Any sound degradation from using a 3.5mm to dual RCA cable? - Page 4 - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #91 of 119 Old 04-29-2014, 08:07 PM
 
gazoink's Avatar
 
Join Date: Apr 2014
Posts: 68
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 16
Quote:
Originally Posted by amirm View Post

My question to you was regarding this comment from you: "Yes, all analog recording devices have scrape flutter to some degree. And no, that has nothing to do with this tread."

First I am puzzled that you brought up the topic of "scape flutter" yet now say it has nothing to do with the thread. 

OMG!  "scRape flutter". I MADE A FREAKING TYPO, ok? Harping on someones error will not get you what you want.  I promise.

Quote:

Originally Posted by amirm View Post
I like to know the spectrum of the distortion and test signal that you and Arny say is clearly audible to you. Surely you measured what you heard. Even if you don't have the measurements anymore, I like to know what you remember it being.

1. Get used to disappointments.  

 

2. My work was to minimize scrape flutter, not determine it's audible threshold(s) for widely varying recorded material and it's masking characteristics.  I was maintaining tape recorders not writing an AES paper.

 

3. Neither Arny nor I ever said anything specific about was clearly audible to either of us as pertains to scRape flutter, except that if it's bad enough, it's audible.   

 

But, you see, I get it.  Actually, several of us get it.  Even if I had that data you're asking for, I wouldn't be sharing it with you, because what you're really looking for is something to pick at, find fault with, and discredit, so you can "win", even if only in Amir-land.  If you REALLY wanted to "know", you dust off that Otari 5050, slap a hunk of tape on it, lash it to your $20K AP rig and get your own answers. But it's not the answers you seek, you seek something with which to discredit others and make darn good and sure that Amir is Audiogod.

 

Now, I'm going to type this next bit very, very, slowly, and, very, carefully, so you'll be able to Google the term correctly, and hopefully won't find some new error of mine to harp on.

 

Google this: "superiority complex".  And don't come back here asking why you were told to do so.

 

It's actually sad, what you're doing.  You do have a lot of valuable knowledge to offer.  But as long as winning the contest and p****** farther is more important than learning a concept, you will likely continue these tactics of sarcasm, down-talking, baiting, and misquoting (actually quote fabricating would be more correct) with intent to discredit.  

 

Sorry, I'm not playing. There's simply no point in helping you maintain that fine thumbs-up ratio.

 

And yes, this entire discussion is off topic.  Just check the thread title. 

Tack and 67jason like this.
gazoink is offline  
Sponsored Links
Advertisement
 
post #92 of 119 Old 04-29-2014, 11:42 PM
AVS Special Member
 
67jason's Avatar
 
Join Date: Jan 2011
Posts: 2,870
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 61 Post(s)
Liked: 778
Quote:
Originally Posted by gazoink View Post

Quote:
Originally Posted by amirm View Post

My question to you was regarding this comment from you: "Yes, all analog recording devices have scrape flutter to some degree. And no, that has nothing to do with this tread."


First I am puzzled that you brought up the topic of "scape flutter" yet now say it has nothing to do with the thread. 
Quote:
I like to know the spectrum of the distortion and test signal that you and Arny say is clearly audible to you. Surely you measured what you heard. Even if you don't have the measurements anymore, I like to know what you remember it being.

big +1...but somehow I am sure he will be back nitpicking this post. rolleyes.gif best just to ignore him...I do.smile.gif

I don't need snobs to tell me how to think, thank you!
LOL!
Why you wouldn't want to join this forum
67jason is offline  
post #93 of 119 Old 04-30-2014, 03:54 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,916
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 462 Post(s)
Liked: 1069
Quote:
Originally Posted by amirm View Post


Here is the Linn measurements again:



Can you please point to where you see evidence of FM modulation? And how you determined that those sidebands at 900 and 1100 Hz are audible? Thanks in advance.

