[Brian Cheney]

Wednesday Jan 7, 9am

We arrived Tuesday afternoon at the Zeus Ballroom of the Alexis Park to find a rather isolated, largish and very echoey room with our rental piano (thanks to Southern Nevada Music, who lend us a Yamaha 7'6" C7 grand for only $1100 the week) standing in exactly the wrong place.  Repeated requests to the hotel for help go unheeded, so we all simply push the unwieldy instrument across the room into place, no easy task.  Our van full of gear pulls in after a 10 hr ride and we spend hours unloading.  A five o'clock rehearsal with percussion, piano, and Lesley Olsher gets pushed back to 9pm and doesn't end until 10:30.  Having enjoyed little sleep the past week I am in a fine mood and looking forward to a grueling Wednesday of setup.

So here we are on Wednesday morning with rehearsals starting at 1pm.  To my great relief a cadre of volunteers (mostly owners who have made the journey to LV on their own dime to help out) set to work furiously unpacking and hooking things up.  Some eight people scurry like angry mice and, since we're biamping and have four VSS subs, much muscle and effort is required.  We attach Sonex to plywood panels and screw footers onto the panels so they stand free and can be moved around.  We're not close to ready when Ray Kimber's crew of five shows up with the recording gear.  Now we have 13 people in a small space trying to do many disparate tasks simultaneously.  Jeanine del Carlo, a stunning 25 year old blonde (daughter of RM40 owner and Metropolitan Opera star John del Carlo) arrives for her rehearsal.  We tell her to come back at 5pm.  There is, however, no putting off the 2pm rehearsal with bassist Bill Bailey, clarinet/sax/flute player Fred Haller, pianist Christopher Salocks, Lesley Olsher on vocals and me at the percussion table with suspended cymbals and brushes.  There's problems with missing and incomplete sheet music and each number must be gone over bar by bar, with changes and notations made nonstop.  All the while we attempt to finish speaker setup, room setup, and recording gear setup.  Ray's crew has two sets of mics (an ORTF Schoeps pair, plus a Sennheiser 800 pair in XY) and must decide what to do based on headphone monitoring, since the speakers are still silent.  At one point 18 people are all doing their thing, trying to get our project off the ground.

By the time we're finished 5 days later nearly 25 individuals participate, all unpaid except the two Union musicians Bill and Fred who each command $1000.  I am dripping gratitude but there's no time for it now.

Ray's guys leave at 5pm to go to their other setup at the Venetian, where they will toil till early morning.  Jeanine arrives for her postponed rehearsal to no recording and no playback.  She gives us an a capella vocal which blows everyone away.  What a talent!  What a voice!  This woman is going to be a star!

Jeanine leaves and it's up to me to fire up the speakers, but I'm exhausted and intimidated.  The sound of Fred's alto from 2ft away has convinced me we'll never bring this off.  Gear simply can't reproduce such sound, to say nothing of Jeanine's overpowering voice.  I make the fatal bad decision to stop and restart early Thursday morning on speaker sound checks, even though the booth opens at noon for performance and playback.

This fatigue-induced miscalculation would nearly ruin our show, and it is entirely my fault.

Part II:
We suck.

[cliffy]

More please, and soon

The suspense is killing me.

I can't believe there has not been more postings in the media about this huge undertaking.

Keep 'em coming.

[Brian Cheney]

Thanks, there will be more tomorrow.

[Brian Cheney]

Thursday Jan 8, 9:15am, first day of show

I arrive early and find two stalwarts already waiting for me: John Casler and Dave Housteau, both of whom would prove to be strong right arms our entire time here.
I'm anxious to hear the playback of yesterday's three hour rehearsal, which should allow us to apply the 7 bands of parametric EQ afforded by the SDE system to the V60's.  Without this flexibility I would never have attempted live-vs-recorded in an unfamiliar venue with sources unheard before. 

We power up the system and hear only loud hum.  It takes nearly an hour of precious setup time to track it down and fix.  Fifteen minutes later on playback, we experience a "loss of lock" on the computer and a long transient reminicent of 50 caliber machine gun fire runs through the system.  The damage is considerable.  In all we lose two preamps, two power amps, and two mid panels to various mishaps.  Some problems do not manifest themselves immediately and we are forced to interrupt playback several times to perform repairs.

