Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[David Satz]

Klaus, cables came up as a side topic in the "high frequency transient fallacy" thread, and were followed up later in the other thread that I mentioned.

I've never heard any differences among different brands or types of professional-quality microphone cable, and wouldn't expect to do so under ordinary circumstances. If I had to run extremely (1000'+) long cables that might well be different. But I honestly believe that for any one brand or type of 50- or 100-foot microphone cable to "sound" different from another must either be a trick of the mind, an unsolved engineering problem of some kind, or both.

Even when real differences can be heard, what is heard should never simply be accepted as "the sound quality of the cable." Nor should cables be bought or sold on such basis. In any case of audible difference, one or the other cable must be incurring significant losses and/or compromising the immunity of the system to interference. A good engineer would insist on finding out which is which, and then choose accordingly. Does that point of view make sense to you?

I generally use transformerless microphones with very low output impedance (ca. 35 Ohms). They are less dependent on specific cable properties than a 150- or 200-Ohm microphone would be; thus says Ohm's Law. But I also use transformer-equipped microphones sometimes (Neumann fet 80 series, Schoeps M 221 B and CMT series). For those I generally keep the transformers strapped to their 50 Ohm output impedance setting. I don't run into any preamp noise problems as a result, and again it helps immunize the microphones against cable capacitance problems.

Good cable shielding is important, but even unshielded balanced cable can reject interference effectively if the rest of the equipment setup is as it should be. Thus I consider that it's probably even more important for microphones and preamps or mixers not to have unresolved "pin 1" issues--see the just-released AES48 standard which is a free download for AES members. Cables can't be relied upon to overcome engineering mistakes in other items of equipment!

[natpub]

David Satz wrote on Mon, 04 April 2005 21:02

--I've never heard any differences whatsoever among different brands or types of professional-quality microphone cable, and wouldn't expect to do so under any of the circumstances in which I ordinarily make recordings...But I honestly believe that for any one brand or type of 50- or 100-foot microphone cable to sound different from another must either be a trick of the mind or else an unsolved engineering problem of some kind--

David, how odd. I clearly hear notable differences in only a change of cable, being quite certain that the cables are relatively new and correctly wired. This is true for both instrument and mic cables.

Please let me respectully add that I'm not sure the "trick of the mind" thing is all that constructive, because it suggests the converse response that "maybe your ears just dont work well?"

I deeply respect your superior knowlege of these things, and your obvious wealth of work in a very pristine field of music and art.

Also, please note, I am not trying to take Klaus' side here. I am just saying I have read literally hundred of cable threads over the years. And, while potentially valid, I think easy comments about mental "abberations" are mildly insulting.

To me, those asserting that there is no viable difference in top end cables "SEEM" to lean to being more aggressive, and less responsive to the quieter claims that there is a meaningful difference. After all, while we are surely reliant on certain technolgies, the differences that result are only useful if they are meaningful in a musical way.

Beyond that, technicians are supporters, but not the fundamental creators. Period. I do not say that to seem harsh or obtuse--I say it because I am tired of mathematical orientation edging to obscure art. Techinicians serve the artists in this business, not the other way around. I play both roles, and I honor both.

I'd certainly remain open to "engineering" (to quote you) differences being an issue, but that appears to be the whole point of the thread??

I sent a note to Jim Williams about this thread, as I hope he will comment. He has been one of the few technicians in the art that has come forward over the years and asserted that cables are indeed significant and can make a meaningful change in artistic results.

Thank you, and forgive any verbal stumblings of mine.
Best regards,

KT

[3D Audio]

natpub wrote on Tue, 05 April 2005 04:56

P.S Klaus, this is an emotional reaction, but I think you were unkind to Lynn Fuston on your replies. I see why he "cried Uncle." His comparrison data may be indeed commercial in nature, but I am pretty sure he is not benefitting much if at all financially. I do not see his comments as input from a retailer here, and don't think they need such sharp rebuke. Indeed, Lynn's methodology may be the most accomodating to the very thing you say you seek. I don't think I am wrong on that, hmm...pretty sure Im not. I don't even know the guy, but Lynn's most current methodology is pretty decent from an experimental or clinical psychologist's point of view, shrug...just one opinion. There could be some improvements, but it isn't my place to say, obviously. Not my product.