Here is the above with some relevant annotation:



The above shows multiple sidebands about every 11 Hz. 11 Hz is a common tone arm resonance frequency. This paper may shed additonal light:

http://www.theanalogdept.com/images/spp6_pics/TT_Design/MechanicalResonances.pdf - gets really interesting around page 7

Here is the spectrum of an AM-modulated signal configured to resemble Vref modulation in a DAC:



Note the number of sidebands: exactly 2. The above is also relevant to the amount of modulation shown in the Linn example: 0.1%

Here is the spectrum of an FM-modulated signal configured to resemble the Linn signal posted above:



Note the number of sidebands > 2

In fact the number of sidebands due to FM modulation can be even numbers from 2 on up. Generally the higher the modulation index, the higher probability of more pairs of sidebands When there is more than 1 pair of sidebands, there is a strong indication of FM.

A stronger and more positive indication can be obtained when phase measurements are available.

Here is the X/Y vector plot of an AM-modulated signal, again configured to resemble the small amounts of AM characteristic of Vref modulation in a DAC:

The graph of the signal is the purple diagonal line. The amount of modulation is too low to have very visible effects.



Here is FFT plot of an AM-modulated signal, again configured with very high amounts of AM modulation to show the generality of the 2 sideband rule:



Here is X/Y vector plot of an AM-modulated signal, again configured with very high amounts of AM modulation. It is prretty striking but it shows that phases of the 2 sidebands are up to 180 degrees apart. The phase of the modulated signal and its amplitude are constantly varing.



Here is the X/Y vector plot of an FM-modulated signal configured to resemble the Linn signal posted above:



The graph is not the open ellipse we see with static phase shift. The center of the graph is actually filled in showing the constant amplitude but constantly varying phase relationships within the FM signal.

Homework assignment for discussion: Is the distortion in the Linn signal predominantly AM or FM? Why?

Here is the corrected plot of the Linn and 2 AVRs:



Homework assignment for discussion: Are the AVR signals predominantly showing AM or FM distortion? Why?
arnyk is offline  
post #94 of 119 Old 04-30-2014, 04:15 AM
 
gazoink's Avatar
 
Join Date: Apr 2014
Posts: 68
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 16
Thank you Arny! Brilliant!
gazoink is offline  
post #95 of 119 Old 04-30-2014, 06:58 AM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,836
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 554 Post(s)
Liked: 341
Quote:
Originally Posted by arnyk View Post

Here is the above with some relevant annotation:

Top of the morning to you Arny! I see that you are fully engaged in this thread again which is cause for joy as we are going to learn a ton I am sure.

As to the rest of your answer, while I normally welcome others to answer questions I put to someone specific, in this case I had not seen you agree or disagree with the question. Hence the reason I was asking specific questions from Frank to better understand these posts:
Quote:
Originally Posted by Frank Derks View Post

Wong again Amir. using the corrected graph the sidebands are now at -69dB. Also it's fm modulation which differs from the traditional two tone masking.

http://www.isa-audiology.org/periodicals/1971-2001_Audiology/1979,%20%20Audiology,%20%20Vol.%20%2018/No.%204%20%20(265-352)/Nelson,%20%20Audiology,%20%201979.pdf
Quote:
Originally Posted by Frank Derks View Post

A distortion component 100Hz under and above a 1Khz test signal isn't masked. a distortion component 25Hz under and above a 12kHz test signal will be masked.

I hope it is not too much trouble to first ask whether you agree or disagree that -69 db sidebands are audible relative to 1KHz main tone.

If you don't mind, I will come back to a few things you said in your response later. For now, it would be good to establish if we are on the same side of the table or not. If we are, then there is no reason for long replies smile.gif.

Thanks in advance.

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #96 of 119 Old 04-30-2014, 07:46 AM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,836
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 554 Post(s)
Liked: 341
Quote:
Originally Posted by gazoink View Post

Thank you Arny! Brilliant!
Arny is so knowledgeable that often he talks over our heads. So if you don't mind, seeing how you understood his post, could tell us what was brilliant and why so that we can follow along.

