Modérateurs: Staff Haute-Fidélité, Staff Univers Casques, Staff Juridique • Utilisateurs parcourant ce forum: ALAIN B 33, BRUFA, Goan, indien29, PAT 94 et 54 invités

Discussions sur le matériel Haute-Fidélité

Signatures sonores ?

Message » 23 Mar 2022 10:57

Esscobar a écrit:
Jean-Luc67 a écrit:
Mais clairement croire qu'il existerait un son anglais, japonais ou autre, là non ! :wink:


Pourtant j'ai lu pas mal de fois sur ce forum ou ailleurs... J'ai toujours trouvé cela surréaliste !

Encore l'autre jour je regardai une video d'un vendeur de Hifi qui disait dans sa vidéo "J'adore le son anglais pour les électroniques"...

JL
Jean-Luc67
 
Messages: 26
Inscription Forum: 25 Aoû 2021 14:30
Localisation: Haguenau
  • offline

Annonce

Message par Google » 23 Mar 2022 10:57

 
 
Publicite

 
Encart supprimé pour les membres HCFR

Message » 23 Mar 2022 11:03

Esscobar a écrit:Intéressant ce que tu dis :bravo:


Effectivement, selon la conception des enceintes, on peut avoir une signature sonore particulière et il est vrai que plusieurs fabricants anglais conçoivent la caisse de leur enceintes comme un instrument de musique en considérant qu'elle vibrera en fonction des fréquences...

Je pense sincèrement que cette voie n'est pas idéale et je préfère les constructeurs essayant justement d'avoir le moins de vibrations possibles en construisant leurs parois le plus inerte possible....

Sinon, je ne suis pas vraiment certain que l'écoute principalement de musique Métal soit le meilleur moyen de juger de la qualité des timbres naturels d'un instrument puisqu'il s'agit de musique amplifiée... A part pour juger de la profondeur du grave sur la basse électrique... :grad:

La configuration dans mon profil


lecteur Luxman D08U, serveur SOtM sMS 1000 Eunashu + sPS 1000 alim, Dac B Audio, ampli Ypsilon Phaeton SE , enceintes JEAN MAURER 370 F, câbles Plénitude évidence + Ypsilon , meubles Woodlink.
marius30
Membre HCFR
Membre HCFR
 
Messages: 520
Inscription Forum: 01 Mar 2010 19:51
  • offline

Message » 23 Mar 2022 11:17

marius30 a écrit:Sinon, je ne suis pas vraiment certain que l'écoute principalement de musique Métal soit le meilleur moyen de juger de la qualité des timbres naturels d'un instrument puisqu'il s'agit de musique amplifiée... A part pour juger de la profondeur du grave sur la basse électrique... :grad:

Ah tu serais surpris :)
J'ai quelques morceaux qui mettent en PLS un système qui passe parfaitement bien de la musique classique / tradi / ... et pas que sur le grave (loiiiiin de là).
Et je trouve même que de tout les morceaux que j'ai dans ma playlist de test ce sont les plus discriminants ;)
Mais ce n'est pas le sujet.

D.

La configuration dans mon profil


La bougie de ton intelligence n'éclairera ta vie que le jour où tu arrêteras toi-même de souffler dessus !
On ne peut pas donner à boire à un âne qui n'a pas soif
Avatar de l’utilisateur
Dagda
Staffs DIY & Haute-Fidélité
Staffs DIY & Haute-Fidélité
 
Messages: 12242
Inscription Forum: 22 Déc 2005 14:53
Localisation: Fontenay sous Bois
  • offline

Message » 23 Mar 2022 12:28

Dagda a écrit:
marius30 a écrit:Sinon, je ne suis pas vraiment certain que l'écoute principalement de musique Métal soit le meilleur moyen de juger de la qualité des timbres naturels d'un instrument puisqu'il s'agit de musique amplifiée... A part pour juger de la profondeur du grave sur la basse électrique... :grad:

Ah tu serais surpris :)
J'ai quelques morceaux qui mettent en PLS un système qui passe parfaitement bien de la musique classique / tradi / ... et pas que sur le grave (loiiiiin de là).
Et je trouve même que de tout les morceaux que j'ai dans ma playlist de test ce sont les plus discriminants ;)
Mais ce n'est pas le sujet.

D.


