Audyssey MultEQ XT, XT32, PRO, Applet, MultEQ-X (màj)

[Hugo S]

Bjr,

Une petite explication complémentaire sur le mode de fonctionnement du Audyssey LFC (Low Frequency Containment). c'est issu de du site US TWICE:

Audyssey Readies "Do Not Disturb" Mode
By Joseph Palenchar -- TWICE, 5/17/2011


Los Angeles - A new post-processing technology developed by Audyssey Laboratories for AV receivers and HTiBs will let consumers enjoy bass late at night through their home theater without disturbing the sleep of people in other rooms.

The technology, called low-frequency containment (LFC), will appear in AV receivers later this year from Onkyo, Denon and Marantz, Audyssey told TWICE. Consumers will press a button on a remote or on a product's front panel to engage LFC mode.

The technology addresses the ability of bass waves below 200Hz to penetrate walls more easily than high-frequency waves, said Audyssey co-founder and CTO Chris Kyriakakis. To keep the peace with family members and other apartment dwellers, LFC dynamically limits bass output according to volume changes in program material. At the same time, however, LFC artificially synthesizes harmonic tones based on the lost fundamental bass tones "to give consumers the impression of improved bass," Kyriakakis explained. These higher frequency harmonics, he noted, are less capable of penetrating walls than fundamental bass tones.

The synthesized harmonics are added on top of the existing harmonics of the program material's bass tones.

Walls muffle high-frequency sounds by 20dB more than they muffle bass frequencies, he noted.

LFC is one of three audio post-processing technologies that Audyssey plans to announce in the coming months.

C'est une système qui malgré son nom, risque de faire pas mal de... bruit...

Hugo

PS: http://translate.google.fr/

[djdactylo]

Grosso modo, çà va analyser les fréquences de résonances de la pièce.
Les filtrer de manière bourinne et booster les adjacentes pour pas avoir l'impression d'avoir perdu ses graves.

Cà peut marcher pas mal mais bon, espérons que çà ne va pas déformer tout le spectre.

Aussi, je suppose que c'est pour ceux sans EQ pro, car dans notre cas, les fréquences de résonances sont déjà filtrées....

DJ

[Hugo S]

Bjr DJ,

Grosso modo, çà va analyser les fréquences de résonances de la pièce.
Les filtrer de manière bourinne et booster les adjacentes pour pas avoir l'impression d'avoir perdu ses graves.
Cà peut marcher pas mal mais bon, espérons que çà ne va pas déformer tout le spectre.
Aussi, je suppose que c'est pour ceux sans EQ pro, car dans notre cas, les fréquences de résonances sont déjà filtrées....
DJ

Si on relit le complément précédent on voit que LFC est une surcouche posée au dessus du MultEQ, que celui-ci soit XT, XT32, voire l'un des précédents calibré (avec) Pro.

Ce qui y est indiqué en fait, est que LFC limite dynamiquement, donc en temps réel, toutes les fréquences en dessous de 200Hz. Donc une bande de fréquences 20 -> 200Hz qui a déjà été "égalisée" par MultEQ et ceci en particulier pour les diverses fréquences de résonance de la pièce.

Sachant que dans ce process LFC, toute cette bande 20-200Hz qui a été atténuée, est ensuite "harmoniquement" crée, recrée ou réajustée au dessus de 200Hz.

Très concrètement on écrête, probablement à partir d'un certain niveau, donc pression sonore, toutes les fréquences jusqu'à 200Hz. Puis on décale au-dessus de 200Hz toute cette partie décalée, en synthétisant "adroitement" (et c'est là que doit se trouver tout le savoir faire d'Audyssey)-, leurs harmoniques à partir de 200Hz.

Est ainsi limitée, l'énergie, donc la pression sonore liée aux fréquences < 200Hz, mais subjectivement ou psychoacoustiquement d'un point de vue "audible", l'écrêtage de ces fondamentales est faible...

Sauf peut-être dans toutes les install où la partie sub est adéquate et où ces fondamentales sont également ressenties dans tout le corps. Mais là encore il est possible de restreindre cet impact négatif en affinant la valeur du seuil d'écrêtage, qui pourra(it) être finement adaptée, spécifiquement aux besoins de chaque install.

Maintenant pourquoi disais-je que ce système allait faire du "bruit"?

Tout simplement parce que le LFC est à l'opposé de la tendance actuelle qui consiste à s'équiper du plus gros sub pour reproduire toutes les fondamentales, y compris les plus basses. Avec ce système LFC, ce type de sub n'est en réalité plus forcément aussi indispensable...

Car ici, la question qui est en réalité posée et pour laquelle le LFC se pose comme réponse est: qu'est-ce qu'une fidélité sonore?

Est-ce la stricte reproduction de l'original "vivant" ou bien une reproduction finement adaptée à ce que cette même "réalité" soit effectivement perçue comme telle...