Thanks,

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #97 of 119 Old 04-30-2014, 09:14 AM
Member
 
Riffmeister's Avatar
 
Join Date: Jul 2003
Location: Indianapolis
Posts: 187
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 31
Quote:
Originally Posted by amirm View Post

Arny is so knowledgeable that often he talks over our heads. So if you don't mind, seeing how you understood his post, could tell us what was brilliant and why so that we can follow along.

Thanks,
He was obviously talking about the brilliant colors in the meaningless graphs.
amirm likes this.
Riffmeister is offline  
post #98 of 119 Old 04-30-2014, 10:04 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,916
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 462 Post(s)
Liked: 1069
Quote:
Originally Posted by amirm View Post

Quote:
Originally Posted by gazoink View Post

Thank you Arny! Brilliant!
Arny is so knowledgeable that often he talks over our heads. So if you don't mind, seeing how you understood his post, could tell us what was brilliant and why so that we can follow along.

Why not at least try asking me those questions?
arnyk is offline  
post #99 of 119 Old 04-30-2014, 10:05 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,916
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 462 Post(s)
Liked: 1069
Quote:
Originally Posted by Riffmeister View Post

Quote:
Originally Posted by amirm View Post

Arny is so knowledgeable that often he talks over our heads. So if you don't mind, seeing how you understood his post, could tell us what was brilliant and why so that we can follow along.

Thanks,
He was obviously talking about the brilliant colors in the meaningless graphs.

I'm always interested in improving the readability of my posts. Please tell me about which graphs you find particularly meaningless.
arnyk is offline  
post #100 of 119 Old 04-30-2014, 10:15 AM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,836
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 554 Post(s)
Liked: 341
Quote:
Originally Posted by arnyk View Post

Why not at least try asking me those questions?
OK, what is brilliant about your post?

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #101 of 119 Old 04-30-2014, 10:43 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,916
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 462 Post(s)
Liked: 1069
Quote:
Originally Posted by amirm View Post

Quote:
Originally Posted by arnyk View Post

Here is the above with some relevant annotation:

Top of the morning to you Arny! I see that you are fully engaged in this thread again which is cause for joy as we are going to learn a ton I am sure.

As to the rest of your answer, while I normally welcome others to answer questions I put to someone specific, in this case I had not seen you agree or disagree with the question. Hence the reason I was asking specific questions from Frank to better understand these posts:
Quote:
Originally Posted by Frank Derks View Post

Wong again Amir. using the corrected graph the sidebands are now at -69dB. Also it's fm modulation which differs from the traditional two tone masking.

http://www.isa-audiology.org/periodicals/1971-2001_Audiology/1979,%20%20Audiology,%20%20Vol.%20%2018/No.%204%20%20(265-352)/Nelson,%20%20Audiology,%20%201979.pdf
Quote:
Originally Posted by Frank Derks View Post

A distortion component 100Hz under and above a 1Khz test signal isn't masked. a distortion component 25Hz under and above a 12kHz test signal will be masked.

I hope it is not too much trouble to first ask whether you agree or disagree that -69 db sidebands are audible relative to 1KHz main tone.

If you don't mind, I will come back to a few things you said in your response later. For now, it would be good to establish if we are on the same side of the table or not. If we are, then there is no reason for long replies smile.gif.

Thanks in advance.

The question above seems to be poorly stated.

The frequency that modulates the 1 Khz tone in the Linn turntable test appears to be approximately 11 Hz, and due to tone arm resonance.

This is an old document by Zwicker (1974) but it appears to me to still be relevant:

https://ia600501.us.archive.org/21/items/nasa_techdoc_19740015158/19740015158.pdf

Page 12:

"Even the acoustic impression shows parallels. For instance,
the ear can still follow the variations of the frequency up to
some 4 Hz, while at modulation frequencies up to some 20 Hz the
frequency variation seems harsher, so that the ear no longer
seems quite able to follow the frequency change. The sensitivity
decreases again. Above 20 Hz the tone becomes uneven and bubbling,
until the side frequencies become audible at about 125 Hz. Here,
too, these are affected only by the masking and are perceived up
to the highest modulation frequencies."

