Post by: Barry Hufker on June 20, 2007, 01:34:05 PM
I just finished auditioning a new microphone cable (a new product) from Company X. It's a good cable, well-built using a proprietary "system" (my word) of silver and copper. The XLRs on the end are Neutrik.
I have heard cable differences in the past but usually attribute them to poor quality of the materials used or the amount of capacitance. I tested "Cable X" (my name for it) as well as I could, comparing its resistance with others I own: Gotham "star quad", Gotham "regular" (2 conductors and a shield) and Canare Star Quad. I purposely used different lengths of wire to see how the resistances might vary. I measured pin 3 only for each.
Gotham Star Quad (50'): 13.2 ohms
Canare Star Quad (15'): 16 ohms
Gotham "regular" (25'): 2.4 ohms
Cable X "proprietary" (20') 3.2 ohms
Using an a capella concert as the music test, During rehearsal I used a pair of Sonodore RCM-402 microphones (omni) in an "A-B" configuration. They have mogami wires going from the multipin output of the mics. to the power supply. The mics are solid state but have their own dedicated supply. From the supply I ran two 100' audio runs to a Gordon Instruments Model 5 preamp (two pres in a unit). From there, I used two, 3'Canare cables to the input of a Sound Design 722 recorder (using it's built-in converters). The rate was 96/24.
The sound for this set up was good and as expected. For the concert, I kept everything the same except substituted two 40' runs of Cable X. The difference -- an improvement -- was quite noticeable. The sound was more open and "smoother". I don't know how to expand on the idea behind that word. Now you could say it was the different resistances between the Gotham Star Quad and Cable X. And/or you could say it was the different lengths of the cables. But I believe it to be more than that.
Adding Quickly: I did a similar voice test using a U-47 FET, Cable X and the "regular" Gotham cable. The lengths and resistances not being exact but close. Again I heard the difference. The setup was FET to Gordon preamp (same channels and gain setting used for the swap) to RADAR 24 (Adrenaline Plus) at 96/24.
I then ran a frequency sweep from my monitor speaker through the FET, using the same two cables, recording on a RADAR at 96/24. I did an FFT of each sweep. Cable X's sweep showed more detail than Gotham's. Gotham "rounded" or didn't show a lot of things that Cable X's sweep did -- slightly sharper at points with the appearance of minor changes simply not in the Gotham sweep.
NOW: If the sound of cables comes from "something being wrong" with a system (and I am willing to accept that). Then which system (wire -- Gotham, Cable X) is "wrong" and what is "wrong" in such a simple setup?
Thanks for reading this. I've tried to say it all as briefly as possible, so it may be that I need to clarify things to you as this thread continues.
Barry
Post by: Markus Sauschlager on June 20, 2007, 03:47:09 PM
It struck me hard when there was a difference already in the headphones while doing some test recordings, even before actually A/Bing the recordings. I absolutely dislike the terrible handling properties of the Monster. I listened over and over again that evening and finally found the results with another cable "better" (interestingly from my first impression in the phones I liked the Monster best).
I listened to my test again time after time with fresh ears and unfortunately liked the Monster best each time…
I did further extensive tests including Gotham GAC-3 and Mogami 2534 and 2549 in the hope one of these may beat the Monster. None did. The Gotham was closest but the upper mids seem to be a bit scooped and smeared while offering similar top end resolution. Monster sounded most true in a certain way. (I tested with two acoustic guitars, mandolin and voice, and repeated the test using various mics. Only with an SM7 I liked the GAC-3 better).
I ended up buying two SP1000 cables, because at least for stereo mainmic applications I found myself finally wanting the extra "truth" and immediacy they offer. (I’m still feeling like a fool because I paid such an amount of money merely for cables).
However I still think that a cable cannot "sound" better than another, but that some are degrading the signal less or different than others and in popular music close miking applications it may not necessarily make a significant quality difference in the end.
Barry, I doubt that the differences are caused merely by resistance differences.
I neither think one can say that one is "wrong". But I think it’s best to simply use your ears. In case the "truer" sounding one was "wrong", would you use the "right" one even you don’t like it as much?
Could you say what "CableX" is? I’m quite interested, as I’m still looking for an equivalent substitute for the monster (that is more flexible, less expensive and doesn’t have that weird plastic mesh around it. They make a shockmount almost useless because of their stiffness)
Best regards,
Markus
Post by: Barry Hufker on June 20, 2007, 04:56:50 PM
Now that I read your comment I'm sure you're correct. It is not an issue as to which cable is better but as to which one degrades the signal least.
I'm sure it is not resistance as a sole factor. Resistance, capacitance, quality of materials, etc. are certainly factors. And I would always let my ears be the judge. That's why I think Cable X has some advantages in certain situations and maybe few (or none) in others. But I am really eager to have someone propose why two wires sound different when the electrical difference in the circumstances between them appears to be so small. Some people will propose there can't be audible differences without some technical variance. I don't know...
I'd prefer not to tell the brand for now. For now, I'd like the discussion to be rooted in "the real" (an actual, true example) without bias.
Thanks,
Barry
Post by: Schallfeldnebel on June 20, 2007, 05:54:40 PM
What also plays a big role is the material of the dielectricum. Capacitance is one thing, but what is the material isolating the inner wires from the shield. That can be plastic, it can be Teflon or as in video distribution, it can be even air.
The cables using Teflon sound better than those using plastic, and the cables with airpockets, sound the best, only they are very hard, and used only in permanent set-ups. Also I believe that cables with a massive core, sound better than using thin single wires to make the cable flexible.
Most funny, although all my experience since 1989, I quitted using special cables when I placed converters on stage in 1995, and went AES out to my machines. The effect of only using 10m of cable weighted up against 50m or more highend cable. It was also out of practical reasons. Most of the time the best sounding cables are the worst in handling. Also single cables sound much better than snakes even when made from the same material, so I started to use single wires, until in found out rolling out, and cleaning up cables took about 2.5 hours of work altogether for every session.
Most cables of Van Den Hul I used in those days were in a sort of experimental stage, and never designed for the tough life on the road, so often these cables got problems. Broken wires inside causing me nightmaires. My worst sounding cables are still the ones which are still alive, most good ones are unuseable.
I use Gotham since it is flexible and good shielded. I do not care about the sound anymore because the time I win, I can put in balancing instead of rolling out several cables.
Erik Sikkema
Schallfeldwebel
Post by: Barry Hufker on June 20, 2007, 06:26:53 PM
Investing in outstanding cables is a tough thing. We use so many of them that if we purchased and used only the best, the financial cost would be huge. As you know, I do a fair amount of remote recording and I've had people paint my cables when I wasn't around. Yes, that's true. I was recording a stage show of a chorus with a small band. All my cables were run and the show was set. When I returned I found the group had painted the stage floor black -- and my cables (grey already) as well. I was furious and made a stink about it. But the point is, the best cables are possibly not the best choice, as you say.
Oddly tho' the cables I am testing seem to make the sound smoother. That can be wonderful for some things such as a capella chamber chorus or maybe a nice, "smoky" jazz vocal, but at the moment the cable makes me wonder if it is somehow robbing the signal of its transient response. I haven't tested for that and I need to, but that is how this wire makes me feel, although I've no proof of that.
Barry
Post by: mwurfl on June 20, 2007, 06:58:01 PM
Did you have the Gordon's input impedance set for 2 M ohms, or the low Z setting? I would expect that the 2 Mohm setting would minimize the effects of cable reactances (C and L), and would thus minimize any audible differences between the cables. To that end, it would also be most interesting to have you repeat the whole test (if you have the time and inclination!), only changing the input Z switch to the other position, and then see if the audible difference between cables is more apparent, less apparent, or the same. If you judge it to be the same, it would be most fascinating!
Thanks for the report!
Mark Wurfl
Post by: Klaus Heyne on June 20, 2007, 08:28:11 PM
After hearing from yet another of my clients (a highly reputable classical engineer) who was wondering why he gets RF in certain environments with his cables, I decided that from now on I will supply the interconnects as part of any major microphone restoration/upgrade/custom modification I provide.
It is plain ridiculous to spend thousands of dollares on such work, where I try to squeeze the last bit of available performance out of a mic system, only to give a considerable amount of that back over to (to me) unknown cable choices the client makes, or defaulted to by coincidence.
I will ask the client to use the cable I provide for several weeks/gigs, then, if he wishes, he can go back to the choices he would rather like to make.
The cable, by the way, is Gotham's GAC 3 (the old Austrian-made, thicker, doodo brown version) terminated to my instructions and equipped with Neutric goldplated connectors.
Post by: maxdimario on June 21, 2007, 05:32:10 AM
Post by: Barry Hufker on June 21, 2007, 11:41:35 AM
No, I haven't. This wire supposedly is a combination of silver and copper, or silver-clad copper. I haven't opened it up to look (lazy). I should investigate more but other projects are taking my time.
Mark,
I'll have to check into this with Grant Carpenter. I know for the Sonodore experiment the Model 5 input impedance was 2M ohms. I wanted to get a signal straight from mic to cable to preamp for the next test so I powered a U47 FET directly from the preamp. This is the part I have to check: pressing the phantom power button lowers the input impedance (I believe) to 1k ohms. But in this test the difference in quality was noticeable as well.