A cela il ne peut y avoir une seule réponse, celle-ci ne peut qu'être individuellement personnelle, donc sujette à discussions, débats.

Sorry d'avoir été un peu long et d'avoir débordé sur des terrains un peu particuliers, mais n'est-ce pas ça aussi, le rôle de HCFR?

Hugo

[palm]

Oui ce sera un peu polémique car a l'opposé de la fidélité sonore.
Perso pour une écoute a faible volume ça me choque pas

[Hugo S]

Bjr,

Intéressant post sur les objectifs de correction acoustique du MultEQ XT(32) (+ DSX) et le résultat perçu recherché (par opposition à du statiquement mesuré):

Hi Larry,

Timbre matching is a tricky thing. First of all let's start with a little experiment. Take two identical speakers and place them next to each other so that they are touching. Now play pink noise that alternates from one to the other. You may be shocked to find out that they sound completely different. That's because even a slight variation in placement changes the response. In a real setup with three front speakers that most often do not include an identical center things get even more unmatched.

Even if you have free field measurements, once the speakers are placed in the room the free field data becomes less useful. The timbre of the speaker is determined by the balance of axial response vs power response. This balance changes dramatically because of placement, listening distance, room acoustics, and most importantly: speaker directivity. In an anechoic chamber one typically measures a listening window response that involves 7 measurements on an ellipse (or circle depending on your school of thought). These measurements provide a certain standardized weighting of on- and off-axis responses that can be used to compare different speakers. It has been found to have reasonably good correlation to the perceived timbre although there are other factors as well.

In a real room we don't have the luxury of listening window measurements for each speaker. But, we can extrapolate a quasi-listening-window response if we spatially sample the listening area where the speakers are pointing. That is done by selecting the appropriate weighting for each measured response and is something that simple averaging can't do.

MultEQ treats the speaker and room as a combined system. When it measures impulse responses it's not trying to make the anechoic speaker responses match. It's trying to match them in the room they are in. This involves taking the effects of reflections into consideration. Once the measurements are collected a kind of target curve is imposed on the time domain impulses. Think of it as a perfect impulse: all the energy appears in the first spike and nothing after that. Of course, that's not going to happen in a real room. But, we can show that the filtering process pushes each measurement closer to a perfect impulse than before. Doing so reduces the energy in the blips that happen after the first impulse and therefore the effect of the reflections.

It's not true that this type of filtering reduces the low frequency resolution. The frequency resolution is determined by the length of the impulse response (the time that it extends over). The longer the response the higher the low frequency resolution. This is a fundamental property of the Fourier Transform. If you truncate an impulse response by cutting off its tail (and shorten it) then you will get what Markus is talking about.

So, getting back to timbre matching: If your goal is to do the first experiment I mentioned and expect to hear no difference then you will not succeed. It's not possible to equalize the directivity of a speaker. The focus has to be on a solution that takes into account the in-room balance of direct and reflected sound and produces filters that bring each speaker closer to the target response.
__________________
Chris

et

A time-domain response (or impulse response) is corrected by MultEQ so that it more closely resembles a perfect impulse. In doing so, the reflections (copies of the original signal that arrive at various later times) are suppressed. This correction is done by creating an FIR filter that is applied to the impulse response. When the signal in each speaker plays through this FIR filter it is convolved with it as it plays and the corrections are applied continuously.

You can see some examples of before and after time domain responses here. Click on Seat 1, 2, etc.

The effects of the reflections (that show as blips later in time) are reduced. But, that is done by the filtering operation and not by going in with "tweezers" to kill each reflection. The amount by which each is reduced depends on the spatial distribution of the effect they have on the response.

Hi Chris,

Thanks for the explanations regarding correcting the effects of reflections.

It is interesting to note that this discussion of reflections started from a question about the value of electronic timbre matching. (Refer to the postings quoted below.) I was hoping that we could get your thoughts on whether you share Markus' point that inorder to do timbre matching that you need to take free-field measurements. Obviously we're not really interested in removing all reflections to do timbre matching, but does filtering the impulse response hurt timbre matching if we lose frequency resolution in the bass region? Any general explanation of how Audyssey does timbre matching would be appreciated.

Thanks.

Larry

Hi Markus,

Do you ascribe any value to electronic timbre matching?

Larry

You're talking about speaker matching? I think this is very important. I also think that free-field measurements are hard to do if you're not in a free field but a domestic living room. For example, there's typically a reflection from the floor within 2ms. If you filter the impuls response so this reflection doesn't skew the results, the frequency resultion of the magnitude response decreases and results are only valid from 500-1000Hz and up. Earlier reflections (e.g. from the back of a seat) will make matters even worse.

Hi Markus,

Thanks for the response.

I am referring to electronic techniques that Audyssey and THX market that claim to match the timbre of speakers so that they sound similar particularly when panning the same sound to different channels. Why would that require deriving free field measurements? Doesn't matching actual in-room responses to the same target curve still assist in regard to timbre matching?

