Considering starting a Speaker Cable company - thoughts? - Page 4 - AVS Forum
First ... 2  3  4 5  6 
2-Channel Audio > Considering starting a Speaker Cable company - thoughts?
Mr.SoftDome's Avatar Mr.SoftDome 09:14 PM 06-12-2012
I love this thread.... I'm sorry but these cable threads do add a bit of spice. I'm on the side that cables can make an improvement.

So let me ask this. Is it possible, and even taking science in to account, that to really hear a difference in cables, that your entire rig may have to be at a higher end mid level and up. Not just speaker cables but your power cords, ICs, etc are at that same level.

Your electronics are not a BB AVR and you use separates . Not just entry level separates but above that. Some room panels in place. Doesn't have to be insane.

But, is it not possible if you have went thru all those steps that higher end cables can help to pull it all together?

Why is it always the excited owner with the system in his own house and most familiar with all of thoses pieces of the puzzle slammed as just stupid and have too much money for their own good?

I am one of those different guys. Not rich but above average. Spent years building ( not rich) and consider myself sane and don't do things just because but I have maybe $9k, maybe more in speaker, power and ICs. Power conditioner on top of this.

Do you think if these cables were plugged into an AVR from Best Buy and your fav ID speaker is it possible that you wouldn't notice a difference? Does it take the weakest link and move down for it all come together?

I am thrilled with my set-up. It took several years to get there. I research and test the best I can in my own home without this "put a towel over my head" and let me listen and pick to see if I can tell a difference. I know my set-up.

Give some credit to owners even if you don't agree. I do think it takes it all to come together. That may not include a Yamaha AVR. This may be your weakest link.

PS I do love this thread ....

Rick

MrAcoustat's Avatar MrAcoustat 09:37 PM 06-12-2012
Cables Cables Cables if you want to sell expensive cables, i'm sorry but you will not succeed until you have a BRAND NAME on them many DO IT YOURSELFERS make very very good cables BUT just try to RESALE them and you will see how good your cable realy is, in the cable world you have those that believe ( i'm part of that group ) and there are the non believers what ever you say or do they simply wont believe but one thing is SURE if you believe no one will be able to change your mind BUT if you don't someday someone will make a believer out of you.
diomania's Avatar diomania 09:38 PM 06-12-2012
Quote:
Originally Posted by Mr.SoftDome View Post

I'm on the side that cables can make an improvement.



I do think it takes it all to come together. That may not include a Yamaha AVR.
What evidence are you basing that on? Note, opinion and evidence are not the same thing.
MrAcoustat's Avatar MrAcoustat 09:48 PM 06-12-2012
Just one example, one 10 foot pair of Tara Labs Temporal Continuum Retail Price in the 90s $2,500.00 price paid on Audiogon shipped from Italy $550.00 yes this cable was very good in the 90s and cables have come a long long way in 20 years plus BUT the only difference for me between 1990 and 2012 is that in 2012 i could afford them and they sound just as good as in the 90s.


337
Mr.SoftDome's Avatar Mr.SoftDome 10:04 PM 06-12-2012
Fair question. I have never had two set-ups in my house before. I moved a great 55 inch display into my bedroom when I replaced it with a 70 Sharp Elite. Loved it in bedroom but couldn't stand the TV speakers for such a nice secondary display.

I did not want to replicate my living room set-up in my bedroom. It's my bedroom! Not rich!

I use Dynaudio in Living Room. I went to Magnolia and checked what they had. Gosh darn I bought Bower and Wilkins CM5 monitors , CM center and a Denon 3312 AVR.

I admit the AVR is fun. Haven't owned one since Yamaha 2400. It has a sleep timer on remote!!! Seriously don't have this on my pre-pro and since this is bedroom it's great.

No sub or rear speakers....

This is a nice little bedroom set-up. For the first time I had two different level of gear. I was on vacation last week and played. I tried my Audioquest Comet speaker cables and Synergistic Research Power Cord. Not much of a gain. Didn't try changing out the ICs as a bit of a hassle for my time off. Maybe another time. I didn't notice much of a change.

Nice little set-up but my living room has been far more sensitive to change. Can I expect the same improvement from the $1k Denon and my $20k Classe SSP-800 and Simaudio Titan Amp?

It was fun to play last week. It made me appreciate my living room that much more. It was very easy to try these cables in my bedroom. Different gear same cables. To myself and what counts is the entire chain makes it come together and allows improvement.

Rick
diomania's Avatar diomania 10:14 PM 06-12-2012
Quote:
Originally Posted by Mr.SoftDome View Post

Fair question. I have never had two set-ups in my house before. I moved a great 55 inch display into my bedroom when I replaced it with a 70 Sharp Elite. Loved it in bedroom but couldn't stand the TV speakers for such a nice secondary display.
I did not want to replicate my living room set-up in my bedroom. It's my bedroom! Not rich!
I use Dynaudio in Living Room. I went to Magnolia and checked what they had. Gosh darn I bought Bower and Wilkins CM5 monitors , CM center and a Denon 3312 AVR.
I admit the AVR is fun. Haven't owned one since Yamaha 3800. It has a sleep timer on remote!!! Seriously don't have this on my pre-pro and since this is bedroom it's great.
No sub or rear speakers....
This is a nice little bedroom set-up. For the first time I had two different level of gear. I was on vacation last week and played. I tried my Audioquest Comet speaker cables and Synergistic Research Power Cord. Not much of a gain. Didn't try changing out the ICs as a bit of a hassle for my time off. Maybe another time. I didn't notice much of a change.
Nice little set-up but my living room has been far more sensitive to change. Can I expect the same improvement from the $1k Denon and my $20k Classe SSP-800 and Simaudio Titan Amp?
It was full to play last week. It made me appreciate my living room that much more. It was very easy to try these cables in my bedroom.
Rick
So, there is an answer somewhere in your reply? I don't see it. Or maybe you were replying to someone else...
Mr.SoftDome's Avatar Mr.SoftDome 10:19 PM 06-12-2012
I told you what I did. Still AVS issues with reply to post on my iPad.

