[kwkarth]

Quote:

Originally Posted by Kees /img/forum/go_quote.gif I think he is right. He is measuring the sound as you would hear it from your phones and looking for the differences in situation a and b. If you take away the phones and measure the output of your amp you would get the discussion about the difference you measure are audible or not. If a mic picks it up from your phones it is safe to assume it is audible.

non-sequitur

 

[digger945]

I think I understand what your saying kw, and I am working now to see if the inputs on the 0404 interface will handle the output of my amp, don't want to fry anything. I may have to use the Predator as source and the 0404 to record the measurements. Any advice before I begin would be much appreciated.

 

[Kees]

Quote:

Originally Posted by kwkarth /img/forum/go_quote.gif non-sequitur

I thought it was pretty straight forward.

You must have some different logic.
I think I'll leave it to you then.

 

[digger945]

The 0404 can indeed be used for playback and recording simultaneously (loopback?). The problem I have now is that the recording is then returned to digital and thus, when magnified, looks pretty stairstepped, even at 16bit/192kHz(max sample rate for usb 2.0). I'm still playing around with it, but thinking the analog signal should be captured before re-digitizing. I am looking at the Picoscope, but $ I don't have right now. I did read on Audioholics that the Adobe Audition software will reproduce sine waves pretty accurately on screen. Maybe I will sell some stuff in my sig to fund the research

I really would like to know more about this.

 

[kwkarth]

Quote:

Originally Posted by digger945 /img/forum/go_quote.gif The 0404 can indeed be used for playback and recording simultaneously (loopback?). The problem I have now is that the recording is then returned to digital and thus, when magnified, looks pretty stairstepped, even at 16bit/192kHz(max sample rate for usb 2.0). I'm still playing around with it, but thinking the analog signal should be captured before re-digitizing. I am looking at the Picoscope, but $ I don't have right now. I did read on Audioholics that the Adobe Audition software will reproduce sine waves pretty accurately on screen. Maybe I will sell some stuff in my sig to fund the research I really would like to know more about this.

Although I admire your efforts, I don't think you'll see anything revealing in the way you're approaching this. The things that are audibly different, cable to cable are very subtle, and probably are found in the realm of subtle phase shift, or maybe if you're lucky very minute FR differences, 0.1db here, 0.1db there. I would suggest approaching this by first, clearly identifying the differences that you do hear, then "tune" your measurement devices to focus in on those differences, otherwise the subtle differences will be "hidden" by everything else that you try to measure.

 

[wavoman]

Shouldn't we build an A/B box (or software) instead of an A/B/X/ box.

Continuing my rant against A/B/X --

For listener-blind testing:

Being able to identify X as either A or B is a different skill, a harder skill, than being able to listen to two cables, or two resolutions, whatever, presented as "A" and "B" but with the assignments, the presentation order, randomized, and reliably tell us which you prefer (if either).

The statistical significance does not even have to be high -- we are talking about subjective things here, and precision is not the order of the day -- if I most of the time pick one cable over the other as the one I prefer, well that's it.

Test what we are trying to achieve -- and we are trying to achieve the better sounding cable, NOT trying to prove we can correctly assign a third sample to one of the other two (that is what ABX does, and who cares?).

Memory for sound makes ABX way harder than saying "I prefer B to A".

 

[digger945]

kw,
The main difference I hear...
Alo solid silver... very bright, almost harsh, makes me want to turn the volume down a little.
Alo cryo... good overall... even across the board.
OK, I can hear that one is definitely brighter than the other.

I realize that the differences are going to be subtle, just as the amount of time that separates two instruments only ten feet apart, sound wise, is only .0088 seconds.

I bought these IC's just for this forum, to test and listen to, and compare.
Soon, we will see the differences we hear. The only way I have to learn is by doing, even if it is wrong to begin with. By eliminating a bunch of wrong ways, I eventually get to the right way.
Thank you for nudging me in the right direction, kw.