It's impossible to electronically alter a speaker's indirect sound field without compromising the directly radiated sound. While timbre is not only a property of the direct sound, compromising the direct sound will ultimately prevent accurate sound reproduction. The real challenge is to match speaker directivity to the room and application.

__________________

Hugo

PS: http://translate.google.fr/

[Woody78]

Avons nous une vague idée du % d'information qui sont reproduites sur les canaux DSX W et qui devraient normalement se retrouver sur les frontales ?
Je pose la question car pour des soucis de budget je ne pourrais offrir la même qualité d'enceinte et d'amplification a ces canaux qu'a mes frontales

[Hugo S]

Bjr,

Avons nous une vague idée du % d'information qui sont reproduites sur les canaux DSX W et qui devraient normalement se retrouver sur les frontales ?
Je pose la question car pour des soucis de budget je ne pourrais offrir la même qualité d'enceinte et d'amplification a ces canaux qu'a mes frontales

Il est quasi impossible de répondre précisément à ta question. Sachant qu'aujourd'hui apparaît également un autre contexte type 9.x ou 11.x: le DTS Neo X.

Alors concernant le DSX, ce qui est connu est que les canaux W et H servent en fait à reproduire des réflexions latérales et/ou hautes, "adroitement" synthétisée en fonction d'un algorithme propre à Audyssey. Ceci sous-entend donc un complément de signal/pression sonore qui va participer au niveau sonore global. Un niveau qui (pour une même référence donnée) demeure constant d'un point de vue de la perception sonore au niveau de l'oreille.

Du coup et toujours pour un même volume sonore global (perçu par l'oreille), la charge individuelle tant sur les L/R (voire C), que sur le WR/WL et HL/HR sera moindre. Puisque la même pression sonore (niveau global perçu par l'oreille) précédemment répartie sur 3 enceintes frontales, sera désormais répartie sur 5 (contexte W ou H), voire 7 enceintes (contexte W+H). Ceci sans parler ou prendre en compte une répartition plus homogène entre les enceintes Frontales et Arrières...

Donc dans un contexte DSX (W ou H, ou W+H), la "charge" qui sera globalement demandée à chacune des enceintes sera probablement individuellement moindre, que la charge qui leur aurait été demandée/attribuée dans un contexte standard 5 ou 7 (donc hors contexte W, H).

Ceci sous entend que pour le contexte DSX W et/ou H, on pourrait y utiliser des enceintes plus petites que des L/R. Ceci à timbralité équivalent de ces enceintes W/H par rapport aux L/C/R. Le but étant de ne pas créer un déséquilibre tonal, qui pour le coup pourrait être trop perceptible au niveau de l'oreille.

Maintenant concernant le DTS Neo X, ce qui transparaît est qu'il s'agit d'un système qui se comporterait comme le DPL IIx voire DPL IIz. Donc comme un système de synthèse de canaux. Le même raisonnement que celui tenu dans le cadre DSX, quant à la répartition de la pression/niveau sonore final perçu par l'oreille, prévaut donc ici.

SAUF que s'agissant de synthèse de canal/canaux, il est possible qu'à un moment donné, dans la répartition du signal global, l'algorithme DTS Neo X décide d'individuellement attribuer à l'un de ces "nouveaux" canaux ( WL/WR ou HL/HR), la totalité du volume sonore... Auquel cas, il vaudrait mieux avoir partout en Frontal, des enceintes équivalentes, "adéquates" à leur tâche, donc (de préférence) identiques à celles L, (C), R...

Pour ce qui me concerne, c'est dans cet esprit que je possède un ensemble Frontal totalement identique avec 7* Klipsch THX Ultra2 KL650. C'est très probablement LARGEMENT trop, mais ceci permet de ne plus jamais se poser une quelconque question, quant à une possible limitation de la dynamique sonore finale pouvant être reproduite.

Sachant que le même raisonnement vaut pour l'amplification, où la puissance individuelle nécessaire à chaque canal sera d'autant moindre dans un contexte 11, que dans un contexte 5, voire 7...

Voilà me semble-t-il, ce que l'on peut dire aujourd'hui sur le sujet.

Hugo

[Woody78]

Merci pour ta réponse complète Hugo

J avais testé le DSX W avec des petites Klipsch B2 en complément de mes RF63 et cela ne m'avait pas choqué
Pour le DTS NEO X la question ne se posera pas puisque je ne prévois pas de changer mon Onkyo 3007 pour le moment .. et que de toute façon il m'est impossible de placer des colonnes en W .. ca sera donc surement des Klipsch B3

Reste la partie amplification .. soit je conserve mon UPA5 pour les FR/C/SR et laisse les SRB+futur DSX W sur le 3007.. soit je place toute la scène frontale sur l UPA5 et SR+SRB sur le 3007
J'aurai tendance a choisir la 1ere solution car je pense qu'il y a plus d'information et de soin a apporté aux SR qu'aux DSX W

[Hugo S]

re Bjr,

...
Reste la partie amplification .. soit je conserve mon UPA5 pour les FR/C/SR et laisse les SRB+futur DSX W sur le 3007.. soit je place toute la scène frontale sur l UPA5 et SR+SRB sur le 3007
J'aurai tendance a choisir la 1ere solution car je pense qu'il y a plus d'information et de soin a apporté aux SR qu'aux DSX W

Perso je possède 2* UPA-7. L'un est dédié aux WL/L/C/R/WR, l'autre aux HL/HR/(SL/BL/BR/SR).