Barry
Post by: rodabod on June 21, 2007, 11:58:17 AM
Post by: Barry Hufker on June 21, 2007, 01:42:14 PM
I respect your opinion, please tell me what you "don't get" as I'd be very interested.
I use star quad all the time as I do a lot of recording in theatres as well as "downtown" where there's a lot of RF. I know (from Klaus) about "reussening" for RF and so I use Gotham Star Quad. I guess I am not yet in a financial or aethetic position to own poop colored tho'.
Barry
Post by: Barry Hufker on June 21, 2007, 05:53:18 PM
I've just received a reply from Grant Carpenter:
"Yes, by the P48 standard, input XLR pins 2 and 3 are each pulled up to the phantom supply with a 6.81k ohm resistor, setting the common-mode Z, or a differential Z of 13.62k ohms. At 2M ohms input Z, the input stage does little to further load the mic."
So to sort of answer your post, I've tried the experiment at 2M ohms for music and at 13.62k ohms for speech. In both cases there was an audible difference. I don't know exactly how similar because of the two difference sound sources. But the essential character of the change was there in both cases.
Barry
Post by: mwurfl on June 21, 2007, 07:38:02 PM
I can see that my near future may just see me doing this very sort of test for myself, using my own Gordon. Although, it would be nice to know what "cable X" is so I can try the test with it too....
Mark Wurfl
Post by: Grant Carpenter on June 21, 2007, 08:06:22 PM
Regarding audio paths, I prefer the least current through the shortest conductor, thus the 2M ohm input Z. The P48 phantom supply is electrically part of the mic, is a part of the output Z of the mic, but is physically located outside the mic, typically in the pre. The resulting signal and supply currents must pass through the mic cable and will be affected by the cable (and supply), becoming part of that mic's signature.
My shortest answer to the frequent "which mic cable?" question is Teflon is the best dielectric (between all conductors, but not the jacket), silver is the best conductor, and copper is close. Better yet, when possible, lose the cable and put the pre at the mic.
Grant Carpenter
Gordon Instruments
Post by: Barry Hufker on June 21, 2007, 08:15:46 PM
Of course... Grant.... you need to design a "preamp" stand for that....!
Barry
Post by: maxdimario on June 22, 2007, 05:36:43 AM
oxide, pressure, tightness of winding make a difference, especially in old wire.
a good example are 'crackly' guitar cords from the 60's..
so to me ultimate sound is solid wire...
Post by: rodabod on June 22, 2007, 07:39:31 AM
| Barry Hufker wrote on Thu, 21 June 2007 18:42 |
Roddy, I respect your opinion, please tell me what you "don't get" as I'd be very interested. |
Hi Barry,
Well, for me, I'm not convinced it adds up on paper. We have a typical mic preamp with say, 1.2K input impedance and a mic source of say, 200 Ohms. Looking at the DC resistance values firstly, we are not going to experience any significant losses. As for impedance, do we have capacitance values for these example cables? I'm kind of doubtful that the resultant frequency response is going to make a huge difference.
Having said that, I've not had a chance to use any of these exotic cables. I think I have probably been put off the whole cables argument by the HiFi world where there is a lot of snake oil!
You could always analyse a reel of cable by measuring its frequency response at a given source and load impedance. Otherwise, a blind listening test could work, although I've never been too sure about that method for various reasons.
Post by: Barry Hufker on June 22, 2007, 11:18:25 AM
I have been as skeptical as you. I am no believer in the mystical wires of hifi. Yet, I've heard real differences with these two microphone cables (and actually with other wires before). I've been ready to explain it with the usual technical reasons -- capacitance, etc. but can't in this setup. I don't think the wires have to be exotic to hear some difference for better or worse.
I suppose the point of this thread for me is not are there wires providing a different sound, but I don't believe our current (pun) technical understanding can explain the reasons why the differences exist. Decades ago I heard the sound of different capacitors, etc. in audio circuits. I suppose it's only reasonable to hear it in wires...
Barry
Post by: David Bock on June 22, 2007, 01:41:34 PM
| Quote: |
one thing I've noticed with multi-strand wire is that the individual strands do touch each other and therefore make contact but not always in the same fashion, oxide, pressure, tightness of winding make a difference, especially in old wire. |
Absolutely true.
You probably also noticed irregular oxidation of that wire.
It reminds me of how many broken (or a molecule away from being broken) wires I've found in U47's and M49's. Almost always at a bend in the wire or solder junction, sometimes still in the jacket.
| Quote: |
so to me ultimate sound is solid wire... |
Not too practical, kind of like a Lowther. Good in theory but doesn't play out in practice.
The other option is Litz wire, and several people make audio cable based on this. You get the use advantages of stranded without any of the oxidation.
Some makers go to great lengths to maintain exact conductor vs. insulation through the run of the cable as well.
And then there is the issue of the insulation itself, which is too large a debate for me to get involved in, other than to say that in products I've built, manufactured & tested it CAN make a difference in both mic and line level applications.
BTW I'm also not debating solid vs stranded as I have NO (!) experience with solid core cable as far as I know, other than.....phone lines?
regards,
David Bock
Post by: Mike Cleaver on June 22, 2007, 03:07:37 PM
However, it was only used where it would not be subject to any movement.
In later years, stranded wire became more common before everything moved to pc boards.
Solid copper open pair often was used to connect the telephone central office to the transmitter site and sometimes from the studio direct to the transmitter, often strung along with telegraph wires beside the railway tracks.
You'd pump out enough db at the station to end up with plus 8 at the transmitter site.
At my first station, we had an RCA BA6A putting about plus 24 into the transmitter line to get plus 8 at the site.
That solid copper sounded great back then, a lot better than the minuscule stuff used today along with amplification and equalizers to get the required line frequency response.
About the only thing that was ever added to the open copper pair were WE 111C repeat coils at each end.
Post by: Barry Hufker on June 22, 2007, 10:55:28 PM
http://www.hufkerrecording.com/microphone%20cable%20test.htm l#Cable_Sound_Samples
The first is me speaking into a U47FET, first using one cable and then the second. Outside of that, all is identical. The file consists of both samples. See if you can tell at which number I made the edit between the two. I will say that when the files were converted to MP3 some of the distinctions were lost. But there is a subtle change.
The second is an excerpt from a live choral recording. You can read the setup for this on the webpage. The ticks or pops you'll hear are the conductor's shoes! The building's air conditioning blower is on and it begins to rain (although the rain isn't in this excerpt).
After that I've posted a frequency sweep from 20 Hz to 20 kHz. This setup was monitor speaker to U47, then as for the speech test. The frequency plots are from Spectrafoo. I've done them in sections in an attempt to get the clearest picture for each frequency band. The yellow plot is of the Gotham GAC3 (the two conductor with shield). The red plot is "Cable X". Gee I love the intrigue of writing "cable X". It is so mysterious!
See and hear and let us know.
Barry
Post by: Klaus Heyne on June 23, 2007, 03:59:13 AM
Arguments about objective criteria or their lack were put forth. Next came the point that every cable sounds the same, and it was all audiophile hogwash/hubris.
Then there was speculation as to the best/worst conductor/sleeve/shield material and shape, and how that was supposed to be the reason for the superior/inferior performance of the cable in question.
In the end, not much was gained by the reader- irrefutable facts he could take to the bank - so the whole exercise did never live up to the level of discussion seen here when it comes to microphones.
I see two ways to avoid a repeat show of that murky past:
1. Try to be extra respectful of someone else's differing opinion about a cable's sound when you post.
2. If you cannot cite a scientifically acceptable measurement correlation (f.ex. a line capacitance of 1500pf in a 30ft cable cannot possibly yield a frequency balanced cable) or if you cannot cite empirical, commonly agreed to data (of the subjective type we seem to readily accept in mics like a U47, CMC5 or 251, but somehow deem below the pale when discussing cables) then don't bother trying to convince the reader of the validity of your opinion. Instead, simply and specifically state what you like or do not like about a cable.
That last bit is much more valuable to me than trying to give respectability to your subjective opinion with citations of pseudo science. It gives me a starting point- I then can compare your results to what I have tested in the past.
So let me try to live and post by my own rules:
I just posted about starting to supply my restored/modified mics with Gotham's GAC3 (same as Neumann's IC3) as interconnects.
Why?
1. I found them to be always top of class in RF suppression, when comparing them to any otehr commonly available balanced line cable.
2. I found them to be least objectionable in what they do to the sound of a mic. A bit too slender in the bass, not as robust in the lower mids as my dream cable would be, but the rest I can very well live with: very little congestion in the upper mids, and smooth transitions from one frequency band to the next, throughout the audio spectrum.
Despite my past loyalties to Gotham's GAC 3 and GAC7 (tube mic) cables, I am always looking, and quite willing to replace Gotham tomorrow with a brand and type of cable that does the job better still.
Let me know what in your opinion that superior cable might be (hopefully readily available, and in a price range a mortal can afford!) and I trust my instinct to recognize when the light starts flashing...
