Christian Heil

The icing on the cake? Christian Heil granted us an interview. We reviewed the essence of the L-ISA project, the IT tools needed to implement it and, finally, his vision of modern amplified sound.
More than exciting… Essential.

SLU: A persistent legend would have it that you had the idea for L-ISA one day after a concert where you, like 90% of the audience, were poorly positioned.

Christian Heil (President L-Acoustics…): It didn’t happen in one day; it’s been a problem for a very long time. When I founded my company 30 years ago, I didn’t know what sound reinforcement or live production were, I was only interested in the technology that we were helping to move forward and that satisfied us for many years.
Little by little, I put together teams of sound engineers and application engineers.
I worked closely with friends in show business, and I noticed that they were often happy with the work they were doing because they occupied a privileged place in the theater, but the people in the audience weren’t enjoying that same level of satisfaction. So I put myself in the shoes of the spectator, and I realized that I was frustrated. I had to concentrate to connect the sound with what my eyes were seeing, a process that spoiled the artistic experience of the show for me.
Unlike a technician, a spectator is looking for enjoyment, spontaneity, sensation and realism. He doesn’t try to understand if the treble is clean or if the phase is coherent, and that’s what made me gradually think about refocusing on sound and image, to merge them. Another factor which pushed me on is the enlargement of stage-fronts to provide more room for set and lights, which in turn relegates the sound to the sides and disassociates sound and image.

Jean-Claude Casadesus and the Orchestre National de Lille in fusion with the L-ISA system

SLU: We try to fight this outward drift by using small central reinforcements…

[private]

Christian Heil: We try to do that but it doesn’t work very well. So I told myself that we had to try to enter the multidimensional and hyperrealistic realm, correcting the historical error of putting speakers on the left and right, while the people are in the center.

SLU: At first, we used to hang a central cluster…

Christian Heil: It was a very good idea! Then we had to fall into the illusion of stereo, while we never stopped using dual mono. The solution to try and change the game could only come from a respected, innovative and well-established manufacturer. But it’s not going to be easy. We’re not going up against other speaker manufacturers – instead, we’ll be opposed by lighting designers and scenographers who won’t want arrays of speakers in the field of vision.
I’ll tell you a story. I was invited to the London concert of a world-famous artist. The show starts and I can’t see him. The huge video wall projects psychedelic images. It ends up showing him, very small and at the center of the stage. Is this really doing justice to the artist, who was returning to the stage after a long absence?

A review of how L-ISA works. The “spatialized” zone offers any listener within it an immersion in the mix with a precise correlation between the location of the source on the stage and the amplified sound it generates.

Here is what a standard left/right would provide in the Lille Arena. The light green area represents the zone where the stereo image and the phase are optimal, the dark green area is where the listener is covered in dual mono, interfering also with the out-fills and, finally, the yellow area is where only the out-fills are heard. The portion where the audience receives a stereo signal is no more than 8%.

The same prediction, but with the L-ISA system. The dark green here is the interfering zone between L-ISA and K2 side hangs. 26% of the public receives a signal in phase and with a spatialized image.

The experience of a concert is not “sonic” enough anymore. L-ISA is a gamble that needs the support of sound engineers, but the ones who undertake this choice are still rare. François Gabert, Stéphane Evrard and the entire ONL team took the plunge – let’s call that a perfect alignment of planets.

SLU: When did L-ISA take shape?

Christian Heil: We started in 2012 with Sherif El Barbari – our field man, who specified the project and participated actively in the development – and with Guillaume le Nost, who took charge of the development and built the necessary team for that. For a few years we worked in the shadows of a laboratory, doing algorithmic tests and compatibility tests to create sound designs.
We had the systems, but switching from left/right to multichannel meant changing their size. That said, we have the necessary models, between Kara, Kiva II and Syva. We should have had Kiva II for today’s production, but unfortunately it is too recent a product, and we couldn’t find the number of enclosures necessary in time.

François Gabert’s FoH, with his Lawo and his pair of PCM96s. At the top, the seven arrays of the L-ISA system, extended beyond the stage.

SLU: I must say that in the extended version, with seven arrays, it really starts to perform! How many are needed for a typical deployment?

Christian Heil: L-ISA has two parts. The first, called the Scene system, uses five or more arrays to cover the width of the performance area where the musicians are. The second, called the Extension zone, offers a wider panorama in which we will put reverberations, ambiance, some synths and choruses. A kind of outer panorama. It’s not necessary for the Scene and Extension systems to have the same density.
If you were touring with a big rock band in arenas like the AccorHotels in Paris or the O2 in London, it wouldn’t be unthinkable to hang K2, which opens very wide, naturally has more contour and would immediately reassure the sound engineers, who would be in their comfort zone. The major disadvantage is that, instead of hanging two big classic arrays, we would put five slightly shorter ones. Add the Extension and the budget will no longer be feasible. We work according to a scale ratio and a number of specific components, whether they are K2 or Kiva II.

L-ISA. Lots of small enclosures or fewer large ones…

SLU: Agreed, as far as the drivers are concerned, but for the low frequency transducers and the possibility of some models, like K2, to be used in full range, the game changes with Kiva…

Christian Heil: Absolutely. Tonight, for example, this is a configuration where the need for classical music contour is not very high, but even in case of more demanding music, the amount of equipment would not drastically change. To better understand, let’s go back a bit to basic acoustics. When you put the hundred or so 15-inch or 18-inch membranes of a traditional system into an audience, there’s a little chaos, sort of like what was done before the V-DOSC when you multiplied the number of boxes.
Suspended bass sources combine only in the plane that bisects the line between them, and when you get outside of that, you have problems. When you add some subs, which are occasionally arranged frontally for a more distributed result, you don’t gain much energy because of the losses in efficiency due to the linear configuration.

Eight KS28 flown centrally, cardioid and endfire. It may seem like little, but their placement, their impact and their incredible sensitivity make this number more than enough, especially for classical music.

The concept of L-ISA specifies that all of the bass and the infra be located in the center, because everyone knows that if we could place all sources of bass in the same place, within a distance of one half-wavelength, we would have the punch throughout the whole room. Ideally, even if it’s not going to please the lighting people, hanging the subs in a central point, when technically possible, is good for the sound. When this isn’t possible, one solution may be to decouple the sources to avoid interference, but for that you have to convince sound engineers who naturally prefer their comfort zone.

Madje Malki, the hyperrealist immersion adventurer and Christian Heil’s long-time accomplice!

Even a friend and adventurer like Madje, who used L-ISA for a long Renaud tour (a well known and highly appreciated French pop singer – ed.), sometimes feels the need to return to a more conventional mix. Finally, it seems to me that, in the low end of the spectrum, we’re reaching overkill. A well-known heavy metal band used 64 subs at the AccorHotels Arena recently. I think it works, in a bit of a….. chaotic way. (he laughs).

SLU: Fortunately, otherwise it would be intolerable! If I understand you correctly, you want to entrust the bass and the infrabass to a single central point of emission, and the rest of the spectrum would go to L-ISA arrays.

Christian Heil: Everything below 100 Hz can be grouped in one place, which allows you to use smaller arrays. That’s why there is no K2 in the central array. We connect everything in the center with subs, which also work as bass cabinets.

SLU: Do they have a specific preset?

Christian Heil: They have to go up to 100 Hz. We did some outdoor tests, and even if I take the outermost array in Kara in relation to a single source point of bass and infra, it still works better than if I hang K2 and I add some subs in the center, because they interfere. This, of course, can’t be extended infinitely because of the relative distance between subs and heads. It would create problems of delay and level. So I’m not saying that in a stadium, with a central system and a stagefront of 70 meters, we would place Kara 30 meters from the bass point, but in many other places, we would have a better result than with the left/right.

SLU: Kara is therefore the reference enclosure for L-ISA. A single 3” driver is not a problem?

Christian Heil: The number of drivers doesn’t matter, it’s a line-source system, we use as many enclosures as it takes.

SLU: How does one determine the number of enclosures with L-ISA, exactly?