On ne parle de la même chose en fait.... Je parle de timbre, de naturel du grain d'un violon, d'un cuivre, d'un piano.... Et tu me parle de mettre un système en difficulté.... Ce sont des choses totalement différentes ....

Mais tu as raison, ce n'est pas le sujet... :grad: , Par contre cela me permet de mieux comprendre certaines interventions au sujet du matériel, donc cela reste intéressant..

La configuration dans mon profil


lecteur Luxman D08U, serveur SOtM sMS 1000 Eunashu + sPS 1000 alim, Dac B Audio, ampli Ypsilon Phaeton SE , enceintes JEAN MAURER 370 F, câbles Plénitude évidence + Ypsilon , meubles Woodlink.
marius30
Membre HCFR
Membre HCFR
 
Messages: 520
Inscription Forum: 01 Mar 2010 19:51
  • offline

Message » 23 Mar 2022 12:40

Ah oui d'accord :lol:
Bon là on rentre aussi dans la problématique de savoir comment ça a été enregistré ;)

La configuration dans mon profil


La bougie de ton intelligence n'éclairera ta vie que le jour où tu arrêteras toi-même de souffler dessus !
On ne peut pas donner à boire à un âne qui n'a pas soif
Avatar de l’utilisateur
Dagda
Staffs DIY & Haute-Fidélité
Staffs DIY & Haute-Fidélité
 
Messages: 12242
Inscription Forum: 22 Déc 2005 14:53
Localisation: Fontenay sous Bois
  • offline

Message » 23 Mar 2022 13:02

Jean-Luc67 a écrit:Je pense que nous sommes tous d'accord pour dire qu'il y a des différences de qualité de son entre plusieurs ampli ou enceintes. Il y déja la hiérarchie des prix et pleins d'autres critères...

Mais la question de départ était :

Y a-t'il une signature sonore en fonction des marques ? Seriez-vous capable de le décrire ? De le reconnaitre à l'aveugle ?

C'est quoi le son anglais ? Japonnais ? ....

Merci
JL

Si cela peut t'aider , dans son livre ""Sound Reproduction: The Acoustics and Psychoacoustics of Loudspeakers and Rooms", Floyd TOOLE chapitre 17 répond précisément a ta question ..(ici seulement une petite partie car le chapitre est très long ) :(