We have now just an hour left before the noon piano recital starts.  I make some basic  assumptions some of which will later prove spectacularly wrong: #1 is stacking the VSS subs in parallel and in phase with each other.  In spite of midroom placement, they excite powerful room modes we will struggle with for days.  We go with a "narrow" dispersion pattern from the V60's (waveguides off) to minimize room reflections and, at the suggestion of our recordists, place two 4x8' Sonex panels at a 45 degree angle at the back of the room to control echo and provide dispersion for the audience sitting directly in front of them.  Ray's engineers go with the Sennheiser pair in XY with 90 degree coverage, also to minimize the room. I play back Lesley Olsher's rehearsal, make some adjustments, and quickly discover I must change every setting I had used in my home LEDE environment, including crossover frequencies, overlap, EQ points  and slopes.  It's frantic work and we must stop when the pianist arrives.

Playback of the piano is off, but encouraging.  The 1 o'clock set with Jeanine del Carlo sounds OK, but her chest voice is too prominent and her prerecorded background music  anemic.  Still, we are floored by her presence and style.  With Lesley Olsher's high lyric soprano and a full house of listeners problems become very apparent: voice is veiled, cloudy, and rolled off, sax is too loud and shrill, string bass weak and thuddy, piano colored.  Brushed cymbals sound fairly good, if way too far back in the mix.  Overall playback does not  sound like our live sound very much at all, and things appear to get worse, then better, intermittently.  The 5pm instrumental is unsatisfactory on playback, with a real lack of transparency and considerable coloration.  A general lowbass prominence and thickness are particularly puzzling, since we should not be exciting the 30Hz area with our music but we are.  All in all, playback quality is dismal and I will be up most of the night considering remedies.  I also must confront the possibility we just aren't good  enough to bring this off.

Part III
Major surgery

[Brian Cheney]

Friday Jan 9, second day of show, 9 am:

A night of intense review has given me some insight into our meltdown.

When faced with problems in all ranges of audio reproduction it's best to start at the top.  The dull trebles we heard on playback yesterday were a consequence of setting up using rehearsal playback recorded without audience.  Human bodies absorb high frequencies.  The speakers were set to "flat" (no waveguides, treble controls at max) but the large space was soaking up highs, and the audience was making them even duller.  So I start by sweetening the trebles.  I apply 2.9dB of boost  at 5.08kHz with a Q=2.4, which gives a gentle boost from pole frequency out to 20 kHz.  This immediately cures the treble dullness.

 I must point out now that EQ has been compared to "pushing on jello".  Boost or cut in one area affects spectral energy levels elsewhere.  In this case, we are strengthening the overtones of just about every frequency below 5 kHz.  I must now be cautious about further boost in lower ranges. 

The next and most obvious problem is heavy, soggy bass.  This bloat appears to be centered around 33Hz, the floor-to-ceiling room mode of the Zeus Ballroom.  However, there is a lower component as well.  Careful listening in the empty room with no signal reveals a near constant air conditioner hum at around 30Hz.

Now, when recording and playing back in the same enviornment you must deal with the Problem of the Two Acoustics (discussed in a previous thread).  I did not expect to have a 30Hz (2nd octave) bass problem since our music was not exciting it.  The air conditioner hum did excite this mode, however, and recording/playback was doubling the effect.  Indeed all room characteristics including reverb time, late arrivals, primary resonant modes and slap echo were all doubled.  So I introduce two remedies: turning the top woofer in each stack of two around for bipole operation (which would alter room bass loading) and applying bass EQ in the 30 to 33Hz range with a Q =.25 which would cover both the vertical mode and the air conditioner.  There was nothing I could do about airliner takeoffs we experienced every half hour or so, but at least these were brief.  Playing back the previous day's recordings (with audience) and using my two main helpers John and Dave at the bass amp level and EQ controls, I repeatedly go through the settings until we settle on an optimum 30Hz cut (-6dB) and woofer output level (lower than before).  This cleans up the bass considerably.

There is still some congestion in the midrange, and some loss of transparency vs. live sound.  But things overall are much better.  Imaging is spectactular: each instrument precisely localized in space, the proper distance from the mic, and completely unwavering.  Timbral accuracy is very good, particularly on piano.  CD playback of familiar material is just fine.  It is only on live sound that I can detect any problems.