Christian Heil: We begin the L-ISA design from the base, the left/right. Here in Lille we looked at the 2016 design, the SPL mapping, and we proposed the “same thing ” but with another system. We provide the same energy, distributed in the room.

An indiscreet view of a place long kept secret, the famous auditorium of Marcoussis, where the trio of Heil, El Barbari & Le Nost spent long days in full immersion. Many stacks of six Kara each and four SB28 have served the cause for many years…

L-ISA is providing the same distributed energy as a traditional system. But better.

SLU: So, I’ll ask the question to the designer and engineer: is the 60-120 Hz octave reproduced by a system of small speakers equivalent in impact and in qualitative feeling to that delivered by a system with an equal emissive surface but using, for example, 12” cones?

Christian Heil: We asked ourselves this question, too. The answer is somewhere between yes and no. I think that, first and foremost, it’s about mixing references, but at the same time I’m not a mixing engineer, and if Madje were here, he might not agree (he laughs!). With everything between 60 Hz and 100 Hz, I will always prefer the reproduction from a 15-inch and even an 18-inch transducer, which has less distortion. Once you go above 100 Hz, I don’t see the difference anymore.
Now look at the problem in reverse: do I not prefer to divide the sound energy in the 80 to 120 Hz band over a series of 8-inch transducers, rather than concentrating all that energy on two 15-inch speakers? This time, the answer is not so clear. If I put a single signal on an 8″ membrane and a 15″ membrane, I would prefer the 15″.
But it’s not the same thing at all with music. When I combine a lot of different signals and I drive them all into a 15″ membrane, their combination is not ideal; I prefer more and smaller membranes. When we talk about Madje and Renaud (Madje and the sound company Potar Hurlant have used L-ISA on a long Renaud tour– ed.), as wel as all good pop and rock mixers, they all tend to bring their mix back to the center.

Renaud’s stop in Paris at the Zenith. Six KS28 in a cardioid configuration and a central array of 14 Kara.

SLU: Which has always provided the punch…

Christian Heil : Exactly, except that, if we put kick drum, bass, in short, the fundamental elements in few Kara and only in the center, the system will suffer more than, for example, in a line of K2. We must learn to expand and better exploit the number of enclosures in the air, which the classical sound engineers are doing very well. That said, if Madje doesn’t agree, I’m sure he’d be happy to express his opinion here!

SLU: He’s welcome whenever he wants and he knows it (Madje, if you’re listening…). Is that why you prefer L-ISA for pop shows that are a little more “calm”?

Christian Heil: Certainly, it’s more appropriate for a style of music that takes advantage of spatialization and not so much of power in the way that we have just mentioned. I’m thinking of Jean-Michel Jarre, for example. The very nature of his music would work excellently with L-ISA.

SLU: And could benefit from automation of the movements…

Christian Heil: Of course, there are plenty of possibilities with the movements that, in some music, could add a lot.

Which enclosure for L-ISA?

SLU: Don’t you think that the current limit of L-ISA is Kara, because of its rather limited impact in the bass, its dynamics and, especially, its dispersion, which at 110° doesn’t have the appropriate coverage for your multichannel technology… in other words, wouldn’t it be better to have a specific enclosure for L-ISA in order to help it succeed?

At least 140° of dispersion from Syva up to above 10 kHz. Just what is needed for L-ISA and for a lead vocal amplified only from the center array to reach the entire audience.

Christian Heil: Of course Ludo, but all in good time. The answer is clear. If L-ISA and multidimensional systems in general catch on, we will have to reinvent sound systems to make them lighter, quicker to install and with wider coverage.
If we meet with success beyond the enthusiasm of the public – namely the willingness of the productions to take the leap –we’ll likely have to go back to the drawing board. I think now it’s still premature. In any case, it’s the sound companies who will demand systems from us that are designed specifically for multidimensional deployment.

SLU: There is already Syva…

Christian Heil: Yes, and it’s an excellent enclosure for L-ISA. One could easily amplify a classical or jazz ensemble in a room of 1,500 seats with this Syva. (This was actually a premonition of Christian Heil. Alain Français a highly respected live sound engineer did it at the Théâtre des Champs Elysées in October – ed.)

SLU: With its very wide dispersion, it is the only one in the L-Acoustics catalog that perfectly corresponds to the L-ISA principle, which requires that each group of the frontal system optimally cover a common area representing more than 75% of the audience area.

Christian Heil: This enclosure was a gamble and the results are very good, in residential applications and in live. Its nominal horizontal dispersion is 140° and effectively guarantees excellent coverage, which is the key to L-ISA.

The software side of L-ISA

SLU: Where are you in the development of the IT tools to manage L-ISA?

Christian Heil: We are advancing in parallel. Today, the latest generation of Soundvision is able to qualitatively analyze a multidimensional design because we know that it is different and necessarily complex. We have been doing left/right for 40 or 50 years without asking ourselves very often the question of where it overlaps, how far the central stereo – which is actually closer to being a dual mono – goes.. In fact, no one knows what the percentage is of each category.

A view of the L-ISA gauge. The number of criteria is such that it would be difficult or impossible to make it “stick to the walls”. The GUI of this gauge is still provisional.

SLU: Yes, you can see it with Soundvision.

Christian Heil: Right, we now have a layer in Soundvision that is practically a certification of our own designs, whether they are left/right or multidimensional, and this by means of a score.
This new layer is the result of various criteria identified as the percentage of the audience covered, the one to which all signals converge. There is an SPL distribution criterion. We also check – and this is new – the temporal impact related to different acoustic paths, depending on where the listener is positioned in the room. We’ve gotten used to it, but in dual mono, as in stereo, we live with comb filtering and the sonic pollution of the second source. We also have a criterion that is related to the resolution of the sources: the ability of a spectator to recognize and separate the sources he hears.
It’s a kind of spatial resolution, like a pixel in an image. We will also look at the angular deviation that exists between the real source and the sound source. A score is given to the horizontal and vertical deflection. Due to the placement of the arrays, often where it errs the most is in the vertical deflection, which pushes us to negotiate the lowest possible hangs. But this is difficult because a show combines sound, lights, images and scenery, and everything must coexist for the entire audience. All of these aggregate scores provide a final score. No more lying to ourselves… we know exactly and objectively where we stand from a qualitative point of view.

Christian Heil with, on the right, Stéphane Evrard, the Technical Director of the ONL.

SLU: L-ISA demands real know-how from the operator…

Christian Heil: Yes, but I’m doing what I used to do with the V-DOSC twenty years ago. L-ISA works on principles that we establish almost like school rules – rules made to be outdated later on. The technicians who use this system will go outside of the box and will invent countless other uses. The Puy du Fou, with its circular stage, and other theme parks loved the possibilities offered by L-ISA.

L-ISA, work in progress ?

SLU: Will the L-ISA processor algorithms evolve further?

Christian Heil: Guillaume le Nost, who develops it, and Sherif El Barbari, who utilizes it, have spent nearly three years working on the way audio is manipulated in the processor.

Sherif El Barbari assists François Gabert and answers all his questions about L-ISA, which he knows like the back of his measurement microphones.

Sherif, who, much like you, is extremely attentive to details, often brings me suggestions for improvement, but micro-details, especially compared to what we are used to hearing from good old left/right, whose interferences are inherent to the very principle of the two arrays. But the answer is yes, it will certainly evolve. If you had seen the first preset that I created for the V-DOSC at a time when I didn’t know the particularities inherent to shows, it was…

SLU: Not good? (Laughter)

Christian Heil: There have been subsequent generations of presets – as I recall, there have been seven – and now we understand what needs to be done.

SLU: How do you harmonize what the hanged and stacked enclosures reproduce?

Christian Heil: The lower ones play a mono reduction. This is logical, because if you are at the end of the stage and very close to it, you will hear only your lip-fill enclosure. If I don’t play a mono reduction and the soprano is at the opposite end of the stage, you will miss her song. With a mono lip-fill, you will hear it. They say it is possible that in the future, and depending on the circumstances, it will be composed of mono and non-mono mixes. We are working on this aspect, and maybe we will go to stereo

SLU: What should you do for the sides, where a lot of people tend to be positioned? At the moment there is K2 here in Lille, and even on the Renaud tour. Could you not attempt to create even a reduced effect of spatialization?

