LM4562 vs OPA2134 vs [ ? ]
Apr 20, 2020 at 3:29 AM Post #16 of 31

439598

100+ Head-Fier
Joined
Nov 25, 2015
Posts
332
Likes
76
I asked a simple question, and your reply is quoting somebody's book -with over "200" tests, yet you can't point to ONE
that shows your assertion that: " ...If a person can hear a difference in electronic circuits, there are tests that can easily
measure that difference."
?

Name one.

pj
allhifi, it really isnt worth trying to discuss this sort of thing here, it never leads anywhere.
Even on diyaudio where you will find some of the most experienced engineers, not just armchair scientists, in recent years many have begun to change their stance on the conclusive of measurements to audiblility. At this point the ASR mindset only delays true progress in the field of audio electronics
 
Last edited:
Apr 20, 2020 at 4:32 AM Post #17 of 31

bigshot

Headphoneus Supremus
Joined
Nov 16, 2004
Posts
23,407
Likes
4,933
Location
A Secret Lab
They're changing their stance based on what? Controlled listening tests that show that things that should be inaudible are clearly audible? Or ones that show that two DACs or amps that measure alike sound clearly different? I'd like to see those tests if they exist.

I've had dozens of CD players, DAPs, amps and DACs over the years ranging from $40 Walmart DVD players to an Oppo HA-1, and since the introduction of oversampling, I haven't found a single one that sounds clearly different than the rest when used for the purpose it was designed for. I'd really like to know whether something like that exists, and it seems to me proving that unmeasurable differences are clearly audible in controlled tests would be revolutionary and we would have heard about them from the AES.

If you have this kind of information, please share it. If you present convincing evidence and don't just dismiss us as "armchair scientists", we won't dismiss you as someone who is just talking smoke and mirrors. Deal?
 
Last edited:
Apr 20, 2020 at 8:30 PM Post #18 of 31

castleofargh

Sound Science Forum Moderator
Joined
Jul 2, 2011
Posts
9,533
Likes
4,927
allhifi, it really isnt worth trying to discuss this sort of thing here, it never leads anywhere.
Even on diyaudio where you will find some of the most experienced engineers, not just armchair scientists, in recent years many have begun to change their stance on the conclusive of measurements to audiblility. At this point the ASR mindset only delays true progress in the field of audio electronics
It rarely leads anywhere because most of it is obviously and demonstrably false, while the rest is typically claimed without bringing any evidence, so there is nothing to do with it. Don't blame skeptics for demanding controlled experiments and methods, that's where true progress and fact based knowledge comes from.
 
Apr 21, 2020 at 4:08 AM Post #19 of 31

bigshot

Headphoneus Supremus
Joined
Nov 16, 2004
Posts
23,407
Likes
4,933
Location
A Secret Lab
I don't think he cares what anyone says. He just wants to come in here and make unsubstantiated claims and then duck out again. He probably is a sock for the other guy. That is why most trolls arrive in self validating pairs.
 
Apr 21, 2020 at 4:16 AM Post #20 of 31

castleofargh

Sound Science Forum Moderator
Joined
Jul 2, 2011
Posts
9,533
Likes
4,927
I don't think he cares what anyone says. He just wants to come in here and make unsubstantiated claims and then duck out again. He probably is a sock for the other guy. That is why most trolls arrive in self validating pairs.
Present evidence or I'll treat this as an unsubstantiated claim.
😈


Wait a minute! are you me with another account?
 
Last edited:
Apr 21, 2020 at 3:56 PM Post #21 of 31

bigshot

Headphoneus Supremus
Joined
Nov 16, 2004
Posts
23,407
Likes
4,933
Location
A Secret Lab
Ooo! I forgot and used punctuation and capitalization in that last post!
 
Apr 23, 2020 at 8:56 AM Post #23 of 31

SoundAndMotion

100+ Head-Fier
Joined
Jul 15, 2015
Posts
297
Likes
191
What does ASR (in "the ASR mindset") stand for?
ASR is Audio Science Review(link), a heavily measurement-oriented audio forum.

WRT Echoic Memory: I suspect @Speedskater may know (so don't answer, @Speedskater ), but I'm not sure others understand it, especially those with an opportunistic relationship with science. Not sure they should be elevated to "armchair scientist".
Since you brought it up, please answer this: imagine you run out for supplies. When you return, you see a phone message. You play it: "Hey, it's me. Came up with a new riff... what do you think?" followed by 10s of guitar. You forget to hit save, so didn't avoid an auto-delete, and didn't take notes. After the echoic memory time runs out, what will you remember and what will you forget? How do you know which parts of @allhifi 's description fall into the forget-group or the remember-group? Surely you don't believe the entire message is forgotten. Why not? You assume @allhifi has forgotten all...
 