If you can't except what I said fine. I have performed my own tests. What level is your chain?

Have you been able to try similar in your own home for a prolonged period of time? Two different levels of rigs and same cables.

Please share...

PS if I was truly clueless and in denial I would have said there was a nice change with the Denon and Bowers. Cables change all systems right? I said there was none. The system is great for what it is. It's not my living room rig where I have went thru a couple different levels of cables and I know this system well. It was fun to try this out.

What I hate most when talking about this stuff is I'm the farthest from a snob or whatever as there is. I usually stay out of these threads. But for me, I have answered my own questions.
It's the entire chain that makes it come together. And a certain level of gear. My opinion of course.


Rick
mcnarus's Avatar mcnarus 07:33 AM 06-13-2012
Quote:
Is it possible, and even taking science in to account, that to really hear a difference in cables, that your entire rig may have to be at a higher end mid level and up.
Not unless at least one of the two following propositions is true:

1) The laws of physics change when you spend more on audio gear.

2) Human biology changes when you spend more on audio gear.

Which of these do you want to believe?
AJinFLA's Avatar AJinFLA 08:07 AM 06-13-2012
Quote:
Originally Posted by Mr.SoftDome View Post

I'm on the side that cables can make an improvement.
I'm on that side too.
Quote:
Originally Posted by Mr.SoftDome View Post

So let me ask this. Is it possible, and even taking science in to account, that to really hear a difference in cables, that your entire rig may have to be at a higher end mid level and up. Not just speaker cables but your power cords, ICs, etc are at that same level.
Sure, that's not only possible, but probable. It is far more likely that "higher level" (translation - higher cost) "audiophile" injuneered equipment, would be susceptible to cabling effects, than competently/sane well engineered <"Mid" (which is really "pricing") level stuff, which could lead to actual changes to the soundfield into human auditory thresholds. But you said to "take science into account". What about the science of you, the hearer? Why are you focused only on the rig as the source, when seeking out audio "effects", while ignoring the science of the you/human perception? Do you consider audiophiles to be human, or have they transcended such annoying limitations?
Quote:
Originally Posted by Mr.SoftDome View Post

That may not include a Yamaha AVR. This may be your weakest link.
Absolutely agree. The thing that makes the cables and widgets work, is the very same thing that makes the Yamaha AVR not work.

cheers,

AJ
arnyk's Avatar arnyk 08:45 AM 06-13-2012
Quote:
Originally Posted by Mr.SoftDome View Post

I love this thread.... I'm sorry but these cable threads do add a bit of spice. I'm on the side that cables can make an improvement.
So let me ask this. Is it possible, and even taking science in to account, that to really hear a difference in cables, that your entire rig may have to be at a higher end mid level and up. Not just speaker cables but your power cords, ICs, etc are at that same level.

Which part of the excluded middle are you interested in? ;-)

Is it possible that there is equipment that is so badly designed that it will only work with certain cables? Not only is this possible, but there is an example of it that is documented on said manufacturer's web site by one of his employees: http://forums.naimaudio.com/displayForumTopic/content/3819189062263093 This piece of equipment is arguably poorly designed because there are numerous examples of power amps that lack thespecial feature mentioned on the above web page, but that still work well with ordinary cables and speakers. To mention a specific brand name: Adcom. However, I've found many examples of power amps that lack the purportedly offending feature mentioned above, and one turned out to be an inexpensive power amp for automotive audio.

More to the point which seems to be more typical of your question: Is there some general rule of science that says that as a general rule, audible differences between cables can't be heard unless they are used with equipment that performs at exceptional levels that generally exceed that provided by your typical middle-of-the road AVR? The answer is no. Here's the reason why: Your AVR is exceedingly unlikely to be the major source of masking in your system. The role of primary masker of audible differences in your system is reserved for(and with all due respect) your ears, followed by your listening room, and finally followed by your speakers.
diomania's Avatar diomania 09:21 AM 06-13-2012
Quote:
Originally Posted by Mr.SoftDome View Post

I told you what I did. Still AVS issues with reply to post on my iPad.
If you can't except what I said fine. I have performed my own tests. What level is your chain?
Have you been able to try similar in your own home for a prolonged period of time? Two different levels of rigs and same cables.
Please share...
PS if I was truly clueless and in denial I would have said there was a nice change with the Denon and Bowers. Cables change all systems right? I said there was none. The system is great for what it is. It's not my living room rig where I have went thru a couple different levels of cables and I know this system well. It was fun to try this out.

What I hate most when talking about this stuff is I'm the farthest from a snob or whatever as there is. I usually stay out of these threads. But for me, I have answered my own questions.
It's the entire chain that makes it come together. And a certain level of gear. My opinion of course.
Rick
My question was, "What evidence are you basing that on?" and now it's clear that you are basing your opinion on your opinion.

Once again, "Note, opinion and evidence are not the same thing." Try basing your opinion on evidence such as DBT (double blind test), it may shock you.
Mr.SoftDome's Avatar Mr.SoftDome 11:55 PM 06-14-2012
Quote:
Originally Posted by diomania View Post

My question was, "What evidence are you basing that on?" and now it's clear that you are basing your opinion on your opinion.
Once again, "Note, opinion and evidence are not the same thing." Try basing your opinion on evidence such as DBT (double blind test), it may shock you.

Hey, reply by IPad now works. I don't need DBT. Like I said I was able to test two set-ups at two different levels with same cables and with my own gear.

You guys wanna throw DBT but I also ask if you have tried the test in your own home with two systems at two ends of the spectrum with gear you know. Most of the doubters no but want to pretend they are experts and follow in what they read. I'm okay with that but there is another side.