L'idée est de privilégier au maximum l'homogénéité Frontale. Sachant que la sonorité arrière est par définition de "nature" différente, elle permet donc une éventuelle "hétérogénéité".

Aussi si j'avais dû être dans ton contexte, j'aurais utilisé le UPA-5 en WL/L/C/R/WR et le 3007 pour le reste.

Mais ce n'est que mon humble avis.

Hugo

[slyn]

Bonjour,
J'ai deux caisson de basse JBL 3635 filtrés en actif avec un behringer DCX 2496.
Je viens d'acheter un DENON 4311 qui possède deux sorties SUB (audyssey XT32).

Savez vous si le DENON (audyssey XT32) peu gérer directement les caissons sans passer par le DCX ?

[Hugo S]

Bjr,

Bonjour,
J'ai deux caisson de basse JBL 3635 filtrés en actif avec un behringer DCX 2496.
Je viens d'acheter un DENON 4311 qui possède deux sorties SUB (audyssey XT32).

Savez vous si le DENON (audyssey XT32) peu gérer directement les caissons sans passer par le DCX ?

Tout dépend comment tu utilises ton 2496...

S'il est utilisé pour "tweaker" la courbe de réponse basses @ 25Hz/30Hz des 3635 au travers de corrections/filtres spécifiques type "B6", la réponse sera probablement non. MultEQ XT32 ne saura pas faire exactement la même chose, à savoir étendre électroniquement la linéarité de la courbe de réponse vers le bas.

Sachant que si ton 2496 est utilisé en tant que "simple" coupe haut (passe bas), MultEQ XT32 seul, fera largement l'affaire.

Maintenant si tu es dans le premier cas, ce que je pense, je te conseillerais d'utiliser MultEQ XT32 par dessus le 2496 utilisé uniquement dans le contexte d'extension, linéarisation @25/30Hz. Ceci devrait te permettre de retirer l'optimum des 2 systèmes.

Ayant moi-même utilisé un ensemble similaire, tu peux en lire un CR, ici.

Hugo

[Hugo S]

Bjr,

Le DTS Neo X venant compléter l'offre 9.x, 11.x,... je reproduis ici (avec les liens qui vont bien pour d'éventuelles questions de copyrights), les singulièrement intéressants échanges sur les fonctionnements comparés entre le DSX et DTS Neo X... et pour tous ceux qui seraient impatients et ne souhaiteraient pas tout lire, il suffit de se rendre sur le tout dernier post cité.

Sachant que Roger Dresler est le précédent directeur technique de Dolb, aujourd'hui à la retraite... aussi ses opinions tout en étant particulièrement instructives, sont à remettre dans le contexte où dans tout ceci (contexte DSX et DTS Neo X), c'est Dolby qui n'ayant rien de tel à proposer, a du coup, le plus à perdre...

With 5.1 or 7.1 source material, do the Neo:X wides play back sounds that would have phantom imaged between the fronts and sides, or is it extracting some other ambient information?

With 5.1 sources Neo:X primarily draws from the surrounds, with the following conditions:

a) When the surrounds are decorrelated, or
b) When the L and Ls (or R and Rs) are correlated.

In case b), a good example is the DVD Video Essentials "walk around the room" with voice or test signal. Movie mode pans this quite smoothly around the room, using the wides as a new speaker in the operation once the L and Ls channels carry signal. The result is much the same in Neo:X Music, except the wide outputs are 3 dB louder at the apex of the pan, causing them to stick out noticeably. The assumption may be that movies carry correlated pans, like this test signal, whereas music does not so need the extra jolt. While probably true for classical music, not sure it is a universal truth for pop music. Foreigner 4 DVD-A pans sounds thru the whole space, and these are affected. Even so, the end result is that with 5.1 music, Neo:x is more subtle than DSX. With some recordings, that can be a huge difference, as with Sailing to Philadelphia, DVD-A. It's not the voice of God that is invading my room with DSX. Neo:X is much more consistent with the original recording, and that's good.
__________________
Roger

Thanx for following up Roger; appreciate the feedback.

The result is much the same in Neo:X Music, except the wide outputs are 3 dB louder at the apex of the pan, causing them to stick out noticeably.

Just to confirm, it is Movie mode with the 3dB boost to the wides, not Music mode? Have you tried the 'Mission to Mars' test? (That's the first disc I tried with DSX.) If so, which mode sounds less distracting: Movie or Music?

