So I have to ask "You think you've got purity?" Its a fancy notion that with the perfect room, drivers and loudspeaker system would make sense but the days of purity in the traditional audiophile sense are over and indeed never existed in a true sense. There's dinosaurs out there that continue to condemn such digital correction methods but these are so backward in their thinking that they can't comprehend nor understand how to implement it correctly to reap the benefits. Make no mistake I condem no one based on how they run their own setup but to cite antiquated audiophile scripture even whilst time ticks on shows they have no place in the future of audio and seek to hold it back for their own sake. Digital audio is still in its infancy with a bright future and much is obviously going to be forthcoming in the next decade or two. On the flipside, analogue' days are very much numbered and its not hard to see why, its inflexible and almost inferior already but to many its become the norm, what they grew up with and have seen mature - afterall better the devil you know.
Some had luck with EQ'ing and DSP'ing their system. Some had luck creating a perfect data source to conversion. Some had luck to dial in a speaker that complies 100% to phase linearity, time alignment, and flat amplitude.
We have to see all this in the same context, and in the next reply here I have taken the liberty to use a DEQX as an example for my point of view:
1. To really compare which approach would be the best requires several things. It must be the same speaker. It must be the same room. It must be the same amps etc. It must be the same music source and so on. I can't see any chance that anything can be concluded unless such a comparison is done. There is no science, to cry it out loud, that red is a better colour than green. It's nonsense.
2. One stance is: To me a DEQX or whatever really improved a lot to what I was used to before, dialling everything in over time.
The Other stance is: My system is faultless and any mess with the bits would kill my sound.
If we look at 1 and 2 not much can be learned from that.
If we look a way from that a DEQX could cure many things on imperfect made speakers, and to an extend also would be able to present a "Better listening room", then my next question would be this:
Scenario one:
1 You poses a decent listening room.
2 You have a bit perfect source
3 You have decent amps
4 You have perfectly time aligned, phase correct and dialled in flat amplitude response using passive components to achieve it.
5 The room it self manipulate with the amplitude of natural courses. A speaker is a smaller box in a bigger box (room).
6 You find your best speaker position and the best listening position, the sweet spot for that set-up.
Then: No doubt that such a system just delivers the music unspoiled. Let it be a bad recording or a good one. Such a system will always sound incredible good if the used drivers allows for it. OK?
Scenario two:
1 You poses a decent listening room.
2 You don't have a bit perfect source due to the DEQX in the game
3 You have decent amps
4 You don't have perfectly time aligned, phase correct and dialled in flat amplitude response using passive components to achieve it.
5 You find your best speaker position and the best listening position, the sweet spot for that set-up.
6 You use the DEQX to create the time alignment, 0 degrees phase turn, EQ, flat response and room correction.
Then: No doubt that this can sound good too. Let it be a bad recording or a good one. Such a system will always sound incredible good if the used drivers allows for it. OK?
Now to my 1000 $ question:
What if we assume you take the analogue perfectly dialled in speaker. Threw out the passive x-over and dialled that in via the DEQX changing slopes, EQ'ing and dragging a piston speaker with a phase turn, into a piston speaker with 0 phase turn. Let the amplitude response be flat in even way as the ordinary filtered speaker. Corrected the room amplitude response.
Will this harassment of data manipulations through everything corrected DEQX to speaker route stand against the bit perfect, speaker perfect, non room perfect set-up?
How can we know? How can we know if we don't try it? And have anyone of us really tried what I describe here?
We can have our religious standpoints, but will that prove anything scientifically?
So?
I would not know it, unless I have 4 physically equal speakers. Let 2 be one set and the other 2 be the other. Diall them in each way. Then compare and conclude for myself.
And this will not happen in my room simply because I don't want to invest in the experiment in time and money.
I have very little chance to evaluate if one is better than the other. And even my few aquatinted handshakes with TACT/room processing sessions, and similar attempts, does not allow for a conclusion as this is not what you can achieve using a DEQX based solution.
The only thing I can derive from say the TACT/room processing is what I heard:
With out room processing: A normal good sounding system. Lalala...
With room processing: Wow...the room "disappeared", the speakers turned to 0 phase (not time aligned drivers of course). Only the summarised amplitude showed 0 phase turn and flatness.
