R/E/P Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: 1 2 [3] 4   Go Down

Author Topic: What's with all the fuss about M/S?  (Read 16765 times)

Andrew Hamilton

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 573
Re: What's with all the fuss about M/S?
« Reply #30 on: May 01, 2010, 03:40:17 am »

Studer list guys have told me that voltage matching _always_ works best for music signaling and precludes the need for tranny termination.   Ampex list guys have said always to terminate the tape machines with what they would have seen in a power matching setup (e.g., DC Ω (aka 600 Ohms)), regardless of load.

The A80 R manual has a lot of specificity about using 600 Ohm termination for NAB, and _200_Ohms_ for CCIR.  I have heard that the machine has a 0 Ohms output impedance.   But there is no mention I can see about using the machine outside of a power matching-esque context, actually, in spite of contemporary European studio practices circa 1970...


If you for some sinister purpose wanted to insert the ATR overbridge in the output path of the machine, and you were (balanced) signaling to a device that had a 100k input Z, should you still flip the 600 Ohm switch on the back of the ATR for best handshake?  Try it both ways and pick the winner?  Or forget the switch for modern studio signaling?  

Interestingly?, also, the ATR has no 200 Ohm switch, even though it _does_ do CCIR.




 




Thanks,
    Andrew
Logged
www.serifsound.com
premastering for CD and DVD-A.  Featuring FTP load in and delivery as well as analog tape transfers.

johnR

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 923
Re: What's with all the fuss about M/S?
« Reply #31 on: May 01, 2010, 09:18:22 am »

Andrew Hamilton wrote on Sat, 01 May 2010 08:40

Studer list guys have told me that voltage matching _always_ works best for music signaling and precludes the need for tranny termination.   Ampex list guys have said always to terminate the tape machines with what they would have seen in a power matching setup (e.g., DC Ω (aka 600 Ohms)), regardless of load.

The A80 R manual has a lot of specificity about using 600 Ohm termination for NAB, and _200_Ohms_ for CCIR.  I have heard that the machine has a 0 Ohms output impedance.   But there is no mention I can see about using the machine outside of a power matching-esque context, actually, in spite of contemporary European studio practices circa 1970...


If you for some sinister purpose wanted to insert the ATR overbridge in the output path of the machine, and you were (balanced) signaling to a device that had a 100k input Z, should you still flip the 600 Ohm switch on the back of the ATR for best handshake?  Try it both ways and pick the winner?  Or forget the switch for modern studio signaling?  

Interestingly?, also, the ATR has no 200 Ohm switch, even though it _does_ do CCIR.


Transformer termination depends on the individual transformer design and there is no universal rule. Many high quality transformers have a strong high frequency resonance which needs to be damped by a suitable termination. Even if the resonance is well above the audible range it can still result in audible intermodulation products.

If in doubt, one way to test this is to run a 10kHz square wave through the (in circuit) transformer, look at the resulting waveform on an oscilloscope, and adjust the termination for the cleanest square wave. There is often some compromise necessary to accommodate a range of load impedances, cable capacitances etc. You might find that a transformer needs no additional termination in a particular use case, but leaving this to luck isn't a good way to guarantee optimum performance. (I've ignored the loading effects of termination on the circuitry driving the transformer, which is another issue).
Logged

Jerry Tubb

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2761
Re: What's with all the fuss about M/S?
« Reply #32 on: May 01, 2010, 12:10:09 pm »

compasspnt wrote on Tue, 27 April 2010 19:19

I might use M/S 1% of the time.



About the same here. Use M/S very sparingly.

This last week I used M/S on one song.

Pulled the Sides back a 1/2 dB to help a centered vocal pop out front more. That's it.
Logged
Terra Nova Mastering
Celebrating 20 years of Mastering!

dcollins

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2815
Re: What's with all the fuss about M/S?
« Reply #33 on: May 01, 2010, 02:48:56 pm »

Andrew Hamilton wrote on Sat, 01 May 2010 00:40


The A80 R manual has a lot of specificity about using 600 Ohm termination for NAB, and _200_Ohms_ for CCIR.  I have heard that the machine has a 0 Ohms output impedance.   But there is no mention I can see about using the machine outside of a power matching-esque context, actually, in spite of contemporary European studio practices circa 1970...



All Transformers have a termination impedance that gives the lowest overshoot/flattest response.  This may or may not be the value that sounds the best.

I have no idea why the record curve would have anything to do with the transformer termination in the Studer, it doesn't make sense to me.


DC

Andrew Hamilton

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 573
Overthinking 102... Was: Re: What's with all the fuss about M/S?
« Reply #34 on: May 01, 2010, 11:23:03 pm »

I should have thought that the optimum sound quality would only truly be possible if the preferred sound was in agreement with a flat transfer, and with no aberrant behavior at the trafo (i.e., no overshoot...).  That one might prefer the mojo version to a dull, if pure, version is a familiar claim.  But the purer version which is also exciting would be superior to either of the other two.