Best regards,
Post by: Schallfeldnebel on June 23, 2007, 07:50:22 AM
Der Schallfeldwebel
Post by: Barry Hufker on June 23, 2007, 10:19:20 AM
Thanks for the comments. I'm not trying to convince anyone of anything, in case I've confused the issue. I've been trying to spark an interest in this discussion by providing some graphs and sound samples for general interest.
My comments have hopefully been: "Here's a cable. I think there's a sonic difference. Come listen for yourself. Take a look at the graphs and see if you can determine anything significant from them." I've used a well-know microphone for the speech example -- the Neumann U-47FET, as I don't have a U-47. I know there is a huge difference between the two microphones but it is possible people are familiar with the FET.
For me, the difference between the Gotham wire and the test cable is an issue of "smoothness". And I'm not sure how I feel about this. For me, I can see the quality as another flavor when miking something but I'm not sure I'd choose it all the time.
With regard to specs, I've posted what I can. I don't have access to the capacitance data for all the wires.
It may be that this thread has now run its course -- and that's fine.
The "test cable" is Accusound's new "Silver Pro". I've avoided the brand to this point so there wouldn't be any bias going into this discussion. With regard to price, the retail is: $140.00 USD for a 20 cable, with a reasonable discount available.
Again, my hope has been to say "here's something interesting". I don't care about convincing people one way or the other. In the interest of "let's expose ourselves to all we can, here is this."
Erik,
You've identified where the edit is. Great work! The numbers 1-6 are Gotham. 7 through 11 are Accusound. In MP3 it's a tough call. The difference is more noticeable at 24 bits and 96kHz.
What I find about the Accusound is that it sounds maybe a bit less nasal and possibly more open. But I feel as tho' it smears sounds together. Again, I'm not convinced these qualities show up in the MP3, but that's the best I could post. The MP3 is at the highest quality the encoder allows me to produce.
Barry
Post by: maxdimario on June 23, 2007, 11:13:34 AM
the last part of the numbers is clearer to me
choral from 57 seconds onwards is also clearer
Post by: Barry Hufker on June 23, 2007, 11:26:26 AM
B
Post by: Schallfeldnebel on June 23, 2007, 12:02:49 PM
for anyone who still has doubts if cables have a sonic influence on a recording, my detection of the right spot prooves there is more than L, C and R.
Der Schallfeldwebel
Erik Sikkema
Post by: Klaus Heyne on June 23, 2007, 12:46:42 PM
You did a fine job with your approach to cable comparisons/evaluations. I have no complaints.
However, the identification of the Gotham cable you used is a bit mysterious to me. You wrote:
| Quote: |
I tested "Cable X" (my name for it) as well as I could, comparing its resistance with others I own: Gotham "star quad", Gotham "regular" (2 conductors and a shield)... |
and then:
| Barry Hufker wrote on Fri, 22 June 2007 19:55 |
... The yellow plot is of the Gotham GAC3 (the two conductor with shield). |
GAC 3 is a three conductor + Double layer Reussen shield configuration. And, as far as I know, the company does not make a star quad (2 twisted pairs + shield) configuration. Please clarify what cable you actually used in your tests.
Thanks,
Post by: Barry Hufker on June 23, 2007, 01:40:05 PM
I can't find in my emails the discussion I had with Lewis Frisch so we must have talked by phone. The cable we discussed is GAC-4/1. This was quite some time ago so I know the moment I say "well Lewis said it could be used that way" is the moment you and he both correct me. So I won't.
Gotham does not list GAC4/1 as a star quad cable. EDIT: Yes it does. See post below.
This however was the cable I used in the music comparison. There is no music example of the Gotham cable as I didn't record that. I listened to the Gotham during a Saturday rehearsal and switched to the Accusound cable for the concert. It was a that time I immediately noticed the difference.
Thank you Klaus for making me do better research. But to be clear I stand by my sonic conclusions and the graphs I produced. The graphs are of the Accusound cable and Gotham GAC-2.
Erik,
Of the two examples I find the Gotham cable to be more to my liking, although I think the Accusound has its place.
Barry
Post by: Klaus Heyne on June 23, 2007, 02:41:13 PM
In any event, I understand that Gotham regards its GAC 3 as the flagship of their balanced mic/line interconnects.
Therefore it may be interesting for you to try a similar length of it and compare it to the GAC2 or 4/1. I at least heard a difference to the better...
Post by: Barry Hufker on June 23, 2007, 03:11:36 PM
EDIT: Here's where I got the idea Gotham makes a Star Quad cable: http://www.gotham.ch/products_en/gac/11002.htm
Thank goodness I found that. I was *sure* I was right (but doubted because I so seldom am!).
Barry
Post by: Schallfeldnebel on June 24, 2007, 06:23:28 AM
I have had nothing but trouble using this kind of connection method you describe above; radio interference was the result. Especially when cables were connected to snakes or when using extension cables having the same wiring method. I believe it is a nice method using one wire to a microphone preamp, but not in complicated set-ups.
I use GAC 3 cable, but only two inner wires are connected. The third wire is used with 4 pin XLR using active powering to some microphones.
Erik Sikkema
Schallfeldwebel
Post by: maxdimario on June 24, 2007, 11:35:59 AM
Post by: Schallfeldnebel on June 24, 2007, 02:39:46 PM
| maxdimario wrote on Sun, 24 June 2007 23:35 |
when you connected the third wire as ground and had r.f. problems did you connect the sheld only on one side? |
In our set-up the shield was only connected on one side.
Schallfeldwebel
Post by: maxdimario on June 25, 2007, 04:33:02 AM
when you consider that cables are like antennas with internal capacitance etc. there are a lot of issues with rf.
in my tube mic I have shielded wires for the power supply as well as the audio, and shunt caps throughout at the connector on the mic itself.
Post by: Kees de Visser on June 25, 2007, 09:43:48 AM
| Schallfeldwebel wrote on Sat, 23 June 2007 18:02 |
for anyone who still has doubts if cables have a sonic influence on a recording, my detection of the right spot prooves there is more than L, C and R. |
. Apart from that it takes many trials to make a (statistically) valid double blind test.Analog sources will always be different by nature, in contrast to digital ones. This makes comparing transducers e.g. so difficult. It has to be proven or at least made very plausible that the variable under test (here: cable) is the only dominant one.
When looking at Barry Hufker's graphs I wonder if the differences wouldn't have been similar when doing two sweeps with the same cable. There is likely to be some variation in every test. Can it be assumed that we're not just comparing random differences in the stimulus ?
Klaus already pointed at the difficulty of discussions about audibility. I think that one of the major problems is the individual "proof threshold". Schallfeldwebel apparently finds justification in one positive result. I prefer to find evidence in (time consuming) double blind (ABX) tests, which are IMNSHO the only valid way to prove audible differences.
Post by: Barry Hufker on June 25, 2007, 11:22:43 AM
I can't disagree with one word you say. There is nothing statistically that "proves" anything in what I've done. In fact, of the two graph responses (Gotham and Accusound), there is nothing to say which one of them is more accurate.
But I also want to make clear my goals:
1. I did hear a difference between the two wires.
2. I wanted to present sound samples for other people to decide for themselves.
3. I wanted to present graphs so there was some "objective" data.
4. I don't want to "prove" anything. I want to spur interest so people will explore.
Erik certainly doesn't need anyone to speak for him, but as long as I'm writing, I'll say this. I've known Erik for quite a while. I have *no doubt* that with circumstances he is comfortable with he would reliably be able to pick out a difference. He has remarkable ears and is a keen listener.
Except for the remarkable ears and keen listener, I would say the same for me. I feel I could reliably choose either cable in a blind test between the two. And I'm not saying I could do more than that.
I value your comments Kees and thank you for making them. You bring us back to an important "ground" for us to have in this discussion. Time permitting at some point, I'll try to make another test or two. (and please include the GAC 3 which, to my ears sounds quite different than the 2-conductor series! K.H.)
That won't be any more statistically significant but it may help us clear things a bit.
Barry
Post by: Kees de Visser on June 25, 2007, 05:48:24 PM
| Barry Hufker wrote on Mon, 25 June 2007 17:22 |
I've known Erik for quite a while. I have *no doubt* that with circumstances he is comfortable with he would reliably be able to pick out a difference. He has remarkable ears and is a keen listener. |
. We both love music and audio and have spent lots of time together listening and discussing while we were still living at a reasonable distance.The (pro-) audio world is full of enthusiastic people and I love that. At the same time I often notice a lack of scepticism and objectivism. From time to time I try to blow a whistle when I feel conclusions are being drawn based on insufficient or wrong facts. IIRC it was Bob Katz who once wrote: "If jumping to conclusions were an Olympic event, we would have some gold medal winners!"
In a recent discussion about scepticism, Bruno Putzeys (who also has his forum on REP) wrote:
"When confronted with a new phenomenon, the first thing you do is organise a controlled trial to see whether it's real or not. When that trial confirms its existence, you can start looking for an explanation. When someone comes to you saying he can find water using a bent stick, the real question you need to answer is not "How can he find water with a bent stick," but rather "Why does he believe he can find water with a bent stick". Only when the answer is "because he can", you would move to the first question. Failing that, the question becomes a psychological one."
So first one has to prove that a difference is audible (in a DBT).
Klaus, do you know of any standardized DBT for microphones ? It's probably not easy to set up.