What makes your claims any better than mine since many of the non- believers try their tests on one system and not two at different levels? And we wont go into the they all sound the same. That's for a different debate. Let's throw $3k cables on a $399 AVR. Check out the analog section on a Classe SSP-800 and compare to an AVR. I have both. No comparison when I use the dacs, software and yes probably BOM component quality in both. Not the same stereo experience.
It's a fact. Your welcome to come over and demo. I still believe in the weakest link theory and that's my audio religion. I have performed many changes in my own systems in my own house. That makes me an expert with my own set-up. No blindfold required.

Your claims are no better than mine. All you have is do a DBT that is repeated like parrots.

We will call the debate even for now, if some folks are happy and love their system with lamp cord awesome! I'm happy with my own tests! We can all debate this forever.

I hope the point is enjoy your system! I have friends who are content with an iPod. Folks on this forum you would think would appreciate the hobby weather you are into all aspects of equipment down to cable but there is a difference (-:

What scientist invented the blindfold?

Peace!

Rick
jwatte's Avatar jwatte 01:26 AM 06-15-2012
It's interesting how you answer a point by quoting me -- and I already address that answer right after the bit that you quoted
Of course cymbals and triangles and other instruments ring. After you hit them! The problem is that a perfect reconstruction filter needs infinite pre-ringing for signals close to Nyquist. This is why attack for percussion is so hard to do right. (well, one reason) Putting your filter at Nyquist/2 significantly improves this phenomenon, and if you then want the filter to be flat up to 18 kHz, you need the corner at 24 and sampling at 96.

In my youth, I did exactly the blind A/B sweitch test you suggest, and ended up able to tell 16 bits from 20 bits (or, rather, consistently prefer 20 bits
. At the time, we didn't have 96 kHz material to test with. I can accept that not everyone is as sensitive, but I know that I am

Regarding 24 dB/oct or higher filters, note that I didn't say you can't build them; I listedd the compromises you need to make when you build them. People do make compromises all the time. Especially for live audio- -- a noisy crowd and UHF mics will raise your noise floor right up. I fully admit that you have more experience in that department (I' was an AES voting member, but as equipment maker, not recordist/mixer,) but that doesn't mean that your "good enough" is mine.
Regarding noise floor, even if noise floor is 30 dB, you will hear things at 10 dB in another range
jwatte's Avatar jwatte 01:31 AM 06-15-2012
People have done DBTs on speaker cables, and the conclusion is that when the same gauge and length of wire iss used, there is no detectable difference
If you test a high end cable against, say, 28 gauge phone cord, then yes, there will be a difference.
arnyk's Avatar arnyk 06:32 AM 06-15-2012
Quote:
Originally Posted by Mr.SoftDome View Post

Hey, reply by IPad now works. I don't need DBT. Like I said I was able to test two set-ups at two different levels with same cables and with my own gear.

But odds are excellent that level-matching, time synching, fast switches, and bias control wasn't invited to your so-called tests. If you did your homework, do tell. How did you match levels, and how did you accomplish the prerequisite gain adjustments? What did you do to control listener bias?
Quote:
You guys wanna throw DBT but I also ask if you have tried the test in your own home with two systems at two ends of the spectrum with gear you know. Most of the doubters no but want to pretend they are experts and follow in what they read. I'm okay with that but there is another side.

What makes your claims any better than mine since many of the non- believers try their tests on one system and not two at different levels?


I've tried the test in my own home with my own equipment, and I've done many tests in other people's homes and with other people's equipment.

One of the best demos around is to get invited to a true believer's home, set up a comparison that according to the high end ragazines should be a slam dunk, and actually let people listen with their ears and not just their preconceived notions.

Been there, done that first starting over 30 years ago. The sonic accuracy of home audio gear has only improved since then. If sonic accuracy is the goal, then logic says that sound quality should converge, and experience says that it has. In 2012 a Pioneer is much more like a Classe than a Dyna was like a MacIntosh in 1975. That's what the test bench says, but even the 1975 comparison was too close to call in many cases.
Quote:
And we wont go into the they all sound the same. That's for a different debate

No, its your straw man. Nobody with a brain is saying that *everything* sounds the same. What we are saying is that the audible differences between the average piece of mid fi (e.g. $400 AVR) and a megabuck high end system composed of the finest separates may be non existent. Or not. Often are.
Quote:
Let's throw $3k cables on a $399 AVR.

Predictable results.
Quote:
Check out the analog section on a Classe SSP-800 and compare to an AVR. I have both. No comparison when I use the dacs, software and yes probably BOM component quality in both. Not the same stereo experience.


Here are the innards of a ca. $8K SSP-800:

424

And here are the innards of a < $700 Denon AVR3311:

564

What's wrong with the AVR?
Quote:
It's a fact. Your welcome to come over and demo. I still believe in the weakest link theory and that's my audio religion. I have performed many changes in my own systems in my own house.

The above is a gross abuse of principle that that you are claiming. Once audio equipment reaches a certain now readily achievable level, it stops being the weakest link. The weakest link in most systems is the listener, because we can prove as many times as we are allowed to, that there can be substantial differences in measured technical performance, and all listeners cannot detect them. We can prove that the next weakest link is the room/speaker and the interface between them. We can prove that the inherent imperfections in the original recording are far more audibly signficiant than those that exist between an AVR and the best separates in many if not most cases.

Finally, what you don't seem to understand is the fact that in many cases the same basic chips end up in high end gear and middle-of-the-road AVRs. The production volume of high end AV gear is insufficient to warrant the production of a special line of signal processing chips for them. The high end designer must either use the same parts as the AVR designer, or resort to the use of an unmanagable array of discrete parts that he probably lacks the technology to even conceive of, let alone design. In the two pictures above, there is a high probability that the largest chip with the 4 big dots on it is the same chip in both products.
Quote:
That makes me an expert with my own set-up. No blindfold required.

Your desire to make your personal biases the focal point of any comparison are noted. It's good to be chauvinistic about your own AV system, but preferring or conforming to your biases is not something that the rest of us are interested in doing. If you haven't noticed, you don't rule us! But, science rules us all.
Quote:
Your claims are no better than mine. All you have is do a DBT that is repeated like parrots.

