REVIEW INCL. Just got monoprice speaker wire in.....why do you guys reccomend this so much? - Page 13 - AVS Forum
Forum Jump: 
 34Likes
Reply
 
Thread Tools
post #361 of 372 Old 10-04-2014, 02:37 PM
AVS Addicted Member
 
Ratman's Avatar
 
Join Date: May 2002
Location: Collingswood, N.J.
Posts: 14,643
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 229 Post(s)
Liked: 314
Buy 10 AWG. That way all the bases are covered. Until the next revealing/comparing speaker wire test is performed.
Ratman is online now  
Sponsored Links
Advertisement
 
post #362 of 372 Old 10-04-2014, 03:37 PM
AVS Special Member
 
CharlesJ's Avatar
 
Join Date: Jan 2006
Posts: 3,408
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 188 Post(s)
Liked: 117
Quote:
Originally Posted by Ratman View Post
Buy 10 AWG. That way all the bases are covered. Until the next revealing/comparing speaker wire test is performed.
I am waiting for the transfer functions to be measure and compared to the 22 ga wire or that 25 ga wire.
CharlesJ is offline  
post #363 of 372 Old 10-04-2014, 03:54 PM
AVS Addicted Member
 
Ratman's Avatar
 
Join Date: May 2002
Location: Collingswood, N.J.
Posts: 14,643
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 229 Post(s)
Liked: 314
If you're referencing the locked (Frozen) thread... it's probably best to "let it go".
Ratman is online now  
post #364 of 372 Old 10-04-2014, 04:18 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,373
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 980 Post(s)
Liked: 546
Quote:
Originally Posted by CharlesJ View Post
I need to revisit this as I just read it. Where does that .5 dB audibility threshold come from?
See the text and references in my WSR article on audibility of small distortions: http://www.madronadigital.com/Librar...stortions.html

Quote:
And, specifically at 100 Hz per your specific example since JND is not linear across the audio spectrum, right?
It doesn't quite work that way in this case. Pls read the references.

Quote:
Also, that dip to 2 Ohms is a rather narrow band at 100Hz and specific to that example.
This is true although this is a statistical problem as I explain in the article (which is summary of research/listening tests). The dip is always audible but the narrower it gets, the less the chance that the listener can pick it up.

Quote:
http://home.provide.net/~djcarlst/abx_crit.htm
This graph from a JAES paper would not agree as top audibility of .5 dB at 100Hz and so narrow in scope. Your link in the previous post was for distortions. I don't think a volume change is a distortion so how would that apply?
Just curious to understand.
No problem . That graph is misused in that reference and in general in forums. Its foundation is not remotely as strong as the research I cite. First let me show the graph you/david reference:



That comes from Ethan's reprint of Clark's SynAudCon which provides more clarity on how they were created: http://ethanwiner.com/phase.html

And this important text:

FREQUENCY RESPONSE AUDIBILITY

Figure 2 [above] is a graph of various bandwidths of minimum phase response aberrations or level mismatches which have proven to be just inaudible in double-blind testing. For most of this data the subject was myself. Spot checks showed that I am fairly typical in my ability, but additional testing would only serve to lower these curves. As they stand, they indicate that very small ripples in the frequency response such as those generated by a slight time offset could be audible.

So as you see, the graph is based on ad-hoc testing to create gross data, not any search for minimum audible threshold, i.e. Just Noticeable Difference (JND). Even there it confirms my data by the mention of small ripples being audible (which are kind of hard to see in his small graph).

The reference I use from Toole/Olive is a much more rigorous analysis where it searches for JND both narrow and wide resonances and with different content. It obsoletes the above graph and provides much more accurate metrics.

I can't find my copy of Clark's AES paper but the language there also explains that the graph is based on ad-hoc testing.

Amir
Retired Technology Insider
Founder, Madrona Digital
"Insist on Quality Engineering"
amirm is offline  
post #365 of 372 Old 10-05-2014, 12:42 PM
AVS Special Member
 
CharlesJ's Avatar
 
Join Date: Jan 2006
Posts: 3,408
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 188 Post(s)
Liked: 117
Quote:
Originally Posted by amirm View Post
See the text and references in my WSR article on audibility of small distortions: http://www.madronadigital.com/Librar...stortions.html


It doesn't quite work that way in this case. Pls read the references.


This is true although this is a statistical problem as I explain in the article (which is summary of research/listening tests). The dip is always audible but the narrower it gets, the less the chance that the listener can pick it up.