You may recall that the A80 R meter bridge has termination switches on the front (similar to the ones on the rear of the ATR I/O module, except the Ampex ones are slide switches and the Studer's are flip switches†), and these seem to be 600 Ohms only, just like the ATR's...  

The Studer manual cites the maximum output with a 600 Ohm load.  Possibly this is just for reference, due to the historical place 600 Ohms has in studios, and possibly it may be in fact a specification that this load _must_ be used in order to achieve the maximum possible output for the 30 Ohms output, _without overshoot or re-eq.  

Again, it does say that the load seen by the A80 has to change if the eq standard changes.  

But through e-conversations with some heady techs (ex-CBS), and, evidently, DC seems to agree, I have gathered that the A80's compensation networks should already be buffered from load effects external to the machine's output connector by design.  So, why, indeed, would the manual ever say to feed a multimeter the output with a 200 ohm resistor in parallel, if measuring the CCIR performance - or a 600 Ohm one, if NAB?  How does it know?  If it's just a matter of level, can't that be adjusted adequately at the trimpots?

The ATR man does not mention a need for changing the load based on equalization.  But I believe that it _does_ indicate that when _not_ driving devices which are already 600 Ohms, to flip the switch, thereby rendering the net load, close enough to the 600 target (Rpt =

                   1
             _____________
             1/Rp1 + 1/Rp2).  


At least that's in the section for calibration of the I/O module.  

What to do with the ATR overbridge in Helsinki?  Besides throwing it in the Gulf.  (;   CCIR works better with modern formulae, according to Jay McKnight, btw.

Please vote:

1) Don't terminate the ATR (or A80, be specific, please, as they do, indeed, use different trafos - Thanks, JohnR) if voltage matching, and forget it?;

2) epoxy the termination switch in the On position?; or,

3) make two pairs of short, extension cables (labeled NAB and CCIR, respectively) which have the target loads soldered across the pins... always using them with the matching standard, unless the given load is already at the target Z?



Andrew      






† Does anybody know a good source for the little plastic sleeves that go over metal flip switch throws?   Such as on an electric guitar?  I'm looking for black, but I might be persuadable, if the color is good...   Mood sleeves?


* (are there any? ...like a working museum studio?   I know of one that is in a incubation phase right now, locally, but it hasn't reopened, yet)
Logged
www.serifsound.com
premastering for CD and DVD-A.  Featuring FTP load in and delivery as well as analog tape transfers.

johnR

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 923
Re: Overthinking 102... Was: Re: What's with all the fuss about M/S?
« Reply #35 on: May 02, 2010, 06:54:46 am »

Andrew Hamilton wrote on Sun, 02 May 2010 04:23

I should have thought that the optimum sound quality would only truly be possible if the preferred sound was in agreement with a flat transfer, and with no aberrant behavior at the trafo (i.e., no overshoot...).


Unfortunately the only way to eliminate overshoot and have a flat, extended high frequency response is to avoid transformers altogether.

Choosing the correct termination is a trade-off between ringing/overshoot on one hand and reduced bandwidth on the other. Typically the best that can be achieved is a slight overshoot with a minimal amount of ringing.

Sometimes additional termination isn't necessary because the input impedance of the following device is sufficient to damp the HF resonance, but there's still the same trade-off.
Logged

Andrew Hamilton

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 573
Re: Overthinking 102... Was: Re: What's with all the fuss about M/S?
« Reply #36 on: May 02, 2010, 11:30:45 pm »

johnR wrote on Sun, 02 May 2010 06:54

Andrew Hamilton wrote on Sun, 02 May 2010 04:23

I should have thought that the optimum sound quality would only truly be possible if the preferred sound was in agreement with a flat transfer, and with no aberrant behavior at the trafo (i.e., no overshoot...).


Unfortunately the only way to eliminate overshoot and have a flat, extended high frequency response is to avoid transformers altogether.


So, Burgess was right not to use trafos, then.   And I recently saw an old Sontec MEP on eBay that was allegedly retro-fitted, by Inward Connections, with their own trafos.  

I too should like to be able to demonstrate to the client that when the measured distortions are at their lowest is also when the same settings produce the preferred sound.  How better to enjoy every little crevice of the funkily inharmonic strains of Hendrix's upside down guitar, than through a clean channel that doesn't, itself, distort, jah?


MDI have a comment on the page about the I/O modules they sell for ATRs that trafos are euphonically advantageous on mic pres, but not really beneficial at the line inputs of tape machines, or something to this effect.   Some will say that it's the program, stupid.  But if the program relies on the mojo of the machine it was recorded on, can a different trafo's mojo really net improve on that in a way that doesn't simultaneously step on the microtoes of any subfeatures of the sound?  I think yes and no.  No, it may not be better than the original machine's mojo, but it might still be better with _some_ iron in the chain if it sounds anemic or naked without it...  I guess it's a mastering call, at that point, rather than a strictly technical one.    Shocked
 

johnR wrote on Sun, 02 May 2010 06:54


Choosing the correct termination is a trade-off between ringing/overshoot on one hand and reduced bandwidth on the other. Typically the best that can be achieved is a slight overshoot with a minimal amount of ringing.