As it turned out during the course of the day, I was optimistic about having solved all problems.  But now people were commenting on the excellent sound quality of playback, and our live sound is well balanced if rather reverberant.

The DSD recording medium is truly amazing.  There is so much headroom that, with full ensemble singing and playing loudly and huge cymbal crashes going off, we still have 18 dB of headroom below digital zero. This would not be the case on redbook or even 24/96.  The audience sing-along in "Wrong Note Rag" plays back  extremely clear and natural.  The perspective on the audience is interesting: on playback its sound emanates from behind the speakers, the same distance the audience is located behind the microphones.  This is remarkable for cardioids in XY and attests to Ray's expert setup.

By day's end I know there is more to do tomorrow, but I go home early and get some rest.  I am absolutely wiped.

Part IV:
Full dipole, anyone?

[Brian Cheney]

Saturday Jan 10, third day of show, 9:30 am

During the course of yesterday's playback I found myself unhappy with bass reproduction, particularly of Bill Bailey's string bass.  On playback his instrument sounds somewhat fat and slow, and it's too far back in the mix. I move his music stand closer to the mic, then decide to go with a setup first suggested by John Casler and also used by Earl Geddes and others here on AC, a dipole configuration of the stacked woofers.  To do this we reverse polarity of the rear firing (top) woofer so that it works out of phase with the woofer below it.  This gives a figure 8 pattern of bass dispersion which eliminates much side wall reflection and provides a leaner, cleaner bass at the sacrifice of a little first-octave output.

Many hobbyists favor dipole bass but achieve it with a simple open baffle enclosure, or lack of enclosure.  While this works, you must boost bass considerably at lower frequencies to compensate for cancellations of the Dipole Effect.  You quickly run out of amp power and woofer linear excursion with this method.  Having stacked woofers allowed us to have two low Q, sealed enclosures which maintain extension and sensitivity without any LF boost.

The dipole woofers work very well.  The dry, tight quality of the instrument is well captured, and transients improve.  The problematic floor-to-ceiling mode is tamed with less EQ.  John, Dave and I go through all EQ settings and optimize each, taking into account our new, improved, better defined dipole lowbass.   After about an hour I am quite pleased with the reproduction quality.

One of our singers, Jeanine del Carlo, who had wowed our audiences with her original songs (particularly "Angel", sung a capella), goes home early due to family emergency.  We fill in with two guest artists,  Jason Serinus (master whistler, "Pavarotti of Pucker", voice of Woodstock on the Charley Brown cartoons), and gospel singer Marcus Cornelius.  Both are well received by audiences.  Jason's whistling is the most astounding display of  unique talent I have encountered in decades.

Still, during this day I find an area where playback quality is still lacking.  There is a "soft spot" in the lower mids quite apparent on Lesley's chest voice and as a thickening of texture on instrumentals.  It nags at me, and I go home determined to locate and correct the problem.  I'm just not sure how to go about it.

[Brian Cheney]

Sunday Jan 11, last day of show, 9am

Two incidents on Saturday remind me there is still work to do on our sound.  It is frustrating that one big problem is beyond my ability to solve.

While demonstrating my battery of percussion toys I strike the woodblock for someone.  There is an immediate loud, undamped, eight cycle buzz from the far upper corners of our room: slap echo, and very strong slap echo at that.  I would have to climb a 20ft step ladder and attach huge absorbers in those corners to stop it.  It is coloring every transient we hit.  Next time in a big room like the Zeus I'll bring long sections of plastic pipe with throw cushions or some such impaled upon them and damp the ceiling corners that way.

The other sobering occurrance is more positive: CD playback sounds wonderful, the best I've heard in a public venue from speakers regardless of size or price.  A friend of recording engineer Kavi Alexander hands me a CDR of his most recent work, Mahler 8 done in Budapest with huge orchestra, triple chorus, six soloists and two organs.  The work is so complex Kavi has abandoned his usual two mic Blumlein setup and gone with 3 mics set about 12' back from the front row of performers.  This has to be the greatest orchestral CD ever made, based on my listening here.  Look for it when it comes out.  My system has no problems reproducing it with perfect clarity and fidelity and without strain.  There is no sign of the soft spot I hear on live playback.

