that they usually rely on the foobar software to do it. And this tool has major problems in the implementation, I've done some fossicking around in it, and I don't like it one bit - just using this program for playback degrades the sound too badly for it to be truly useful, at least for my PC.
It the tool or measuring device is sloppily or poorly constructed then any results using it don't have much value - mention that around here and all the usual suspects go dead quiet, not something they probably want to hear ... 🙂
Here you go - its already nearly 1200 posts and still going strong - Conclusive "Proof" that higher resolution audio sounds different
And the title is rather ironic, considering how the thread shows that whatever it is, "conclusive" it isn't... 🙂
So proof is only proof if it is on the internet? I see. A new variation of "I read it on the internet, so it must be true!"
If you can hear it sighted, you *should* be able to identify it from *only* the sound. It doesn't matter what technique you use, as long as you make sure the listeners don't know what they are listening to. Double-blind ABX provides the extra benefit of statistical controls, but ABX is just one method.
The funny thing s that in many cases, people who can hear differences sighted can't tell them when listening in a double-blind situation (doesn't need to be ABX).
ABX is a tool. Evidence-based is a mindset. Science is evidence-based.
Note that the word "Proof" is in quotes in the title of that thread - where human perceptions come into play there definitely ain't no such animal as proof ...
So then is it not conclusive because it wasn't properly conducted ? Or because there's still bickering going on over what it means? Or because you've found a fatal flaw in Foobar's ABX?
No, the blockiness you hear is the fact that the keyboard cross-blend between a small number of samples for the acoustic grand at different levels. The "lack" of resolution can only be heard as added background noise.
Remember that a large part of the works of artists like Peter Gabriel and Kate Bush were done on 8-bit and 12-bit systems (synclavier and fairlight). I stil remember how the Yamaha TX16W sampler was considered almost over-the-top in using 12 rather than 8 bits when it was introduced in the 80's.
Hahahaha
This one?
Conclusive "Proof" that higher resolution audio sounds different
"There has been very little formal testing but what there is, such as Meyer and Moran point to no audible difference (yes, the test was flawed but this thread obsoletes what they did anyway)"
It's not like someone like NwAv with come forth and say they were wrong, in referencing Meyer and Moran all the time, let alone Meyer and Moran themselves.
In the light of "current understanding", you can attend to the distinction that you are never wrong.
That is where I take issue though, "very highly plausible" is plain and simple better / more accurate than evidence, in some times.
Last edited:
- Status
- Not open for further replies.