Post by: Schallfeldnebel on June 25, 2007, 06:40:09 PM
Barry asked us if we possibly could hear where the recording with the counting jumped from one cable type to the other. When I listened the first time I immediately felt on the 7 something changed in the character of the sound. When I relistened a couple of times, I knew it for sure, so I wrote my reply.
But..... I was wrong in which was which. I found the 1-6 sound better than 7-11. So my conclusion was simple, I heard a difference even with MP3, but what was the best for me, was pure a matter of taste.
I have lots and lots of experience with dedicated microphone cables, and it has happened several times in recordings I started making balance using my multcore-snake, and then switched over to a highend cable, and asked the musicians if the they liked the result better. Then when I explained I only swapped cable, they could not believe a cable could make such a difference.
I would just advice, use your own ears, but keep in mind not to invest a fortune, and cables should be practical in use, and stay alive.
Erik Sikkema
Schallfeldwebel
Post by: Barry Hufker on June 25, 2007, 11:51:57 PM
I can be as skeptical (and even as cynical) as anyone. But I don't know this area (sound of cables) needs a technical explanation. I am not claiming anything scientific. I am stating a personal observation and offering a little bit more (graphs) to see if anyone else can tell a difference, no matter how that difference is produced. And for all purposes, the sound example I present is blind. It can't be double blind because I know the answer, but it can be blind to you. And that ought to be good enough for people to judge. If you think the division I made between the two samples was too intuitive, that can be changed.
I would suggest this kind of thing doesn't need science as a foundation any more than a good meal does. How is one going to do double blind experiments to find out who makes the soup?
When recording I put up a mic I think will do the job and then listen to it. If I don't think it sounds good, double blind experiment or not, I change it. The same would be true for a cable. Personally I don't care if a group of people can always tell which cable is which. It matters that *I* hear a difference and employ it or not. That's the nature of the business -- and the nature of so many businesses.
Barry
Post by: Klaus Heyne on June 26, 2007, 12:18:49 AM
| Kees de Visser wrote on Mon, 25 June 2007 14:48 |
...Klaus, do you know of any standardized DBT for microphones ? It's probably not easy to set up. |
To my knowledge, there is no standardized double blind or any other type of listening test that is universally accepted by the professional audio community at this time.
The reason seems to me pretty straight forward:
In addition to the considerable logistics of developing a standard, the professional audio community, in difference to the audiophile community, seems still rather unsure when it comes to the acceptance of anything but hard data for evaluating audio gear.
As hard data are hard to come by and often not very helpful in the evaluation of sensual perception, the whole field of (subjectively) evaluating audio information is woefully stunted and lacking in my opinion.
Few "professionals" have invested time and energy into their own hearing education, and consequently are unsure of what they hear, and how to file in their brain what they heard. Many of them would not risk being exposed as tin ears and amateurs, possibly ruining their reputation and source of income.
I will continue to defy those who are unwilling to first and foremost use their ears to judge audio. But I will also continue to stress to improve our perceptions, and start to work towards a universal language and protocol of evaluation that is broadly agreeable.
Post by: Barry Hufker on June 26, 2007, 12:55:30 AM
At an AES convention of some years ago, a manufacturer was trying to convince people cassettes sounded as good or better than CDs. I decided to take the challenge. They had a Nakamichi Dragon ultra-tweaked. They switched between music on CD and on cassette while I wore headphones. I got the test completely correct, able to tell which was which every time. As a prize for having done so well, the manufacturer gave me a t-shirt. It read,"I took the Cassette-CD Challenge and couldn't hear the difference."
Barry
Post by: maxdimario on June 26, 2007, 05:25:33 AM
Post by: Markus Sauschlager on June 26, 2007, 06:05:05 PM
I recorded some concerts during the last 2 months performed on great historic instruments of the V
Post by: maarvold on June 26, 2007, 09:05:20 PM
| Kees de Visser wrote on Mon, 25 June 2007 06:43 |
I prefer to find evidence in (time consuming) double blind (ABX) tests, which are IMNSHO the only valid way to prove audible differences. |
It seems like one potentially valuable 'analog' source (although Stephen Hawking might not agree that it is truly analog*) for double blind A/B/X testing might be a piano equipped with Yamaha's Disklavier recording/playback system.
* For example, he says that if you were able to take an extremely high speed movie of a child's spinning top, you could watch the movie in slow motion and observe that the top's spinning slowed down in a series of discreet steps rather than linearly.
Post by: Barry Hufker on June 27, 2007, 12:36:54 AM
Again I can agree with all you say.
Michael, I find your comment about the top quite interesting. That is one of the things I'll file away for later use.
Barry
Post by: Kees de Visser on June 27, 2007, 06:10:09 PM
| Klaus Heyne wrote on Tue, 26 June 2007 06:18 |
To my knowledge, there is no standardized double blind or any other type of listening test that is universally accepted by the professional audio community at this time. |
| Quote: |
Few "professionals" have invested time and energy into their own hearing education, and consequently are unsure of what they hear, and how to file in their brain what they heard. Many of them would not risk being exposed as tin ears and amateurs, possibly ruining their reputation and source of income. |
| Quote: |
I will continue to defy those who are unwilling to first and foremost use their ears to judge audio. But I will also continue to stress to improve our perceptions, and start to work towards a universal language and protocol of evaluation that is broadly agreeable. |
Post by: Barry Hufker on June 27, 2007, 06:28:08 PM
Post by: Kees de Visser on June 27, 2007, 06:30:43 PM
| maarvold wrote on Wed, 27 June 2007 03:05 |
It seems like one potentially valuable 'analog' source for double blind A/B/X testing might be a piano equipped with Yamaha's Disklavier recording/playback system. |
Post by: Schallfeldnebel on June 27, 2007, 08:08:31 PM
| Kees de Visser wrote on Thu, 28 June 2007 06:30 |
Any other ideas for a reproducable source ? |
Yes, a very simple way. You take a line signal from an analog or digital tape recorder, which you attenuate to microphone level, and boost to line level again by using your favourite mike-preamp.
This gives a very reliable source, and the output level can be varied. It is interesting to hear that the higher the output level is, the less sensitive is the cable. Those cables which were on microphone level not that good, sounded much better on line level.
The only thing which is hard to test the way I suggested is the influence of phantom powering.
Erik Sikkema
Schallfeldwebel
Post by: maxdimario on June 28, 2007, 06:41:16 AM
so depending on if it's a transformer mic or not and if the mic pre has a transformer etc. etc. you will get different results.
to test cables I would suggest using an unusually high source impedance, as this is probably the best way to hear the defects.
Post by: Kees de Visser on June 28, 2007, 09:08:40 AM
| Schallfeldwebel wrote on Thu, 28 June 2007 02:08 | ||
|
Mechanical instruments come to mind, like the automated B
Post by: Schallfeldnebel on June 28, 2007, 02:26:13 PM
Most of the time my testing instrument was the organ, and I have been lucky to work with great players like Harald Vogel and Jacques van Oortmerssen, who have such a good touche on the keyboard, that the sound they produce is equal to the ear.
Organs with electro pneumatic action have no different sound caused by a different touche, those might be even better test objects.
Schallfeldwebel
Post by: Klaus Heyne on June 28, 2007, 02:31:58 PM
There is only one, very rare, and probably unobtainable, player piano machine which can express the dynamic and analog behavior of a real piano played by hand. All others are so far away from simulating what a real piano sounds like that test results would be unreliable.
Furthermore Erik's idea of using a pre-recorded sound source suffers from the uncertainty of what really happens electrically during the impedance up- impedance down conversion that's necessary in simulating the impedance behavior of a real mic/preamp interaction.
If that issue could be solved, then the much larger issue of: "what does the mic really sound like"? and therefore: "what are the cable's imprints on the sound of the mic?" could be answered.
But that is not ever how we generally approach music recording, or is it? A good engineer never chooses supposedly neutral components for recording. The vast variations in microphone performance alone attest to the creative choices the engineer/performer/producer must make to bring out the music in the way he/she intends for the recording. That choice extends to the cables used as well.
Except, that, in my experience, there aren't any good, clear choices available (yet!), only more or less unsatisfying compromises.
Unlike the audiophile/HiFi (line level) cable world, cable design for microphone is still in its infancy. Here, the handfull of makers compete mainly on hype, endorsements and price, rather than catering to and satisfying a sophisticated clientele. That clientele does not yet exist in large enough numbers to make a difference in choices and quality of the products that are available.
Post by: Barry Hufker on June 28, 2007, 03:19:24 PM
I used a Telefunken-USA U-47M (with VF14 tube and K47 capsule) for a test involving spoken word only. The setup was otherwise the same as previously posted: microphone (its cable to power supply) to TEST CABLE to Gordon Instruments preamp to generic cable to RADAR 24. The recording again at 24/96. The result was a greater disparity than before. The GAC-2 cable from Gotham was etched and thin, while the Accusound wire was smooth and full. Quite frankly it is the only cable of those two to use. The mic sounded "fine" or "OK" with Gotham but sounded warm and rich with Accusound.
Frequency sweep: taking advice posted here, I made two consecutive frequency sweeps through the U-47M. Interestingly, the Accusound cable showed significant differences from one sweep to another although nothing bizarre, just noticeably different. On the other hand, the Gotham cable's traces -- both of them -- were exactly the same.