This clearly states that FM modulation with modulating frequencies up to "about 125" Hz is not based on side frequencies (or sidebands) and thus sideband masking is not an appropriate criteria to apply.

This graphic accompanies the above text:



Among other things it along with accompanying text makes the clear point that the audibility of FM distortion is highly dependent on the SPL of the tone being modulated. Since none was given in the question, no relevant answer can be given.

That all said, I believe that it has been established that the 11 Hz modulation of a 1 Hz tone found in the Linn test had an approximate deviation of 1 Hz. The chart above puts the threshold of audibility far an 11 Hz modulation of a 1 Khz tone as a deviation of between 1.5 and 2.5 Hz at all sound levels tested. The modulation is probably not all that easily heard. I created some samples and tried to ABX them using FooBar2000 and my old ears which may not be representative of the most sensitive possible listener were reduced to random guessing.
arnyk is offline  
post #102 of 119 Old 04-30-2014, 10:46 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,916
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 462 Post(s)
Liked: 1069
Quote:
Originally Posted by amirm View Post

Quote:
Originally Posted by arnyk View Post

Why not at least try asking me those questions?
OK, what is brilliant about your post?

The work of the people whose findings I based it on.
arnyk is offline  
post #103 of 119 Old 04-30-2014, 12:04 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,836
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 554 Post(s)
Liked: 341
Quote:
Originally Posted by arnyk View Post

The question above seems to be poorly stated.
My apologies as always smile.gif. Here is the question simply put by Frank: "A distortion component 100Hz under and above a 1Khz test signal isn't masked.". The 100 Hz sideband is said to be -69 db.

Please help us understand Frank's comment. Is it correct and if so, why?

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #104 of 119 Old 04-30-2014, 12:07 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,836
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 554 Post(s)
Liked: 341
Quote:
Originally Posted by arnyk View Post

The work of the people whose findings I based it on.
Ah, you are being modest as always Arny. There is a lesson in that for sure smile.gif.

Something tells me that Gaz was not celebrating those references but rather, you showing how feeble my understanding of audio is relative to yours. Don't you agree and if so, what was the core message there that got him excited that much?

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #105 of 119 Old 04-30-2014, 12:09 PM
AVS Addicted Member
 
Ratman's Avatar
 
Join Date: May 2002
Location: Collingswood, N.J.
Posts: 14,308
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 93 Post(s)
Liked: 260
Quote:
Any sound degradation from using a 3.5mm to dual RCA cable?
Remember the opening post?

This bickering is sad. Take it to Twitter.
Ratman is online now  
post #106 of 119 Old 04-30-2014, 01:53 PM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,916
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 462 Post(s)
Liked: 1069
Quote:
Originally Posted by amirm View Post

Quote:
Originally Posted by arnyk View Post

The question above seems to be poorly stated.
My apologies as always smile.gif. Here is the question simply put by Frank: "A distortion component 100Hz under and above a 1Khz test signal isn't masked.". The 100 Hz sideband is said to be -69 db.

Please help us understand Frank's comment. Is it correct and if so, why?

I cannot understand the comment for the reasons previously given.

Perhaps Frank may have something to say in the light of the evidence that has been recently provided if it was new to him. Or not. He speaks for himself!
arnyk is offline  
post #107 of 119 Old 04-30-2014, 03:33 PM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,916
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 462 Post(s)
Liked: 1069
Quote:
Originally Posted by amirm View Post

Quote:
Originally Posted by arnyk View Post

The work of the people whose findings I based it on.
Ah, you are being modest as always Arny. There is a lesson in that for sure smile.gif.

Something tells me that Gaz was not celebrating those references but rather, you showing how feeble my understanding of audio is relative to yours.

I'm sure that you have heard of unintended consequences.
Quote:
Don't you agree and if so, what was the core message there that got him excited that much?