Even so, the end result is that with 5.1 music, Neo:x is more subtle than DSX.

That's why I asked about adjustablity. On my friend's Denon 4311, there is an A-DSX Soundstage parameter, where Soundstage Width and/or Soundstage Height can be set from a -10 to +10 range (those numbers aren't dBs).

With some recordings, that can be a huge difference, as with Sailing to Philadelphia, DVD-A. It's not the voice of God that is invading my room with DSX.

That DVD-A must have had the lead vocals mixed to the L/R channels rather than the centre channel.
__________________
Sanjay

Thanx for following up Roger; appreciate the feedback. Just to confirm, it is Movie mode with the 3dB boost to the wides, not Music mode?

Opposite. Movie mode sounds even across all speakers, Music mode has the 3 dB goose in the wides. Let me clarify, this is in reference only to Neo:X, not DSX. DSX has a 1 dB boost plus 4 dB bass kick as mentioned elsewhere.

Have you tried the 'Mission to Mars' test? (That's the first disc I tried with DSX.) If so, which mode sounds less distracting: Movie or Music? That's why I asked about adjustablity. On my friend's Denon 4311, there is an A-DSX Soundstage parameter, where Soundstage Width and/or Soundstage Height can be set from a -10 to +10 range (those numbers aren't dBs).

I did not try Mars. I tried the DVE disc and some music. Even though the Music mode does not sound distracting, I prefer the Neo:X Movie setting for 5.1 content (of any type), and the Music setting for 2-ch music due to the way it treats the center signal (shared across L/C/R as opposed to only in C as in Cinema).

Back to DSX:

That DVD-A must have had the lead vocals mixed to the L/R channels rather than the centre channel.

Exactly. Quite common in 5.1 music.

I found the DSX Soundstage adjustment. It goes -3 to +3. I'll try it.

ETA: The wide outputs do change in 1 dB steps (acoustic measurement of that speaker), and have no affect on the mains.
__________________
Roger

Movie mode sounds even across all speakers, Music mode has the 3 dB goose in the wides. Let me clarify, this is in reference only to Neo:X, not DSX.

Understood (since DSX doesn't have music and movie modes).

The wide outputs do change in 1 dB steps (acoustic measurement of that speaker), and have no affect on the mains.

OK, so you can get rid of that 1dB boost (and then some). Should help make the effect more subtle or at least let you dial it in for your tastes (assuming it's your cup o'tea to begin with).
__________________
Sanjay

Understood (since DSX doesn't have music and movie modes). OK, so you can get rid of that 1dB boost (and then some). Should help make the effect more subtle or at least let you dial it in for your tastes (assuming it's your cup o'tea to begin with).

Yes, by reducing Soundstage to -3, the overall loudness of the L/R channels is restored to normal. The elevated bass is not affected by the control.
__________________
Roger

Yes, by reducing Soundstage to -3, the overall loudness of the L/R channels is restored to normal.

So every channel is back to being the same level, even the ones being reproduced by two speakers. Does the soundstage still remain noticeably wider, even at the -3dB setting?

The elevated bass is not affected by the control.

Why mess with a good thing. Reference shmeference!
__________________
Sanjay

So every channel is back to being the same level, even the ones being reproduced by two speakers.

Yes, regarding L/C/R. But when DSX is turned on not only are the surrounds decorrelated, their level is dropped 3 dB (colorisé par moi). Not sure why, unless the intention was to focus listeners on the wides...

Does the soundstage still remain noticeably wider, even at the -3dB setting?

That'll have to wait 'till the morrow.
__________________
Roger

Yes, regarding L/C/R. But when DSX is turned on not only are the surrounds decorrelated, their level is dropped 3 dB.

Really, more gain-play? So the fronts AND surrounds are dropped 3dB. Guess they're really trying to emphasize those Wides.

BTW, I try to do something similar (increase ASW) acoustically by using diffusion at the ipsilateral first reflections and absorbtion at the contralateral first reflections. But the reflections coming off those diffusors aren't 3dB louder than the source speaker.

BTW, I've learned more about how DSX operates from your last few posts than I had in all the time since the processing had been introduced. So thanx again.

Now in the very beginning of this (private) use, my initial listening impression was that there was some sort of lack of "message" in the original configuration of the BS made by MultEQ...
Though I increased the BS trim by 2/3dB (I don't remember the precise value). But then something didn't feel/sound right anymore, there was a feeling I would qualify as a lesser envelopment impression (in this DSX configuration)...
So I ended up returning the trim values to their original position, "untouched" BS values that I still use today.

I had always noticed that turning off DSX used to bring the surround field back to life, but didn't know why until Roger posted details. I think the processing would have sounded better had it left the surrounds alone.
__________________
Sanjay

Roger, what do you mean by "decorrelated"?