Kurt,

Thank you for defending me. However I don't feel it is necessary. Klaus and I are friends and, regardless of how he feels about me or my CDs, I have enormous respect for his work and knowledge. I did not take his response as unkind or mean-spirited. (I've been attacked by many people far less knowledgeable than Klaus and I recognize now when someone is being harsh but ignorant.) Quite the contrary, I think his points were extremely valid and if I ever do another Mic CD, I will certainly consider them. Just the same way I considered input from dozens of preamp designers when setting up the Pre CD and many of the microphone designers when setting up the Mic CD comparisons. I would have welcomed Klaus' input at that time just as I welcome it now. But I didn't know Klaus back then.

My philosophy has always been to involve the smartest, most respected people that I can find and to get them in on the front end. That way I can avoid (or try to) criticism after the fact that "if you'd only done this, then it would be valid." I've weighed all the options presented to me, including some that would have required months of work in a laboratory facility, and made my own choices concerning the validity of the testing methodology and it applicability to the real world. It was even suggested to me that I use a spark plug to test preamps for impulse response by Greg Mackie. While it would provide interesting data, I'm not sure a guy recording in his bedroom in Egypt or Iceland would find it that useful or informative.

I have done the best I can with the knowledge I had at the time. I can count among my pool of testing analysts names like Dan Kennedy, Tim Farrant, George Massenburg, Hutch, Dave Hill, Greg Mackie, Wes Dooley, Gene Lawson, and many more than I can recall at this time. (This is for Pre and Mic CDs.) Hopefully next time I can include Klaus' name on that list.

My intent for the Mic CD was simple. Many amateur and beginning recordists have read names like U67, C12, 251, C800G, 44BX, SM7, Coles 4038 and have no idea what they sound like compared to the mics they own. My goal was to allow them the opportunity to "sit in the hot seat" and hear a comparison of some of these classics just as if they had lined them up themselves in their own studio. Many people around the world may never have access to a C12 or an original 251, but they are curious about what they're missing. My goal was to allow them to hear a C12 and a 251 right next to a 57, under testing conditions far more rigorous than most would ever accomplish themselves. (Calibrating to within .1 dB? How many people do that in a mic shootout in the real world? None that I know.)

I think I did an honorable job accomplishing that goal.

Can I do better next time? Likely.

Do I know more now than 4 years ago? Fortunately, yes.

And it is thanks, in part, to people like Klaus and David Satz who inform and enlighten me by sharing here. That, in turn, I share with others and hopefully the circle of enlightenment spreads around the globe.

Therefore, I am not offended or in any way feel unkindly treated by Klaus. I welcome his input.

Additionally, I would propose that our intelligent discussion be expanded to include possible testing methodology, including statistical input from KT, for establishing whether or not differences in cabling can be 1) heard and 2) recorded with meaningful results. If the differences are not obvious at 16/44.1, then how important are they in the real world?

So how would we establish a single variable test that would allow us to evaluate cables? I have heard here that low level signals (mic) probably show off the differences more than line level signals. I would consider that a theory, not fact. But, if true, could we not use a reproducible signal at -60 for the signal source instead of a live performance? I agree with Klaus that variables on the order of performance differences would pretty much negate any valid or repeatable observations.

Let's talk, with Klaus' permission, about how we would go about establishing facts, instead of just adding to the rumor mill.

[Lars Farm]

natpub wrote on Tue, 05 April 2005 05:56

Please let me respectully add that I'm not sure the "trick of the mind" thing is all that constructive, because it suggests the converse response that "maybe your ears just dont work well?"

"trick of the mind" or a persons hearing ability seems to me not to be sonic properties of a cable... The variable "trick of the mind" is safely eliminated by double blind testing.

Lars Farm

[David Satz]

Kurt, many thanks for your posting. I take self-deception as a basic fact of human existence. The greatest self-deception of all would be to imagine that we can exempt ourselves from this problem. Perceptions and memories based on emotional reactions are deeply involved in sonic decision-making; they are subject to all kinds of external and internal influences--everything from ego to indigestion. I don't see how anyone's belief system can change that fact.