Christian Heil: We thought about it; we tried a certain form of duplication but, for the moment, we haven’t found the right solution because the pollution that it causes far outweighs providing quality sound to more people. Returning to L-ISA in general, we bring substantial improvements over what was done before, but we can’t remove some defects.

The need to see the stage and the video screens for the spectators in the stands, and a certain obsession with the “speakers being ugly” on the part of the scenographers, force the systems to be moved far out to the sides and, in the case of L-ISA, to be hung very high – a little too high. A difficult problem to solve.

Verticality, that is, the distance from the audience, will always be a problem. A musical where the speakers are placed very high doesn’t work as well as it should with L-ISA, but no worse than with current systems. We are simply used to it and, essentially, we will always find more defects in a new system than in the one we know.

Syva during one of its first outings, in the expert hands of Alain Français, just a year ago in the heart of the Louvre in Paris.

SLU: So we are almost at the end of the development phase of L-ISA…

Christian Heil: No, that would be a lie, but we are pretty damned far along. We started in 2012 and we didn’t just do theory, we put our ideas into practice in the field – so L-ISA is quite ready to be marketed. The second phase, which will start in parallel, is its adoption by sound engineers.

SLU: Can we say that L-ISA is the latest news from and the future of L-Acoustics?

Christian Heil: The latest from the audio system manufacturer L-Acoustics is Syva. First and foremost, we are introducing a different branch in the group that deals with sound design, which focuses on what the product does when it is put into a situation and not just what it delivers in terms of coverage or SPL. I would say that L-Acoustics is more and more interested in the application and the implementation of its products.
L-ISA is at the head of what we would like to do in multi-channel FoH sound design that’s easy to deploy. In this respect, the merit of this L-ISA installation in Lille goes to Fred Bailly and Sherif. My contribution was basically the #1 and #7 boxes, the ones that extend it (he laughs!). You have to cheat – stretch the orchestra to fill a room this big. The L-ISA Processor knows how to generate 32 outputs, of which six are pre-assigned to downmixes.

And the secret of L-ISA is…

One of the flaws pointed out by Christian and visualized on this rendering, the vertical localization, namely the gap between the stage and the coverage. Note that, for many seats facing the stage, it reaches and exceeds 60°, which is too similar to a shower, and that the lip-fills provide only imperfect coverage since they don’t integrate the particular distribution of L-ISA.

SLU: What distance should you leave between each cluster? It is four meters here, it seems to me…

Christian Heil: There are four meters for a question of fluidity. There could be a little more since we are operating over quite large distances. Beyond a certain distance, the sound does not couple well between one cluster and another and especially if, for example, it is necessary to follow the movement of an artist with a tracker, you lose fluidity in the transition between the clusters and it becomes a “magnetic” effect of jumping from one to the other. It also depends on the dispersion angle of the speaker enclosures but, anyway, it is purely for acoustic and design reasons.

Some small flaws also in the horizontal localization. Without a doubt, a speaker enclosure with a dispersion that opens more laterally, positioned a few meters farther back (if its rear attenuation allows it), could easily solve this problem and partly that of the lip-fills…

SLU: L-ISA will still require a lot of education to explain the new fundamentals…

Christian Heil: That’s right! We are re-shuffling the cards. I was recently told: “but, you always said that one shouldn’t put two line arrays within seven meters of each other!”. That’s true, but specifically when you have the same signal in both, with the whole band. L-ISA works in separation mode, the same signal is never in two adjacent speakers… (I interrupt him)

SLU: Yet, we can open the same signal up onto several arrays via the controller!

Christian Heil: Indeed, there is a mechanism in the algorithm that attempts to minimize interference issues and corrects the tone to ensure that the two systems sound like one, which is otherwise not the case. It is also necessary to do the off-axis coloration and the phase rotation to avoid interference. We spent years developing it with sharp people who listened and split hairs. The algorithm is good. What could be improved even here today is the vertical distance and therefore the realism for the front rows.

The L-ISA plugin, which can be incorporated into VST or AAX environments and operates dynamically in the project and with the automation of the digital editor. This plugin then delivers its instructions to the L-ISA Processor via a simple network cable.

SLU: I agree completely. The coupling and the sound in the first rows is not optimal. It is better to be further back and in the coverage of L-ISA alone.

Christian Heil: Whoever invents near-fills that work and are effective for the first five rows, I’ll take my hat off to him. I attend many shows in London, where I live, but very few shows where the near field is convincing.

SLU: I imagine that integrating part of L-ISA into DiGiCo consoles will help encourage people to adopt it…

Christian Heil: Of course. Sherif and Guillaume have worked to make this agreement with DiGiCo, and this makes the system even easier and more convenient to implement. L-ISA is an adventure that began five years ago and whose maturity will interest productions. I am not convinced that the pace will pick up all at once. It is necessary for people who have confidence in me and in the brand, and those who have power, to take the first step… people, for example, like Stéphane Plisson in France but also the sound engineer of Adèle or of Radiohead.

From left to right: Christian Heil, François Gabert, Erdo Groot and François Bou, the general manager of the ONL. The concert has just ended and everyone is sincerely delighted.

Contemplations as Conclusions

Let us first render unto Caesar the things that are François Gabert’s. The most perfect sound reinforcement system is not much use if one pushes it into the bizarre. Fortunately, this man is as serene and balanced as his mix. We feasted for an hour, savoring in great detail the work of Jean-Claude Casadesus and the Orchestre National de Lille. Also, hats off to the architects who designed the “showbiz” version of the Pierre Mauroy stadium. The versatility of this room is a staggering reality. Especially regarding the acoustics. The capacity, the accessibility, the rigging possibilities, but especially the solid acoustic architecture make it a remarkable venue, an Arena with a capital ‘A’.

With its closed roof with multiple rigging points, its floor as smooth as a line array waveguide, and its wonderful acoustics, this half stadium is THE answer to the needs for show venues. Remember that a resident football team plays, at most, only 25 to 30 matches at home, the rest of the time…

What is L-ISA? To put it simply, I would say that it is the best compromise possible in a field that is all compromises. L-ISA is the most economically viable and acoustically efficient way to obtain an immersive, coherent and spectacular sound front in terms of coverage area, frequency response, phase, dynamics and SPL. The idea and the technology behind it are not new, but their integration into a powerful processor, driven by a very sophisticated algorithm and controlled by an object controller, are quite new. When a juggernaut like L-Acoustics gets involved in a project, it’s never taken lightly, and L-ISA is no exception. Several questions then arise.

Christian Heil, here with Fred Bailly, responds to numerous requests from the technicians present, and prepares for the performance of the star system. It’s not enough to have given the world the V-DOSC, one must assume responsibility for it ;0)

Is L-ISA good for everything? Theoretically, yes. Offering a vast sound field and precise localization of sources to more than half of the audience, and in some cases even more, it is good for all applications and musical styles. In addition, the simple and remotely controllable management of objects, which can also be automated, opens up an infinity of creative possibilities.
Faced with soloists that were too bright, too forward in the mix and probably too “amplified” sounding, François Gabert, for example, made the choice during the concert to move them back in space, a simple way to make them blend more into the mix and give them more authenticity. L-ISA is a key that technicians, but also artists, can appropriate. Let’s not forget that it was through the influence of Pink Floyd that quadraphonic sound existed on a large scale.

Kiva II, an enclosure that can be found often in L-ISA deployments. Passive, 16 ohms, 138 dB SPL max, all in a 14 kg cabinet and with 100° dispersion.

