Purifi Audio by Bruno Putzeys and friends.

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I trust my ears, but I don't trust my mind. It's way to easy to fool yourself. That's why the protocol is so incredibly important. When testing, you are not allowed to know what's been tested when. This is the only way to prevent fooling yourself.

Iow if you know your listening to the modified device, your brain will make assumptions not soly based on the input from your ears.
Bill, I have given you suggestions of things that you can try and have already given you too much information. In the meantime I am not interested in the slightest in debating the issue of who can hear what, you might be fooled easily but that does not mean everybody.
BTW just what does this mean -
Subjective measurements with the proper protocol is the only thing that matters.
Something lost in the translation maybe?

Dan.​
 
Hi Dan,

A subjective measurement is a measurement with your ears. And because this site is about audio, this is what matters. Iow can you hear it or not.

An analogy: Think of a complex electronic circuit with power and a signal. Now you want to know what the voltage is at point A. You put one pin of the multi meter at point A. But where do you put the other pin? If you put the other pin at a random place in the circuit, do you think you get a value that has any merit? Imo the chance of that is slim. You need to place the other pin at a reference point. Usually this is signal earth. Now you can get a value that you can use. Iow if you don't use the proper protocol to measure, your measured value will have no merit.

Exactly the same goes for a subjective measurement. If you don't have a fixed reference point, your measured value will have no merit. And the only way to get a fixed reference point is to only use your ears as input for your subjective measurement.

Best Bill.
 
.....Exactly the same goes for a subjective measurement. If you don't have a fixed reference point, your measured value will have no merit. And the only way to get a fixed reference point is to only use your ears as input for your subjective measurement.
You can have two speakers one internally modified and one standard driven by the same signal. From outside the two speakers look the same but sound different, ears can decide preference. Perhaps you can try my suggestions and decide subjective preference for yourself.

Dan.
 
diyAudio Moderator
Joined 2008
Paid Member
These should be re-drawn superimposed, referenced to the basket.
 

Attachments

  • roll.PNG
    roll.PNG
    58.3 KB · Views: 680
I remember seen somewhere laser interferometry which has shown that a standard surround behaved different to the cone. This is not only a problem at x-max. At some frequency the surround is completely out of phase and especially with larger high x-max drivers the surround surface is quite large. So a more convoluted shape could lead to a more complex vibrational behavior which may reduce this interference but I can not see how it eliminates it.
 
You can have two speakers one internally modified and one standard driven by the same signal. From outside the two speakers look the same but sound different, ears can decide preference. Perhaps you can try my suggestions and decide subjective preference for yourself.

Dan.

This only works if you don't know what speaker is modified.

As soon as you do know what speaker is modified and what speaker is not, your brain will use that as extra input for its perception and thus you don't use the same fixed reference point for your measurement.

Speakers just looking the same doesn't mean anything, your brain is not allowed to know what speaker has the mod. In fact it goes even further: No one in the test room is allowed to know what speaker has the mod, because they will unconsciously give clues.
 
I don't have a clue at the moment what could be the cause of this.
But there is definitley some form of noise that can be caused by magnetic circuits like loudspeaker drivers or filter coils. It is called Barkhausen noise and it is caused by tiny jumps in the magnetisation curve.
John Watkinson for instance claims that Speaker drivers with ferrite magnets wouldn't have an adequate SNR to make full use of the resolution given by 16 bit audio, let alone any of the hi-res formats.
It could be possible that the Barkhausen effect is also a limiting factor in Class-D amps as soon as the rest of the amp is of very high quality. IIRC did the TacT Millennium amp use foil-wound air coils for the output filter. One would not only get rid of saturation effects this way but also the Barkhausen noise wouldn't be an issue. The downside are increased EMC issues.

Regards

Charles

the claim is that the magnetization curve is now independent of the signal and voice coil position
 
This only works if you don't know what speaker is modified.

As soon as you do know what speaker is modified and what speaker is not, your brain will use that as extra input for its perception and thus you don't use the same fixed reference point for your measurement.

Speakers just looking the same doesn't mean anything, your brain is not allowed to know what speaker has the mod. In fact it goes even further: No one in the test room is allowed to know what speaker has the mod, because they will unconsciously give clues.
Who gives a toss, what works works, what doesn't walks. You could try this experiment/advice for yourself, in fact it would take less time than you have spent in front of your PC arguing about the suggested mod experiment. Maybe I have spent too much time dealing with/repairing audio but I have no problem picking significant mods from the standard item, if you tried you might find that you can do so also. This is why I say trust your ears but this is dependant on experience and practice, how old are you and what is your experience with audio electronics ?.

Dan.
 
Not the same Bill - you assumed that one speaker was modded, but unknown which. To really make it DB, you should also not disclose whether any speaker was modded at all, or that both are unmodded, or both modded. Be devious :cool:

Jan

Yes.

But to be completely honest, this specific tests is doomed from the start.
Speakers placed at different points in a room will certainly sound different.

True honesty is difficult.
 
Who gives a toss, what works works, what doesn't walks. You could try this experiment/advice for yourself, in fact it would take less time than you have spent in front of your PC arguing about the suggested mod experiment. Maybe I have spent too much time dealing with/repairing audio but I have no problem picking significant mods from the standard item, if you tried you might find that you can do so also. This is why I say trust your ears but this is dependant on experience and practice, how old are you and what is your experience with audio electronics ?.

Dan.

If you truly trust your ears, why do you use other inputs to your brain as well?

And you are asking me for an argument from authority, that is a logical falacy on top of a logical galaxy.
 
If you truly trust your ears, why do you use other inputs to your brain as well?
And you are asking me for an argument from authority, that is a logical fallacy on top of a logical galaxy.
What ????.....my inputs to you are encouraging and kindly but you want to argue and continue arguing FFS.

You are arguing about cases where a difference is extremely minor and the subject is a novice.
In the case of obvious difference the 'blindness' of the test does not matter except for when formal records are required.

I think you should try some On The Fly modification sighted testing for yourself and learn to discriminate fine differences as a first step.


Dan.
 
AX tech editor
Joined 2002
Paid Member
Yes.

But to be completely honest, this specific tests is doomed from the start.
Speakers placed at different points in a room will certainly sound different.

True honesty is difficult.

That's why Harman developed the Speaker Mover. These guys were dead serious about DB tests. And some of their results are interesting.

Jan
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.