If anyone takes that as an insult, there isn't much I can do about it, except to point out that it applies as much to myself as it does to anyone else. On the other hand, people who refuse to admit this, and who try to confuse others for commercial gain (e.g. those who market high-priced cables based on supposedly special sound qualities, and audio reviewers who act as if their every perception is surely based in fact, who make no attempt to find out whether their proclamations are reality-based or not, who act as if they possess a special divine dispensation), that's another thing entirely--I see no reason for any kid glove treatment there, since that's outright dishonesty for profit.

I allow for the possibility of real, repeatable audible differences between cables under some circumstances, but I say that when that happens, those circumstances may need improving. The cable could be what needs improving; there are definitely desirable vs. undesirable properties and electrical parameters, especially when the cables are long. But they should never be taken out of context. Another rampant audiophile disease (besides acting as if one is above human foibles) is making a near-fetish out of individual components without considering the system in which they are being used. That's why I keep mentioning Ohm's Law and controlled testing.

[Bob Olhsson]

It's important to understand that double blind audio tests always come out random if there is no way to quickly train the subjects in precisely what artifacts to listen for. Just handing somebody an ABX box switch is not a sufficient level of training to produce meaningful results.

Passing an ABX test is meaningful but not passing it proves little other than the fact that the listener was probably untrained. Most people can't tell the difference between a CD player and a cassette machine in an ABX test if the cassette machine has been eqd flat.

A lot of excellent research literature is available today related to the development of listening tests for lossy audio coders.

[natpub]

Oh man, I didnt want to cause such a wave....blush...I try to be as transparent around here as I can, hehe. I am not exactly a mic expert.

On the other hand, I did a thesis some years ago on cognitive difference in audio perception, with some fairly rigourous testing methods. This thread recalls much of that experiment to me, so I commented.

Now, those who have endured the folly of grad school (sorry, Zappa was one of my teachers--don't ask, lol) will know that all research papers end with a typical statement, which is really an inside joke---that statement is a call for MORE research, and a conclusion that nothing is really conclusive!! HAHAHAHA!!!

Bottom line, we are dealing with physics to a point here, but as our dear colleague K.K. Proffit so gently puts it, psychoacoustics is not really an exact science (she didnt say that exactly, but I am sifting some of her old posts-I will see if she can perhaps stop by and add her far more clear thoughts). We ain't able to say 1+1=2 and be solid on it. Even without subjective perception, we would have so many variables as to make any absolute answers impossible. I am fairly sure most of us realize that.

Nonetheless, that said, psychoacoustics indeed exists as a science, as does experimental psychology. And, despite the lack of ultimate and final equations, we can point to really useful and meaningful trends. That was my only suggestion here, and none other. I certainly do not mean to imply that anyone here does not fully realize this, because I see well that you all do.

BTW, Lynn, I was not defending you, hehe, I was defending your methodology as practical and useful:-) Sorry to burst any bubbles,--yer cute but not my type!  yer far too....hmmm....male???  /grinz.

Anyway, the entire thing is about subjective preference, IMO. As I recall, Klaus began asking that if you saw no difference in cable, to disregard comment? If you did see difference, then there would be, as Bob kinda points to, no use in measurement, since how can we possibly ask any population to say X or Y is better or worse? Wouldn't that simply be a form of hmm..what do they call it on Madison Ave? "Focus Groups?" (hehe, Bob's pet peeve!).

Alas, I fear we wander so far from Klaus' intent on the thread. It is probably my fault, sigh. Still, I am not sure any meaningful conclusions other than anecdotal and very limited opinions could ever otherwise be shown.

Again, the whole topic was dangerous ground to begin with, as cable has been a mondo topic from time immemorial on all such forums as these.

I return to the topic then, if you DO see a difference in any such mic cables, which do you prefer and why? That seems to be the original question.

As far as testing if such a thing exists in hearing, I am glad to assist any who care to set up a genuinely rigourous test as best I can. However, I can't imagine such a thing has not already been subjected to some kind of research in the academic literature. Alas, being rather preoccupied with writing and recording music these days, I have to make do with my own subjective thoughts. I gave up shrinking heads some time back.