Can L-ISA also excel in styles that are demanding in terms of speakers and SPL? Once again the answer is “yes”, taking advantage of the enclosures already in the catalog, but it would probably be even easier with a dedicated model: a speaker enclosure with specific forms and specific functions able to reproduce, with modern power and dynamics, the spectrum from 100 Hz to 16 kHz, with a very wide dispersion and maybe adjustable waveguides, a speaker able to deliver more SPL but with a limited number of boxes, in order to facilitate its visual integration.
Finally, it would have to be an enclosure with a cardioid dispersion to allow it to be placed vertically, without the risk of impacting the stage, which would offer a wider L-ISA coverage and which would also serve the front rows. A punchy and modern enclosure that would help sound engineers get away from such a monophonic vision. What is true for the high end is also true for the low end of the spectrum. For the lows, it would ideally be a very powerful box, combining bass and sub reinforcement, able to fully reproduce the first two octaves with a native cardioid directivity, in order to concentrate a factory of bass in a small and central space.

The full-range mapping of the L-ISA deployment in Lille. A maximum SPL of 103 dBA and a minimum of 97 dBA with, of course, all the speaker enclosures operating. There are people there, you still have to feed them…

Is L-ISA creative and easy to use? The answer to the first part of the question is clearly “yes” but, even more, it is also a mixing tool in its own right that has its place in a console and not just relegated to its screen and mouse. Its integration into DiGiCo consoles will be very beneficial and should be emulated. Simple to use… maybe not if you are on the design side.
The need, for example, to concentrate the bass cabinets in a single point makes it difficult to deploy outdoors or anywhere there are problems with rigging point weight limits. It will also take some time for the sound engineer to get a feel for the set of speakers and their layout to open their mix and, especially, to understand that we must use more enclosures to get the pressure that any show requires.

Does L-ISA sound good? Here the answer is once again “yes”. The strategies deployed within the processor make it possible to combine the same signals as constructively as possible in two or more spaced lines. It isn’t obvious on paper. Yet it works much better than what is currently given by two main line sources, which already don’t love each other very much, and to which we add the down-fills, out-fills, lip-fills, front-fills, the flown bass cabinets and the subs on the floor. Let’s be honest. We are used to it, but it is very often “interferenceland”… So yes, sometimes we perceive some artifacts, but the whole is infinitely more faithful and generous in coverage and audio quality than the usual left/right.

A 4000-seat amphitheater, the perfect place for L-ISA. Here the left/right version with 12 K2s, providing a comfort zone to 25% of the audience.

The same shed in Santa Barbara, this time with an L-ISA deployment of five Stage arrays of 12 Kara each and two Extended arrays with eight Kara each. The comfort zone reaches 87% with a true spatialization.

Finally, is L-ISA a good answer to the new standards for levels? Without hesitation, yes. By the fullness of its performance at reasonable pressures, and by playing away from the audience with a single group of point-source speakers the part of the spectrum that has just appeared on the screens of analyzers henceforth touchy to dBC, L-ISA will feast the public, as well as the public authorities. In fact, the Requiem at its magnificent fortissimi reached only 98 dBA and just above that in C-weighted. At 4 dB from the limiters. Another advantage that makes life easier for the mixer is that the output at FoH is not boosted +6 dB as usual, since heads as well as subs distribute the SPL better.
In fact, we could start asking ourselves a simple question. What if left/right is a thing of the past?

JJean-Claude Casadesus and the Lille National Orchestra.

[/private]

Present at the Pierre Mauroy stadium – and as sought after as Christian Heil – was Guillaume le Nost, head of R&D at L-ISA, who answered our questions as well as he possibly could – not that he doesn’t know, but that he knows too much… far too much.
The continuation of our triple report on the L/R-killer – episode 2 is here.

Guillaume le Nost

SLU : How does one go from video games to L-Acoustics?

Guillaume le Nost : The common theme is sound. Sound synthesis. I was working on sound generation algorithms while I was a part-time musician. The drummer in my band also played drums for another band with one of the acousticians of the L-Acoustics team.
I loved what he told me about his work and I joined the company on the same team. I got to participate in the design of the ARCS WiFo and the SB18. My past, which was more oriented toward signal processing, allowed me to propose ideas for algorithms, especially since software and signal processing have gained a lot of importance in recent years.

SLU : Will L-ISA remain as it is now – with few controls and controlled by specific rules – or will it evolve with more possibilities and flexibility for the user?

[private]

Guillaume le Nost : We try to make it easy to use, without too many configurations while, at the same time, there are enough possibilities to let the user achieve his goals. All of this is done within a framework of design rules that are put in place and refined with time and experience. The fundamental principles of the algorithms are established and validated.

Up at the top it’s good; what about down below?

SLU : Will the algorithms allow a smoother transition between the flown system and the various reinforcements on the floor?

Guillaume le Nost : Unlike some other multichannel systems – which have algorithms that are very well adapted to fills, but less to main systems – we took a somewhat opposite approach by throwing all our efforts into the main system.
We have plenty of ideas for everything regarding the fills, but we can’t forget the old saying “the best sound comes from a single source”. We know it and we see it very clearly in simulation and measurement, the addition of speakers immediately degrades the performance of the main system.

SLU : But you have to do it…

Guillaume le Nost : We know that and we’re working on it. L-ISA is anything but fixed and will continue to evolve. The platform is very scalable and, without any problem, we can make improvements without changing the hardware.

A view of the KIVA II lip-fills with the difficult task of coupling in phase with the flown system, as well as in tonal and impulse response.

SLU : Wouldn’t it be possible to do better than mono for the fills?

Guillaume le Nost : This room is large and therefore begs the question: what sound-image width do we want to have in relation to the visual width. Typically, we could have started deploying multichannel fills here. We have LCR downmixes, which could, for example, have been used for the flown laterals, but with two constraints.
The more points of diffusion we hang, the more it pollutes the reproduction – this is true in all designs and we don’t pretend to have a magic wand that will fix it. Plus, it increases and complicates the installation. Our mission is to provide sound to 100% of the audience while maximizing the L-ISA coverage zone. Whatever happens with the fills, at the very least, we are at the same level as what is done traditionally.

SLU : In any case, we fall into the “philosophical” discussion: “I’m seated on the side and I see from the side. Should I also hear from the side?”

Guillaume le Nost : Exactly. It wouldn’t necessarily be coherent, being 80° off the side of the stage, to have a spatialized sound image. No doubt, it could work for a musical, much less for rock.

A view of the stage during the rehearsals. The L-ISA zone is large, but the organizers also need to study a way to offer even more seats facing the stage and fewer on the sides.

SLU : Is L-ISA rather conceptual and open, benefiting from the development cycles of any manufacturer?

Guillaume le Nost : Of course. We are working on a pipeline of features that we will keep us busy in implementations over the next two years. Our job is to prioritize which features will be the most important and fastest for users, so we are constantly listening to feedback from the field. As a result, the product has evolved significantly over the last 12 months, with the improvements being developed very quickly.

L-ISA, a small team in London

SLU : You mentioned a very small team… how many of you are there?

Guillaume le Nost : There are six of us in R&D, plus Sherif El Barbari and Christian. We also have a team dedicated to Private applications, namely the design and installation of exceptional private facilities.

An image taken from the official website of L-ISA, in the auditorium at Marcoussis with Sherif at the controls – a place that used to be forbidden to enter, especially to journalists. There, L-ISA took its first breaths…

SLU : I imagine that the difficulty in this, apart from the legitimate cost of such a private installation, is having the discrete sources to be able to exploit it fully.

Guillaume le Nost : Think again (he smiles). We are experimenting with a lot of. I can’t say any more, but I am confident that, even in residential installations, it will be possible to have fun with L-ISA. We already have upmix solutions to convert from stereo to more channels. These are very convincing, except perhaps with classical, and can be heard in our showroom London. L-ISA has huge potential, even in residential applications.

SLU : Tell us about how L-ISA was born, from within?

Guillaume le Nost : It started from a three-person team in a small space next to the warehouse in Marcoussis, where we spent a lot of time, a whole lot… (he smiles).

SLU : And commandeered a package of Kara!

Guillaume le Nost : We’ve definitely got the full support from within the company in general and from the applications department in particular. L-ISA may not be appropriate for all applications or for certain aesthetic choices of shows. We are used to hearing rock shows with two big systems, it will take time to change the situation. We will have to mix while asking ourselves “how do I use the natural properties of the brain to distinguish different sources in space”.
These properties are usually very under-exploited in sound reinforcement, yet they work very well. Some sound engineers, who have tried it, almost no longer touch their faders since, spatially, we localize the instrument without the need to highlight it by level or other means.