Some might ask why I wasn't using sine wave sweeps and microphone measurements to set up my demo.  These were available from the Sony Sonoma workstation if I wanted.  Yet the idea never occurred to me.  Microphones and ears perceive things differently and have different transfer functions.  The mic which most closely mimics the polar response of the human ear, the Schoeps Sphere, is an omni quite useless in our environment, where we are trying to minimize room contributions.  I will have to apply my engineering skills to identify and fix the problem area.

The soprano voice lives in the range from around middle C (256Hz) to two octaves above it. I hear weakness in the lower range of Lesley's voice which suggests an anomaly in the half octave just above middle C.  It's not a large suckout or I would hear it on CD playback.  When I arrive there is no one around, it will be me, the speakers, and the DCX.  We have opted for full dipole operation of the V60's which allows me to hear their sound equally well from the equipment table behind the piano.  I start with a 1dB boost at 266Hz and advance the frequency selector wheel one step at a time, or 6Hz increments.  By the time I reach 290 Hz I hear an improvement in Lesley's playback, but it's too much--I'm now "hypeing" her chest voice.  I reduce the boost from 1 dB to .5dB.  I am in an area where measurements would be completely useless.  The standard for speaker linearity is +/-3dB, for microphones it's +/- 1dB.  Raw data looks very bumpy and it's impossible to know what the bumps mean: are they FR errors of the transducers, late arrivals, cancellations from mic placement, or other measurement artifacts?  I will have to trust my ears.

I now progress back down from 290 Hz in 6Hz increments, this time with .5dB boost.
Again I hear too much "improvement" and cut the boost to .3dB.  This is not an amount of change sufficient to be reliably audible, according to textbooks.  Still there is no doubt about what I hear.  I land on 284Hz over and over as the optimum pole frequency with a low Q setting for bandwidth.  I check my work by going back and forth around this pole while listening in front of the speakers.  I try CD playback and it too is improved.  Eureka, I have found it!

The best listening positions are on the right side of the room and at least 3 rows back.  This is completely explicable: one side of the room is more absorbent than the other, and in the narrow dispersion mode of the V60's, wavelengths need space to reflect and randomize. 

The live sound and playback have little to set them apart now.  Indeed playback, which I keep at a lower level than live, is clearer than live, less reverberant.  The mic favors detail the ear ignores from further away than the 10ft distance from mic to closest instrument: you can hear more fingerboard from the bass, valves and keys from the sax, hammer and strings from the piano on playback than live.  The live sound of sax and piano is a little too loud for our room, actually.  Brushed cymbals are very clear and contribute substantially to the emotional impact of Lesley's songs, even at low levels barely audible live.

Late in the day I ask Fred Haller to play his instruments solo so we can record them and play them back individually.  Same with Bill Bailey's bass.  It pains me to realize Fred needs to be further back, and Bill about another foot closer to the mic for best balance.
Oh well, lesson learned. 

In 2channel pure DSD there is no mixing, no EQ, no editing.  However, due to the great flexibility of the SDE system (where I can even make a custom crossover just for this room and equipment), I am able to perform the first live-vs-recorded demo in history where room and speaker correction are in the playback.  And that with complex program material, not just a drum kit or a string quartet. 

We have accomplished what we set out to do. At this estate of the art, it doesn't get any better.

[Brian Cheney]

Some 350 people heard our live vs recorded demo over the four days of CES.

If you were one of them feel free to post to this thread.

Some audience members filled out Comment Cards which I will post here starting tomorrow.

[John Casler]

From "day one" of this undertaking Brian knew what an immense challenge it was.

Much of this challenge is cloaked in the "unknown" elements of intersecting interactions of all the variables.

In a "LIVE" performance, you have the several "point source" positions of the performers, and their immediate and secondary reactions in the room.  They all have psychoacoustic properties.

Now then couple this with the numerous seating (room) positions of the audience and the differing combinations to each of those positions and you have the first stages of a remarkable task at hand.

Now step into playback, and you have a "single point" in space where the mikes reside.  It is this point that each performers position and sound is transferred to the recording.  This is the perspective from which any and all recording have limitations.

So when you are sitting in the audience, and listening to the "playback" you are NOT hearing what YOU heard, but more what the MIKE heard from ITS recording position.