Barry
Post by: maxdimario on June 28, 2007, 05:18:12 PM
could also be that the u47 has more there to begin with and the 'good' cable is letting it shine for what it is, as opposed to masking it.
a more complex fet mic might have less clarity to shine through so the difference between cables may also be less noticeable..
Post by: Klaus Heyne on June 28, 2007, 06:22:53 PM
You continue to compare the same two cables with different mics/pres/settings.
This could lead readers to the impression that one of them is bad, the other one good, at least as far as your subjective reporting so far is concerned.
The impression you create is that the Accusound is a good cable, and the GAC2 not so good (the latter of which I agree with, from my past testing.)
Before anyone goes running out and buys the Accusound, it could very well be that, in a different, expanded comparison test, the GAC2 ends up being pretty lousy, and the Accu a bit less lousy. Or, that indeed the Accusound may be the new gold standard.
Post by: Barry Hufker on June 29, 2007, 01:13:46 AM
I take your point but I don't think you've stated the case correctly.
1. Like any "good" scientist, I changed only one variable: I added only a new microphone this time. Everything else was as before -- same preamp, same generic cable between preamp and recorder (which is all I can get) and the same recorder. Same gain settings, same room, same "me", same recording rate and bit depth. The only thing that changed this time was the microphone. By changing the microphone (only) I have demonstrated this sonic change is not relegated to only one microphone or microphone type. The effect may indeed be related to the cables themselves. That should explain this part of the most recent test.
2. As was suggested from one of the posts here, I made two tone sweeps (20-20k) this time for each cable, just to see if graph differences were due to something real or are just "accidental". That's why I posted that the Gotham cable traces were identical in both sweeps and the Accusound traces had noticeable differences.
3. It is not my goal to determine how various cables compare. My one and only goal has been to learn how two cables I own compare. That leads me to a specific artistic goal. Testing a variety of cables to learn how they compare does nothing for me. I can't afford them and I don't have the time/will to make such tests.
So what's the purpose? It is the one every engineer makes for himself/herself. "Here is what I own. Here is what I've run across. How do these two things (and these two things only) compare? Should I make a change or am I happy with what I have?" This then is the story of my finding a new microphone cable that seems superior to what I own.
People who want a larger experiment should look elsewhere. I am only reporting my experience. I've probably done that to the limit I care to for now. If another circumstance arises, then I might want to reconsider this issue for myself. If anyone is interested I'll post that finding too. Of course I am hoping others will make some sort of comparison for us to learn from.
4. I'm sorry (for myself) to learn GAC-2 cable is considered to be a poor wire. I purchased it decades ago when it was sold as a Neumann product. It was the best I could afford then and the best I knew about. I now know about other cables as this thread has encouraged people to mention wires they've had experience with. In no way do I think Accusound is a superior wire to anything other than the cable I've compared it with. It is up to others to make further judgments.
Barry
Post by: Kees de Visser on June 29, 2007, 04:51:13 AM
| Barry Hufker wrote on Fri, 29 June 2007 07:13 |
Testing a variety of cables to learn how they compare does nothing for me. I can't afford them and I don't have the time/will to make such tests. |
Perhaps the test is flawed from an electronical point of view. Comments much appreciated. (e.g. would the order be important ?)
It could also be interesting to daisy-chain 2 identical cables (A+A, B+B) and see/listen if that changes anything.
Post by: Schallfeldnebel on June 29, 2007, 08:52:38 AM
Erik Sikkema
Schallfeldwebel
Post by: Barry Hufker on June 29, 2007, 11:11:18 AM
Barry
Post by: Ralf Kleemann on June 30, 2007, 04:02:36 PM
When you used SpectraFoo ("Complete" version) for creating the frequency response charts, did you also take a look at the Transfer Function feature (part of the instrument list)? This allows for a "power vs frequency" as well as "phase vs frequency" chart, which is most interesting for this type of experiment, especially when combined with the built-in function generator. Try using short bursts/sweeps and you will get a nice impulse response graph for the system you are examining.
Best regards,
/Ralf
___________
P.S.:
| Barry Hufker wrote on Fri, 29 June 2007 17:11 |
Every algebra problem involves solving for "X" and that frightens me. |
Why? It's so easy:
Post by: Klaus Heyne on June 30, 2007, 07:13:40 PM
| Ralf Kleemann wrote on Sat, 30 June 2007 13:02 |
...Try using short bursts/sweeps and you will get a nice impulse response graph for the system you are examining. |
..but how does this examination help me understand why one cable sounds better than the other?
Post by: mdemeyer on June 30, 2007, 11:27:04 PM
Would you elaborate on how you terminate the GAC3 cable you favor? Wondering how you are using the extra conductor and shields.
Thanks,
Michael
Post by: Barry Hufker on July 01, 2007, 01:11:18 AM
No I didn't look at that, but will.
Thanks!
Barry
P.S. Thanks also for finding X. I was worried it would be under my bed tonight while I'm sleeping.
Post by: Ralf Kleemann on July 01, 2007, 05:37:30 AM
Sorry if I didn't elaborate in my previous message.
With the power/frequency and phase/frequency transfer functions (no matter if you use the one provided in SpectraFoo or create one yourself), you have a very good tool to see the overall response of a system, or how it changes the original signal. I wrote "see", not "hear", because while this transfer function gives you a clue about the most vital parameters (power, frequency, phase [as in: signal over time]), it is never meant to replace an actual listening test or a direct comparison.
However, I believe that applying this methodology can yield useful results and goes beyond the simple frequency response chart. Definitely worth a try, especially since Barry is using an application that has this functionality built-in.
Having said this, I would never base a final claim about a cable's musical capabilities on such an experiment alone.
Best regards,
Ralf
Post by: Klaus Heyne on July 01, 2007, 07:41:33 PM
| mdemeyer wrote on Sat, 30 June 2007 20:27 |
Hi Klaus, Would you elaborate on how you terminate the GAC3 cable you favor? Wondering how you are using the extra conductor and shields. |
Gladly.
I use the same termination scheme that Neumann has used in the past, and is still using.
Both connectors on a three conductor + shield mic cable, like Gotham GAC 3, are terminated like this:
Ground conductor and shield go to pin 1 of XLR and to the metal connector's solder lug provided for this purpose. On most connectors this ground/shield lug is a metal loop which has conductance to the connector's shell.
Where there is no dedicated third conductor for ground return, I still will connect the shield (which in these cases also acts as ground wire) the same way as above.
I start out the cable/connector terminations in line level applications between pres, processors, or recorders the same way, but may have to remove the shield termination on one end of the cable if I detect ground loops.
Post by: maarvold on July 02, 2007, 02:14:11 PM
| Klaus Heyne wrote on Sun, 01 July 2007 16:41 |
Ground conductor and shield go to pin 1 of XLR and to the metal connector's solder lug provided for this purpose... but may have to remove the shield termination on one end of the cable if I detect ground loops. |
This was my experience when I made my own cables with Gotham GAC-4/1. I had some mics that exhibited RF-related problems in several studios in Los Angeles so I decided to try Gotham's cable with double-Reussen shielding. If I connected the shield to both Pin 1 AND the connector's shell at both ends it created ground loop problems in some studios. When I disconnected the shield from the connector shell at the male XLR end, the grounding problems disappeared; the RF-related problems still were cured with this configuration. On an impulse, I removed the RF traps (I would call them)--simple inductor & capacitor circuits in the microphones--and the purity of the mic's sound improved as well. It was as if there was a tiny [audible] 'hitch' that had been in every waveform that was no longer there. FWIW, GAC-4/1 has even 15 dB better RF rejection than GAC-3. GAC-4/1 features both double-Reussen shielding and a star-quad configuration in one cable. Quoting from the Gotham USA website:
"Gotham "Star-Quad" cables are the most advanced microphone cables presently available. "
Post by: Barry Hufker on July 02, 2007, 02:24:06 PM
I agree with you. I believe grounding connectors has been largely disapproved of for years just for the reasons you mention.
Barry
Post by: Klaus Heyne on July 02, 2007, 02:45:43 PM
| maarvold wrote on Mon, 02 July 2007 11:14 |
...If I connected the shield to both Pin 1 AND the connector's shell at both ends it created ground loop problems in some studios. |
| Quote: |
Mike, I agree with you. I believe grounding connectors has been largely disapproved of for years just for the reasons you mention. |
Your reasoning is flawed, in my opinion, and keeps contributing to the perpetuation of an unexamined myth that seems impossible to get rid of.
How can there be a ground loop in a microphone line, when the microphone is the beginning of the recording chain? Unless the mic's body or its connector touch another electrical device directly (which it never should!), there is no opportunity for a ground loop.
This is of course in stark difference to interconnecting self-powered units (mic pre/processor, etc.), where, as mentioned before, ground loops may occur, and removal of one of the two cable shields from pin one or the connector case may be necessary.
I have equipped "properly" terminated mic cables for more than 20 years for probably more than 1000 clients. Never once did I get a call with a ground loop issue afterwards.