Klaus, I fear when you named this thread as you did, you invited folks like me to hijack it:P I return you to your regularly scheduled program.

My 2 cents. Yes, I tested some of the cables you mention. Yes, I heard a differnce. No, I have not done any extensive tests of high end LDC to high end cables across any significant gammut. So, I guess I have no useful comment


Regards,

Kurt

[Barry Hufker]

Two quick things.

A few years ago at an AES convention, I had the opportunity to take such a test (CD versus cassette in an ABX test).  The test was sponsored by a major cassette company.  They were using an extremely tweaked Nakamichi deck versus an ordinary CD player.  The difference between the two was always apparent without any previous explanation as to what to listen for.  I got every test right.  My prize? A T-shirt that read "I Took The CD/Cassette Challenge And Couldn't Tell The Difference!"

Dyslexia set in for me a few posts back.  Bill Whitaker recommended Belden 8142F and not whatever I said before.

Barry

[maxdimario]

I can see there is a lot of interest in getting the wire right for mic cables. there are some issues that are not strictly capacitance based etc.

Regarding the wire that goes from the microphone to the power suppy, different factors, including stray R.F. from the high tension and low tension filament supplies, can sum-up within the shielded cable and create distortion.

This is due to capacitance towards the shield (ground) or capacitance between adjacent wires, as well as uneven packing of the wires, partial or total oxidization between the strands (which causes some intermittent electrical fenomena as well as changing characteristics depending on how the cable is twisted or in what position etc.). Cheap guitar cables 'crackle' for the same reason.

In fact I have noticed a major improvement noise wise by using a cable with many independently shielded wires inside to keep P.S. noise away from the output wires.


but let's not forget that the U67 for example uses capacitors of 1600 pf to ground on both audio leads and on every single wire that goes into the mic chassis.

I would say that this is a higher pF than any pro cable is likely to have for normal length of cable.

The U47 does not have this feature and suffers from R.F. modulation distortion in some areas with nearby transmitters.

I believe a part of the 'sound' we hear are the side effects of r.f. energy getting into the audio path and interacting with the electronics in the mic and within the mic cable itself.

the shape and orientation of the wires inside the shield along with insulating material etc. will create a system that has a distinct effect on any radio waves that creep into the wire.


There is a major difference in guitar cables because guitars are consumer products that were invented with anything but hi-fi in mind.
guitars have a variable source impedance from 5.5k to 250Kohms depending on where the volume knob is set at and the value of the pot.

in this case you REALLY need extra low-noise low pF cable because the source impedance is so high and the gain involved is also high.

let's not forget that Fenders and the like have built-in AM antennas (single-coil pickups)and this will influence the sound as it interacts with the cable etc.


As far as the short wires that Klaus mentioned, I can understand that the insulating material and orientation of the wires can make a difference. What I do not understand though is how 3 inches of wire in a power supply can change the sound of a 200 ohm source.

how does it change?

[Klaus Heyne]

Kurt wrote:

Quote:

Klaus, this is an emotional reaction, but I think you were unkind to Lynn Fuston on your replies. I see why he "cried Uncle." His comparrison data may be indeed commercial in nature, but I am pretty sure he is not benefitting much if at all financially. I do not see his comments as input from a retailer here, and don't think they need such sharp rebuke.

I am sorry, but after rereading my post, I cannot find any hint of commenting about Lynn's commercial aspirations or legitimacy.

As to being unkind to Lynn- unless you regard my criticism of one aspect of his methodology as unkind per se and therefore verboten, avoiding discussion of possible flaws will not get  us towards a better methodology of testing.

If the tone of my post struck you as unkind, please accept my aplogies, both of you.


Onward: I found the last few contributions very interesting.

David cannot hear any differences between properly functioning and not too long cables. (He may change his mind once a proper test is set up)

Kurt had a brilliantly diplomatic come back to David (I should study that. for my own betterment)

Lynn is right in asking for concrete ideas for setting up a test.

Here are some:

1. Use only live audio inputs from high resolution microphones- maybe three different  types, to cover the  range. Why live? Because mic cables need to be tested with mic levels from real microphones.