""Loudspeakers I: Subjective Evaluations
337
In a book about sound reproduction, it may seem strange that it has taken
16 chapters to get to the topic of loudspeakers. Well, the reason is it took
that long to understand what it is that loudspeakers need to do; they are a
means to an end. Loudspeaker systems and rooms have traditionally been
designed in isolation. As if by magic, the loudspeaker is expected to sound
good in a small refl ective space of uncertain size and acoustical properties,
with it and listeners positioned with minimal understanding of the acoustical
consequences.
Of course, people know that the room is part of the process, but the details
of the relationship have been hard to tie down. In the trial-and-error process of
traditional loudspeaker design, countless hours of concentrated listening have
been devoted to “voicing” the loudspeaker. The result? It might sound “good”
for the program material favored by that listener in that particular room, but
change the room, and parts of the experience change. Change the recording, and
still other parts of the experience change. It has been a great problem, causing
much confusion about what is responsible for the sound quality we are judging.
Historically, the loudspeaker has carried most of the burden. We now know that
above the transition frequency the responsibility is justifi ed. But in the very
important bass region, the room and the arrangement of listeners and loudspeakers
within it dominate, so no loudspeaker design has any chance of sounding
the same at all frequencies in all rooms.
Add to these uncertainties the notion that what we like—taste—is a personal
thing. One man’s meat might be another man’s poison, so the saying goes.
Ask people at an audio show which loudspeakers they think are best, and you
will get many different answers. Are our preferences in sound quality as distinctive
as our preferences in “wine, persons and song,” to paraphrase another
saying?
While listening to products in stores and the homes of friendly audiophiles,
people are participating in informal experiments. The experimental variables—
loudspeaker, room, music, mood, price, size, brand, and whatever verbal chatter
is happening at the time—are mingling and merging. Whatever opinion emerges
from the amalgam of infl uences might have a relationship to the inherent capabilities
of the loudspeaker . . . or it might not. It is impossible to know. These
exercises set out to assess personal preference in sound but end up being infl uenced
by personal susceptibility to many nonauditory infl uences. No malicious
intent need be assumed; it is just what happens. Any competent audio salesperson
knows that most customers can be “persuaded” by the right kind of
presentation and a well-orchestrated demonstration.
Along with the personal preference argument is the one asserting that people
who live in different parts of the world have distinctive needs. It extends also
to beliefs that different kinds of music may need different spectral balances. If
these were so, there would be additional controls on audio components or loudspeakers,
as shown in Figure 17.1.
As discussed in Chapter 1, the word reproduction implies that somewhere
there was an “original,” and it is the task of the sound reproducing system to
emulate that original. It is well understood that a perfect three-dimensional
acoustical replica of a live performance is simply not feasible, and recordings
have gone on to create their own artistic and abstract interpretations of reality.
As discussed in Chapter 2, our real goal is to “connect with some of the key
underlying perceptual dimensions” so artists and listeners can go beyond the
limitations of small playback spaces. In the intervening chapters, it has become
clear that most of us fi nd considerable pleasure in impressions of direction,
space, and envelopment. This much we have in common. Let us now examine
the matter of sound quality.
17.1 THE GENESIS OF A LIFE’S WORK
This is where the story becomes personal because much of the content that
follows derives from research done by the author and his colleagues over the
past 40 years. Acoustical measurements, and the devices to do them, were well
developed by the time I became a research scientist at the National Research
Council of Canada, in Ottawa, as a fresh PhD EE graduate. It was April 1965.
My PhD research had been on the topic of sound localization and, in particular,
the manner in which sounds at the two ears were processed by the brain to yield
perceptions of direction (Sayers and Toole, 1964; Toole and Sayers, 1965a,
1965b). All of the experiments to that point had been done with headphones,
which allowed signals to each ear to be controlled independently.
A thrilling prospect of the new job was that there was an excellent anechoic
chamber, within which the research could be extended to include listening
under natural circumstances, starting in a refl ection-free environment. For
this, loudspeakers were needed. When anechoic measurements were made on
some highly-rated audiophile loudspeakers of the time, the results were depressing.
Most of the frequency responses were far from fl at, and these were
simple on-axis anechoic measurements made for the purpose of performing
anechoic listening tests. Up to this point, the author had only seen “specifi
cations” for frequency response, and if it were not for the unimpeachable
pedigree of the measurement circumstances, it would have been possible to
think that there had been a tragic error in making the measurements. Suddenly,
claims that the loudspeaker was the “weakest link” in the audio chain
rang true. But could these products really sound as bad as some of the curves
looked?
A logical “Friday afternoon” experiment was to do a simple comparison listening
test in one of the laboratory rooms. Having learned the basics of experimental
psychology for the thesis work, it was obvious that this test had to be
somewhat controlled. So cotton sheeting was hung up to render the experiment
“blind.” The sounds being compared in the monophonic A/B/C/D comparisons
were adjusted to be equally loud. There was no statistical imperative for listening
in groups of four; it just seemed convenient. Interestingly, four-way multiple
comparisons have remained the norm in our subjective evaluations ever since.
A supportive technician built a simple relay switch box. After that, I and a few
interested colleagues took turns sitting in, forming opinions and making notes.
A “Gestalt” impression, a summarized overall rating, was required: a number
on a scale of 10.
The results surprised all of us. The audible differences were absolutely enormous,
but there was general agreement about which ones seemed to sound good.
It remained a topic of discussion for days. The need for loudspeakers for my
anechoic sound localization experiments remained, and the winner of this
simple test showed promise. It was dismantled and some improvements were
made. The experiments proceeded.
Many months passed before another listening test was staged in early 1966.
By then I had learned that bed sheeting is not acoustically transparent and that
music passages needed to be short and repeated, which meant that we needed
a disc jockey (remember this was the LP era). That would be the agreeable technician.
Word had spread and audio enthusiasts from within the organization
lined up to participate and in some cases to bring their personal loudspeakers
to be evaluated. This test went on for several days and yielded enough subjective
data to warrant rudimentary statistical analysis. Again, there was good agreement
about the products that were preferred and those that were not. The
winning loudspeaker was the redesigned unit that was being used in the anechoic
chamber tests. It also had the best-looking set of measured data, assuming one
puts any value in smooth and fl at frequency responses on- and off-axis (see
Figure 17.3b).
Figures 17.2 and 17.3 show several loudspeakers that were used in listening
tests in that period of time. My archives still have some of the handwritten listener
notes from these tests, done 42 years ago! It is interesting to look them
over and, with the benefi t of hindsight, to comment briefl y on these products.
First, it is obvious that they sounded very different from one another. These
were among the fi rst loudspeaker measurements I ever made, and a standard
format had not yet been established, so some of the measurements go to 45°
off axis and others to 60°.
I discovered KEF at an audio show in London (Figure 17.2a) while I was there
as a student. I liked what I heard and purchased two kit versions of the KEF
Concord to be the nucleus of my fi rst postgraduation stereo system. In these
tests it was judged to have good overall balance but some midrange coloration.
This was found to be caused by fl exure along the long dimension of the woofer
diaphragm (a breakup mode) above 1 kHz, occurring in the crossover region to
the tweeter. The large (1.5-in. diameter) tweeter became directional above about
4 kHz, and this was noticeable as a slight dullness.
The Acoustic Research AR-3 (Figure 17.2b) was famous for its novel acoustic
suspension woofer, and it came to be one of the reference loudspeakers of that
generation. Its acoustic performance was well documented in the literature
(e.g., Allison and Berkovitz, 1972), which was a great credit to the company. A
major design goal was to achieve constant directivity, and they did well; the only
minor exceptions are the woofer beginning to beam as it approaches the crossover
to the midrange and the tweeter at very high frequencies. The essential
issue with this product is its frequency response, which signifi cantly rolls off
toward both low and high frequencies. Low-frequency output would be aided by
boundaries (see Figure 12.5 for boundary interaction data). Almost all listeners
found it to be slightly “dull” sounding, and some identifi ed coloration around
1 kHz.