No problem . That graph is misused in that reference and in general in forums. Its foundation is not remotely as strong as the research I cite. First let me show the graph you/david reference:



That comes from Ethan's reprint of Clark's SynAudCon which provides more clarity on how they were created: http://ethanwiner.com/phase.html

And this important text:

FREQUENCY RESPONSE AUDIBILITY

Figure 2 [above] is a graph of various bandwidths of minimum phase response aberrations or level mismatches which have proven to be just inaudible in double-blind testing. For most of this data the subject was myself. Spot checks showed that I am fairly typical in my ability, but additional testing would only serve to lower these curves. As they stand, they indicate that very small ripples in the frequency response such as those generated by a slight time offset could be audible.

So as you see, the graph is based on ad-hoc testing to create gross data, not any search for minimum audible threshold, i.e. Just Noticeable Difference (JND). Even there it confirms my data by the mention of small ripples being audible (which are kind of hard to see in his small graph).

The reference I use from Toole/Olive is a much more rigorous analysis where it searches for JND both narrow and wide resonances and with different content. It obsoletes the above graph and provides much more accurate metrics.

I can't find my copy of Clark's AES paper but the language there also explains that the graph is based on ad-hoc testing.
Yes, the paper had a rather small number of subjects. And I think it tested audibility of level differences in those 3 octave categories.
Now, that 2 Ohm dip considered a resonance condition? Is that why Tool's test more relevant here? Did he take into consideration of the frequency bands and the varying sensitivity? I don't have that paper, thanks.
CharlesJ is offline  
post #366 of 372 Old 10-05-2014, 03:07 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,373
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 980 Post(s)
Liked: 546
Quote:
Originally Posted by CharlesJ View Post
Now, that 2 Ohm dip considered a resonance condition? Is that why Tool's test more relevant here? Did he take into consideration of the frequency bands and the varying sensitivity? I don't have that paper, thanks.
He has a summary of it in his book. But briefly, no the testing was not resonance. The goal was to understand the effect of resonance but the test itself just checked frequency response variations of different amounts. Since low frequency resonances cause frequency response variations, that was a good method to tease out its audible effects.

Amir
Retired Technology Insider
Founder, Madrona Digital
"Insist on Quality Engineering"
amirm is offline  
post #367 of 372 Old 10-05-2014, 03:09 PM
AVS Special Member
 
CharlesJ's Avatar
 
Join Date: Jan 2006
Posts: 3,408
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 188 Post(s)
Liked: 117
Quote:
Originally Posted by amirm View Post
He has a summary of it in his book. But briefly, no the testing was not resonance. The goal was to understand the effect of resonance but the test itself just checked frequency response variations of different amounts. Since low frequency resonances cause frequency response variations, that was a good method to tease out its audible effects.
Thanks for both responses.
CharlesJ is offline  
post #368 of 372 Old 10-06-2014, 03:41 PM
AVS Special Member
 
krabapple's Avatar
 
Join Date: Apr 2003
Location: in a state bordered by Kentucky and Maine
Posts: 5,356
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 175 Post(s)
Liked: 227
Quote:
Originally Posted by CharlesJ View Post
I need to revisit this as I just read it. Where does that .5 dB audibility threshold come from?



Quote:
The "just noticeable difference," or JND for amplitude — that is, the minimal perceptible change in amplitude — varies by the starting amplitude and frequency, but in general it ranges between 0.2 and 0.4 dB.
http://www.indiana.edu/~emusic/etext...loudness.shtml


NB, the lower JND applies more to test signals like pure tones or pink noise, than music, where the JND is higher, so sometime's you'll see figures like 1 dB or even 3dB cited. But to be rigorous, matching to <=0.2 should be unassailable.

Last edited by krabapple; 10-06-2014 at 03:46 PM.
krabapple is offline  
post #369 of 372 Old 10-06-2014, 04:08 PM
AVS Special Member
 
CharlesJ's Avatar
 
Join Date: Jan 2006
Posts: 3,408
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 188 Post(s)
Liked: 117
Quote:
Originally Posted by krabapple View Post
http://www.indiana.edu/~emusic/etext...loudness.shtml