A process is applied to diffuse the sound between the two channels, so it is not able to create a focused phantom image. It also prevents comb filtering from multiple speakers. I think it is a good thing.
__________________
Roger

Really, more gain-play? So the fronts AND surrounds are dropped 3dB. Guess they're really trying to emphasize those Wides.

At least in the fronts the drop in the mains is compensated by the same sounds coming from the wides, so it's not altering the balance so much. As I mentioned, the net effect is actually a gain of 1 dB in the Ref setting. Changing Soundstage to -2 dB removes the net gain boost, but also reduces the DSX widening effect.

BTW, I've learned more about how DSX operates from your last few posts than I had in all the time since the processing had been introduced. So thanx again.

Well, I have no insights into the psychoacoustics of what it's trying to do, only some "black box" observations, FWIW.

I had always noticed that turning off DSX used to bring the surround field back to life, but didn't know why until Roger posted details. I think the processing would have sounded better had it left the surrounds alone.

I ran an old Dolby film trailer called "Train" today which starts with 4 clangs in the 4 corners. DSX messes them up, goosing the fronts and dropping the rears. Not at all how it's supposed to sound.
__________________
Roger

Hi Roger,
I also think it is a good thing for ambiance sounds not intended to be a focused image.
So here's a leading question ...
Is it a "good thing" for highly directional sounds that are intended to be localized?

Larry, you're exactly right. It's not a good thing. I am just playing Toy Story 2 where the toy thief puts Woody in the trunk and drives off, ch8. All sorts of cues from the rear speakers. They are diffused with DSX to the point of localizing nowhere. PLIIx/z and Neo:X handle them as directional sources. (colorisé par moi).
__________________
Roger

Hi Roger,
Thanks for your observations.
Have you had an opportunity to readjust the surround levels, Soundstage and bass so as to compensate for these "non-reference" effects and then obtain subjective impressions? Or would this be a futile exercise that depends too much on the nature of the content?

I did drop the Soundstage to -2 which reduced the level elevation of the main L/R signals, but also reduced the widening effect to essentially nonexistent. Not a bad outcome from my perspective, as I perceived no compelling benefit when it was audible. It was clearly different, but not better, since the original was not lacking in any respect.

Even though I believe the bass elevation is not an intended part of DSX, it would be futile to try to offset it, since it is conditional on DSX activity -- it only exists when wide effects are happnin'. Having said that, one way to essentially eliminate the problem is to change the wide speaker settings to "large" so that they will not contribute to bass management. If you are lucky enough to have wide speakers that roll off the same point as the high pass used for the L/R mains, then it will all work nicely.

Re the surround levels, it would be inappropriate to raise them because that appears to be a deliberate aspect of DSX design. To attempt to undo that would be to armchair redesign their technology.
__________________
Roger

Hi Roger,
And your point is?...
If I understand correctly the main premise of DSX is to increase Apparent Source Width which is a scientific term that describes a general improvement in listener preference. So to quote Gary we are already heavily into "Preference Land". If we are dealing with preference, hey why not see if I prefer my surrounds at the reference levels set by MultEQ?

I already know I prefer my surrounds at the calibrated level. That's why I think DSX has made an error in fiddling with them.

If I were to do anything to offset them with DSX, like adjust the speaker trims up 3 dB, that would throw off all other surround modes. My preference is to not have to jockey trims on an ongoing basis.
__________________
Roger

If there were a focused phantom image from a discrete soundtrack, wouldn't it have been intended?

Most likely. In this recent post I mention how it goes pear shaped when spot effects in Toy Story 2 are decorrelated.

In the debate between what film mixers heard vs. intended, I have faced this matter several times over the years. When movies were released in 70mm 6-track Dolby Stereo, eventually with stereo surrounds, that was the pinnacle of the day. The same movie was also released on 35mm Dolby Stereo optical using what we call Pro Logic, a mono 7 kHz bandlimited surround channel. Which version represented their intention?

Mixers may well intend a spot effect in the surrounds, and do what they can to make it happen. The surround matrix, the optical soundtrack, and arrays hamstring their art, and so the mono array was divided into two, then three (EX), and now 4 smaller arrays (7.1). D-Cinema specifies options for single rear corner speakers for even greater control of spot effects. In future, every speaker in a theater will be individually driven with its own signal, allowing the maximum specificity possible from the system.

Even if it makes it sound more like it's supposed to?

I have to assume Audyssey designed it the way they felt it is supposed to work. I want to hear the mode just as they intended, so I can try to hear their vision undistorted.
__________________
Roger

True, he was trying to mimic theatrical surround arrays, but the fact remains that precise surround imaging could not be achieved at the time he made his initial recommendation.

Still can't, at least not theatrically. I saw 'Transformers 3' in 7.1 recently, and noticed that there were a half-dozen surrounds on each side wall and eight speakers on the back wall. So each surround channel was being reproduced by either 6 or 4 speakers. No precise surround imaging there, even with a discrete 7.1 soundtrack.

