I don't believe cables make a difference, any input?

[analog_sa]

Re: A Question Then....

My question for you is at this point would you trust your ears?

Nice story but completely unrealistic. Objectivists by definition do not compare similarly measuring equipment. I've known a couple and no force on earth could make them listen to the same piece of music twice in sequence through diffent gear, let alone cables. This total confidence in numbers and lack of audiophile neurosis makes me suspect they lead much happier lifes than us.

 

[ravon]

Re: A Question Then....

..
Being the good objectivist that you are, you first verify that this amp measures well ---{fact is it measures almost the same as your amp & any differences are insignificant and would be below the threshold of hearing}--- for all intents and purposes, just like your amp, this amp is properly-designed, functioning properly and compatable with your other equipment.

So you exchange it for your amp in your system fully expecting to hear absolutely no differences, but amazing you do hear differences! When your friend leaves you re-insert your amp into the system only now it's just doesn't sound quite right. So you start by measuring everything once again. Much to your surprise everything checks as being ok! Still compared to the other amp it doesn't sound quite right. So one by one you replace your amp, CDP, speakers, ICs and speakerwire with the exact same model just in case they're somehow not functioning properly but, the sound doesn't change. Every new piece measures like the old piece did and they all measure well, are obviously properly-designed and are functioning properly.

My question for you is at this point would you trust your ears? Or would you start to believe you're fooling yourself, hearing ghosts, under hypnotic suggestion or possible even start believing the test equipment was suspect? What if the test equipment checks out as ok? Would you trust your ears then?

Why would "good objectivists" not just listen to that amp? Simply listening to music can be very nice.

If the friends' amp sounds different, it's always possible to perform measurements or to perform an ABX listening test.

BTW, how are these imaginary differences heard? In a double blind ABX test with carefully matched signal levels?

I'm sure you are aware of the fact that many heard differences are the result of small differences in signal strength.

 

[Andre Visser]

Oh yes, it will be very difficult to measure or calculate every example audiophiles can think of, on the other hand it is very, very easy to prepare test signals which make audiophiles stumble in a test.

Listening to test signals??? What do you expect to hear / learn from that?

 

[nigelwright7557]

Listening to test signals??? What do you expect to hear / learn from that?

I find test signals excellent for setting up the amp.

If I put in a sine wave then I expect a sine wave out.
If its not a sine wave out then something is wrong.

I set up my crossover distortion bias using a sine wave and a scope.

 

[ravon]

Listening to test signals??? What do you expect to hear / learn from that?

One could imagine that the ABX test itself may be a subject of test.

 

[SY]

what I find "troubling" is the fact that I've found no other individual who questions the fact that the source material being used to test human perceptions is one not found naturally occuring in nature.

Really? Most published tests I've seen (and most of the ones I've done) use a mix of test signals and music. Sometimes recorded sound effects. My first tests of a CD phase correction algorithm I invented some years ago used the sound of a breaking glass; the differences with musical material were all but inaudible, but I could get the difference with the sound effect nearly every time. If you want to maximize the probability of detection, the test material needs to be chosen appropriately.

 

[nigelwright7557]

Really? Most published tests I've seen (and most of the ones I've done) use a mix of test signals and music. Sometimes recorded sound effects. My first tests of a CD phase correction algorithm I invented some years ago used the sound of a breaking glass; the differences with musical material were all but inaudible, but I could get the difference with the sound effect nearly every time. If you want to maximize the probability of detection, the test material needs to be chosen appropriately.

A good test could be white noise as this has all frequencies in it.

Then you would need to look at the output on a Spectrum Analyser to interpret the result.

 

[Andre Visser]

I find test signals excellent for setting up the amp.

If I put in a sine wave then I expect a sine wave out.
If its not a sine wave out then something is wrong.

I set up my crossover distortion bias using a sine wave and a scope.

Yes, no problem when testing with instruments, but LISTENING to test signals have very limited use.

 

[GlidingDutchman]

Yes, no problem when testing with instruments, but LISTENING to test signals have very limited use.

Listening to sinewaves have absolutely no purpose in assesing the sonic signature of audio gear.

Yes, you can check speaker rattle and bandwidth but that is it - you need an oscilloscope and other sophisticated test instruments when playing with generated signals.

A good test disc for LISTENING is NAIM's "True Stereo" album. It is unprocessed analogue recordings using two AKG 414EB in the ORTF stereo array feeding a NAGRA IV-S stereo r-to-r recorder.

The recording is so good it is bloomin scary!

Catalogue number: naimcd080

I have used this disc to quickly audition one of Andre's amp systems and it works great (both CD and amps 🙂 )

D

 

[jneutron]

Really? Most published tests I've seen (and most of the ones I've done) use a mix of test signals and music. Sometimes recorded sound effects. My first tests of a CD phase correction algorithm I invented some years ago used the sound of a breaking glass; the differences with musical material were all but inaudible, but I could get the difference with the sound effect nearly every time.

Hey there old guy...

In those test signals, what were the mathematical relationships between the desired location of the image, the ITD of the two channels, and the IID of the two channels. Was the source material capable of duplicating the exact sound stimulus in all regards. IOW, was the human presented the exact soundfield that would have been created by the object physically being there? Or, was the stimulus artificial..

If you want to maximize the probability of detection, the test material needs to be chosen appropriately.

That has been my point all along...

A good test could be white noise as this has all frequencies in it.
Then you would need to look at the output on a Spectrum Analyser to interpret the result.

While staying away from the light, of course.

For your test, how would we detect an interchannel time shift?

Cheers, John

 

[Andre Visser]

A good test disc for LISTENING is NAIM's "True Stereo" album. It is unprocessed analogue recordings using two AKG 414EB in the ORTF stereo array feeding a NAGRA IV-S stereo r-to-r recorder.