2. Include sound sources which are already at the brink of annoyance under the best of cricumstances: Belted female vocals, soprano saxophones, fortissimo piano, complex musical events with emphasis on upper midrange frequencies.

Why? Because of what I mentioned earlier-  The most noticeable qualitative differences in cables seem to be in their handling of challenging mid range information. The rest is icing on the cake.

3. Use a length of cable most commonly used in mic cable application: 25ft-30ft

4. Use the same brand and type connector (needs to be agreed to by all participants) on all cables. Terminate all cables the same way, with the same type of solder. Use connector pins of similar alloys as the inputs on the mics they will mate to.

5. Initially, at least, avoid multi-conductor tube mic cable. Why? Because too few manufacturers offer it, and because its application varies too much- the range of voltages and/or current running in designated conductors (and potentially causing an effect on the audio conductors next to it) varies too much from mic to mic.

6. Include external artificial disturbances - RF, AC-motors, fluorescent lighting, etc. to test the basics of the cable.

7. Find a way to familiarize listeners enough with the test set up so that they feel comfortable making qualitative comments. If that means that they bring their own headphones or other devices that can be included in the set up, let them.

8. Keep the listening sessions very very short. From 30 years of experience I can tell you that any single audio event under test lasting longer than a few seconds is useless.

9. Repeat the tests at different times of the day/night, before and after meals, etc.

10.. Discourage conversation about the test objects during the test at all, including remarks of personal favorites, etc., and use anonymous ballots:
I have witnessed and been part of the psychology of insecurity in listening tests, where everyone is struggling not to lose face and be found out as the only tin ear in the group.

Much more, I am sure.

P.S.: Lynn's comments about 16bit as acceptable release format for a cable CD:
As I said- I am sure that differences will be readily audible, but I am not sure whether the relative aesthetic/emotional merits of cables will come through. One would need to test that.

[JamSync]

natpub wrote on Tue, 05 April 2005 15:31

Bottom line, we are dealing with physics to a point here, but as our dear colleague K.K. Proffit so gently puts it, psychoacoustics is not really an exact science (she didnt say that exactly, but I am sifting some of her old posts-I will see if she can perhaps stop by and add her far more clear thoughts). Kurt

Hi Kurt,


Thanks for the heads-up email. What I have always said is that testing in the audio engineering realm has been consistently marred by the prominent researchers never having a handle on proper methodology. The tested populations are hardly random (mostly white guys over 40). Rarely is anything done as a factor analysis--the experimental designs are never complex enough to grasp the depth of the subject matter.

Most of the research I've read for the past 30 years boils down to anecdotal reports with weak attempts at correlating badly defined variables.  The populations tested are usually so limited in scope as to be infinitesimal when trying to generalize to a large, culturally diverse population.

Not that anecdotal research isn't useful in some limited regard, but it's hardly the holy grail that audio engineers make it out to be.

Audio engineers seem still to believe in the trickle-down theory of technology and  audio decision making. That changed several years ago to the bubble-up situation we have today.

I was at a seminar today where some guy from Motorola was talking about ring tunes and how great his stereo cell phone sounded. Note the word "great" here. Do you think he could tell the diff between his surround sound home theatre and the sound of his cell phone? Sure.

BUT, he still thought the cell phone sounded "great"...ie, good enough to listen to whole albums of music. So I ask you, who cares if the consumer can hear the difference? What we need to be asking is "why does the consumer still think the cell phone sounds 'great' if s/he can tell the difference?" What's up with that?

Figure that one out and you have a much bigger piece of the perceptual puzzle, and one that discrimination trials will NOT give you.

[mwurfl]

I've been reading this thread, and other similar ones on the PSW forums, with great interest.

As a college-trained EE, I, like many others I suspect, am uncomfortable with the notion that even with all of today's technology, we still may not be able to fully characterize the sound quality contribution of various cables based solely on measurable technical parameters.  My engineering training has biased me that way -- I really want to have a logical, repeatable technical explanation for any & all differences that we can hear.  However, I have resolved to be open-minded, and accept that we just don't yet necessarily know how to measure everything that may matter.