 

[kwkarth]

Quote:

Originally Posted by digger945 /img/forum/go_quote.gif kw, The main difference I hear... Alo solid silver... very bright, almost harsh, makes me want to turn the volume down a little. Alo cryo... good overall... even across the board. OK, I can hear that one is definitely brighter than the other. I realize that the differences are going to be subtle, just as the amount of time that separates two instruments only ten feet apart, sound wise, is only .0088 seconds. I bought these IC's just for this forum, to test and listen to, and compare. Soon, we will see the differences we hear. The only way I have to learn is by doing, even if it is wrong to begin with. By eliminating a bunch of wrong ways, I eventually get to the right way. Thank you for nudging me in the right direction, kw.

If you don't mind, digger, a couple more questions;

1. What music or program material are you listening to when you hear these diffrences?
2. What exactly do you plan to measure, and how? What instrumentation?

Thanks!

 

[digger945]

Deleted, explained in post #167.

 

[kwkarth]

In looking at and evaluating time domain related data, you may find meaningful differences revealed with square wave data.

From the snippets you shared, I would guess the upper trace in the first graph is the silver interconnect and your trace correlated with faster rise times and maybe a tendency to ring or overshoot or at least sound brighter. You might also want to check the capacitive loading of each cable. Are you showing left and right channel or a single channel for both cables in each graph?

An interesting thing to do would be to superimpose input and output, as well as cable to cable waveforms. Superimposing input and output would show you the absolute fidelity or faithfulness of the signal propagation.

 

[digger945]

Quote:

Originally Posted by kwkarth /img/forum/go_quote.gif In looking at and evaluating time domain related data, you may find meaningful differences revealed with square wave data. From the snippets you shared, I would guess the upper trace in the first graph is the silver interconnect and your trace correlated with faster rise times and maybe a tendency to ring or overshoot or at least sound brighter. You might also want to check the capacitive loading of each cable. Are you showing left and right channel or a single channel for both cables in each graph? An interesting thing to do would be to superimpose input and output, as well as cable to cable waveforms. Superimposing input and output would show you the absolute fidelity or faithfulness of the signal propagation.

How do I check the capacitive loading of a cable?
Each graph is right channel(upper waveform) and left(lower) for just one cable, you have to do some fast scrolling to campare the yellow graph(Monster cable) to the orange graph(cryo). Superimposing will be just as soon as I can find suitable software. That's actually what I wanted from the onset, but I am impatient.
I can show another graph with the "original recording", but I am gonna do the whole test again, with a song that has "real instruments". The first song I listened to that had the most notable difference was "Forget me nots" by Patrice Rushen, but I am sure the song is full of artificial synths also.

 

[JadeEast]

Digger

Could you do a number of recordings of the same cable and see if the wave form data is the same? Multiple recordings may show how much noise and randomness is in the displayed recording waveforms.

P.s. I can do a quick super impose if you would like.

 

[digger945]

Jade, if I give you the photos or images, you can superimpose them all together? How about the "original" plus cable a plus cable b (three images)?

 

[kwkarth]

Quote:

Originally Posted by digger945 /img/forum/go_quote.gif How do I check the capacitive loading of a cable? Each graph is right channel(upper waveform) and left(lower) for just one cable, you have to do some fast scrolling to campare the yellow graph(Monster cable) to the orange graph(cryo). Superimposing will be just as soon as I can find suitable software. That's actually what I wanted from the onset, but I am impatient. I can show another graph with the "original recording", but I am gonna do the whole test again, with a song that has "real instruments". The first song I listened to that had the most notable difference was "Forget me nots" by Patrice Rushen, but I am sure the song is full of artificial synths also.

You check capacitance with a capacitance checker.

Sorry. Many DVM's these days can check capacitance usually @ 400Hz.

So by your graphs, the top trace in the yellow graph looks quicker and the bottom trace looks quicker in the orange graph. Are you sure you have captured the exact same moment in time?

Try a square wave at say 10kHz and focus in on the rising and falling knee in the waveform.

 

[digger945]

Quote:

Originally Posted by kwkarth /img/forum/go_quote.gif Are you sure you have captured the exact same moment in time?

LOL! I am 99% sure that I am mostly correct

You simply cannot understand just how hard it is to pick out a specific point only 1/100th second long at the same spot in two different recordings, but I think I got it right this time.
If I am doing it right the cryo is .012 nF for both signal and ground, and the Monster is .018 nF signal and almost nil on the ground.

I'm also looking for some software to generate some test signals.