Anyway, my point was that people often associate Holman's dipole recommendation with the mono surround channel of the Pro Logic era, when that's not the reason he recommended dipoles. It had to do with recreating surround arrays, irrespective of the number of surround channels.
__________________
Sanjay

Is DTS Neo X going to blow away Audyssey DSX and Dolby IIz ?

No.

In my opinion, DSX is more detrimental than useful, tilting the front/rear balance forward and adding extra bass; and Neo:X is not as good as PLIIx for existing content -- narrow surround effect and weird timbral changes to certain instruments.

This is when using wide speakers in a 9.1 setup. I have not tried height speakers yet.

An as yet untapped capability of Neo:X and to a lesser extent PLIIz is the ability to encode specific height/width cues and recover them. This is something DSX cannot do. Until someone makes such newfangled soundtracks, we won't know if that blows away anything, or just blows.
__________________
Roger

(Roger) I'm a bit confused by your results. I wonder if you would be willing to summarize your DSX experiments.

I have a 5.1 system with Wides. I looked at relative levels using the AIX test/demo disc balance signals (these are broadband pink noise). I found a -2dB change in the surrounds but no change in the overall front levels (Chris says that's how it should work). I also ran a couple of LF front sweeps and while there was an offset it's not large or consistent with frequency.
First I calibrated the system with 8-position MultEQ XT. I made no changes to the trims except to raise the subs 9 dB, back to “normal.” DSX Soundstage was at Ref.

First I played some DVD 5.1 movie clips, and when obvious L/R pans occurred I sensed not only the wider effect but an emphasis in the sound coincident with those wide effects. In order to see if I could see why that happens, I used a speaker test signal DVD with filtered pink noise. With that continuous noise parked in the L (or R) channel I set the volume to 70 dB for the SPL meter in the MLP. Then I turned on DSX and saw the SPL increase 1 dB as the wide speaker became active.

The same DVD has bass noise signals in each channel. I did the same SPL sequence and this time when switching on DSX the level increased 4 dB. That seems perfectly consistent with what happens when an additional speaker is added to the bass management pool. Easily avoided in the design, but there it is. At home, one can switch the wides to Large so that bass will not be part of the equation.

I also tried the same wide speakers with Neo:X, and panned a continuous noise signal (Digital Video Essentials) Ch7 “walk around the room” and it drove that speaker to exactly the same SPL as the adjacent ones as the pan proceeded. So that is at least some evidence that MultEQ got the levels set right. Do you have DVE? Run it with and without DSX and see what happens for your setup.

I fully recognize that Audyssey suggests broadband noise and sweeps are not appropriate tools for assessing the efficacy of their products.

I agree, not appropriate for evaluating proscenium reflections. But that’s only part of the story, and not the effect I heard with movies (or music). Hard pans to L/R and moving pans around the space happen every day in movies. Sometimes even in music (Foreigner 4)!
__________________
Roger

Hi Chris,
You seem to be suggesting that a flyover is a gimmick, yet you state that DSX is better at handling pans from front to back than conventional 5.1. Isn't a flyover the same as pans from front to back?

Hi Larry,

There are three components that a rendering system must respect in order to be valuable: (1) acoustics (2) psychoacoustics and (3) the desires of the content creators

To do #3 in a way that doesn't require a new speaker every time someone wants to put a special effect somewhere we had to examine the statistics of the available content. Sure there are times when the sound designer calls for a height effect, but the vast majority of content does not require height panning. #1, #2, and #3 above are greatly enhanced by the addition of these channels starting with wide, then height.

However, since in a 7.1 configuration there are physical speakers placed to the sides of the listener isn't this issue of not being able to hear phantom images at 90 degrees moot?

The ITU recommendation for surround speaker placement (and the way content is mixed) requires the main surround speakers to be at 110°. That is too far back from the front speakers that are required to be at 30° to form a coherent stable phantom. Putting the wide speaker at 60° fills that gap nicely and we have content where a person speaking starts from the surround speaker and they can walk to the front seamlessly. If the gods granted us more speakers, we would have one at 90° to get a stable center image on the sides, but the best optimization of #1 and #2 above directed us to 60°.

Won't a DSX configuration still need surround back speakers, in addition to the Wides, to fully support a complete front to back pan?

Sure. Remember, we are not advocating to eliminate the back surround speakers. It's a matter of priorities. If you are only given 7, I think you are missing out by using two of them in the back.
__________________
Chris

I wonder if it would significantly degrade the DSX Wide signals were they to be split between L and Ls and R/Rs, thereby not needing another pair of speakers. Virtual wides, if you will. Then use the new speaker/amps for heights. 11.1 effect over 9.1 speakers.