L-ISA. Controller or mixing console?

A view of the L-ISA controller with the points highlighting the different microphones picking up the choirs. This automates a number of functions, such as “width”, the width of the sonic image of each point, or “distance”, the distance of the sound from “behind” in the mix determined by its attenuation in level, its brilliance in the high end, and the increase of the amount of reverb compared to the direct signal. It is therefore quite easy to intervene creatively in the mix.

SLU : How would you reinvent the mixing console for L-ISA?

Guillaume le Nost : For us, the L-ISA processor is just a spatializer. All the functions of a mixing console – EQ, gain, dynamics and effects – remain fully relevant and necessary, but inside the console.

SLU : But, for example, couldn’t the linear faders become knobs and leave this space for functions specific to L-ISA?

Guillaume le Nost : Yes, for example it would be convenient to map the distance function of L-ISA onto the faders. We are currently working with DiGiCo to integrate and map the controls. We are in the testing phase and, by the end of 2017, DiGiCo should release an update with a L-ISA mode for the entire SD range. It turns out that we chose communication protocols that are very similar, so the link is natural, especially since both teams are based in London. Sherif has always had a good connection with DiGiCo.

SLU : Could you give some examples of functions that will be integrated into the console?

The implementation of L-ISA on a DiGiCo console with the specific controls immediately available.

Guillaume le Nost : For each channel on the console, the user will be able to operate in normal mode, with a pan, or in L-ISA mode with, in the small slice of screen, a small object that can be directly controlled with a zoom.
All the parameters will be directly accessible, the pan, the width, the distance and the elevation, plus an output dedicated, for example, to the sub or, if you choose, to drive the outfills from this auxiliary. It will be possible either to drive the subs individually from this output or to select the outfills.

SLU : But what do you manipulate, the groups or the channels?

Guillaume le Nost : We started from the idea that “1 input/microphone = 1 object”. This provides perfect and very direct control, and the visualization will be very elaborate, without the need to go look on another screen – our controller screen – except when you want to have an overview of the signals in the space.

SLU : You could use one of the screens of the console for that…

Guillaume le Nost : Ideally we would, but it would require integration of the L-ISA controller software into their console. For now, we have chosen to integrate a kind of small L-ISA plugin in each channel.

How the “objects” work

SLU : Who controls the movements of the objects?

Guillaume le Nost : First of all, we put the package on the movements to provide flexibility and ease of implementation. For example, movements can be implemented over a network, and not just simply within the controller. You can have musicians on stage that trigger interaction sequences via a dedicated plugin. Working with objects greatly simplifies the use of spatialization, since the position of these objects and the diffusion configuration are completely independent of one another.

SLU : Is the movement of objects – which requires a lot of processing resources – limited in speed or number? I’m thinking of an artist who would be given this command and who would make a very intense use of it.

Guillaume le Nost : We’ve anticipated everything and conducted a lot of stress tests. We have crossfade times that make moving from one situation to another very smooth. A sort of a compromise between precision and absence of clicks.

SLU : So, the “Elevation” parameter has appeared…

Guillaume le Nost : Yes, this is one of the latest features, where we achieve truly 3D setups. L-ISA is much more flexible compared to Atmos – which is nothing more than 7.1 plus a few objects that move, which also remains formatted for cinema and designed for rooms with well-defined shapes.

The 360° L-ISA installation with many elevation points, installed for the LAB, a domed venue at the New York Panorama Festival.

We can go from a very limited and very simple flat deployment to what we did this year for Coachella: a half-sphere like the Géode, but much larger and with countless video projections.

SLU : Quite the opposite of this Requiem…

Guillaume le Nost : Here, instead, we wanted to achieve an L-ISA Live reference deployment, with a guarantee of results and respect for the sound and the audience. We’re looking to achieve a qualitative standard, but creative designers are also interested in the possibilities offered by L-ISA. This is the third applications branch called Creative – after Live and Private – and it will include museums, events and contemporary art, for example. The users love the workflow based on objects, since the design and pre-production phase is facilitated by a configuration, for example, using nine points, before switching to 20 points, simply changing the setup of the speakers.

An isometric view of the L-ISA installation at Lille. To allow the ONL to discover the principle of hyperrealistic immersion, small stacks of three Kara each were placed in the Nouveau Siècle auditorium, the resident hall of the Orchestra, for a full-scale test.

You can simulate a 360° space in the studio and work on spatialization with only four points. You will have more precision and latitude of movement in the final system, while a four-point deployment only works if you are right in the center. L-Acoustics’ advantage is also the know-how in large scale installations. It’s not just about coming out with a clever processor, but about setting up rules, training and supporting users so they can benefit fully from its advantages, especially in Live.

L-ISA can be learned

SLU : Yet, for the moment, there is no specific training for L-ISA…

Guillaume le Nost : There is no formal, planned training at the moment but, of course, we train all those who will be working on L-ISA configurations. At the same time, we are developing a specific training center, especially since it is an aspect that is very important for Christian and, on in addition to knowing how to operate the controller, the sound design of the system must also be understood.

A new way to conceive mixing. The singer (green) will be amplified by three arrays – 4, 5 and 6 – and this is shown on the controller screen where the green object ‘C’ has a “Width” that extends beyond the three lines representing the three arrays of speakers. He will also be in the foreground. The other musicians will be reproduced by other arrays and also by those that amplify the singer. Remember that the seven arrays each open at 110° (or more with Syva enclosures) and overlap.

SLU : There should be two modules: one for the sound engineer and the other for the system engineer…

Guillaume le Nost : Mixing is quite simple and intuitive. I think that we’ll start with the training on the deployment of the system, especially since this is where we can do instructive demonstrations for touring system engineers, installers and decision makers, and we will show that L-ISA is no more expensive than a traditional system, only a different logic of energy distribution at equal SPL. I am also convinced that we have to forge closer links with sound schools to reach younger generations, who are more flexible and open to new things.

SLU : Do you think we should also talk about sound aesthetics?

Guillaume le Nost : Actually, it’s essential that we do. With L-ISA we can localize the amplified source directly above where the acoustic one is visually, which generates a result that is both natural and transparent, but we can also choose another path, where the sound becomes immersive and localized off-stage.
We can oscillate between an aesthetic that is almost imperceptible to another that is obviously very amplified. Or, we can perform technological feats that go almost unnoticed, like in the show “Le Dernier Panache” at the Puy du Fou theme park, where the scene turns but the sound remains with the audience, thanks to the use of a grandstand tracker and 24 active channels. The funny thing was that, on the first day, we had a bug that turned the system in the opposite direction of the scene. It was a sign problem in the code that we settled in less than half an hour (he laughs).

And my subs, they won’t budge…

SLU : How do you manage the subs with moving elements? Part of the spectrum must be sent to them…

Guillaume le Nost : We have very high level parameters in the software to define the sections of the space that will contribute to this signal, in order to preserve temporal alignment with the heads. If a source is completely in the back, it will not be at all in sync with the sub, and we may perceive two sounds, which can be disturbing. For example, you can decide that only the frontal sphere contributes to the sub, and the setups will manage this in a very gradual way for the group of objects. We try to limit the number of parameters for each object, since they multiply the number of processing channels by 96 each time.

A pop/rock application of L-ISA, on the Renaud tour. Number 19 is the “Renaud object” (I don’t think he would like being called this – ed.), placed alone and completely in front of the others. Behind him, we see the different stems. No Pan and very little Width. Notice also how Madje (Malki, FoH for Renaud – ed.) made the choice to modify the spatialization for each song.

SLU : On the other hand, you don’t change the spectral response of what goes in the background. The sound will then be clean of any sub contribution and will “resume” when it comes back into an acoustically compatible area.

Guillaume le Nost : That’s a choice. We don’t know if this is the best approach, but it works well.