loudspeaker was preferred over all the preceding
designs at the time. The implication is that
smooth fl at axial frequency responses and wellbehaved
off-axis performance combine to yield
reproduced sounds that many listeners like.
That is a conclusion that has only been reinforced
in subsequent years.
This happened in the late 1960s. It remained
a background activity for several years. I read
what I could fi nd on the topic (not much), and
a lot of that was more anecdotal than scientifi c.
The BBC Research Department had done some
fi ne insightful work—for example, Shorter
(1958)—but budget cuts slowed further progress.
Professional psychoacousticians, most of
them in universities, nibbled around the edges
of the real problems in sound reproduction, but
none chose to engage the topic directly. Phone
calls and visits to some prominent loudspeaker
designers found that all were very cooperative
and candid. There were more questions than
answers and more opinions than facts. This was
a fi eld that could use serious scientifi c investigation,
but the industry was too fragmented to be able to mount such an effort
on its own.
On the measurement side, there were three distinct camps:

1. The “on-axis” school of design, based on the thought that the fi rst
sound to arrive at the ears had a dominant role not only in localization
but also timbre. The important information was thought to reside in
the on-axis frequency response curve. Refl ected sounds in rooms,
according to this idea, would be perceptually suppressed.

2. The “sound-power” philosophy took the opposite view, making the
assumption that the listening room is a highly refl ective space with a
sound fi eld that is suffi ciently diffuse that all of the sound radiated
from a loudspeaker in any direction is fully integrated. According to
this idea, an integration of all sound radiated in all directions, a single
curve, would be the perceptually dominant factor.

3. The “room curve” viewpoint took into account the frequency response
and directivity of the loudspeaker, as well as the refl ective properties of
the room. The logical argument was that this combination was what
we heard, and therefore this single curve must be the principal metric.

FIGURE 17.3 On- and off-axis anechoic measurements
on a mid-1960s loudspeaker and on the redesigned
unit that was created for use in anechoic chamber
experiments. Dotted curves at low frequencies indicate
uncertain accuracy due to anechoic chamber errors.