Hi Roger,

This is the exact experiment that Tom and I did in the lab at USC several years ago. The answer is: it is impossible to create a phantom between L and LS for a listener facing forward. Not even close. It's quite easy to prove with a pair of stereo speakers in front of you. Sit in the middle and listen to the nice phantom. Then turn 90° and listen again with the speakers to your side. Phantom is gone. The same thing basically happens when you try to create a phantom between front and surround speakers. Not to mention, that even if it did work, it would only work for one listener.

Wides are needed and work well to make the front-back more seamless. Furthermore, in experiments by Y. Ando in Japan it was shown that 60° (actually 55°) is the most important direction selected by listeners as "enhancing auditory source width". So, the Wide channels serve two purposes: (1) fill in the perceptual imaging gap between front and surround and (2) provide ASW enhancement.
__________________
Chris

Hugo

PS: http://translate.google.fr/

[jazzliberator]

Bonjour,

je ne suis pas en salle dédiée et mon équipement est dans ma signature. Sur mon Onkyo 876 (MultEQ XT) je me pose la question du réglage (en manuel)de mes Front.

A savoir si je pouvais caler mes Mission 792 (biblios avec plage de fréquence de 58-20 000Hz) en pleine bande? Quid du Double Bass (activé ou coupure 80Hz)? Ou alors je suis de manière stricte les réglages/calibration?

Mes autres enceintes (un cran en dessous) restent elles calées à 80Hz et mon caisson avec le potard calé à 120Hz.

Merci.

[Hugo S]

Bjr jazz,

Bonjour,

je ne suis pas en salle dédiée et mon équipement est dans ma signature. Sur mon Onkyo 876 (MultEQ XT) je me pose la question du réglage (en manuel)de mes Front.

A savoir si je pouvais caler mes Mission 792 (biblios avec plage de fréquence de 58-20 000Hz) en pleine bande? Quid du Double Bass (activé ou coupure 80Hz)? Ou alors je suis de manière stricte les réglages/calibration?

Mes autres enceintes (un cran en dessous) restent elles calées à 80Hz et mon caisson avec le potard calé à 120Hz.

Merci.

Je ne connais pas spécialement les Mission 792. Toutefois ce que je peux t'affirmer est que vouloir pousser le fonctionnement des enceintes en "Full" est juste à l'opposé du résultat que tu cherches à obtenir.

En effet, passer des enceintes en full amène l'amplificateur à travailler en peine bande, donc potentiellement à des fréquences largement inférieures à 80Hz. Ce type de fonctionnement est très gourmand en courant, il peut facilement mettre à genoux un ampli multi canaux, qui n'est pas prévu pour ça.

D'autre part, les enceintes elles-mêmes vont devoir reproduire des fréquences dans une plage de fréquences où la distorsion qu'elles vont produire sera élevée, du coup la "fidélité" globale s'en ressentira.

Ton install est clairement représentative d'une approche "HiFi", dans l'esprit British, c'est à dire à des niveaux sonores potentiellement peu élevés. Cette install comprend le très bon sub Rel Q100. Un caisson conçu pour descendre très bas en fréquences, mais pas forcément en niveau. Il est capable de réellement "magnifier" les Missions en complétant symétriquement vers le bas, la "précision" et le "ciselé" (identificateurs de la sonorité) des Missions dans le haut du spectre.

Très concrètement tu devrais couper tes Missions à 80Hz (voire un peu plus si ceci devait s'avérer nécessaire, il n'y aura pas de perte de focalisation image) et le Rel pourra être coupé à 120Hz. Ce qui ne limitera pas la reproduction des fréquences du canal .1 (sub) qui peuvent monter jusqu'à 120Hz, sachant qu'il ne reproduira les fréquences des autres canaux que jusqu'à leur fréquence de coupure basse, ici validée à 80Hz.

Tu auras ainsi un système optimisé d'un point de vue fonctionnement tant en HiFi, que HC, mais en gardant à l'esprit que les niveaux sonores maxi possibles ne seront pas ceux... possibles dans une salle de cinéma...

Sachant que dans tous les cas je te conseille d'acquérir un appareil intégrant le Audyssey MultEQ XT32, ce type de correction acoustique étant en effet BEAUCOUP plus évoluée que le MultEQ XT "simple", l'écoute (avec) prend une autre dimension...

... et je sais, les "conseilleurs", ne sont pas les "payeurs"...

Hugo

[jazzliberator]

Mon installation se trouve dans un séjour de 30m² et occupe1/4-1/3 de celle-ci donc le niveau maxi se trouve vite atteint (à mes oreilles, pour endurer un film de 2H); quant à l'ampli ADA, il tient bien la route (1500VA et 250 000µf).

Simplement, avec le pleine bande j'ai une sensation d'impacts plus francs et un peu plus profonds; à moins que cette impression ne soit que flatteuse? Je vais faire avec ce que tu me conseilles.

Il faudra que j'envisage le remplacement de mon 876 par un 5008-5508 (mais il y a un Mitsubishi 4000 qui me fait des signes ). Les principales différences XT-XT32 se situent à quels niveaux?

Merci de la réponse Hugo.