The recording is so good it is bloomin scary!
D

Thanks Dewald, I will get me one of those.

André

 

[nigelwright7557]

Hey there old guy...

In those test signals, what were the mathematical relationships between the desired location of the image, the ITD of the two channels, and the IID of the two channels. Was the source material capable of duplicating the exact sound stimulus in all regards. IOW, was the human presented the exact soundfield that would have been created by the object physically being there? Or, was the stimulus artificial..


That has been my point all along...



While staying away from the light, of course.

For your test, how would we detect an interchannel time shift?

Cheers, John

A good test would be to record a complex sounding instrument.
Compare the output of this with the output of the amp for differences using FFT to get the signals into a PC.
The FFt woudl show up any differences in frequencies and also any added frequencies or subtracted frequencies.

 

[ravon]

Yes, no problem when testing with instruments, but LISTENING to test signals have very limited use.

Listening to sinewaves have absolutely no purpose in assesing the sonic signature of audio gear.

Oh, it can be very useful to use sine waves (among other signals) in listening tests.

Please don't bother, it seems you already know everything you need to know.

 

[SY]

In those test signals, what were the mathematical relationships between the desired location of the image, the ITD of the two channels, and the IID of the two channels. Was the source material capable of duplicating the exact sound stimulus in all regards. IOW, was the human presented the exact soundfield that would have been created by the object physically being there? Or, was the stimulus artificial..

That would depend on which paper, eh? Many test signals have no reality; they're a mathematical construct. That would include the signals used by the Japanese guys who came up with that (still questionable) 1us figure.

My breaking glass signal was recorded in mono with a B&K 4002 onto a custom lab recording system (Nicolet). This was 1985, I didn't have anything better.

Lest I sound too negative, I agree with you philosophically. Once you've reduced a soundfield to a discrete number of channels, it's all illusion. That's far and away the biggest error. If there's a better way to do it, that strikes me as a much more fruitful line of research than (say) wire directionality.

 

[Andre Visser]

Oh, it can be very useful to use sine waves (among other signals) in listening tests.

Please explain.

Please don't bother, it seems you already know everything you need to know.

Now where does this come from?

 

[KSTR]

In those test signals, what were the mathematical relationships between the desired location of the image, the ITD of the two channels, and the IID of the two channels.

For those who can read german I'd recommend http://sengpielaudio.com, he's a top notch authority in the field. Bottom line on the ITD/ILD issues is that there is not exact (as such) mathematical relationship between ITD/ILD and localization ("position" and "size" of the phantom source), it' all very empirical, different researcher got different results. It depends on many secondary factors and of course is bound to the 2-ch playback in the equilateral triangle setup. When I tested "Trinaural/OLM" and other rematrixing playback setups with 3 speakers (L-C-R) I found (with music) and verified (with specific test signals, for that matter @GlidingDutchman) some interesting differences in the way ILD/ITD relationships a percieved, especially in the "size" parameter. "Direct" (mainly ILD) sources get more "direct", "diffuse" (more ITD) sources get more "diffuse", which partly accounts for my faible for this reproduction variant (which still is very compatible, execpt for HRTF-based localization trick -- Q-Sound -- which cannot and does not work).

- Klaus

 

[Andre Visser]

Once you've reduced a soundfield to a discrete number of channels, it's all illusion. That's far and away the biggest error. If there's a better way to do it, that strikes me as a much more fruitful line of research than (say) wire directionality.

"Illusion, biggest error", maybe so but it is possible to recreate it remarkably well.

 

[jneutron]

A good test would be to record a complex sounding instrument.
Compare the output of this with the output of the amp for differences using FFT to get the signals into a PC.
The FFt woudl show up any differences in frequencies and also any added frequencies or subtracted frequencies.

The mere act of conversion from time varying waveform to a magnitude spectra loses ALL timing information.

There are an infinite number of waveforms that can have the exact same FFT spectra.

There must be correlation between the input and the output.

And don't forget, it's gotta sample sufficiently fast to get into the 1 uSec domain, as well as below 1dB differences.

Cheers, John

 

[nigelwright7557]

The mere act of conversion from time varying waveform to a magnitude spectra loses ALL timing information.

There are an infinite number of waveforms that can have the exact same FFT spectra.

There must be correlation between the input and the output.

And don't forget, it's gotta sample sufficiently fast to get into the 1 uSec domain, as well as below 1dB differences.

Cheers, John

There is modern equipment that woudl have no problem doing this.

DSP's run at many times 1uS.

I did a project a few years back to convert a piezo device under a car to work out its engine RPM through vibration.
That just needed a bog standard PIC microcontroller.

 

[jneutron]

That would depend on which paper, eh? Many test signals have no reality; they're a mathematical construct. That would include the signals used by the Japanese guys who came up with that (still questionable) 1us figure.

Yah, I question 1uSec as well.. I'd limit the region of interest to the 2 to 5 uSec regime...I use 1 uSec as a convienience in typing, and would actually use a system capable of 100 nSec just to remain sufficiently accurate.

My breaking glass signal was recorded in mono with a B&K 4002 onto a custom lab recording system (Nicolet). This was 1985, I didn't have anything better.

As a mono signal, it would only be useful to detect the balance between channels for a centered image.. If you limit yourself to pan pots, when you shift the image to one side via the knob, you move different frequencies unequally.. we humans don't follow the cosine law exactly.

Lest I sound too negative, I agree with you philosophically. Once you've reduced a soundfield to a discrete number of channels, it's all illusion. That's far and away the biggest error. If there's a better way to do it, that strikes me as a much more fruitful line of research than (say) wire directionality.

You negative???pshaw...neva happened..

Cheers, John