So for example, when you heard Leslie from the audience you were maybe 15-20 feet away from her, but during playback, you were sitting at the MIKE, and were only 3-4 feet from her.

Or if you listened for Brian's brushed cymbals during the live performance, you likely heard little because he was now at least 30ft away.  Yet the sensitive MIKE was 1/2 that distance and the playback revealed just the right amount of brushed cymbals in the mix.

Now this then doesn't begin to scratch the surface, since we also have in the reproduction chain all signal blended and sent to two points in space, and the reassembly into a cohesive sonic tapestry is dependant on the correct balance to the listener.

Now the CDWG's can provide an excellent tool to this assemblage, BUT . . . in a room this size, the SPL's needed are reduced by the Wave Guides and even with them off, the amps would (as Brian mentioned) have been overloaded.

So the best that one could do under these circumstances is approach the tones and frequencies of the original LIVE performance.

Secondly this type of application creates a very UNIQUE playback anomalies.

In this room you have specific acoustical interactions, such as reverb, echo, RT-60, modes, nodes, etc, etc.

So the listener hears these during the live performance as natural.  These sonic artifacts are "recorded" onto the Hard Drive.

During playback these Sonic elements are played back through the speakers, and as they are scattered throughout the "SAME" room, they again create sonic artifacts, of the artifacts .  This seldom happens in "normal" playback circumstances since the dimensions and acoustic properties of your livingroom are substantially different than the "recording venue".

Now that doesn't make the acoustic additions of your room any "better", but it does mean that they will not "double" those of the same venue.

So I mention these as just a small portion of the challenge of doing a LIVE versus VMPS Recorded type of comparison.

In light of these elements and the several hundred more, I was quite pleased at the abilities the system and recording effort yielded.

If you would hear the same recording on MY system which is BEYOND HEAVILY DAMPED, it would likely yield an even much closer approximation.

Or if you get a chance to hear the recording on Headphones, which subtracts the second round playback room, you would hear a VERY Lifelike rendition, but no soundstage.

All in all, one would have to say that under the circumstances it was rather remarkable how close B, came to shaping the sound toward reality.

[Brian Cheney]

You should have heard playback Sunday, John.  That was very close to live sound!

[Brian Cheney]

See positivefeedbackonline's show report for an accurate description of our Saturday sound.

[John Casler]

See positivefeedbackonline's show report for an accurate description of our Saturday sound.

here: http://www.positive-feedback.com/Issue41/ces_gordon.htm

[John Casler]

And our Ball roomate James Bongiorno

http://www.positive-feedback.com/Issue40/ces2009.htm



AMP Legends

[Housteau]

Some might ask why I wasn't using sine wave sweeps and microphone measurements to set up my demo.  These were available from the Sony Sonoma workstation if I wanted.  Yet the idea never occurred to me.  Microphones and ears perceive things differently and have different transfer functions.  The mic which most closely mimics the polar response of the human ear, the Schoeps Sphere, is an omni quite useless in our environment, where we are trying to minimize room contributions.  I will have to apply my engineering skills to identify and fix the problem area.

The soprano voice lives in the range from around middle C (256Hz) to two octaves above it. I hear weakness in the lower range of Lesley's voice which suggests an anomaly in the half octave just above middle C.  It's not a large suckout or I would hear it on CD playback.  When I arrive there is no one around, it will be me, the speakers, and the DCX. 

Unfortunately I had a flight Sunday morning and just wasn't able to be there, but hoped that someone would be there early to assist.  But, with the speaker placement so far into the room and the fact that they were dipole nearly top to bottom, I noticed that I could get a good idea of the room sound for the audience when sitting back at the control area behind the piano making adjustments, so I knew it would be OK.

It really was quite a progression and a passage of sorts making it through Wednesday and Thursday to reap the rewards of Saturday.  Even Friday had shown the possibilities enough that several of those attending came back on Saturday.  You could see it in their faces.  They were having fun.



And I am surprised that we have not seen that written anywhere yet.  After all isn't that why we indulge ourselves in this hobby?  It should be about enjoyment and having some fun.  As serious as it was to stick ones reputation out there on a limb, Brian still made it a fun learning experience for those that attended, even on the days when things were not going so well.  I think some did get that and that is why we had the crowds we did.