It was as if the walls framing the room just fell down and left me floating in the air together with the sound. Fantastic, I thought.
But 5 minutes after I simply got a headache to listen to this. My brain were so disturbed by anti-processing against it, I think. It did not sound of music at all as you find it in a concert hall.
I could only conclude that DSP'ing a speaker to technical perfection did not rival even the sound of the non-corrected speaker and the untreated room.
But this conclusion doesn't count for my believes that a i.e. a DEQX'ed system might appear differently to me? I simply don't know.
At that moment I concluded. Don't harass with my brain with DSP'ed sound please.
Now having said all that,which doesn't exactly lead to any conclusions, and is derived out of a dinosaurs mindset, who has no posssibility to see into the future imagining through clairvoiance skills, that digital manipulated sound will beet the dinosaurs back to the past of B.C., let me gently quote a friend of mine who has a more hostile attitude to DSP'ed sound than I have, and furthermore is the designer of one of the bit perfect players available:
Quote:
Let me first stress on how very very much the stage (width, depth, height) already varies with bit perfect playback over different XXHighEnd versions, those versions only differing in jitter.
Other means (with the phenomenon consistency as a key word) determine what would be the best jitter signature, and from (a.o.) that follows the stage ...
That stage, varying per album/recording has -to my determination- one recognizeable property only : very flat is never good.
Or to my subject : very flat will not have been reality (from in-room normal recording, or from mixing). But :
Everything else but flat is IMO not determined. Instead, I must just trust that where the jitter signature is ok, what is perceived is reality.
What is the message ?
The message is that a wider stage says completely nothing about "good" or reality. The message is also, that with DSP stuff id is dead easy to create e.g. a wider stage. Is that good ? no.
The only thing you would be doing is masking the other anomalies in your chain.
As often, things can be compared with photography;
When you have an unsharp picture, you can sharpen it just by adding noise. It really works. Sadly, now there's noise in the picture and you now see *that*.
With (computer) music playback too, you can add noise (Foobar has a plugin for that somewhere). The result ? exactly the same as with images : a more crispy sound.
But be honest ... *knowing* that this is achieved by adding (audible) noise, would you go for that ?
I put myself to the task of achieving the best SQ possible by means of one thing only : as much 1:1 playback as possible. This is applied in XXHighEnd to my best knowledge, this is applied in the nos-DAC I use, this is applied in the (speed of the) CrazyAmps I use, this is applied in the "distortionless" representation of the Orphean horn loudspeaker I use.
Do not take the direct or underlaying messages in the latter as some blahblah and "I have the best playback system on the planet". It is *THE* thing to hunt for (1:1 playback), and it just BRINGS you the best system on the planet. Don't believe it ? hop over again.
Guys, I and a few others have been there for the past 24 months or so, where indeed in all directions the 1:1 playback principle was applied (mind you, this is the opposite of DSPing). The improvements by it are not known in this Milky Way, if you only think about the short time where it happened.
A lot of you share these improvements, although for a small part only : XXHighEnd. I can tell you that at least the same efforts were applied to the Orphean-MKII, as well as the development of the CrazyA amplifier. This is (except for a few) what you NOT have, which makes you blind in judging.
Ehh ... about what ? right, about applying digital means in the DSP area, which only WILL destroy.
Oh, it might come to you as better allright, and in your situation it might even be completely legit. As legit as a masking cable to cut sharp highs ...
Digital fooling has an impact that surpasses analogue fooling by far. Think of this one small example only :
Not bit perfect playback, in most of the (well tweaked PC) systems emerging from (unnecessary) dithering only, implies unintentional (not on the recording) volume changes of the smallest digital step of 1/65536. You wouldn't even be able to twist the volume knob that little !. But it's audible anyway ... (various anomalies are the result, but think of the stage changing).
Btw, did you know that the best way to perceive stage changes is to listen at "sweet spot distance" but right in front of one of the speakers ? This is the best means to perceive sound coming from the left speaker only (when sitting at the left side), or the sound coming still from the middle(ish). The more the latter happens, the better it is.