I did the DBTs myself. If it were feasible I'd be happy to come over to your house and show you what you don't seem to want to know.

You seem to be telling me to salute your outdated ideas about what constitutes a valid test, when I've known better for over 30 years.
Quote:
We will call the debate even for now,

No, in any factual evaluation of the relevant facts, your viewpoint suffers from not conforming to reality.
Quote:
if some folks are happy and love their system with lamp cord awesome!

The above is just more high end audiophile chest-thumbing with a little excluded middle dropped in. "Lamp cord" is usually restricted to 18 or 16 gauge wire. We recommend the use of 12 gauge, which is never used for powering household lamps. You appear to be distorting the relevant facts in order to please your own ego and demean others.
Quote:
I'm happy with my own tests!

Good for you. Because they are inherently biased they are relevant only to your preferences.
Quote:
We can all debate this forever.

This appears to be because of your lack of willingness to be factual and honest. You seem to just want to stand up, moon us all with your ignorance, and tell us that you are right and we are wrong. That is very insulting and self-serving.
Quote:
I hope the point is enjoy your system!

The point is that I can enjoy my system without demeaning your system, but you don't appear to be able to enjoy your system without insulting me, my intelligence, that of my friends, my ability to work with audio technology, etc., etc.
Quote:
I have friends who are content with an iPod.

Right up to the headphone jack, and in accordance with an extant Stereophile review, IPods are technically very good. Since they can play uncompressed .wav files quite accurately, they are within their obvious limits about as sonically effective as the most expensive 2 channel media players. Why do you feel compelled to be insulting about other people's audio gear when by many reliable sources, it is actually quite good?

Quote:
Folks on this forum you would think would appreciate the hobby weather you are into all aspects of equipment down to cable but there is a difference (-:

The difference is about perceived value, not actual sound quality.
Quote:
What scientist invented the blindfold?]

Irrelevant and insulting.
Quote:
Peace!

The many insults and apparent false claims in the post above make a sham out of the word peace. :-(
arnyk's Avatar arnyk 07:11 AM 06-15-2012
Quote:
Originally Posted by jwatte View Post


Of course cymbals and triangles and other instruments ring. After you hit them!

Unfortunately, what follows is not nearly as factual.
Quote:
The problem is that a perfect reconstruction filter needs infinite pre-ringing for signals close to Nyquist.

The appears to be a stealthy excluded middle argument with a factual error tossed in.

First off, nobody designs perfect reconstruction filters. Instead, they design filters with inaudible losses. Therefore the focus of your comment is an imaginary filter that you brought up as a false straw man defense for an indefensible position.

Your position is indefensible because the ringing in musical instruments is always well below Nyquist in a RedBook system. For example, cymbals have their largest resonances below 10 KHz.
Quote:
This is why attack for percussion is so hard to do right. (well, one reason)

This is most likely a false claim. Percussion is routinely done right (reproduced with no audible losses or changes) in modern equipment. By that I mean that we are still searching for the recording and listener that can reliably detect the effects of a good DAC and ADC running back-to-back and selectively interposed in a signal chain.
Quote:
In my youth, I did exactly the blind A/B sweitch test you suggest, and ended up able to tell 16 bits from 20 bits (or, rather, consistently prefer 20 bits

Thousands of people have done this test under what are no doubt far more favorable conditions and have been reduced to random guessing.

My best guess - you alleged test failed on one of more the canonical basics of a good listening test:

(1) Level matched
(2) Time synched
(3) Totally blind
Quote:
. At the time, we didn't have 96 kHz material to test with.

Not a problem for at least the last 10 years.
Quote:
I can accept that not everyone is as sensitive, but I know that I am

How nice of you to put yourself over everybody else. ;-)
Quote:
Regarding 24 dB/oct or higher filters, note that I didn't say you can't build them; I listed the compromises you need to make when you build them.

24 dB/octave filers are totally irrelevant to the discussion of brick wall filters used in digtial audio. Never were used for that purpose, never will be used.

Quote:
People do make compromises all the time.

Of course but many ot the compromises you appear to damn are not actually audible problems at all.
Quote:
Especially for live audio- -- a noisy crowd and UHF mics will raise your noise floor right up.

My UHF mics have noise floors that are 100 dB down. They aren't even the best around or even just close. You are again exposing your ignorance of the relevant facts.
Quote:
I fully admit that you have more experience in that department (I' was an AES voting member, but as equipment maker, not recordist/mixer,) but that doesn't mean that your "good enough" is mine.

Right. My "good enough" is easy to prove that it is good enough. Your high-handed view of good enough doesn't stand up to real world testing.

Quote:
Regarding noise floor, even if noise floor is 30 dB, you will hear things at 10 dB in another range

So what?


Here's your challenge, which is actually very easy to set up if you are as resourceful as you claim:

Do some proper listening tests (that conform to my 3 points above) that show that you are actually able to hear what you claim.
Brownstone322's Avatar Brownstone322 08:45 AM 06-15-2012
Quote:
Originally Posted by jwatte View Post

In my youth, I did exactly the blind A/B sweitch test you suggest, and ended up able to tell 16 bits from 20 bits (or, rather, consistently prefer 20 bits. At the time, we didn't have 96 kHz material to test with. I can accept that not everyone is as sensitive, but I know that I am

Oh, man! Only in the wacky world of audiophilia will you read stuff like this!
arnyk's Avatar arnyk 09:16 AM 06-15-2012
Quote:
Originally Posted by Brownstone322 View Post

Oh, man! Only in the wacky world of audiophilia will you read stuff like this!

Right - what our correspondent doesn't seem to know is that the most dynamic recording around on any format media only has about 14 bit resolution, and most are more like 12 or 13.

What do you call it when a person is listening to a recording with maybe 13 bit resolution, and believes he hears the infinitesimal difference caused by using a 20 bit or a 16 bit DAC further down the line?