The problem with this argument is that two ears and a brain respond
differently to a complex sound fi eld than does a microphone, which
simply adds all sounds together without regard for the direction from
which they arrive or how much time has passed since the direct
sound. This philosophy disregards notions of precedence; binaural
discrimination; the spatial effects of early, late, lateral, and vertical
refl ections; and temporal integration phenomena.
Such is the problem of seeking a single curve that embodies all of the truths;
the real situation is much more complex. An associated problem was that
measurements were often done using 1/3-octave fi lter sets (fi lters with fi xedcenter
frequencies producing a staircase, not a continuous, curve). The poor
frequency resolution limited the ability to see narrow-band irregularities, and
the histogram form of displaying spectral data did not lend itself to revealing
more than the crudest trends. Toole (1986) has a detailed review of this
historical background.
Looking back, the data that were used in those days could not be reliable
indicators of sound quality in rooms. This very likely was the origin of the belief
that “we cannot measure what we can hear.” Indeed, if the measurements lack
adequate frequency resolution and not enough measurements are made, the
belief is fulfi lled.
As a result, a lot of listening went into the design process. Listening tests
tended to be fairly casual affairs, with participants almost always knowing
what was being auditioned. Great weight was placed on the opinions of musicians
or regular concertgoers (in those days, this meant classical, acoustic
concerts). The source material was usually LPs, themselves subject to signifi -
cant imperfections and variability. In such a climate of uncertainty, it could
not be surprising that many loudspeakers of indifferent quality were being
produced.
In the mid-1970s, the project became a foreground activity for the author,
time-shared with other tasks. This was a fruitful period. Around 1986 the
Athena project was established, a partnership between the National Research
Council of Canada and a nonprofi t consortium of fi ve Canadian audio manufacturers,
to help fund the effort. The mission was to explore the interface
among loudspeakers, rooms, and listeners. That project wound up around the
time, in 1991, when I joined Harman International Industries, Inc., where I
established a corporate research group that has continued to add to the base of
scientifi c knowledge. (I retired in 2007. It has been a very gratifying career.)
The following discussion of loudspeakers will be presented in two parts:
subjective evaluations (Chapter 17) and objective evaluations (Chapter 18).
Along the way, there will be overlap and interaction because of the fi nal objective:
to identify the measurable quantities that correlate with listener opinions.
Finally, all of the knowledge is put to the test, to fi nd a means of predicting
The Genesis of a Life’s Work
CHAPTER 17 Loudspeakers I: Subjective 344 Evaluations
listener opinions from an analysis of measured data (Chapter 20). It is an interesting
story.
17.2 SUBJECTIVE MEASUREMENTS OF
LOUDSPEAKERS—TURNING OPINION INTO FACT
It may seem oxymoronic to place the words subjective and measurements in
such close proximity. However, when it is possible to generate numerical data
from listening tests, and those numbers exhibit relatively small variations and
are highly repeatable, the description seems to fi t. It has been this more than
anything else that has allowed the exploration of correlations between technical
measurements and subjective opinions. Technical measurements, after all, don’t
change, but however accurate and repeatable they a may be, they are useless
without a method of interpreting them or a way to process them so they relate
more closely to perceptions. Subjective measurements provide the entry point
to understanding the psychoacoustic relationships. The key to getting useful
data from listening tests is in controlling or eliminating all factors that can
infl uence opinions other than the sound itself.
In the early days, most of us thought that listeners were recognizing excellence
and rejecting inferiority when judging sound quality. As logical as this
seemed, it was soon thrown into question when listeners in the tests showed
that they could rate products just as well with studio-created popular music as
they could with classical music painstakingly captured with simple microphone
setups, sometimes even better. How could this be possible? None of us had any
idea what the studio creations should sound like, with all of the multitrack,
close-miked, pan-potted image building and signal processing that went into
them. The explanation was in the comments written by the listeners. They
commented extravagantly on the problems in the poorer products, heaping scorn
rich in adjectives on things that were not right about the sound. In contrast,
high-scoring products received only a few words of simple praise. People seemed
to be able to separate what the loudspeakers were doing to the sound from the
sound itself. The fact that from the beginning, all the tests were of the “multiplecomparison”
type may have been responsible. Listeners were able to freely switch
the signal among three or four different products while listening to the music.
Thus, the “personalities” of the loudspeakers were revealed through the ways
the program changed. In a single-stimulus, take-it-home-and-listen-to-it kind
of test, this would not be nearly so obvious.
Humans are remarkably observant creatures, and we use all our sensory
inputs to remain in control in a world of everchanging circumstances. So when
asked how a loudspeaker sounds, it is reasonable that we instinctively grasp for
any relevant information to put ourselves in a position of strength. In an
extreme example, an audio-savvy person could look at the loudspeaker, recognize
345