And on the gripping hand...?

johnR wrote on Sun, 02 May 2010 06:54


Sometimes additional termination isn't necessary because the input impedance of the following device is sufficient to damp the HF resonance, but there's still the same trade-off.


I believe I can read this as a vote for using the terminators in most voltage matching scenaria but _not_ using the terminators for feeding, say, the Variable Mu (which already looks like 600 Z), because the load would drop with each extra passively  paralleled 600.   This is how I've been doing it.   And I have firstly been auditioning the ATR without the overbridge when deciding on a transfer path.   When this way sounds best, there is no lingering doubt.  When the other way seems to sound better, I assure you that I at least start unpacking the Ouija board, smudge the room with Sage, and perform several basic asanas before re-evaluating the decision to proceed.   I reckon this is as cautious as it ever really needs to be (especially in light of the client-involved proofing rituals).   Or, would you like to say, if it needs any more color, rather than just correction, it should be remixed!



Thanks,
    Andrew
Logged
www.serifsound.com
premastering for CD and DVD-A.  Featuring FTP load in and delivery as well as analog tape transfers.

dcollins

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2815
Re: Overthinking 102... Was: Re: What's with all the fuss about M/S?
« Reply #37 on: May 03, 2010, 12:51:23 am »

johnR wrote on Sun, 02 May 2010 03:54


Unfortunately the only way to eliminate overshoot and have a flat, extended high frequency response is to avoid transformers altogether.



It's possible to have ruler-flat response and vanishingly low distortion with transformers it's just unusual to see it done properly......



DC

Gold

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1453
Re: Overthinking 102... Was: Re: What's with all the fuss about M/S?
« Reply #38 on: May 03, 2010, 10:38:22 am »

Andrew Hamilton wrote on Sun, 02 May 2010 23:30


So, Burgess was right not to use trafos, then.  


Since they were mainly used for disk cutting, having low end phase shift and/or distortion is a bad thing. It will cause more vertical excursion.
Logged
Paul Gold
www.saltmastering.com

On the silk road, looking for uranium.

zmix

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2828
Re: Overthinking 102... Was: Re: What's with all the fuss about M/S?
« Reply #39 on: May 03, 2010, 11:57:05 am »

Gold wrote on Mon, 03 May 2010 10:38

Andrew Hamilton wrote on Sun, 02 May 2010 23:30


So, Burgess was right not to use trafos, then.  


Since they were mainly used for disk cutting, having low end phase shift and/or distortion is a bad thing. It will cause more vertical excursion.




I call getting out of bed in the morning a vertical excursion...

Gold

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1453
Re: Overthinking 102... Was: Re: What's with all the fuss about M/S?
« Reply #40 on: May 03, 2010, 04:59:33 pm »

zmix wrote on Mon, 03 May 2010 11:57


I call getting out of bed in the morning a vertical excursion...



A single vertical excursion is fine. A double can be embarrassing.
Logged
Paul Gold
www.saltmastering.com

On the silk road, looking for uranium.

bblackwood

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 7036
Re: Overthinking 102... Was: Re: What's with all the fuss about M/S?
« Reply #41 on: May 03, 2010, 05:25:28 pm »

I dunno, I probably use M/S about 99% of the time.

Red cars, blue cars, I s'pose...
Logged
Brad Blackwood
euphonic masters

Bob Boyd

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1133
Re: What's with all the fuss about M/S?
« Reply #42 on: May 03, 2010, 05:39:43 pm »

I use M/S quite often here.

My Weiss DS1-MkIII lives in M/S and whenever I reach for my Massive Passive, 99.9% of the time, I dial it in as an M/S EQ via the Dangerous Master.  A killer combo- especially  for mid definition.
Logged
Bob Boyd
ambientdigital, Houston

http://ambientdigital.com
http://myspace.com/ambientdigital

Twitter: @bobboyd


Look, I know it's mean.  But sometimes the end justifies the mean.

masterhse

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1059
Re: What's with all the fuss about M/S?
« Reply #43 on: May 03, 2010, 10:18:43 pm »

I guess that I should be used to it by now, but sometimes I'm amazed at the polarity of opinion among mastering engineers but the quality of work done by all concerned.

Yellow cars anyone?
Logged
Tom Volpicelli
The Mastering House Inc.
CD Mastering and Media Production Services

compasspnt

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 16266
Re: What's with all the fuss about M/S?
« Reply #44 on: May 04, 2010, 08:41:07 am »

White.



Or silver.
Logged
Pages: 1 2 [3] 4   Go Up