Clearly, he lacks the ability to do the math, not that the relevant math is very daunting.
diomania's Avatar diomania 10:23 AM 06-15-2012
Quote:
Originally Posted by Mr.SoftDome View Post

Hey, reply by IPad now works. I don't need DBT. Like I said I was able to test two set-ups at two different levels with same cables and with my own gear.
You guys wanna throw DBT but I also ask if you have tried the test in your own home with two systems at two ends of the spectrum with gear you know. Most of the doubters no but want to pretend they are experts and follow in what they read. I'm okay with that but there is another side.
What makes your claims any better than mine since many of the non- believers try their tests on one system and not two at different levels? And we wont go into the they all sound the same. That's for a different debate. Let's throw $3k cables on a $399 AVR. Check out the analog section on a Classe SSP-800 and compare to an AVR. I have both. No comparison when I use the dacs, software and yes probably BOM component quality in both. Not the same stereo experience.
It's a fact. Your welcome to come over and demo. I still believe in the weakest link theory and that's my audio religion. I have performed many changes in my own systems in my own house. That makes me an expert with my own set-up. No blindfold required.
Your claims are no better than mine. All you have is do a DBT that is repeated like parrots.
We will call the debate even for now, if some folks are happy and love their system with lamp cord awesome! I'm happy with my own tests! We can all debate this forever.
I hope the point is enjoy your system! I have friends who are content with an iPod. Folks on this forum you would think would appreciate the hobby weather you are into all aspects of equipment down to cable but there is a difference (-:
What scientist invented the blindfold?
Peace!
Rick
Mr.SoftDome, besides post #105, read this carefully: http://www.roger-russell.com/wire/wire.htm It will shed much needed light on your personal database on speaker cables.
jwatte's Avatar jwatte 02:40 PM 06-18-2012
Quote:
Originally Posted by arnyk View Post

Right - what our correspondent doesn't seem to know is that the most dynamic recording around on any format media only has about 14 bit resolution, and most are more like 12 or 13.

We used our own recording, made with middle-end mixing gear and microphones, on 20-bit equipment. The reference monitors were not that great, either -- Mackie HR-824s -- but sufficient at the time.
Quote:
Clearly, he lacks the ability to do the math, not that the relevant math is very daunting.

... wow! Just... wow! Talk about hidden assumptions!
jwatte's Avatar jwatte 02:58 PM 06-18-2012
Quote:
Originally Posted by arnyk View Post

First off, nobody designs perfect reconstruction filters. Instead, they design filters with inaudible losses. Therefore the focus of your comment is an imaginary filter that you brought up as a false straw man defense for an indefensible position.

My argument is "if the perfect filter can't do it, then clearly an imperfect filter will be worse."

You may be more familiar with this from the old debate on FIR vs IIR controls for console EQ. Nobody likes FIR, because it effectively "pre-rings" the signal, which is a very unnatural thing to do for audio. Thus, the phase mashing done by the IIRs is actually preferrable to most recording/mixing engineers. And, the fact that phase mashing is preferred to linear phase when shaping sound should indicate that ears are actually sensitive to phase, too.
Quote:
Your position is indefensible because the ringing in musical instruments is always well below Nyquist in a RedBook system. For example, cymbals have their largest resonances below 10 KHz.

Your make a true statement (the largest (but not only) resonance of cymbals) to back up another statement that's not related. Or are you saying that only the largest resonances matter? However, this argument you made again misses the point. The question wasn't directly about resonances and ringing; it was about transients, where you argued that anti-aliasing and reconstruction filters were good enough, and I countered that a perfect filter needs/assumes pre-ringing to actually work when close to Nyquist. That is the center of my argument: Nyquist / the sampling theorem holds for static signals. Music is not static. The difference, if you design a system assuming only the textbook sampling theorem, are perceptible.
Quote:
Quote:
in my youth
Not a problem for at least the last 10 years.

I wish I was only 10 years away from my youth.
Quote:
How nice of you to put yourself over everybody else. ;-)

I'm just saying there may very likely be people who hear more details than other people, for reasons of genetics, impairment, physics, brain chemistry, or whatever. I'd argue that pretty much everyone who visits this message board is in the top 10% of "people who perceive and are annoyed at poor sound reproduction" in the world.
Quote:
24 dB/octave filers are totally irrelevant to the discussion of brick wall filters used in digtial audio. Never were used for that purpose, never will be used.

Again, you're switching your argument around. Follow the thread. The argument was that analog brick wall filters require a lot of compromise -- even with the best available analog components. And, once you have aliasing in the digital domain, you can't get rid of it. Using 96 kHz sampling frequency for recording means you can make fewer compromises in the analog domain before it even hits the digital.
Quote:
My UHF mics have noise floors that are 100 dB down. They aren't even the best around or even just close. You are again exposing your ignorance of the relevant facts.

Some of us didn't get to do a lot of sitting in with the masters of the domain. I can't but envy your prowess in this area.
Quote:
Here's your challenge, which is actually very easy to set up if you are as resourceful as you claim:

I had much more resources for this in the '90s (not to mention better ears, although luckily I'm still doing okayish on that front.)

To re-cap my argument: "Nyquist" / the sampling theorem holds for static signals. Music is not static. The difference, if you design a system assuming only the textbook sampling theorem, are perceptible. Nobody has actually addressed this argument head on, especially at the math level! And, no, the math at that point is no longer "trivial."
I wonder if there isn't room in some academic institution for actually studying the applied math of this effect in more detail. I've not actually seen any papers that address this -- they *all* fall back on the old crutch assuming static signals.
If there are papers that don't do that, and get into the significantly more messy theory of what actually is going on, I'd love to see a reference. That would be very convincing to me.
jwatte's Avatar jwatte 03:06 PM 06-18-2012
Quote:
Originally Posted by Mr.SoftDome View Post

And we wont go into the they all sound the same. That's for a different debate.