SLU : Tell us about the distance parameter, I was not convinced in rock’n’roll…

Guillaume le Nost : In rock, we don’t necessarily want to correct the space, and this control is not an aesthetic reverb. We conceived it as a tool for merging the system into the global space. This is not very interesting in reverberant rooms but, for example, here in Lille it works well, and François makes very good use of it. It works on the parameters of level, EQ and the ratio between direct and reverberant fields.

Calculating the sounds in the processor

SLU : Are stereo and LCR downmixes mono-compatible, and is the phase of this little world clean?

Guillaume le Nost : Absolutely. We worked a lot on the phase, especially for the main system where an object can be very punctual in one array – and therefore present no danger – but can also be extended and therefore reproduced by several points of diffusion. For this we have our internal recipes – algorithms with a philosophy that has guided all our choices and with the priority to preserve the impact. We have tried different strategies of spatialization, and the one on which we concentrated the most maintains this impact and the line array effect to which all our clients are accustomed.

The L-ISA Processor or, more prosaically, the core where 96 MADI streams are processed and redirected to 32 outputs, still in MADI. The rest is hush-hush and buttoned lips.

One system is very coherent and, as soon as one expands into the two adjacent ones or beyond, there are no problems or very few. We minimize them. This is one of the successes of this development. We spend our time asking ourselves what is gained in becoming more complex and what do we lose by keeping it simple.

SLU : The choice of 96 in and 32 out – of which 7 are dedicated – was this easy to make?

Guillaume le Nost : We thought about this a lot. There are other products on the market that are 64×64 but, as we want to push mainly live applications, we prefer to process more sources and limit the diffusion points, the latter of which also require a lot of time and financial means to implement. Also, putting out a lot of sound, for example in the rear – let’s say 16 points – is not rational because the people in the near field will perceive a high level, but those in front of the stage will have too little. We really want to rationalize multichannel design and, seeing where we want to hang them and how we want the diffusion points to work, we can’t go beyond 32.

SLU : Yes, but we both know how the customers are. Will it be possible, if necessary, to link two processors?

Guillaume le Nost : We can imagine a lot of things having more inputs or outputs. There is now a version of the L-ISA Processor for Creative applications that offers 96 inputs to 64 outputs.

SLU : We understand that you have limited the operation of the processor. Is it therefore not possible to make it sound bad?

Guillaume le Nost : Insanely bad, no; we have been quite restrictive on this point. The strangest thing one can do is open a source to 90°. That’s wide, but we did not want to go, for example, to 360°, although, technically, the algorithms would allow it. Enlargement in width is the most complicated function in terms of calculation. For the time being, it is preferable to limit this possibility in order to guarantee the most qualitative operation possible.

A Renaud tour date with a large number of Kara in five flown arrays. 11 each in arrays 1, 2, 4 and 5 and 13 in the central one dedicated to the voice and, habitually in rock, to bass and drums. These habits will evolve so as not to limit the surface of the membranes moving the air and, therefore, the SPL.

L-ISA, lots of enclosures you don’t have to forget

SLU : But you have to know how to “open” your sources to take advantage of the speaker configuration the device is driving, and exploit the theoretically available SPL. The drum set can’t come out of only the center array or from the one corresponding to its location on stage.

Guillaume le Nost : I see you have understood… hence the need to rethink one’s mix for a certain configuration of speakers and no longer just for a simple left/right. If we hang clusters, it’s so that we can use them. This is an important point.

SLU : What ratio of SPL increase do you achieve by using more than one array?

Guillaume le Nost : I don’t have the formula memorized anymore but it is written down.

SLU : Is L-ISA a possible answer to A/B sound designs for musicals?

Guillaume le Nost : : I am convinced that it is a solution for A/B design. This does not solve the problem of elevation (which also exists in A/B – ed.) but for all the interactions that led to the adoption of that method – which is rather bulky and constraining to implement and control on a daily basis – L-ISA is a solution. We are currently working on this issue specific to musicals but natively, if each microphone feeds one array, the question of comb filtering is settled. However, we can do even better (and his eyes sparkle- ed.).

An excellent salesman of concepts as well as creator of them, Guillaume le Nost, at P. Mauroy stadium.

SLU : Is L-ISA a system whose time has come?

Guillaume le Nost : Clearly. We listen a lot to binaural headphones. Virtual reality is here. Surround and Atmos are becoming more and more popular. Consequently, listeners have expectations that change too, especially when we look at the price of a show. We expect a higher quality sound experience. There is, of course, an SPL below which people do not have fun, a “physiological” rule that we will not bring back into question, but between this level and 10 dB more, we prefer to propose something more interesting and creative.

SLU : You seem passionate about L-ISA.

Guillaume le Nost : This is a very interesting project that touches a bit on sound design, that calls for the ideal speaker enclosure for this mode of diffusion, that requires changing mixing habits… there are very complex themes for which there are still plenty of answers to find.

Here ends the second part of our report on L-ISA. If you missed the first, see the link here..

Dulcis in fundo, we had the pleasure of interviewing Christian Heil, with his exciting and uncompromising analysis, explaining why audio must evolve and offer more and better to take back its importance. Like, for example, L-ISA.
See you soon for this third and last part.

[/private]

The orchestra and choirs during rehearsal, nearly 300 artists for a masterful opus.

L-ISA, Immersive Sound Art is Christian Heil’s latest sonic undertaking, with which he intends to restore the nobility of audio, the link between the artist and sound reinforcement. In this three-part report, we discover the innovation of L-ISA and its great potential.
Verdi’s Requiem, Jean-Claude Casadesus and the Orchestre National de Lille (ONL), and Lille’s Mauroy Stadium served as a showcase for L-ISA, the new multichannel system driven by the L-ISA Controller.
We had the pleasure of exploring this system during a long day of listening, interviews and informal exchanges – each of them fascinating – with ONL management and sound and the L-ISA team, including an appearance – rare enough to be worth the trip – by Christian Heil.

The team who brought color back to Verdi. From left to right: William Duvet, systems engineer and liaison between ONL and L-Acoustics; Guillaume le Nost, L-ISA head of R&D; Fred Bailly, applications engineer, touring for L-Acoustics; Christian Heil, founder and president of L-Acoustics, father of coherent wave fronts etc, etc; Sherif El Barbari, head of application at L-ISA; François Gabert, sound engineer for ONL and mixer for the Requiem performance; Erdo Groot, producer and classical sound engineer in charge of following the score with François; and finally Stéphane Evrard, technical director for ONL.

Half a stadium with no turf can make the crowd even more “vibrant”

[private]

The location is magnificent. Imagine a large, covered football stadium with 50,000 seats. Take the turf off the playing field, lift it, and slide it over the other half of the field using machinery that is as improbable as it is hydraulic. Deploy a heavy hanging partition to divide the two huge volumes, magically conjure up some new retractable grandstands, and you get the “arena configuration” of Pierre Mauroy Stadium. 25,000 seats – a total success with good, viable acoustics, plenty of rigging points, and a concrete floor ready to load everything in.

The other side of Pierre Mauroy Stadium, where they store the half of the playing field that‘s been removed to transform the Stadium into the Arena. We suppose that it’s in a sort of tray supported by legs equipped with wheels. Note the cooling fans and the technical package designed to dress up Lille’s football players playground…

When we arrive in the morning the rehearsals are already in full swing, which allows us to understand the wide coverage area of the L-ISA deployment and how the K2 side hangs and the Kiva covering the first rows would behave with the inevitable zones of interference.

The ONL’s Lawo mc²56 in the background and the pair of PCM96s in charge of wrapping up and presenting 300 artists for an audience that is accustomed to concert hall acoustics…

We’ll come back to this later, but now, let’s take a moment to tip our hat to François Gabert, at the helm of the Lawo mc²56. Between the morning rehearsals and the evening Requiem – a unique piece, uninterrupted and without the slightest pause – his mix went from shadow to light. Enthralling.
A quick reminder that the ONL is well and thoroughly equipped with Lawo consoles and Neumann, Sennheiser and Schoeps digital and analog microphones in order to be able to record its work.
Two mixers reside in ONL’s home auditorium, ‘Le Nouveau Siècle’, in the heart of Lille, and it’s is only natural for them to be here in the Pierre Mauroy Arena for this concert. The L-ISA system comprises five arrays spaced above the stage, with two additional extension arrays flown outside of the stage area. Each of the seven arrays is composed of nine Kara which are reinforced by a single central point of subs hanging vertically above the first rows, consisting of two arrays of four KS28 each, in cardioid configuration.