My true goal is to help create a context in which people receive the best possible answers for their questions. Given the size of this place, that is something that no single person can accomplish by himself.

In some sense every time I have to answer a question to move closer to that goal is a small failure.
arnyk is offline  
post #108 of 119 Old 04-30-2014, 04:33 PM
 
mc2ed's Avatar
 
Join Date: Apr 2014
Posts: 198
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 20
Thyrone!

Are you still following your own thread?
mc2ed is offline  
post #109 of 119 Old 04-30-2014, 04:53 PM
 
gazoink's Avatar
 
Join Date: Apr 2014
Posts: 68
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 16
Quote:
Originally Posted by amirm View Post


Arny is so knowledgeable that often he talks over our heads. So if you don't mind, seeing how you understood his post, could tell us what was brilliant and why so that we can follow along.

Thanks,

I did understand his post, thoroughly, which is why I said it was brilliant.  And yes, I do mind.  I'm not playing the game, Amir, and I told you that.  If you need it explained (which I highly doubt), go Google a while, you'll get what you need.  Otherwise, your request is just another obvious "bait".  I'm not biting.

 

If anyone other than amirm wants me to explain Arny's post, PM me, I'm happy to do so.  Sorry, it's the only way not to get Amired in again.  I'm sure Arny will or has done a fine job on his own anyway (haven't read today's posts yet, working all day).  

gazoink is offline  
post #110 of 119 Old 04-30-2014, 04:55 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,836
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 554 Post(s)
Liked: 341
Quote:
Originally Posted by arnyk View Post

I cannot understand the comment for the reasons previously given.

Perhaps Frank may have something to say in the light of the evidence that has been recently provided if it was new to him. Or not. He speaks for himself!
I hope he does but I am left with you having answered on his behalf Arny. smile.gif So just to be clear, you can't tell if a 900 Hz distortion at -69 db relative to a 1000 Hz is audible?

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #111 of 119 Old 04-30-2014, 04:59 PM
 
gazoink's Avatar
 
Join Date: Apr 2014
Posts: 68
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 16
Quote:
Originally Posted by amirm View Post


Ah, you are being modest as always Arny. There is a lesson in that for sure smile.gif.

Something tells me that Gaz was not celebrating those references but rather, you showing how feeble my understanding of audio is relative to yours. Don't you agree and if so, what was the core message there that got him excited that much?

Nope, I was celebrating the references, and the presentation of relevant data.  I don't celebrate anyone's feeble understanding in anything, that's just not kind, nor do I repeatedly highlight casual errors.

 

As to the core of why I got excited that much...as I said...get used to disappointments. 

gazoink is offline  
post #112 of 119 Old 04-30-2014, 05:33 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,836
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 554 Post(s)
Liked: 341
Quote:
Originally Posted by arnyk View Post

Here is the above with some relevant annotation:



The above shows multiple sidebands about every 11 Hz. 11 Hz is a common tone arm resonance frequency.
That's just shocking! Shockingly good! I am so amazed that you could resolve those spikes down to single frequency Hertz on the X axis by just eyeballing it. I hope this is one of those gifts you said your instructors gave you at school. I am going to commit suicide if you say you were born with such abilities smile.gif. Heaven knows I could not do that.

What I can do however is read the stereophile review that you linked to earlier that went with that measurement: http://www.stereophile.com/content/linn-lp-playing-system-measurements

"The tonearm resonance with the Arkiv lay at 10Hz; the "shoulders" at exactly 10Hz on either side of the central peak are due to this resonance. They lie at -41dB ref. 5cm/s, so it's hard to predict what their subjective effect will be."

So fortunately we don't have to speculate here about the frequency. We know it was at 10 Hz and it is indeed tonearm induced. Which begs the question why you said this when presenting that chart:

"BTW if one measures any of these effects (which I have) the same sideband structure can be seen in the spectral analysis. IOW the sideband structure we see arising from digital jitter are also present in recordings that are contaminated by flutter and wow from analog recordings, FM distortion from loudpseakers, and even vibrato in notes played on certain musical instruments. Case in point: LP playback on a Linn TT:"


Looks like you attributed the distortion sidebands to Wow and Flutter rather than tonearm resonating. Two very different things, right? And as such have no parallels to digital jitter.