No, that is actually exactly the the debate we're having with regards to speaker cable. (I wish I'd moved the relative phase and 24/96 discussion to some other thread, but too late now -- mods, can you split it?)
Quote:
Your claims are no better than mine. All you have is do a DBT that is repeated like parrots.

DBTs are the generally accepted way of settling a debate that can be quantified. If two different products cannot be told apart during the ABT, then the only reason to prefer the most expensive version would be for cosmetics or other non-sound-related reasons.
It's fine to pay $15,000 for speaker cables, if your argument is "I like the purple irridescent color." But, if that's the actual argument, I imagine you can get the same effect for cheaper.
It may also be fine to pay $15,000 for speaker cables, if your arugment is "knowing I paid that much for them, I will enjoy my system more." But for me, it'll actually be worse -- I prefer pay enough to get rid of things that get in the way of unadulterated listening, and no more.

The argument seems to boil down to "placebo works, and I don't need DBTs to show that." Curiously, we could engineer a test where we send out zip cord in expensive jackets, vs metallurgically-engineered cables in expensive jackets, and compare the impact, using a DBT that is "blind" to something other than what is apparent. That might be fun :-)
Jinjuku's Avatar Jinjuku 04:01 PM 06-18-2012
Quote:
Originally Posted by Joe Skubinski View Post


Then I guess there's no hope.

>>>


If you wish to compare performance verses price between your room treatment and our cable products, I'm game.





If we start with a nicely furnished room in a typical 2 channel listener equilateral triangle sitting of 8 feet or less, and in one case you place all the room treatment you deem required, and in another case I place all the cables required, both nearly equivalent in dollars (and of course high enough in value to actually complete the job), I am positive that a majority of listeners would choose the cables over the treatment.


Let me know if you ever find yourself a room(s) in a mutual setting where we could do this. Maybe a CES or RMAF..

Can I get odds? I'd actually show up with cash in hand for this one biggrin.gif

So you are saying take the room and cable with something like mogami and then JPS. Then take the same room put mogami back in and let Ethan setup with equivalent $$ sound treatments?

When you say 'high enough value to complete the job' can you expound on that? What is 'the job'? Since you are obviously making an assumption, what assumption is that about 'the job'?
arnyk's Avatar arnyk 09:23 AM 06-19-2012
Quote:
Originally Posted by jwatte View Post

We used our own recording, made with middle-end mixing gear and microphones, on 20-bit equipment. The reference monitors were not that great, either -- Mackie HR-824s -- but sufficient at the time.
... wow! Just... wow! Talk about hidden assumptions!

So what was the actual resolution of your recording once you finished it?

Do you even know how to properly estimate it?
arnyk's Avatar arnyk 09:42 AM 06-19-2012
Quote:
Originally Posted by jwatte View Post

My argument is "if the perfect filter can't do it, then clearly an imperfect filter will be worse."

Worse in what sense?

Theoretically?

On the test bench?

Audibly?

Only one of those ways is interesting when it comes to sound quality.
Quote:
You may be more familiar with this from the old debate on FIR vs IIR controls for console EQ. Nobody likes FIR, because it effectively "pre-rings" the signal, which is a very unnatural thing to do for audio. Thus, the phase mashing done by the IIRs is actually preferrable to most recording/mixing engineers. And, the fact that phase mashing is preferred to linear phase when shaping sound should indicate that ears are actually sensitive to phase, too.
Your make a true statement (the largest (but not only) resonance of cymbals) to back up another statement that's not related. Or are you saying that only the largest resonances matter? However, this argument you made again misses the point. The question wasn't directly about resonances and ringing; it was about transients, where you argued that anti-aliasing and reconstruction filters were good enough, and I countered that a perfect filter needs/assumes pre-ringing to actually work when close to Nyquist. That is the center of my argument: Nyquist / the sampling theorem holds for static signals. Music is not static. The difference, if you design a system assuming only the textbook sampling theorem, are perceptible.

I guess you didn't understand my statement" "Argument, smargument". The execuitive summary is that abstract arguments don't matter much to me. I'm primarily interested in things that actually sound better. After > 30 years of doing ABX tests, I've heard a lot of what someone said, based on some abstract theory, that "It sounded better" turn out to sound the same. Sound the same = no sale!
Quote:
I wish I was only 10 years away from my youth.

Whatever that means.
Quote:
I'm just saying there may very likely be people who hear more details than other people, for reasons of genetics, impairment, physics, brain chemistry, or whatever.

Yes, and applying that argument, I can say that there are some people who can run better than others. Problem is, that 2 minute mile is very elusive. Listening is at its core athletics. Until someone actually shows me a 2 minute mile or at least provided an adequate documentation of the same, I'm very skeptical. The same goes for people who effectively claim to hear 20 to 50 dB below the known thresholds of hearing. You know, people who say they can sort DACs by ear like I sort apples or potatoes.

Quote:
I'd argue that pretty much everyone who visits this message board is in the top 10% of "people who perceive and are annoyed at poor sound reproduction" in the world.

However, you don't to spin your wheels with the 20th bit or even the 16th bit to understand why that is. Part of it is exquisitely explained in Letvin's book "This Is Your Brain on Music".

Quote:
Again, you're switching your argument around.

Really? If you want to claim that and be credible, you need some relelvant quotes showing me doing that. Seeing none... '-)
Quote:
Follow the thread.

Been there, done that. ;-)
Quote:
The argument was that analog brick wall filters require a lot of compromise -- even with the best available analog components.

Irrelevant. Other than some DIY and maybe some boutique trips down memory lane, analog brick wall filters haven't been used in commercial equipment for almost 3 decades. Well, two decades for sure.

Quote:
And, once you have aliasing in the digital domain, you can't get rid of it.

Irrelevant. Digital brick wall filers can be very effective.

Quote:
Using 96 kHz sampling frequency for recording means you can make fewer compromises in the analog domain before it even hits the digital.

You almost make it sound like you actually think that every ADC has to have an analog brick wall filter! Surely you jest!