The L-ISA system and its seven arrays of nine Kara each – 63 speakers in all – reinforced by a cardioid array of eight KS28 subwoofers and, finally, by two side hangs of eight K2 each, one for each side.

The amplification, at least in part, is flown close to the systems. One of the LA-RAKs is the “II” version and includes three LA12X amplified controllers, which are essential for the KS28 units.

Knowing the power of the KS28 – which the manufacturer rates at 3 dB SPL more than the SB28, but when driven by the LA12X gives much more, not to mention its dynamics – we will not be left wanting for the bottom end of the spectrum.
Two arrays of eight K2 each, one left and one right, provide the connection between the edges of the zone covered by the L-ISA system out to the last seats perpendicular to the stage – seats which will enjoy only a good monaural sound.

On either side of the L-ISA deployment are four LA-RAKs, including an LA-RAK II for the KS28. The amplification for the front fills and side hangs, as well as for the monitors, is deployed on the floor. We’ll only be talking about the front-of-house sound in this article but, of course, carefully placed monitors allowed the chorus in particular to hear the orchestra, not just the timpani, drums and other brass that are positioned right in front of them!

Fred Blanc-Garin, or how to talk about sound…

The first to answer our questions, despite a tight schedule, is Frédéric Blanc-Garin, the technical alter-ego of François Gabert with the ONL. Like Gabert, he is a contract technician and, alternating with François, he takes care of the Orchestra’s digital studio and, inevitably, of its outings like today.

The two accomplices caught in the act. On the left, Frédéric Blanc-Garin and, with his hands on the faders, François Gabert.

SLU : How did you find yourself at the helm of this little-known system and why L-ISA?

Frédéric Blanc-Garin : While on tour with Paul Personne – I don’t mix only classical music – I had the opportunity to work with a Studer Vista that has a very interesting feature: the Virtual Sound Panner or VSP, which allows you to really open up your mix without affronting the audience and without losing energy as you would with a standard pan. I really enjoyed myself with guitars that maintain an understanding of the musical message while preserving a true identification of the sound source.
Back from the tour, I told François (Gabert – ed.) about it, pointing out that our Lawo didn’t have it and that Hervé de Caro (sadly, since deceased – ed.) could do nothing for us. Last November, during a job with William Duvet, he told me that a system was being set up at L-Acoustics, which he called “stereo for everyone.”

My interest piqued, I spoke about it with the orchestra and François went to listen to it in the auditorium at Marcoussis. He came back convinced by the audio but less convinced by the workflow. The next step was – again thanks to William and Fred Bailly of L-Acoustics – to deploy a mini L-ISA system in the auditorium at, Nouveau Siècle, so that François could mix and become familiar with it, and so our management could decide whether or not to adopt it. We mixed an L-ISA version and a stereo version, playing the latter at a higher level but only on the speakers of the 3rd and 5th arrays of the 1 to 7 extended deployment, the same we have tonight.

A few minutes before the show, Fred Bailly speaks with William Duvet.

SLU : What speaker enclosures did you use?

Frédéric Blanc-Garin : Clusters of three Kara, each on pole mounts. We first played the stereo mix for our director who, seated in the balcony and at a good distance, found it quite narrow and a little closed. This was explained by the absence of the orchestra and the fact that we would normally reinforce what is missing. We then switched to the multichannel system.
He then got up and wandered all along the balcony to listen to what would be perceived from each seat, and to verify the absence of any golden triangle as you would have in stereo. He then came back to me and said ” it sounds like the orchestra is here….
We got the green light, and it was technical director Stéphane Evrard who took over.

SLU : L-Acoustics assures that the SPL and driver in the L-ISA system are not sold at a higher price than a traditional system.

Stéphane Evrard between François Gabert (left) – who says “without him, tonight would be left/right!” – and Erdo Groot, who assists François in following the chart, which is nothing more than an annotated score.

Frédéric Blanc-Garin : This is essential because it was necessary not to exceed the 2016 budget, in which we had flown four arrays of K2. The whole system, including the rigging, fits in the budget.

SLU : I guess you’re talking about Alive, the audio vendor tonight. William Duvet, who was the intermediary, works there?

Frédéric Blanc-Garin : Yes, he does contract work for Alive, with whom we collaborate regularly; I do a lot of things with him. He is a big believer in the Heil House (he smiles).

The sound processing in the DSPs

SLU : L-ISA is an idea based on an algorithm. As far as the sound is concerned, do you find that the ratio of benefits to drawbacks is satisfactory or better than satisfactory?

Frédéric Blanc-Garin : Yes. There are some phenomena, but these are slight and it deserves to be listened to with other pieces of music and in other places to get a better idea… and, then again, this algorithm is constantly evolving. This is a first experience that will lead to others. While walking around the room and in terms of emotion, I have a much more satisfying impression when I’m inside the coverage of L-ISA than outside it, and I’d have more to say about the problems generated by two standard arrays and the stereo beast than I do about this configuration. I had the same feeling about the digital microphones when we adopted them.

One of the many digital Neumann condensers, a KMD preamp used with a KM185 capsule sporting the now famous light blue badge and the inscription “AES42”, the standard for transporting digital audio signal stream and a service data stream that allows you to vary the settings directly from the Lawo console, which has also adopted it.

Apart from quantifiably better performance, we’ve discovered that the digital microphones are more defined in the mix and the off-axis pickup is naturally attenuated without having to roll them back, so the sounds become cooperative and the mix comes together better.
With L-ISA, we achieve the sensation of bringing the listener closer to the musician, without the need for opera glasses. We bring him more emotion, and that’s our mission. Finally, the rear attenuation of this system will allow us, if we continue, to bring a little more “air” to our sound by pulling the mics up a bit next year, including on the violins, which is always good in classical.

SLU : Tell us a little about the micing. You and François have a magnificent stock of digital microphones…

A pair of MKH8090 mounted on their MZD8000 preamp/converters… or “how to digitize a great microphone”!

Frédéric Blanc-Garin : We mainly have digital Neumann and Sennheiser mics and, when a record label comes to make a recording of our orchestra bringing their own main pair, these allow us to add some closer-miced signals with different colors. For this Requiem, we’ve added about twenty transducers to our setup.
The sound of the choirs is captured using pairs of MKH8090s mounted on MZD8000 preamp/converters and KM 183Ds for the outer pairs. Initially, we would have preferred to have only the Sennheisers but, since the Neumann has the same directivity and we were a little short on microphone pairs, we mixed them. The rest of the orchestra is also picked up using Neumann and Sennheiser digital microphones, including TLM 103Ds and MKH8040s.

François Gabert, the man who runs 300 artists through 101 enclosures.

We let Fred get on with his job, since François Gabert joins us to spend a few minutes behind the scenes, despite a schedule as packed as a semi on tour!

François Gabert

SLU : How did your collaboration with the ONL come about?

François Gabert : I started doing occasional jobs for them a good decade ago and, after Lille 2004, I left to do other things.
Four years ago, I came back because Fred (Frédéric Blanc-Garin – ed.) – who is a pure technician, while I am much less – asked me to. He had the idea of a digital studio with the Lawo and 60 digital microphones that convinced the ONL because it brought a lot of notoriety through quality recordings.
I’ll take this opportunity to say that we’re going to develop a specific YouTube channel for the Orchestra that will give it even more outreach.

SLU : It’s going to be a change from the make-up tutorials and video-game walk-throughs that abound on the web (laughs)

François Gabert : Don’t talk to me about that, it’s a problem for me too.

SLU : Is amplifying a classical orchestra something natural and does L-ISA contribute to it?