But I have gone to numerous studio sites after hearing "ground loop" complaints that were supposedly microphone-related. Never once was the proper termination of the connectors implicated (although RF- issues stemming from improper terminations could often be improved upon) but either improper wiring to the mic or other components downstream were the cause of the problem.
P.S.: Does anyone really think that Neumann, which equips all of its premium mic packages with a properly terminated 25 footer of GAC3 would choose a termination which, right out of the box, would spell trouble?
Post by: Schallfeldnebel on July 02, 2007, 05:13:34 PM
But in many cases the microphone is connected to a stage box going into some patchpanel, and then sent to the desk. It is that problems occur because of wrong earthing in those patch panels.
Erik Sikkema
Schallfeldwebel
Post by: Klaus Heyne on July 02, 2007, 05:59:44 PM
| Schallfeldwebel wrote on Mon, 02 July 2007 14:13 |
...It is that problems occur because of wrong earthing in those patch panels. |
Then I suggest to fix those secondary problems, instead of sacrificing the audio performance of multi-thousand dollar mics.
Post by: maarvold on July 03, 2007, 03:55:19 AM
This is a nice idea in theory. But, using the Eastwood Stage (aka The Warner Brothers Scoring Stage in Burbank, CA) as the example where I encountered this specific problem, I sense reluctance to change a wiring scheme that has been used to generate hundreds of thousands of dollars successfully just because a couple of people want to bring their own cables--ESPECIALLY when lifting the connector's connection at the male end solves the problem: both ground hum AND RF.
The problem, to me, implies that the mic panel is grounded and 'sitting' at a different ground potential than the audio ground and the difference 'resolves' itself through the audio portion of the line, but am not a 'big enough fish' to feel comfortable asking them to change.
Post by: Barry Hufker on July 05, 2007, 03:29:33 PM
IMNSHO, balanced cable connectors should've never had three pins to begin with. This is what caused the confusion in the first place. Equipment is supposed to respond only to the difference voltage between the hot and cold pins. The ground pin should be used only roughly to equalise the ground/chassis potentials of the interconnected equipment. Had XLR connectors had only two pins, the function of the shell would have obviously been to connect the shield to the chassis.
Unfortunately, a profusion of different wiring practices has spawned quite a few boxes whose performance degrades when current flows into pin 1. In a typical bad case, pin 1 is wired to the circuit board, allowing all current flowing through the cable shield to make its way into the internal ground of the circuit. This is why some people have resorted to the practice of leaving the shield open on the receiving end or connecting it only through a capacitor. At least this cuts out hum in pin 1 sensitive gear, although RFI is still free to come in through the front door.
A new standard, AES48, finally normalises the situation (unfortunately compliance is optional) by requiring pin 1 to be directly bonded to the chassis and the XLR shell. Two pieces of AES48 compliant equipment will always interface correctly because no amount of current into pin 1 can cause currents anywhere in the audio circuit. Ground loops are a fact of life in any sufficiently complex setup and AES48 compliance insures that they are allowed to exist without impacting audio performance.
So, if all equipment were correctly designed, the answer to your question would've been a definitive "both pin 1 and shell". Pin 1 would suffice for low frequencies, the shell would help shield the inevitable pig-tail ground connection. (For best results, Pin 1 ought to be connected to chassis ground rather than signal ground.) But lacking AES48 compliance across the board, the answer is, unfortunately, "depends".
I believe this then to be the answer the people in this thread arrived at. It would be nice to connect to the shield but for those of us who have to work with what we find, connecting to the shell isn't practical.
Barry
Post by: Klaus Heyne on July 05, 2007, 04:37:01 PM
The poster confirms my termination scheme and gives good technical explanations why.
However, I assume from his advocating to eliminate ground in a balanced line that he refers primarily to audio equipment downstream from the mic - you couldn't run phantom powering for mics that way.
Additionally, deleting shield termination at the beginning of the audio chain would bring you the "antenna" effect, nice, hissy RF.
His final comment that one may not find the ideal terminations in the field and in some unknown audio equipment it is connected to, is largely irrelevant in the case of microphone set ups, because of the fact that nothing is fed to the mic (first link in the the chain!) and except for the couple of extreme situations that were mentioned earlier, I cannot imagine any variation from the grounding/shielding scheme I advocated.
Conclusion: always start by terminating mic cables as I described earlier in this thread.
(soon to be a stickie)
Post by: Barry Hufker on July 05, 2007, 05:13:08 PM
Barry
Post by: Andy Peters on July 05, 2007, 11:27:59 PM
| Barry Hufker wrote on Sat, 23 June 2007 07:19 |
What I find about the Accusound is that it sounds maybe a bit less nasal and possibly more open. But I feel as tho' it smears sounds together. |
I don't see how this test can be truly fair. The speaking voice (if that's your test signal) will never be identical for all iterations of the test. You can't clamp the speaker's head into a test jig, and you can't ensure that the voice tone remains constant.
-a
Post by: Barry Hufker on July 06, 2007, 12:51:37 AM
For me it's not about being "fair". It's about what I heard no matter how imperfect the test. This is what I posted earlier. Everyone must judge for themselves when presented with something new. You listen to the new thing. You listen to what you have. You are unbiased to the extent you can be and then you make a decision to buy or not.
I am *not* claiming a scientific test, nor am I aiming at creating a standard for others as to how they should approach their testing and purchasing.
Testing this new cable has been a journey Klaus has been kind enough to let me chronicle here.
Barry
Post by: maarvold on July 06, 2007, 02:21:12 AM
| Barry Hufker wrote on Thu, 05 July 2007 21:51 |
...It's about what I heard no matter how imperfect the test. This is what I posted earlier. Everyone must judge for themselves when presented with something new. You listen to the new thing. You listen to what you have. You are unbiased to the extent you can be and then you make a decision to [use it] or not... |
The analogy for many self-improvement manuals is (loosely quoted) "you are like a guided missile: zig-zagging toward your goal, correcting your course with ever smaller maneuvers as you home in on the target".
This is certainly what I do: attempt to weed out (and never repeat) the negative experiences and try to build upon and/or recreate the positive ones. Isn't this why many here seem to like transformers: because they often like what they hear when a transformer is part of the signal chain?
Post by: HockeyMike on July 13, 2007, 09:50:15 PM
Klaus, sorry if this is a basic question, but how are you connecting the ground conductor and shield to both pin 1 and the solder lug? Dividing up both the shield and the ground conductor wire into two segments? The ground conductor on GAC-3 just seems so thin, being 24 AWG.
Thanks,
Michael
Post by: Klaus Heyne on July 13, 2007, 11:01:40 PM
Gotham suggests to connect both ground lug and connector shell lug by dividing the dual Reussen layer of copper strands into two- one to each lug, and the ground conductor to pin 1. That way, they all have the same ground potential in the end.
I normally don't go through that much trouble, and solder to whatever lug is bigger the fat dual-layer of copper shield, and the cable's ground conductor to pin 1 of XLR.
The important issue is to make sure that you jumper between XLR pin 1 and shield lug.
Post by: Barry Hufker on July 13, 2007, 11:55:39 PM
http://recforums.prosoundweb.com/index.php/t/18221/5149/
Barry
Post by: Andy Simpson on July 15, 2007, 02:19:33 PM
I have generic mic cables in the workshop where I test my mics. Having followed this thread with some interest I decided to go straight to the next stage, which involved putting one of my mics directly into the pre-amp i.e., stuck into the back, with no cable.
With one mic straight into the pre and the other into a gain-matched channel via my usual length of cable for orchestra (approx. 100ft). The mics were basically matched and sound identical usually - which I checked before the test.
I tested with an 8 bell tambourine - at a distance of approx. 10ft - because I always find that basic impulse response is well tested by the individual sounds of each bell and how much natural separation, life & 'dynamic freedom' (the opposite of the sound of compression) there is.
After I gain matched the output of each mic (there was a gain loss from the cable) I did an a/b comparison on loop of a small section.
I was hoping not to find a (great) difference, but I have to admit that it was a disturbingly large difference in this test.
The difference I would describe as extra resolution or clarity.
Further, no amount of eq could bring back this resolution (it reminded me a little of the muddying that tape does).
There was also a gain of ~3dB for going without cable, so noise floor also gained.
I will be making some heavy duty mic stands for my pre's before I set foot in the concert hall for my next recording session and I will run the long runs at line.
This is not a scientific test.
The pre-amp, cable length & mic in question will no doubt affect the results massively.
However, for anyone who can trust their ears, I strongly recommend this test. It is not a subtle thing.
Andy
Post by: Andy Simpson on July 15, 2007, 04:05:45 PM
I just repeated the test with snare drum and then nylon string classical guitar. Both from 5-10ft.
Snare was much clearer & more dynamic and guitar was much cleaner & more alive.
On the guitar especially, I noticed a huge improvement where the resonances of the body/strings were much more clearly audible and (please forgive my 'audiophile' style description) the sound was more engaging & realistic.
I can only imagine what this will do to an orchestra recording. I can't wait to make the a/b because I suspect that in stereo these differences will be multiplied (as has been my experience with other mic resolution issues in the past).
Again, I would urge anybody to try this test (with plenty of pre-amp headroom).
In most cases, the microphone output xlr can easily be unscrewed & pulled slightly out to allow direct connection to pre-amp inputs.