Apr 23, 2020 at 10:41 AM Post #24 of 31

sander99

1000+ Head-Fier
Joined
Sep 15, 2017
Posts
1,302
Likes
569
Location
The Netherlands
I would think: in general, what you remember is something that has already been processed by the brain, partly broken down into certain abstract components.I think you would remember some things from different levels of abstraction. The highest level of abstraction would be the melody and rhythm I guess. At some inbetween level you would have a recollection of the general sound of the guitar.
Compare with this: you have read a message on a piece of paper. What do you remember? You probably remember the words. But not a complete "image" of little black lines on the paper (very low abstraction level). Maybe you remember some details about the font style and size that caught your attantion (some middle abstraction layer).
What allhifi remebers is probably not the actual audio itself, but mainly the conclusions/feelings he had about it. And those conclusions/feelings - to say it politely - could be influenced by expectation bias etc.
 
Apr 23, 2020 at 11:20 AM Post #25 of 31

SoundAndMotion

100+ Head-Fier
Joined
Jul 15, 2015
Posts
297
Likes
191
Oh @sander99 ! You blew it! Well really, I did. I wanted @bigshot to answer.... you seem to know what you're talking about, and I didn't want you to try. Your first 2 paragraphs are spot on, and I don't disagree with anything, but what aspects of the audio message will be forgotten? Nice examples for the written message.
As for your last paragraph, it is almost certain that allhifi doesn't remember the audio itself, but as you point out, he can remember the abstract components that he had processed at the time. As with all humans (as I expect allhifi is), he could have experienced contemporaneous expectation or confirmation biases (or others), and in the last 16 years may have colored his memory of the abstractions due to other experiences.
Since I can't be sure what any of this means exactly:
the effortless (flowing-like-water) clarity, dimensionality, tunefulness/distinctive tonal hues -the sheer musical naturalness of this DAC
I can't really guess what he may or may not remember. He might remember all that with very little bias, or it may have been meaningless gibberish even back then. I'm unable to tell.
 
Last edited:
Apr 23, 2020 at 1:35 PM Post #26 of 31

bigshot

Headphoneus Supremus
Joined
Nov 16, 2004
Posts
23,407
Likes
4,933
Location
A Secret Lab
Since you brought it up, please answer this: imagine you run out for supplies. When you return, you see a phone message. You play it: "Hey, it's me. Came up with a new riff... what do you think?" followed by 10s of guitar. You forget to hit save, so didn't avoid an auto-delete, and didn't take notes. After the echoic memory time runs out, what will you remember and what will you forget?

We tend to remember things we can contextualize- things like stories, patterns, melodies, ideas. We tend to forget information that we can't connect to a context- strings of random numbers, random pitches or letters, and sound fidelity that isn't connected to the context of a direct comparison.

For instance, if I do a controlled listening test between two similar sounds, I might remember that I thought one of them sounded better. But if you presented me with one of those sounds without the context of the other one, I wouldn't be able to tell which one it was.

The length of auditory memory for sound depends on how different it is. We can easily remember what muffled or harsh sounds like in general, but it's impossible for us to discern two reasonably similar quality sounds when presented to us with a more than a few seconds between them.

You can test your own auditory memory pretty simply. Take a music track and alter the response 2 or 3 dB in some part of the range. Listen to it directly next to an unaltered file and prove that you can discern the alteration. Then start inserting 1 second, 2 seconds, 3 seconds between the samples and find out where your accuracy starts to fall off. It's quite difficult, not just because of auditory memory, but because our ears tend to adjust to slight deviations over time. The longer the sample, the less likely you would be able to discern which one it was.

The sorts of potential differences we are talking about with DACs or amps are very small. The only way for us to grade them accurately is a direct A/B comparison. You can't trust your memory for stuff like this.

I can't really guess what he may or may not remember. He might remember all that with very little bias, or it may have been meaningless gibberish even back then. I'm unable to tell.

He's remembering his emotional reaction to listening to music, not precise memory of the fidelity of it. He is just attributing the emotions produced by the music to the fidelity. That likely isn't what's causing his emotional reaction.
 