JA goes on to explain the cause of the other pair of sidebands:

"Small spurs at ±20Hz, the second harmonic of the tonearm resonance, can also be seen, but these are 60dB down in level."

Makes sense, right? So those are not speed modulations either.

The rest of may be attributable to the synchronous motor. This being an analog system, JA correctly speculates that one can't be sure of what is being looked at. He measures a second sample and gets this:

LINrpsFIG2.jpg

Quite a bit better! The whole +-100 Hz distortion is gone. But wait, the story gets better. Here is the only other turntable they had tested, a Pioneer PL L1000-AT 8008 turntable:

LINrpsFIG3.jpg

Man, that looks clean. No correlated distortions at all. The sidebands are down to -90 db just 200 +- of our main tone!!!

Shocker, right? So many decades later our digital systems can't produce a distortion free digital audio experience that good out of HDMI.

I am worried Arny that the other side is going to say we chose to put the worst measurement out there rather than full disclosure. And they might have a point.

I have not had time yet to analyze the rest of your post but IMHO, we are not looking too good.

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #113 of 119 Old 04-30-2014, 06:27 PM
 
gazoink's Avatar
 
Join Date: Apr 2014
Posts: 68
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 16

Huh.  So you really Don't know the difference between distortion products and modulation sidebands.  Interesting.  

 

The word "distortion" doesn't occur even once in the Stereophile article you linked to.

gazoink is offline  
post #114 of 119 Old 04-30-2014, 07:02 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,836
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 554 Post(s)
Liked: 341
Quote:
Originally Posted by gazoink View Post

Huh.  So you really Don't know the difference between distortion products and modulation sidebands.  Interesting.  

The word "distortion" doesn't occur even once in the Stereophile article you linked to.
I am pretty sure JA assumed the reader knows that if the source has a 1Khz tone but what comes out has other frequencies like 900 Hz, that those are distortion products. But maybe you are right and I am the only dummy in these circles. Would you by the same logic say that digital audio jitter is not distortion? You said analog speed modulation is --- how did you call it -- walks like a duck, talks like duck? So if these sidebands in analog are not distortion, they are likewise not in digital jitter. Hopefully you are not saying that. Because if you did, you would be in conflict with industry norm. Here is a very small sampling:

AES Paper: Jitter Simulation in High Resolution Digital Audio
"To reconstruct an audio waveform samples must be located precisely in time. Practical systems have sources of jitter described by both correlated and uncorrelated elements that result in low-level distortion. However, less well known is how different forms of jitter distort an audio signal. Jitter theory is developed to produce a simulator to enable jitter induced distortion to be determined. Distortion spectra can then be observed and time domain distortion auditioned. Jitter induced distortion is compared to a range of errors, including DAC errors and incorrect use of dither. System architectures studied include LPCM with up-sampling and noise shaping and SDM."

AES Paper: Measurement of Sampling Jitter in Analog-to-Digital and Digital-to-Analog Converters Using Analytic Signals
"Sampling jitter is the variation in timing of an audio signal in an analog-to-digital converter (ADC), or a digital-to-analog converter (DAC)[1] . Since sampling jitter causes distortion products during the reproduction and recording of audio signals, methods of measuring and assessing characteristics of sampling jitter must be developed in order to improve the quality of digital audio equipment."

AES Paper: Theoretical and Audible Effects of Jitter on Digital Audio Quality
"Many digital audio systems now use some form of self-clocked digital interface for audio delivery. With the advent of new digital audio systems that use IEC 61937 to convey non-linearly coded audio, the total number of devices using the IEC 60958 interface has substantially increased. The digital interface may contain jitter that translates to distortion in the audio at the point of conversion back to the analog domain. Sources of digital audio, the digital interface, the mechanisms by which errors are introduced, and the effect on DACs are examined."