Andy
Post by: Klaus Heyne on July 16, 2007, 01:19:03 AM
Only problem: The test is not realistic, as you will need a mic cable in real world recording- usually at least 25-30ft. long. (This reminds me of speed records set by "bicycles" which are ridden 3 ft. behind a giant aerodynamic shield mounted to a muscle car- not exactly realistic!)
However, you could use the sound of the cable-less connection as a benchmark: judging the "best" microphone cable as the one with the least amount of deviation from that sound.
Post by: Andy Simpson on July 16, 2007, 04:43:24 AM
Yes, this can be used as a benchmark for cables.
For myself, I can't justify using mic cables again unless I absolutely have to. I have worked very hard to get resolution in my microphones and am happy to find such a large degree more than I expected. Maybe >5dB, if I had to quantify. Highly significant.
Perhaps with some mic designs this is not a feasible thing to do, for physical or even acoustic reasons, but for me I have tasted something much better and will absolutely be working this way at my next session (as long as I can contrive some sturdy stands!).
Andy
P.S.: in any case, since the thread has so far consisted largely of subjective opinions & debate, I would suggest that at least this test will give a repeatable and healthy result which nobody will find to be subjective.
Post by: Tomas Danko on July 16, 2007, 05:22:43 AM
| Andy Simpson wrote on Mon, 16 July 2007 09:43 | ||
While I agree that this 'test' is extreme, it isn't unrealistic. Yes, this can be used as a benchmark for cables. For myself, I can't justify using mic cables again unless I absolutely have to. I have worked very hard to get resolution in my microphones and am happy to find such a large degree more than I expected. Maybe >5dB, if I had to quantify. Highly significant. Perhaps with some mic designs this is not a feasible thing to do, for physical or even acoustic reasons, but for me I have tasted something much better and will absolutely be working this way at my next session (as long as I can contrive some sturdy stands!). Andy PS, in any case, since the thread has so far consisted largely of subjective opinions & debate, I would suggest that at least this test will give a repeatable and healthy result which nobody will find to be subjective. |
I'd be interested in comparing the two recorded tracks by phase cancellation. Hopefully it would be quite enlighting.
Post by: Klaus Heyne on July 16, 2007, 01:50:45 PM
| Andy Simpson wrote on Mon, 16 July 2007 09:43 |
I have worked very hard to get resolution in my microphones and am happy to find such a large degree more than I expected (by removing the cable and plugging the mic directly into the mic pre) Maybe >5dB, if I had to quantify. Highly significant. |
It's news to me that the term 'resolution' is quantifiable and can be attached to dB units. Please give a definition for 'resolution', and how you define a 5dB improvement.
| Quote: |
...since the thread has so far consisted largely of subjective opinions & debate, I would suggest that at least this test will give a repeatable and healthy result which nobody will find to be subjective. I'd be interested in comparing the two recorded tracks by phase cancellation. Hopefully it would be quite enlighting. |
As soon as you can define for us objective parameters of 'resolution', I will follow along.
In the meantime: Phase reversing (and thus pointing out any divergence between the cable and non-cable connections) still will need subjective interpreting in terms of musical meaning of these deviations- rather than assuming that the larger the deviation, the more useless the cable, you would need to decree what amount of deviation in what frequency band is least/most objectionable to the ear!
| Quote: |
There was also a gain of ~3dB for going without cable, so noise floor also gained |
In audio measurements, noise floor improvement and gain increase are not necessarily directly linked. Your cable-less connection may have seen a slight gain increase (though I very much doubt that it was anything more than maybe 0.5 dB), but any noise floor improvement would need to be separately measured.
Post by: ioaudio on July 16, 2007, 02:16:34 PM
Post by: Andy Simpson on July 16, 2007, 03:46:14 PM
| Klaus Heyne wrote on Mon, 16 July 2007 18:50 |
It's news to me that the term 'resolution' is quantifiable and can be attached to dB units. Please give a definition for 'resolution', and how you define a 5dB improvement. As soon as you can define for us objective parameters of 'resolution', I will follow along. |
The definition of gain in resolution which I use is this:
the equivalent increase in perceived detail level from the equivalent gain in volume.
For example, by turning up the monitors by 5db I have a perception of increased detail.
This perception of increased detail for a 5dB monitor gain increment is roughly the same as I experienced in the test.
Subjective it is but I would guess that if we were all tested and asked to guage (in dB) increase of detail against increase in monitor gain, we would likely have similar results. So perhaps we can agree on some pseudo measurement which we will all be able to relate to in some useful way?
| Quote: |
In the meantime: Phase reversing (and thus pointing out any divergence between the cable and non-cable connections) still will need subjective interpreting in terms of musical meaning of these deviations- rather than assuming that the larger the deviation, the more useless the cable, you would need to decree what amount of deviation in what frequency band is least/most objectionable to the ear! |
Alas, if we could make such measurements in a controlled way we could at least measure differences, which would render the subjective aspects as academic.
In any case, I would not expect to gain anything from phase cancelling as it would be impossible to get an actual cancellation. Impulse response tests would be more benefitial perhaps, but like I said, for anyone who can trust their ears, it is not a hard test.
| Quote: |
In audio measurements, noise floor improvement and gain increase are not necessarily directly linked. Your cable-less connection may have seen a slight gain increase (though I very much doubt that it was anything more than maybe 0.5 dB), but any noise floor improvement would need to be separately measured. |
Well, the noise floor issue comes easily as matching the channels audibly reduced the noise-floor. I was listening at varying levels up to very loud to check this.
Regarding the gain issue, my results were repeated many many times without alteration. A significant increase in gain but far more significant, a bigger gain in resolution.
Andy
Post by: johnR on July 16, 2007, 04:08:40 PM
Post by: Andy Simpson on July 17, 2007, 04:05:15 AM
Also, I guess that some ribbon designs would be good for this test since reflections could be controlled with polar pattern - and that kind of mic would probably benefit greatly too.
As I mentionned before, the test is not so subtle and this is why I would urge anybody else to try it. Maybe some acoustic treatment could remove any doubt.
Andy
Post by: Andy Simpson on July 17, 2007, 04:13:20 AM
| ioaudio wrote on Mon, 16 July 2007 19:16 | ||
what microphone was used in that test - phantom powered? some manufacturers were/are buidling line-level microphones to overcome this problem, but many sound engineers feel the need of choosing various preamps for different recording situations - which is not possible with the preamp build in the mic (of course you could still route the line signal back to a mic preamp of your choice) -max |
The mic in question is not phantom powered.
The companies making line-level mics are presumably making compromises & decisions in their designs which not many people are happy with.
Andy
Post by: Tomas Danko on July 17, 2007, 08:04:36 AM
| Klaus Heyne wrote on Mon, 16 July 2007 18:50 |
In the meantime: Phase reversing (and thus pointing out any divergence between the cable and non-cable connections) still will need subjective interpreting in terms of musical meaning of these deviations- rather than assuming that the larger the deviation, the more useless the cable, you would need to decree what amount of deviation in what frequency band is least/most objectionable to the ear! |
In the end I do agree that the subjective interpretation is king. However, I also feel it is a good starting-point doing the phase reverse test before anything subjective happens.
Because if it nulls completely, you'll have a hard time explaining any subjectivity to me. If it doesn't null completely, either the rest of the signal will tell you something of value or just telling you there is a difference so now lets start the emotional evaluation.
Anyone having done the entire DAWsum test blindfolded while taking notes would probably agree it's a sane first-step.
Post by: Andy Simpson on July 17, 2007, 08:47:46 AM
Cable vs cable - well, that could easily be subjective or placebo but this isn't so subtle. And that is the reason I am suggesting that people try it.
Andy
Post by: Jim Williams on July 17, 2007, 11:42:51 AM
Microphones and cables have resolution beyond that. Just because a digital signal phase cancels it doesn't mean the mic or cable is represented, only the converter resolution is.
Post by: johnR on July 17, 2007, 04:39:38 PM
Post by: Barry Hufker on July 17, 2007, 05:03:51 PM
Barry
Post by: Markus Sauschlager on July 17, 2007, 06:28:12 PM
Post by: Jim Williams on July 18, 2007, 10:58:27 AM
| johnR wrote on Tue, 17 July 2007 13:39 |
Another factor here is the interaction between microphone and cable. For example a tube mic with no negative feedback will have a high output impedance and will be sensitive to changes in cable capacitance, but a solid state mic with an op amp in the output stage will have a very low output impedance and will be relatively unaffected. Comparison of cables requires a mic that is (as far as possible) unaffected by the changes in cable characteristics. |
Tube mics use an output transformer to lower output impedance. It is no different than a transistor mic with an output transformer. Linearity can be affected by the primary transformer load on the tube stage, but on the secondary side it's no different than the transistor mics that use an output transformer.
Post by: Andy Simpson on July 18, 2007, 12:22:33 PM
You can only try and see how the cable (or lack of) affects the mics you use.
It is that simple from a practical point of view.
After that, on a mic by mic basis, we can discuss the possible reasons.
Isn't anybody interested enough to try this with a few of the well known standard mics?
I would post some examples but it would be silly to show an example from such an unfamiliar source mic.
People would find it impossible to extrapolate whether there would be any difference with the mics they use.