If you notice in the first picture that all the seats are not full.  It was like that for several performances.  That was due to two different reasons.  First off not everyone arrived on time and some came after the performances started.  The only way to the other side was to cross between the mics and the performers, and thankfully most knew not to do that.  However, that actually worked to our advantage and as Brian explained, the right side of the room had the best sound.  The second reason was that some new that was where the best sound was from a previous performance, or from being told as they came in.  Not all things were accidents .

Speaking of accidents, or incidents, Thursday was stacking up to be one of the less stellar days of The Show .  But, as you have seen and read, we made it well beyond this low point.



What a learning experience it was.  One of the things that impressed me the most was hearing the transformation of the sound within that room and seeing step by step how that was accomplished.  Brian's ears did the listening and mostly John and I did the moving, tweaking and dialing in of the gear under his direction.  As Brian has already mentioned, not one measurement was taken.  The entire process was accomplished just by listening.  I don't have a trained ear like that and even though I have been involved in this hobby for quite a while, I just haven't had the opportunities to hear and experience live acoustic music such as this.  So, for me this process and transformation seemed like magic.  Maybe it was magic, because when it sounded right for the live playback it also sounded right for most of the prerecorded music as well.  Although there wasn't a whole lot of time for that.

In between performances there was a lot of mingling going on with everyone curious on how we did everything.  That was the chance to meet a lot of legends of the industry, reviewers and fellow Audio Circle members.  If things started to slow down too much that didn't last long if James Bongiorno was close by as he felt obligated to fill the room, as he put it, with more pizazz.

[Brian Cheney]

The Zeus is quite isolated and it took some barking in the lobby to get the crowds in. 
Next time we'll do more signage and arrows to drum up business.

It was fun, if very nervewracking.  I look forward to repeating the demo at the next CES.  This time we should get to our goal a lot more quickly.

[doug s.]

The Zeus is quite isolated and it took some barking in the lobby to get the crowds in. 
Next time we'll do more signage and arrows to drum up business.

It was fun, if very nervewracking.  I look forward to repeating the demo at the next CES.  This time we should get to our goal a lot more quickly.

fascinating - i have followed all the threads re: this ewent w/great interest.  even w/your learning curve, perhaps it would be worthwhile, (presuming money isn't a problem), when you do it again, to rent the room a day early, to get a head start on set up?

best,

doug s.

[John Casler]

I wonder if it might make sense to have "two rooms"

Room One "Live sound"

Room Two Heavily treated Recorded or even just "real time" Playback System sound.

The participants could sit and listen to the LIVE room, then walk into the treated Real Time or Recorded Playback Room.

That way we could address the sonic "Tale of Two Rooms" by heavily treating the Playback Room.

I especially like the ability to listen to live, walk to the next room and hear the same performer, "still live", but reproduced via the system.

That way you could actually do it in smaller rooms and use more treatments (actually to each)

Might be something to consider.

[Housteau]

If there had been power available on the audience side of the room, I think it would have been better to switch the band and audience around.  This would have placed a live wall behind the listeners and a more treated one behind the performers and system.  That fabric on the room dividers was not very thick, but I think it would have helped somewhat.  If that same room is being considered, then rerouting power distribution over to that side might also warrent a thought, or two.  I think this could be done fairly easily with pre-made heavy duty power leads terminated to gang boxes.  From there everyday extension cords, power strips and conditioners could finish off the routing necessary.

[Brian Cheney]

I would probably use the same end of the room as before, for two reasons:

1. Power outlets are all located there.  Extension cords tend to get kicked and tripped over.

2. We already know what to do with the performers and speakers placed as we did the first time.  This really cuts down on setup time.

The earliest we can have access to the space is noon Tuesday.  I would likely make that day "setup day" rather than Wednesday in order to have the extra adjustment time.

I also plan to take an extra van full of gear and have more acoustic treatments with me.

[cliffy]

Big B,

I am definitely interested in attending next year's demo.  I would be happy to help set-up, tear-down, etc...
I am sure there are a few other SF Bay Area VMPS fans that would like to lend a hand.

The other item that needs addressing is the "media blackout".  Your project is worth a serious write-up in every self-respecting audio mag and web-site.

I hope your demo at this year's show brings you more business and success in the New Year.

Spencer