Earlier Arny taught us that speakers have FM Modulation. So here is a paper on that:
AES Paper: Modulation Distortion in Loudspeakers
"When comparing 2 loudspeakers, one with direct radiator bass system and the other with horn loaded bass, a subjective judgment was that the one with the horn loaded bass is -cleaner.- Both speakers were by the same manufacturer. Various tests were applied and by process of elimination it appears the difference in listening quality is due to frequency modulation distortion."

AES Paper: New Techniques for Evaluating Audio Amplifiers via Measuring for Induced Wow and Flutter and Differential Phase Distortions

Title self explanatory.

AES Paper: Measurement of Flutter and Wow in Magnetic-Tape Instrumentation Recorders (from 1955!)
"In ANY conventional electrical recording system – be it disc, film, or tape – certain distortions in reproduction are inevitable. It is the purpose of this paper to discuss the properties of flutter and wow and the methods of measuring these disturbances."

I will stop here and have you make the case that additional tones are not distortion even though they add "roughness" as you mentioned in your first post as a result of scape flutter.

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #115 of 119 Old 04-30-2014, 07:23 PM
 
gazoink's Avatar
 
Join Date: Apr 2014
Posts: 68
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 16

Sorry, I'm not playing.  I'm not making a case, I'm not playing your word games, and I'm not debating with you.  

 

It's clear that when you look at a spectrum you don't know distortion from modulation sidebands.  No need to cite references for that, it's pretty obvious. 

 

edit: I saw a typo, and had to remove it lest it be re-quoted for the balance of my adult life.

gazoink is offline  
post #116 of 119 Old 05-01-2014, 05:22 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,916
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 462 Post(s)
Liked: 1069
Quote:
Originally Posted by amirm View Post


Quite a bit better! The whole +-100 Hz distortion is gone. But wait, the story gets better. Here is the only other turntable they had tested, a Pioneer PL L1000-AT 8008 turntable:

LINrpsFIG3.jpg

Man, that looks clean.

Looks to me like some important details got lost in a FFT with insufficient resolution.



Two very easy ways to get the necessary resolution come to mind:

(1) FFT with more points - I seem to recall that Spectra Lab and some others offer up to 2 million points or more for about 25 times more resolution.

(2) Since the test is centered at 1 Khz, just downsample the data, and get more mileage out of the (looks like) 65K points that you have.

True confession time. I may have done something along the lines of err stimulating Mr. Atkinson into doing these tests. I suspect that his heart may have not truely been behind the project.
arnyk is offline  
post #117 of 119 Old 05-01-2014, 06:31 AM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,836
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 554 Post(s)
Liked: 341
Quote:
Originally Posted by arnyk View Post

Looks to me like some important details got lost in a FFT with insufficient resolution.

Thank you for your reply. Alas as before, there was no need for graph analysis as the text points out the tonearm resonances there:

"A narrow-band analysis of its performance is shown in fig.3. Slight shoulders can be seen at ±11Hz, which are due to the fundamental tonearm-mass/cartridge-compliance resonance, but the 65,536-point FFT can't resolve any other wow & flutter spuriae."
Quote:
Two very easy ways to get the necessary resolution come to mind:

(1) FFT with more points - I seem to recall that Spectra Lab and some others offer up to 2 million points or more for about 25 times more resolution.

(2) Since the test is centered at 1 Khz, just downsample the data, and get more mileage out of the (looks like) 65K points that you have.
I am very surprised at your comment here Arny. As noted he has 65536 points in the FFT and the bandwidth shown is 400 Hz. That means that the resolution of that chart is 0.006 Hertz! Our auditory filter bandwidth is over 100 Hz in the 1 Khz range that the test tone was at (see my article: http://www.madronadigital.com/Library/RoomReflections.html). So as it is, we have way, way more resolution than what need. Even if the original measurement was done at full bandwidth of say, 22 Khz, we would still have 0.3 Hz resolution which is way more than we need. There is a reason most applications like Adobe Audition which both you and JA used don't go higher than 65,356 points. We would be splitting atoms as the saying goes.