Quote:
Some of us didn't get to do a lot of sitting in with the masters of the domain. I can't but envy your prowess in this area.
I had much more resources for this in the '90s (not to mention better ears, although luckily I'm still doing okayish on that front.)

To re-cap my argument: "Nyquist" / the sampling theorem holds for static signals. Music is not static.

Except it is piecewise static. Digital technology has worked well for non-stationary signals in my presence since the early 1970s when someone gave me a 200 KHz true 16 bit ADC and DAC to play with.
Quote:
The difference, if you design a system assuming only the textbook sampling theorem, are perceptible.

You're talking in circles. It is well known that even many $2 audio interfaces that are thrown in on cheap motherboards are sonically transparent for music. People who have tried the experiment like me and some friends of mine even know that they are sonically transparent even if you cascade a half-dozen of them.

Quote:
Nobody has actually addressed this argument head on, especially at the math level! And, no, the math at that point is no longer "trivial."

Why worry about the math when this stuff works so well in the real world?
Quote:
I wonder if there isn't room in some academic institution for actually studying the applied math of this effect in more detail. I've not actually seen any papers that address this -- they *all* fall back on the old crutch assuming static signals.
If there are papers that don't do that, and get into the significantly more messy theory of what actually is going on, I'd love to see a reference. That would be very convincing to me.

Maybe you should pay attention to something more important in the real world, like how many angels can dance on the head of a pin?

Do some relevant bias-controlled listening tests, and be ready for a big shock. ;-)

jwatte's Avatar jwatte 02:00 PM 06-20-2012
Quote:
Digital brick wall filers can be very effective.

To do that for anti-aliasing with sampling, they have to sample at a frequency higher than redbook. Agree? And, if so, that's EXACTLY MY CLAIM for why redbook is not good enough for recording for anti-aliasing purposes. If you mean over-sampled ADCs, then we're actually in agreement on this point.
Quote:
Why worry about the math when this stuff works so well in the real world?

Wasn't that what they said about direct-mastered pressed LPs, too? :-)

Seriously, though, do you agree that FIR (linear phase) EQs actually sound less preferable than IIR EQs? Most audio guys I've known have agreed, but I'd like to know where you fall on this actual, perceptible, real-world topic.
arnyk's Avatar arnyk 02:11 PM 06-20-2012
Quote:
Originally Posted by jwatte View Post

To do that for anti-aliasing with sampling, they have to sample at a frequency higher than redbook

Well, that's done to avoid having sharp cut-off analog filters in the system.

Quote:
And, if so, that's EXACTLY MY CLAIM for why redbook is not good enough for recording for anti-aliasing purposes. If you mean over-sampled ADCs, then we're actually in agreement on this point.

If you thoroughly conflate oversampling with upsampling, then you become hard to follow.

Quote:
Wasn't that what they said about direct-mastered pressed LPs, too? :-)

Your writing is now so confused that you need to make a post that says what you want to say in one place..

Quote:
Seriously, though, do you agree that FIR (linear phase) EQs actually sound less preferable than IIR EQs?

First show me someone who can reliably detect the difference between 44.1 KHz filters that are implemented different ways, but have the same frequency and phase response.
Quote:
Most audio guys I've known have agreed, but I'd like to know where you fall on this actual, perceptible, real-world topic.

As a general rule, there is quite a bit of latitude in how 44.1 KHz brick wall filters are implemented, given that certain fairly simple goals are met, This is because the differences if they exist, are concentrated in frequency ranges where the ear is pretty insensitive.

Ever do a DBT relating to trying to flavor test DACs? I have. Usually, a gigantic waste of time given that the DACs are at least halfways decent.
jwatte's Avatar jwatte 04:06 PM 06-20-2012
Quote:
Originally Posted by arnyk View Post

Well, that's done to avoid having sharp cut-off analog filters in the system.
If you thoroughly conflate oversampling with upsampling, then you become hard to follow.

If a ADC runs at X*N kHz for an intended X kHz sampled signal, with a softer analog anti-aliasing filter, and then brick-wall digitally and down-sample to your target sampling rate (decimation,) this is an over-sampled ADC.
up-sampling is when you do sample rate conversion upwards for an existing digital signal, which you may do to send the signal out through an "over-sampled" DAC, or to make the signal compatible with some other processing that requires a higher rate.
This usage of the words has been established in DSP for a very long time, so I fail to see how this would be conflating at all. (In strictly true pedantry sense, an "oversampled" DAC is actually an upsampling DAC, as opposed to the ADC where true oversampling is done before decimation.)

On the other hand, your statement may just be strictly true but irrelevant. Yes, if I (or anyone) thoroughly conflate oversampling with upsampling, then the discussion is hard to follow. But the sentence you quoted does nothing of the sort.
Quote:
First show me someone who can reliably detect the difference between 44.1 KHz filters that are implemented different ways, but have the same frequency and phase response.

My question was explicitly around the phase differences of an IIR filter (which is not phase linear) to a FIR filter (which is _typically_ phase linear.) If they implement exactly the same transfer function, then they will sound the same.
But, as you see if you actually read my original question, that wasn't my question. My question was, re-stated:

Do you, personally, agree that linear-phase EQ sounds worse than commercially available ("typical" and not phase linear) IIR EQ (analog or digital)?

It seems you might do, or at least you think there is a difference based on phase, because you make sure to mention phase response in your qualifications of the irrelevant comparison you suggest above.
The reason I ask this, is because I believe that phase response differences, below 180 degrees of shift, are audible, and I am showing some examples of where this is pretty obviously the case to most engineers in the field. You seem to agree, but I don't want to read too much into your statement one controlling for phase, above.
Once we agree that phase response differences below 180 degrees are audible, I can make the follow-up argument of what happens to dynamic signals/transients when they get close to Nyquist.
I'd like to make that argument, in turn, to show what I think is a part of the mathematical explanation for the differences I perceived back when I was listening into this stuff with even better ears than I have now -- note, the 44/96 difference is not something I've personally blind A/B-d, just openly A/B-d a few years after the 16/20-bit A/B I talked about before.
Once I've made that argument, I can then subtantiate both my claim that I heard it, and the explanation for why that was possible to hear, to finish off my explanation for why redbook audio is not a 100% transparent format for actually available equiment/recordings/ears.
Quote:
Ever do a DBT relating to trying to flavor test DACs? I have. Usually, a gigantic waste of time given that the DACs are at least halfways decent.