Last edited:
Apr 23, 2020 at 3:37 PM Post #27 of 31

SoundAndMotion

100+ Head-Fier
Joined
Jul 15, 2015
Posts
297
Likes
191
We tend to remember things we can contextualize- things like stories, patterns, melodies, ideas. We tend to forget information that we can't connect to a context- strings of random numbers, random pitches or letters, ...
In general, this kind of works. Since you can contextualize to anything you can remember or imagine, I would say "things for which we can create an abstract representation". And since you can create a context and therefore abstract representation for random numbers or letters (that's what I do, if I want to remember), I'd pick different examples, but I understand what you're saying.

...and sound fidelity that isn't connected to the context of a direct comparison.

For instance, if I do a controlled listening test between two similar sounds, I might remember that I thought one of them sounded better. But if you presented me with one of those sounds without the context of the other one, I wouldn't be able to tell which one it was.

The length of auditory memory for sound depends on how different it is. We can easily remember what muffled or harsh sounds like in general, but it's impossible for us to discern two reasonably similar quality sounds when presented to us with a more than a few seconds between them.

The sorts of potential differences we are talking about with DACs or amps are very small. The only way for us to grade them accurately is a direct A/B comparison. You can't trust your memory for stuff like this.
I bolded here. "sound fidelity" is so broad and undefined as to be meaningless.
No "the length of auditory memory for sound depends on how different it is" is NOT true. First, it sounds as though you mean there's a whole range of memory lengths (perhaps I misunderstand), when really we're talking about 2 lengths: echoic memory length and generated abstract memory length (aka long-term memory). But more importantly, it depends on abstract-ability not differentness. And assuming one can actually hear a difference at all (not part of this discussion), the ability to assign an abstract representation to the difference is the key.

You can test your own auditory memory pretty simply. Take a music track and alter the response 2 or 3 dB in some part of the range. Listen to it directly next to an unaltered file and prove that you can discern the alteration. Then start inserting 1 second, 2 seconds, 3 seconds between the samples and find out where your accuracy starts to fall off. It's quite difficult, not just because of auditory memory, but because our ears tend to adjust to slight deviations over time. The longer the sample, the less likely you would be able to discern which one it was.
You're the last person I'd go to for advice on doing a listening test. The last 2 sentences seal the deal.

He's remembering his emotional reaction to listening to music, not precise memory of the fidelity of it. He is just attributing the emotions produced by the music to the fidelity. That likely isn't what's causing his emotional reaction.
Did you interview him? Did he tell you this? If not, quit synthesizing BS to fit your narrative.
 
Apr 23, 2020 at 4:52 PM Post #28 of 31

bigshot

Headphoneus Supremus
Joined
Nov 16, 2004
Posts
23,407
Likes
4,933
Location
A Secret Lab
You're the last person I'd go to for advice on doing a listening test.

Then why the hell are you asking me questions about it? Sheesh! I was respectful to you and answered your questions as best I could. You can be respectful too, or I'll just ignore you and talk past you to the rest of the group. I'm sick and tired of people insisting on pushing everything into pissing matches to feed their own ego. I will blow right past that crap. That's the way I roll!
 
Last edited:
Apr 23, 2020 at 5:41 PM Post #29 of 31

SoundAndMotion

100+ Head-Fier
Joined
Jul 15, 2015
Posts
297
Likes
191
Then why the hell are you asking me questions about it? Sheesh! I was respectful to you and answered your questions as best I could. You can be respectful too, or I'll...[snip]... talk past you to the rest of the group. I'm sick and tired of people insisting on pushing everything into pissing matches to feed their own ego.
I didn't ask you about doing a listening test; I asked why you posted the link to echoic memory. I figure being honest is part of being respectful, which I was being. You threw out the listening test stuff on your own.
According to what you wrote a while ago, talking past people to the rest of the group is your intentional MO. Should I find that for you? It has nothing to do with my ego; it has to do with you playing armchair scientist... sometimes reasonably well, sometimes not so much. Actually, I find you quite anti-science.

I'll just ignore you...
I will blow right past that crap. That's the way I roll!
Groovy. Up to you...
 
Apr 23, 2020 at 6:12 PM Post #30 of 31

bigshot

Headphoneus Supremus
Joined
Nov 16, 2004
Posts
23,407
Likes
4,933
Location
A Secret Lab
Talking past people is what I do when it isn't worth talking with them. Up to you... You're just getting argumentative now. Let me know when you'd like to talk about something.
 

Users who are viewing this thread

Top