NB, the lower JND applies more to test signals like pure tones or pink noise, than music, where the JND is higher, so sometime's you'll see figures like 1 dB or even 3dB cited. But to be rigorous, matching to <=0.2 should be unassailable.
Not sure whose quote is that 2nd one but that .2 and .4 is not totally accurate, certainly not for the full spectrum.
Florentine published a test that actually found it much higher above 10kHz and down lower even though it only went to 250 Hz.
Florentine, Mary, et al 'Level Discrimination as a Function of Level for Tones from .25 to 16kHz. Journal of Acoustic Society of America, 81(5) May 1987, pg 1528-1541.
She didn't take into account the octave range effect. She had more that 1 subject.
CharlesJ is offline  
post #370 of 372 Old 10-06-2014, 04:23 PM
AVS Special Member
 
krabapple's Avatar
 
Join Date: Apr 2003
Location: in a state bordered by Kentucky and Maine
Posts: 5,356
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 175 Post(s)
Liked: 227
Quote:
Originally Posted by CharlesJ View Post
Not sure whose quote is that 2nd one but that .2 and .4 is not totally accurate, certainly not for the full spectrum.
Florentine published a test that actually found it much higher above 10kHz and down lower even though it only went to 250 Hz.
Florentine, Mary, et al 'Level Discrimination as a Function of Level for Tones from .25 to 16kHz. Journal of Acoustic Society of America, 81(5) May 1987, pg 1528-1541.
She didn't take into account the octave range effect. She had more that 1 subject.
The quote is from the link I posted right under it -- to an online textbook.

No one said 0.2-0.4 dB JND was for the full spectrum. As I wrote, those low extremes are for the most discrimination-enhancing probe signals.

For example,
Quote:
Toole and Olive, on the other hand, in their 1988 study used pink noise for their acoustic signal source and determined that a 5 kHz resonance, with Q = 1 was just detectible at .25 dB.

They found pink noise to be the most revealing signal. They did employ other types of acoustic signals in their study and discovered that when using the least revealing of these that just detectible threshold increased by a factor of 5

Last edited by krabapple; 10-06-2014 at 04:27 PM.
krabapple is offline  
post #371 of 372 Old 10-06-2014, 04:33 PM
AVS Special Member
 
CharlesJ's Avatar
 
Join Date: Jan 2006
Posts: 3,408
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 188 Post(s)
Liked: 117
Quote:
Originally Posted by krabapple View Post
The quote is from the link I posted right under it -- to an online textbook.

No one said 0.2-0.4 dB JND was for the full spectrum. As I wrote, those low extremes are for the most discrimination-enhancing probe signals.

For example, "Toole and Olive, on the other hand, in their 1988 study used pink noise for their acoustic signal source and determined that a 5 kHz resonance, with Q = 1 was just detectible at .25 dB.

They found pink noise to be the most revealing signal. They did employ other types of acoustic signals in their study and discovered that when using the least revealing of these that just detectible threshold increased by a factor of 5"
Thanks, just saw it
Did Toole and Olive investigate down around 100Hz that was used in the speaker dip or up at 16 kHz?
So, I would think that if say music is used at tha 100Hz dip, the JND would be higher than the posted 0.5 dB?
CharlesJ is offline  
post #372 of 372 Old 10-06-2014, 05:18 PM
AVS Special Member
 
CharlesJ's Avatar
 
Join Date: Jan 2006
Posts: 3,408
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 188 Post(s)
Liked: 117
Quote:
Originally Posted by amirm View Post
He has a summary of it in his book. But briefly, no the testing was not resonance. The goal was to understand the effect of resonance but the test itself just checked frequency response variations of different amounts. Since low frequency resonances cause frequency response variations, that was a good method to tease out its audible effects.
OK, I went to your link and spent more time re-reading it and another place that briefly mentions Tools finding on this JND.
I would think that in your paper the high Q would be a much narrower band than the medium Qand the low Q. And, I would suspect the Q correlates with Clark's reference to 1/3 octave, 1 octave, and 3 octaves. And, I would think that Tooles find of 0.25 dB for a low Q at 5 kHz would also correlate with Clark's paper.
Your link indicates that it is much easier to detect a low Q than a high Q. That is exactly what Clarks chart shows as well, so I don't see why it is wrong, just because it had 1 subject? Didn't they have similar findings?

I would think that speaker dip at 100 Hz is considered a high Q event, not so easy to detect? Another note is the very sensitive test signals used in both cases, Toole And Clark. And, Tool mentions a 5 fold increase when using much less sensitive signals; perhaps music?
So, I question that 0.5 dB audibility at 100Hz especially since I found Tools paper and he states" With many common sound, under common listening circumstances, it is surprising just how much the measured performance of a signal path can be modified without significantly altering perceived timber."
No specific mention of detection dB at 100 Hz, but may be there someplace.
But, again, I have no standing in this.
CharlesJ is offline  
Reply Audio Theory, Setup, and Chat

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off