The higher number of FFT points you have heard about are only useful if you are say performing Sonar return analysis at much higher frequencies (e.g. 200 Khz) where you are *looking* at the waveform as opposed to worrying about what you hear. It has no applicability to analyzing what is audible.
Quote:
True confession time. I may have done something along the lines of err stimulating Mr. Atkinson into doing these tests. I suspect that his heart may have not truely been behind the project.
No doubt you are right that most of what we all talk about is stuff you have done years before us. And we have proof of that in the text from JA:

""The tonearm was my 1989 Ekos, the cartridge an early sample of the now discontinued Arkiv, and the spectral analysis was performed using Adobe Audition (previously called Cool Edit Pro) set to a 65,536-point FFT size and a Blackman-Harris window."

Let's look at one of your graphs:


So yeh, you used Cool Edit for analysis and so has he which means for sure the idea was yours and not his. What puzzles me though is what you you said below that graph: "Note the number of sidebands: exactly 2. ". You started with this comment in this post: "important details got lost in a FFT with insufficient resolution." So how do we know if "important" details are not lost in your graph including additional sidebands? You too have a wide skirt around your main tone frequency. How is it that detail is not lost there yet it is in JA's graph? You both use the same tool with the same number of FFT points. What should JA have copied from you that he has not?

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #118 of 119 Old 05-01-2014, 07:41 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,916
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 462 Post(s)
Liked: 1069
Quote:
Originally Posted by amirm View Post


Let's look at one of your graphs:


So yeh, you used Cool Edit for analysis and so has he which means for sure the idea was yours and not his.

I don't know about that. One day I opened Stereophile or their web site and saw a plot that was obviously done with CEP and that was that.
Quote:
What puzzles me though is what you you said below that graph: "Note the number of sidebands: exactly 2. ". You started with this comment in this post: "important details got lost in a FFT with insufficient resolution."


So how do we know if "important" details are not lost in your graph including additional sidebands?

I've published independent sources such as a PDF from if memory serves Agilent's web site that said the same thing. It is an example of low levels of AM where the modulating signal is a sine wave. It is a well known result if you study modulation processes for much more than a few minutes.

Here's a typical reference: http://cp.literature.agilent.com/litweb/pdf/5954-9130.pdf
Quote:
You too have a wide skirt around your main tone frequency. How is it that detail is not lost there yet it is in JA's graph?

Knowing the right answer is a good guide to have when preparing a demonstration. ;-)

I didn't have the option to exercise some of the options I described above because the test signal frequency was so high (12 KHz).
Quote:
You both use the same tool with the same number of FFT points. What should JA have copied from you that he has not?

If the test frequency is low enough running the FFT at a lower sample rate is an option that can yield higher resolution in the frequency domain. In the case of Stereophile's TT tests, the test frequency was 1 KHz and simply using a sample rate that is several times lower than 44 KHz would yield several times better resolution in the frequency domain.
arnyk is offline  
post #119 of 119 Old 05-01-2014, 10:48 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,916
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 462 Post(s)
Liked: 1069
Quote:
Originally Posted by gazoink View Post

Sorry, I'm not playing.  I'm not making a case, I'm not playing your word games, and I'm not debating with you.  

It's clear that when you look at a spectrum you don't know distortion from modulation sidebands.  No need to cite references for that, it's pretty obvious. 

edit: I saw a typo, and had to remove it lest it be re-quoted for the balance of my adult life.

The counterpoint is that much nonlinear distortion is the result of either amplitude or frequency modulation.

So called Vref jitter is usually AM distortion and so-called clock jitter is usually FM distortion.

Some spurious responses found in IM distortion tests can often be predicted by amplitude modulation theory.
arnyk is offline  
Reply Audio theory, Setup and Chat

User Tag List

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