I commend Klaus for starting this thread on his forum, as I think it takes courage to publicly stand up for the bottom line of sound quality at the risk of the public scorn of nay-sayers who only think like I had been trained to.  Owning and using a VM1KHE, along with hearing about the almost-universal accolades of those industry professionals who have had opportunity to experience Klaus's services, has taught me to trust that if Klaus says he can hear differences, there are indeed differences.

So, I am hoping that just perhaps, another thing that can come out of this thread would be a new understanding of correlation(s) between measurable technical parameters and what people hear.  I.e., is it possible we are on the brink of discovering either a new way that technical parameters (or combinations thereof) can affect the sound, or perhaps some "new," previously uncharacterized parameter that does?  (I'd really be interested in this last one, since as far as I know, understanding of cable (transmission line) parameters hasn't changed in decades -- it's L, R, & C per unit length).

So I'm wondering if, in addition to the testing methodology that Klaus outlined in his previous post, someone out there, who has access to (and the time to use!) lab gear, can characterize some sample cables, and then send those same cables to Klaus for him to evaluate from his experienced listener standpoint.  I fully realize that many questions arise out of this proposal, such as variations in cable load impedance of the lab setup vs. Klaus's real-world equipment, etc.  At least I'd like to get others thinking this way as well.

On the last point of load impedance, does anyone on this thread know how or if impedance mismatch at the load may be affecting what we hear?  With RF and video systems, this is very important, since unwanted reflections can cause echoes and "smearing" of fast transients, and so load matching is always a fundamentally important requirement.  I've never seen (nor taken a whole lot of time to research, really) whether anyone has done any studies on this for audio purposes.  Maybe the telephone people years ago, who would rightly be concerned since their line lengths are miles, not 20' or so.  But it still seems like with today's high-resolution formats, 20' might be long enough to show (audibly, at least) differences in the signal with different mismatches.

Sorry for the long post.  Keep up the great work, Klaus!  And by the way, I LOVE your phrase, "the brink of annoyance."  I hereby resolve to incorporate it as much as possible in my daily conversations....

Mark Wurfl  

[Barry Hufker]

Again, counting myself as someone who has heard differences at least twice in different audio cables, I will add this from today's ProAudio Digest, Vol.7, No. 4.  And I quote:

"Depending on the microphone, cables may have some influence on the sound, or nearly none at all. Many modern condenser mics have such a low output impedance, and mixers and preamps have such a high input impedance, that the cable just doesn't impact the sound at all. Other characteristics, like how the cable lies on the floor, whether you can hang a mic from the ceiling without its twisting over time, and how heavy or sturdy the cable is, make more difference to the overall usability.

--
David Josephson"

I offer this for whatever it is worth.  Mr. Josephson is certainly an authority on microphone design and operation.

Barry

[Zilla]

After reading various forums which discuss aural differences between cables, I find that there are some general oversights.  People seem to have concentrated only on the wire.  I would suggest that one think of a cable (mic or otherwise) as a system.  

A wire is only a single conductor carrying a single signal.  A cable has more than one wire in its assembly and is usually terminated at each end by connectors.  For example, a mic cable has two conductors and a shield contained in a single jacket, terminated by XLRs.  Each connector has two connections per contact:  an inner connection where the wire is crimped/soldered, and the outer connection which mates with a complimenary contact.  I could go on, but I think the point of a cable being somewhat more complex than first appears has been made.

When one auditions/blind tests/double-blind tests a cable, there is more than wire construction to consider.  You are also hearing the effects of cable termination, connector quality, etc..

[David Satz]

Mark, just to respond to one small part of your post: The electrical wavelength of the highest audio frequencies is on the order of 15 kilometers--even longer if you account for a velocity factor of less than 100%. Audio cables in studios are generally less than 1% of the shortest audio wavelength, and don't function as resonant circuits or "transmission lines." Voltage transfer is the issue rather than power transfer; bridging rather than matching loads are used.

Microphone manufacturers regularly specify an output impedance of 200 Ohms or lower, and a required load impedance of at least 5 - 10 times that, e.g. 1 - 2 kOhms for the input of a preamp, mixer or recorder.