use their scientifi c and technical skills to fi nd ways to deliver rewarding experiences
to more people in more places.
In the category of loudspeakers and rooms, however, there is no doubt that
differences exist and are clearly audible. Because of this, most reviewers and
loudspeaker designers feel that it is not necessary to go to the additional trouble
of setting up blind evaluations of loudspeakers. They believe that their professionalism
can overcome any biases from nonauditory inputs. This attitude will
be tested.
17.3 CONTROLLING THE EXPERIMENTAL VARIABLES
Any measurement requires controls on variables that can infl uence the outcome.
Some can be completely eliminated, but others can only be controlled in the
sense that they are limited to a few options (such as loudspeaker and listener
positions) and therefore can be randomized in repeated tests, or they can be held
as constant factors. Much has been written on the topic (Toole, 1982, 1990;
Toole and Olive, 2001; Bech and Zacharov, 2006). The following is a summary.
17.3.1 Controlling the Physical Variables
The listening room. The discussions of Chapter 13 show graphically and dramatically
how much a room can change what can be heard at low frequencies.
If comparing loudspeakers, listen to all in the same room. As pointed out in
Chapter 11, listeners have the ability to adapt to many aspects of room acoustics
as long as they are not extreme problems.
Loudspeaker position. Listen in the same room, with the loudspeakers each
brought to the same position. If that is not possible, ensure that all loudspeakers
are auditioned in each of a set of standard locations and the results averaged.
Listener position. Listen in the same room, using a single listener in the same
location. If there are multiple listeners, on successive evaluations it is necessary
to rotate listeners through all of the listening positions.
Relative loudness. Perceived loudness depends on both sound level and frequency,
as seen in the well-known equal-loudness contours (Figure 19.3). Consequently,
something as basic as perceived spectral balance is different at
different playback levels. In comparing the sound from audio components,
loudness levels must be very closely matched. If the frequency responses of the
devices being compared are identical—that is, fl at—as in most electronic devices,
it is a task easily accomplished with a simple signal like a pure tone and a
voltmeter. Loudspeakers are generally not fl at, and individually they are not
fl at in many different ways. They also radiate a three-dimensional sound fi eld
into a refl ective space, meaning that it is probably impossible to achieve a
perfect loudness equality for all possible elements (e.g., transient and sustained)
of a musical program. There has been a long quest for a perfect “loudness


meter. Some of the offerings have been exceptionally complicated, expensive,
and cumbersome to use, requiring narrow-band spectral analysis and computerbased
loudness-summing software. A few years ago, when Aarts (1992) suggested
that B-weighted sound-level measurements were adequate, many of us
were greatly relieved. More recently, that option has been challenged; additional
research suggests an even better solution (Soulodre, 2004; Soulodre and Norcross,
2003). Fortunately, it also is simple to implement: a high-pass characteristic
somewhere between that of B and C weighting but with no high-frequency
roll off. Figure 17.4 shows the standard A, B, and C weighting curves, along
with the new proposal, the RLB (Revised Low-frequency B) curve. Nevertheless,
the more different the spectra of the program material being compared, the
greater will be the diffi culty in achieving a loudness match. There is no set
of sound level adjustments that could ensure equal perceived loudness for the
loudspeakers shown in Figures 17.2 and 17.3; a balance achieved with one kind
of signal would not apply to a signal with a different spectrum. Fortunately, as
loudspeakers have improved and are now more similar, the problem has lessened,
although not disappeared entirely. This principle, of course, also applies
to the loudness balancing of the channels in a multichannel audio system. The
typical method of using a midfrequency band-limited noise cannot be reliable,
but, fortunately, perfection is not necessary in that application. In all of these
cases, if in doubt, turn the instruments off and listen; a subjective test is the
fi nal authority.
Dernière édition par fredoamigo le 23 Mar 2022 13:38, édité 1 fois.
fredoamigo
 
Messages: 4322
Inscription Forum: 04 Déc 2010 10:21
Localisation: Aix en Provence
  • offline

Message » 23 Mar 2022 13:36

syber a écrit:
Et la neutralité chez Cabasse, JBL ou Kef ou TAD ... ben, ça sonne pareil ! Comment pourrait-il en être autrement ?

A moins que la neutralité soit subjective ? :lol: Pourquoi pas ? Plus rien ne m'étonne sur les forum HiFi !