A design that already shows what can be improved in future editions, so that even more of the audience is in the L-ISA coverage area. Alas, in classical as with popular music, to be near or on the sides, means not hearing as well…

François Gabert : I’ve toured a lot with bands, 25 years of working with many different systems, and I always make sure that the audience can hear at least as well as I can, especially in the bass and infra-bass. For example, I’m a fan of the arced subs, even if this configuration softens the attack a little bit, but it makes up for it because it guarantees everyone an equal tonal balance.
With L-ISA it’s a bit like that, we can allow 65, sometimes 70%, of the people to hear a realistic spatialization and true stereo. Real definition of the sources in the space. It’s a whole different thing than having a dual mono with two stacks or a stereo that is only effective for 10% of the audience.

SLU : In this case it would have been better not to put seating perpendicular to the left and right of the stage (smile).

François Gabert : We can’t do everything just yet (he laughs), but in an ideal world, it would have been necessary to move the system back to gain some width in the coverage and more seats in the optimum sound zone. However, we will be playing for 14,000 people, so we even left the upper tier empty. We have coverage right up to the ring of sky boxes.

The Pierre Mauroy Stadium in “arena mode”, photographed from the top of the second tier of bleachers. The third (upper) tier was not used for the Requiem, which attracted 14,970 spectators!

SLU : This room is remarkable, by the way, it is almost dry.

François Gabert : Without the audience it’s not, it’s a little live, but once it fills up, it indeed has a very short and clean reverberation time. It’s quite surprising. The first time we came here with the ONL, I almost got taken in and, at the beginning of the concert, I had to send much more reverb than during the soundcheck.

SLU : Do you think that using L-ISA to localize the on-stage sources is a relevant and acoustically interesting solution?

François Gabert : Much more so than with a stereo pan, where the choice of “placement” of the source is valid only for the spectators seated right in the center of the triangle. Physically, there is no ideal solution. With L-ISA we are coming closer to an acoustic solution; let’s say not even amplified, a concert hall where everyone does not have the same sound image, but where everyone has a spatialization related to where he is.

Sound and lighting FoH, entrenched low in the stands facing the stage and the system.

I may be shooting myself in the foot, but amplifying an orchestra is cheating (F. Gabert)

François Gabert : I may be shooting myself in the foot, but amplifying an orchestra is cheating. With L-ISA it is possible to recover some of the natural acoustics that an orchestra can have in a concert hall.

SLU : The ideal would be to put an amp inside each violin…

55 intense minutes and then the release. Since it would be difficult and improbable to do an encore of a Requiem, Francis can also enjoy the applause of a delighted audience.

François Gabert : Yes! It’s sort of what we are doing here, since I distribute the elements of the orchestra in the Kara systems based on their placement on stage. I’m only lacking a little airiness in the sound from the mics to sweeten up the instruments with a natural sound that can be very harsh in proximity. The little harshness you mentioned off the track in the violins, is due to this.
This is not because of the algorithm of the L-ISA processor or the digital microphone, but the difficulty of maintaining a distance from the source while, at the same time, maintaining the level and the precision in the room. To conclude, L-ISA is not perfect, but it’s infinitely better for the audience than the usual left/right.

William Duvet, the good advice man

William Duvet, a technician who played a key role in bringing L-ISA to the ONL. We made sure to get him to tell us about his role as facilitator, without which Verdi would have had only two arrays.

William Duvet with, to his left, the computer that controls the L-ISA Processor.

SLU : Apart from really loving L-ISA, and spreading the word about it, what do you do?

William Duvet : I am a technician, mostly focused on systems, but I mix as well. I work in northern France and in Paris for Audio Intelligence.
For some years, I’ve been using L-Acoustics products and in early January, after surfing around the web and discovering this system, I spoke with Fred Bailly (super application engineer for touring at L-Acoustics – ed.) and I thought it might have been interesting for this show, which had already been done three times with a traditional sound system.
So, with the help of L-Acoustics, I acted as a liaison with the ONL, even though I had never worked for them before.

Sherif El Barbari, the L-ISA field man who, as Christian Heil will tell us later, never misses an opportunity to measure, listen and discuss this system.

SLU : But you are from Lille and are here on behalf of Alive, the audio vendor at this event.

William Duvet : Yes, absolutely. I’ve also worked at Mauroy Stadium with other vendors.

SLU : That still doesn’t tell us how you got into the operation with the ONL.

William Duvet : That was through Frédéric Blanc-Garin and François Gabert, who is a colleague. Frederic found the idea interesting and set up a meeting to discuss it, then he wanted to meet with L-Acoustics to move the project forward. In the meantime, I was trained by Fred Bailly and the wheels were set in motion, up until the final bidding and the choice of the vendor in charge of deploying the L-ISA system in the stadium.

The LA-RAKs, then nine Kara of the number 1 array and the eight K2 of the stage-right lateral array.

SLU : What is your role in the production itself?

William Duvet : I managed installing all the gear, verifying the cabling and rigging with L-Acoustics and, finally, assisting François to allow him to find his space and his console pans in the processor.

SLU : Exactly, why is the controller display screen on the side and not facing the mixer?

William Duvet : François wants it this way. He doesn’t want to see it or play with it during the show. We decided on the placement during the soundcheck. If, in the end, he decides to make changes to, I’ll take care of it for him.
He wants to keep his hands on his faders and not be disturbed, let alone pick up the mouse to do it. Integrating the L-ISA plug-ins into DiGiCo consoles will make it remarkably more comfortable.

SLU : How did you go about setting up the matrix for the show? I know that “matrix” is not an accurate term, but we understand each other (smiles).

William Duvet : We started from what we did in trials last month, during testing. We set the patch of the show and we worked on it like a virtual soundcheck, with the multi-track recordings of last year’s show. This is not the same piece of music, but the microphones, console and stage were very similar to the concert tonight, and it gave us an initial idea. Then we used the recordings of this morning’s rehearsals.

SLU : How many inputs are being processed?

William Duvet : 84, two of which are wireless mics for announcements.

The 84 “objects”, as François positioned them in the controller. At the bottom of the screen are those that are most similar to the original signal; at the back, those that are processed by the ambient algorithm incorporated into the processor, and attenuated in level and presence to simulate the distance. Note also the “wings” on either side of each object, which represent the extent of both sides of the system or systems that reproduce the source and, therefore, the width of the image.

Not such a complicated setup

SLU : Can you explain the setup to us?

William Duvet : The Lawo console outputs MADI to the L-ISA processor. After processing, this sends the signals, still in MADI, to an RME converter that converts the output channels in AES/EBU, and it is these digital outputs that will drive the amplified controllers that feed each array separately. We use the MADI clock.

SLU : Where is the L-ISA processor?

William Duvet : It is at the stage end, near the stage rack of the Lawo. It would be useless to have it here.

Again, the screen of the L-ISA controller. All the way down, beneath the numbers 77, 78, 81 and 82, the reverb effect returns in the foreground: the “distance” is set to zero, but the “Width” is set wide enough to create a very broad and overlapping front.

SLU : At what sample rate is it all running?

William Duvet : 48 kHz. We could have switched to 96 kHz, but the console was deliberately kept on an older software version. Updating just before the show would have been…

SLU : Absolutely agreed. Why are there eight zones on the L-ISA controller screen?

William Duvet : There are eight sectors and, if you look closely, seven dotted lines that correspond to the seven Kara arrays. It‘s a graphical representation that facilitates the localization of objects in the sonic image. But don’t forget that each array opens at 110°. We’ve also used the auxiliary outputs of each object to create the mix for the fills, a mix which is adapted in level and content. The sub channel is simply the mono output of all objects.

SLU : What are your plans after this production?

William Duvet : System engineering for Intelligence Audio.

SLU : Another northerner invading the capital ;0)

William Duvet : (he laughs) Yes, and it seems that there are lots of us! That being the case, the North is also moving and there is plenty of work right here, too.

Jean-Claude Casadesus, the ONL, his four soloists and his 200 singers greet a captivated audience. The power of the Requiem – emotional as well as sonorous – has been perfectly reproduced by the L-ISA system.

After the description of this beautiful production of classical music, we’ll move on to the technical aspects in the second part of our report, with Guillaume Le Nost, head of R&D at L-ISA and, finally, to the interview with Christian Heil in the third and final installment.

[/private]