I agree that all competent DACs sound the same, given the same input signal. Again, that has nothing to do with my argument --- my argument is about how the signal gets to be what it is in the first place.


There was also mention of someone claiming to hear 20 dB or 50 dB below hearing levels. That's not me. I claim to hear things about 20-30 dB _above_ the hearing threshold.
You yourself said you swing with gear in the 120 dB range. Let's assume that's average SPL, for program with average SPL -15 off peak response. (I e, not your over-compressed modern masters.) Let's further go with the 95 dB "perfect dither" S/N ratio of perfectly mastered 16-bit audio. That puts the noise floor at (135-95 == 40 dB.) And that's the noise -- for a flute or triangle to have any kind of real definition, it needs to be 5-10 dB stronger. Similarly, a sound that was softer than 40 dB could still be heard over the noise, but could not be represented by that 16-bit signal.
Or, perhaps you actually meant 120 dB peak response? In that case, we're still talking 25 dB noise floor, which is totally perceptible, especially in a recording that starts with very soft, delicate instruments, and only uses the dynamic peak for the most dynamic bits. "1812 overture" for the dynamics, perhaps? :-)
I think that this argument, in itself, is sufficient to say that there exists listening environments and program material for which 16 bits resolution just doesn't quite cut it. Sure, it's done splendidly better than LPs or tapes or whatever else was the delivery media of choice in the past. But it's not the evolutionary end of resolution in formats -- the step to 96/20 (or, because that's what ended up happening, 96/24) actually makes sense, and I wish more material would take advantage of it.

Regarding estimating the "bit depth" of materials used in testing. We recorded in a quiet/isolated room. We used good microphones. We calibrated average SPL 15 dB off peak, and the peak signal recorded was within -6 dB of top, so the highest bit was in use. We did not do any additional processing (besides dithering,) so there was no compression/processing bit depth loss.
arnyk's Avatar arnyk 08:55 PM 06-20-2012
Quote:
Originally Posted by jwatte View Post

Do you, personally, agree that linear-phase EQ sounds worse than commercially available ("typical" and not phase linear) IIR EQ (analog or digital)?

I've been doing listening tests long enough that some of the DACs involved were linear phase, and some were not. There was not a necessary connection between being linear phase and being sonically transparent.

I know of no competently-executed listening tests that show otherwise.

When you say:

"I agree that all competent DACs sound the same," I am faced with the observed fact that not all competently-designed DACs are linear phase.
Quote:
You yourself said you swing with gear in the 120 dB range. Let's assume that's average SPL, for program with average SPL -15 off peak response. (I e, not your over-compressed modern masters.) Let's further go with the 95 dB "perfect dither" S/N ratio of perfectly mastered 16-bit audio.

The above appears to be in error because does not seem to consider the benefits of shaped dither, which has been in common use since the middle 1990s (over 15 years ago).
Quote:
That puts the noise floor at (135-95 == 40 dB.) And that's the noise -- for a flute or triangle to have any kind of real definition, it needs to be 5-10 dB stronger.

Here we have an example of the common mistake of presuming that the noise floor of recordings are due to converter noise, when in fact the most common source of noise is room noise.

We also see another common error which is the belief that instrumental sounds can't be heard below the noise floor, but this is secondary to the first mistake of not understanding of where the noise floor comes from.

Here's your challenge - find a commercial so-called high resolution recording that is audibly altered by being converted to 16 bits using technology that has been in general use for at least 15 years.

I've been surprised by my analysis of commercial high resolution recordings that show that they do not have lower noise floors than is possible with just 16 bits used wisely, even when their contents bears many of the hallmarks of having a true > 16/44 production history. IOW, if we take commercial high resolution recordings and then discard all of those that show any signs of just being upsampled low resolution recordings, we are still left with recordings that have noise floors that would not tax the capabilities of the humble audio CD.
jwatte's Avatar jwatte 12:10 PM 06-21-2012
Quote:
Originally Posted by arnyk View Post

I've been doing listening tests long enough that some of the DACs involved were linear phase, and some were not.

Why are you answering a question other than the one I asked? If you're actually an experienced audio engineer, it seems "EQ" and "DAC" should be easily distinguishable.
Quote:
The above appears to be in error because does not seem to consider the benefits of shaped dither

I talked about that previously. You can see me describe it and discuss the considerations many posts back.
Quote:
in fact the most common source of noise is room noise.

Again, this is talking about a different question. Unless you claim that it is not possible to record in a room where the difference between peak signal and hearing below the room noise is greater than 95 dB. Is this what you claim?
The math is something like: Assume you calibrate your peak bit to 120 dB peak SPL. Let's assume any sound is masked at 20 dB below room noise. Let's assume best-case 95 dB S/N for 16 bit gear. Then you are saying that it's not possible to record in a room where noise is below (120-95+20) 45 dB. This is on the face of it a false claim. Thus, the fact that room noise is a common source of noise does not actually matter to my argument.
Quote:
We also see another common error which is the belief that instrumental sounds can't be heard below the noise floor

I already talked about that before. I'm talking about resolution in the digital signal, once it's been converter. It is by definition not possible to hear a signal below the *digital resolution* noise floor.
Some might even say that room noise is actually part of the signal you want to record, because it's fidelity to the recording environment.

If your argument is that 16/44 is "good enough" for "most cases," then I have already agreed. But that by definition means that there are some cases where it is not good enough. When I explain how those cases arrive, the only counter-argument I hear is "in most cases, ..." which is actually an implicit agreement.
First ... 2  3  4 5  6 

Up
Mobile  Desktop