Certains experts gloseurs et les audiophiles durs et purs superposent leurs arguments contre ton hérétique expérience :lol:

Les fameux "tests ABX d'enceintes" c'est une vue de l'esprit, seul au monde Harman dans son local y arrive à peu près ,et encore ces tests Harman sont critiquables, because le positionnement en hauteur d'une enceinte lors du "test ABX" change notablement le "son" de ladite enceinte ....

La "neutralité" de l'enceinte à pour limite la "quasi impossibilité" de la comparaison fine ( sauf enceinte mal fichues).

La configuration dans mon profil


Ingénieur du Son
5.3 Local d'écoute traité 130 m3 . Filtrage & corrections FIR, processeur QSC by Ohl + 4 YAMAYA NS-1000x avec 4 médiums Be + 2 tweeters FOCAL Be + 4 SVS + 3 KEF coaxiaux Q 100 + écran LG 88' 8k
Avatar de l’utilisateur
Igor Kirkwood
Membre HCFR
Membre HCFR
 
Messages: 10166
Inscription Forum: 30 Nov 2007 11:17
Localisation: Briare
  • offline

Message » 23 Mar 2022 13:53

Igor Kirkwood a écrit:
syber a écrit:
Et la neutralité chez Cabasse, JBL ou Kef ou TAD ... ben, ça sonne pareil ! Comment pourrait-il en être autrement ?

A moins que la neutralité soit subjective ? :lol: Pourquoi pas ? Plus rien ne m'étonne sur les forum HiFi !


Certains experts gloseurs et les audiophiles durs et purs superposent leurs arguments contre ton hérétique expérience :lol:

Les fameux "tests ABX d'enceintes" c'est une vue de l'esprit, seul au monde Harman dans son local y arrive à peu près ,et encore ces tests Harman sont critiquables, because le positionnement en hauteur d'une enceinte lors du "test ABX" change notablement le "son" de ladite enceinte ....

La "neutralité" de l'enceinte à pour limite la "quasi impossibilité" de la comparaison fine ( sauf enceinte mal fichues).


Pourquoi ça ? si l'enceinte mesure bien en vertical ? (hors mis la réponse en peigne généré par le sol )
fredoamigo
 
Messages: 4322
Inscription Forum: 04 Déc 2010 10:21
Localisation: Aix en Provence
  • offline

Message » 23 Mar 2022 14:14

fredoamigo a écrit:Pourquoi ça ? si l'enceinte mesure bien en vertical ? (hors mis la réponse en peigne généré par le sol )

A cause des réflexions et des modes :wink: .

C'est pour cela que pour comparer 2 enceintes, je les place côte à côte et le tweeter à hauteur d'oreille.
Esscobar
 
Messages: 5173
Inscription Forum: 27 Sep 2016 18:10
  • offline

Message » 23 Mar 2022 14:58

Jean-Luc67 a écrit:Mais la question de départ était :

Y a-t'il une signature sonore en fonction des marques ? Seriez-vous capable de le décrire ? De le reconnaitre à l'aveugle ?

C'est quoi le son anglais ? Japonnais ? ....


A une époque on parlait effectivement de son anglais, français, allemand... mais en fait, le plus souvent, il s'agissait de prendre un modèle d'enceinte connu et dominant le marché, et de dire c'est le son "de tel pays".

Par exemple, la ProAC Response 1 était très populaire en son temps, donc on disait que c'était "le son anglais". Les grosses JBL des années 70, c'était "le son américain", etc.
Pio2001
Contributeur HCFR 2019
 
Messages: 8386
Inscription Forum: 07 Oct 2003 12:50
Localisation: Neuville-sur-Saône
  • offline

Message » 23 Mar 2022 14:58

Esscobar a écrit:
fredoamigo a écrit:Pourquoi ça ? si l'enceinte mesure bien en vertical ? (hors mis la réponse en peigne généré par le sol )

A cause des réflexions et des modes :wink: .

C'est pour cela que pour comparer 2 enceintes, je les place côte à côte et le tweeter à hauteur d'oreille.


Oui ça j'avais bien compris et même avec des enceintes bien conçus avec une réponse verticale correcte tu peut te permettre quelques légères fluctuation en hauteur de façon a écouter dans la fenêtre d'écoute (LW sur le spinorama ) soit 10 dégrée en vertical ..donc pas obligatoirement dans l'axe strict .