Also, it could easily be a function of my mics and maybe it is not so significant with normal mics.
Andy
Post by: Klaus Heyne on July 18, 2007, 01:19:07 PM
| Andy Simpson wrote on Wed, 18 July 2007 09:22 |
Isn't anybody interested enough to try this with a few of the well known standard mics? |
It's already been said here: your idea is impractical, because few would want (or in a practical manner could) mount their favorite mic pre amp on the end of a mic stand.
So, especially with tube mics, which need a power supply- yet another box you would need to mount on the mic stand- your test is not useful for the rest of us who still need to connect our recording mics through cables.
I'd suggest to move on to discuss other, more practical, ways of testing mic cables' sounds, and finding a way to optimize these sounds.
Best regards,
Post by: johnR on July 18, 2007, 06:02:05 PM
| Jim Williams wrote on Wed, 18 July 2007 15:58 | ||
Tube mics use an output transformer to lower output impedance. It is no different than a transistor mic with an output transformer. Linearity can be affected by the primary transformer load on the tube stage, but on the secondary side it's no different than the transistor mics that use an output transformer. |
I was actually referring to the difference that negative feedback makes to output impedance. A mic like the TLM 103, which has an op amp in it (and hence a large amount of negative feedback), is likely to have a lower output impedance than one with little or no negative feedback, regardless of whether it's tube or solid state. I was assuming this would make the op-amp mic less susceptible to cable changes (personally I don't like the idea of an op amp in a mic; I was just making the point).
Post by: Andy Simpson on July 19, 2007, 05:25:31 AM
| Klaus Heyne wrote on Wed, 18 July 2007 18:19 | ||
It's already been said here: your idea is impractical, because few would want (or in a practical manner could) mount their favorite mic pre amp on the end of a mic stand. So, especially with tube mics, which need a power supply- yet another box you would need to mount on the mic stand- your test is not useful for the rest of us who still need to connect our recording mics through cables. I'd suggest to move on to discuss other, more practical, ways of testing mic cables' sounds, and finding a way to optimize these sounds. Best regards, |
I was only hoping that people might _try_ it.
5 minutes spent in the privacy of studio/home with a tambourine or bunch of keys or accoustic guitar should show immediate results.
We might all benefit from hearing the full potential of our favoured signal chains, in order to judge the compromise of cable choice with a reasonable reference point.
Andy
Post by: Barry Hufker on July 19, 2007, 10:24:07 AM
I think it's worth a listen. It's not hard to do and could be revelatory.
I'm very busy at the moment but will certainly try it if only "just to see" -- I mean "hear".
Barry
Post by: Markus Aalto on July 20, 2007, 07:35:30 AM
This is one very interesting high end cable:
http://www.sommercable.com/2__produkte/2__1_meterware/2__200 _0281.html
It's about one year old cable type with very low capacitance. I have only one 5 meter sample piece yet but found it interesting.
Galileo is interesting less expensive "high end" cable too for studio use. I like this cable and use it in the studio. It's also mechanically strong. Quite thick but flexible.
http://www.sommercable.com/2__produkte/2__1_meterware/2__200 _0271.html
Sommer cable is not so old and traditional manufacturer like some others but they have very innovative attitude and interesting products. They make also nice low capacitance and star quad multipair cables.
About Gotham cables:
Because i use cables mostly in live use the sound is not only aspect for selecting the cable. I've used Gotham in the past but found that they weren't long lasting in live use. The jacket is too soft and thin and breaks easily. It's the biggest problem. I don't understand why they doesn't make them stronger. Maybe in studio use it is not a problem. Gotham makes some special type with strong PUR jacket but haven't had possiblity to check it yet. (I've used Gac-2, Gac-1 and Gac-7 in the studio which is very good tube mic cable).
Anyway i try to keep more expensive and fine cables in the studio and use more durable brands and types on stage.... In live use i like very much Belden 1813A. It is about similar diameter cable than Gotham (quite thin. The jacket material is really strong and durable but still very flexible. I've bought 6 rolls yet (100m /328ft) in couple of years... It stays flexible in cold winter conditions too... I don't know why Belden seems to be "out of fashion".
Not much audible opinions in this post but some experiences anyway.
Markus A, Finland
Post by: Barry Hufker on July 20, 2007, 10:49:54 AM
Sommer Cable appears interesting. I would request a sample but don't want an unending supply of catalogs showing up at my door.
Barry
Post by: mdemeyer on July 21, 2007, 02:25:22 PM
ELECTRICAL DATA:
Capac. cond./cond. per 1 m 46 pF
Capac. cond./shield. per 1 m 90 pF
Compare this to Mogami 2549:
K1 (cond to cond): 11pF/m(3.4 pF/Ft)
K0 (cond to shield): 76pF/m(23 pF/Ft)
Not disputing the anything about your experience with the sound, but to claim very low capacitance is not correct.
Michael
Post by: Markus Aalto on July 21, 2007, 03:38:29 PM
| mdemeyer wrote on Sat, 21 July 2007 21:25 |
Not sure I understand the comment about the capacitance of the Sommer cable being low..... |
Ok. Not so sensational numbers but maybe lower than average i've seen. Must say i haven't read every manufacturers datasheets very carefully.
It seems in Finland is currently no importer for Mogami cables. I don't have experience about them. I've been interested about 2549 but haven't had possiblity to get the cable yet. Importer of Neutrik connectors sells here Cordial cables so it's the reason why almost everbody uses them..... It is very usual cable brand in Europe. Basic cables.. nothing special.Markus
Post by: Klaus Heyne on July 21, 2007, 09:06:03 PM
Please continue to share your personal experiences with cables, rather than give a listing of all the cables you have heard about being available and having this or that spec (unless you can correlate the specs to the sound!)
I will be merciless from now on, if that rule is violated- out of respect for the level of discussion that's been the norm and that I'd like to have continue here.
Much obliged,
Post by: Tomas Danko on July 22, 2007, 02:48:02 PM
| Jim Williams wrote on Tue, 17 July 2007 16:42 |
Phase cancellation of two signals is limited by the resolution of the A/D converters which in pro audio is around -100 ~-105 dbu. Some are a bit better but are not commonly used. Microphones and cables have resolution beyond that. Just because a digital signal phase cancels it doesn't mean the mic or cable is represented, only the converter resolution is. |
I'd suggest analog phase cancellation through polarity reversal and summing, then.
At the end of the day however, if the difference won't make it through the A-D converter of choice when I record my music I'm not sure it is of interest to begin with.
Post by: mrdave1981 on July 22, 2007, 10:15:48 PM
For anyone interested in trying out the cable talked about in this discussion. Gotham Audio GAC-3 and GAC-4/1 are currently both available on eBay.
I purchase directly from the US Distributor of Gotham Audio, LLC.
All the best.
DAVE
Post by: Barry Hufker on July 22, 2007, 11:36:50 PM
Barry
Post by: djosephson on July 27, 2007, 05:42:01 PM
There could be a whole separate discussion on cleaning up microphone audio from the influences of local electromagnetic fields. I would suggest reading the AES papers Jim Brown and I wrote on RF susceptibility of condenser mics, which are available on his website www.audiosystemsgroup.com
The RF issue out of the way for the moment, the other discussion is about the interactions between the cable and the devices on either end. Transformerless mics tend to be the least influenced, transformer mics with buildout resistors a little more so, those without buildout resistors the most sensitive. Some mic preamps are well behaved, others have negative feedback applied to the input connectors and behave differently with different mics and cables attached.
Post by: Markus Aalto on October 05, 2007, 02:56:57 PM
| Markus Aalto wrote on Fri, 20 July 2007 14:35 |
About Gotham cables: Because i use cables mostly in live use the sound is not only aspect for selecting the cable. I've used Gotham in the past but found that they weren't long lasting in live use. The jacket is too soft and thin and breaks easily. It's the biggest problem. I don't understand why they doesn't make them stronger. Maybe in studio use it is not a problem. Gotham makes some special type with strong PUR jacket but haven't had possiblity to check it yet. (I've used Gac-2, Gac-1 and Gac-7 in the studio which is very good tube mic cable). |
I know this is slightly off topic but need to update my previous comment regarding Gotham GAC-2
Last time i've bought this cable was many years ago. Yesterday i got a new lot of grey GAC-2 cable and the jacket seems to be different than before. Thicker and different material than what i've had previously. I like this better. I don't know if all the different coloured GAC-2 versions uses the same materials.
Post by: Rich Mays on October 06, 2007, 01:41:21 AM
| Kees de Visser wrote on Wed, 27 June 2007 18:30 |
The Yamaha might be more stable. Any other ideas for a reproduceable source ? |
A readily found source is a pipe organ with MIDI performance recorder, often found on non-mechanical action instruments. This gives you an analog source with absolutely consistent performances. Unless testing would take more than several hours, you can reasonable expect tuning to be similarly consistent.
Whenever I need to run comparison tests I go to a nearby church with an excellent large instrument. The only missing ingredient is percussive transients. Too bad this is a church and not a theater!
Many of us have organist friends/clients-- ask them if they might record some varied pieces for such use. It also gives you the opportunity to get nostalgic, as these MIDI beasts use floppies!
Rich