D'ou ,Ma question initiale a igor qui disait que le son change notablement avec le positionnement en hauteur et pour quoi je suis dubitatif tant qu on restent dans la fenêtre d'écoute .
fredoamigo
 
Messages: 4322
Inscription Forum: 04 Déc 2010 10:21
Localisation: Aix en Provence
  • offline

Message » 23 Mar 2022 15:56

fredoamigo a écrit:D'ou ,Ma question initiale a igor qui disait que le son change notablement avec le positionnement en hauteur et pour quoi je suis dubitatif tant qu on restent dans la fenêtre d'écoute .

Quand on fait une mise au point d'un système, on le fait pour une zone d'écoute. S'en écarter peut considérablement modifier la réponse en fréquence d'où le fait qu'il faut être rigoureux lorsque l'on fait une écoute comparative.

D.

La configuration dans mon profil


La bougie de ton intelligence n'éclairera ta vie que le jour où tu arrêteras toi-même de souffler dessus !
On ne peut pas donner à boire à un âne qui n'a pas soif
Avatar de l’utilisateur
Dagda
Staffs DIY & Haute-Fidélité
Staffs DIY & Haute-Fidélité
 
Messages: 12242
Inscription Forum: 22 Déc 2005 14:53
Localisation: Fontenay sous Bois
  • offline

Message » 23 Mar 2022 16:19

Dagda a écrit:
fredoamigo a écrit:D'ou ,Ma question initiale a igor qui disait que le son change notablement avec le positionnement en hauteur et pour quoi je suis dubitatif tant qu on restent dans la fenêtre d'écoute .

Quand on fait une mise au point d'un système, on le fait pour une zone d'écoute. S'en écarter peut considérablement modifier la réponse en fréquence d'où le fait qu'il faut être rigoureux lorsque l'on fait une écoute comparative.

D.


D’accord avec toi mais Igor parle de changement notable .. quand je fait des écoutes extérieures , la première chose que je fait c'est écoute assise vs écoute debout (je mesure 185 CM ) s'il y a un changement notable c'est que les enceintes on un soucis ..chez moi mes enceintes sont surélevés de 23cm le pavillon est a 110cm et aucuns changement notables sur l’écoute assise ou debout .mais pour une écoute dans le cas d'un test ab oui bien sur le même emplacement dans la même fenêtre d'écoute est recommandable ..
d'autres part , nous sommes beaucoup plus sensible au rayonnement horizontal que vertical vu le positionnement de nos oreilles si nous avions une oreille sous le menton et l'autre au dessus du front il en serait bien sur autrement et le rayonnement vertical serait prioritaire ..raison aussi pour la quelle il n’apporte rien a l'imagerie tant en vertical qu en diagonal .
fredoamigo
 
Messages: 4322
Inscription Forum: 04 Déc 2010 10:21
Localisation: Aix en Provence
  • offline

Message » 23 Mar 2022 16:30

Ca dépend de la dispersion des enceintes.
Si le hors axe (vertical et horizontal) est maitrisé, la sensation de différence en bougeant dans la pièce ne sera que peu marqué.
Je constate quelque-chose de similaire chez moi (même si je ne mesure que 176cm :mdr: ) du fait que j'ai un coaxial dans les enceintes qui n'est pas hyper directif ;)

D.

La configuration dans mon profil


La bougie de ton intelligence n'éclairera ta vie que le jour où tu arrêteras toi-même de souffler dessus !
On ne peut pas donner à boire à un âne qui n'a pas soif
Avatar de l’utilisateur
Dagda
Staffs DIY & Haute-Fidélité
Staffs DIY & Haute-Fidélité
 
Messages: 12242
Inscription Forum: 22 Déc 2005 14:53
Localisation: Fontenay sous Bois
  • offline

Message » 23 Mar 2022 18:43

Dagda a écrit:Ah oui d'accord :lol:
Bon là on rentre aussi dans la problématique de savoir comment ça a été enregistré ;)


Oui tout à fait…. :zen:

La configuration dans mon profil


lecteur Luxman D08U, serveur SOtM sMS 1000 Eunashu + sPS 1000 alim, Dac B Audio, ampli Ypsilon Phaeton SE , enceintes JEAN MAURER 370 F, câbles Plénitude évidence + Ypsilon , meubles Woodlink.
marius30
Membre HCFR
Membre HCFR
 
Messages: 520
Inscription Forum: 01 Mar 2010 19:51
  • offline


Retourner vers Discussions Générales

 
  • Articles en relation
    Dernier message