Another quote from a dinosaur:
As stated elsewhere before, when in the digital domain the sound is already very much different between software players (all being bitperfect) using exactly the same PC, DAC, Amp and speaker route makes me clear that manipulating is changing things in one way or the other. Call it jitter caused by whatever things used but then in the least possible destroying way.
Add some DSP to it, some IC's and more things will be changed for the worse in that respect already.
Is this not the truth? Or is it ONLY getting better by adding a digital EQ to the system?
You all (i.e. pro digital DSP people) only talk about the major advantages but what do you have lost along the way? Why are you not worried about that?
This part is what worries me the most and not without reason...
Perhaps one needs an honest and transparent system to hear those changes. It will already be more difficult if the source is corrupted beforehand (upsampling, jitter, etc.).
About a device that is rather popular for EQ, the DEQX:
The latter is about the 16 bit processing the DEQX would apply which really is nothing to hope for results. Even with 24 bits and a digital volume only, the sound gets so much degraded that there's *really* no hope for a better net result.
The least it would need is 32 bit digital input, hence a player that would output that. So we talked about adjusting XX to that, which just can be done.
Note for those who come up with it : E.g. Foobar can do it.
... But what to do if you don't like the sound of Foobar at all (as how we percieved that at that time).
Now, all together - and that's how complicated it is - my suggestion of just connecting the DEQX *is* legit, because the potential remainder of it all will fail anyway. That is, unless you are able to justify DSP in the 16 bit domain or chose Foobar as your player ...
(please note that this would come to a similar decision as throwing the 6K CDP out, and use a $100 instead to let the DEQX work -> hard to decide for).
What the DEQX discussion comes down to, is that actually the device itself flaws. It flaws for its processing with a number of bits that equal to the input.
This is unrelated to the DSP subject itself ...
Oh well ...
Let me finish with that it may be not the best idea to point out how people should listen (more) to live instruments, because IMHO the gang overhere is too profesional to not do - or interpret like that. Remember, this (work) is not about setting up the best system for our own room ... it is about creating the best elements (for any room). Might you not have noticed, I for one plain buy the instruments in order to compare.
And then the biggest bang : in my room I listen to complete live instruments from the audio playback chain. This by itself is dangerous to say, because 5 months ago I already said it, but it still improved. But mind you, 6 months ago I sure was not. Now :
Anyone being sure that he is listening to live instruments (start with the cymbals !) and using DSP (needed to get there or not) is entitled to tell the other to improve (by whatever means). But be very careful that you indeed listen to these live instruments ... and that your jaw won't drop when listening at my place.
I know how dangerous it is to say all this, but might it be true in *your* eyes (ears hehe), you know that DSP is not needed.
Quote ended
Let me finalize:
It is not for me to say which of the approaches is the right way, and maybe I could even accept both having tried both (at the same time), which I didn't, but only know that there are basic rules out there that sustain my Piston Preach stated and supported by those links I passed.
If no 'Piston' is to be found in your speakers after any attempts done just to throw in drivers in any set-up...I believe it does not matter what **** we apply to distribute frequencies, equalize them, manipulate them in the time domaine, we cannot believe anyhow that the final result is really what was intended by the recording it self.
And the recording itself, namely our source for playback, has certainly it's own problems to deal with, having all of us to believe this is real instruments we should listen to.
My point in all this is just to come to the speaker itself, the final audio device in the playback chain that delivers all what's behind it the sound you should trust and recognize as music.
For my agenda I don't give a shit about how it's done, just it's done. I heard that digital manipulation can make quite some (word not known) disasters to the sound. That's why I keep my fingers out of it. I dare not to do the trial as some of you really think doesn't matter because you made it work for you and maybe it doesn't matter?
Room correction from that camp I heard on several occassions. Huh..that gave me a headache after 10 minutes. Sorry it just did.
Are we deluted?????? Or is it all in our minds that set up the stage for this theater play:
My background and my thoughts has to be interpreted with the passed away guy with whom I discussed sound the most, and who took the oportunity to take a few notes before he passed away:
Quote:
At first I most emphasise that we are human beings. We have
developed a language and an enormous capacity of memory. We can
remember. That capacity alone is what differentiates us from the
animals. But more important, the simple fact that we have survived the
harsh nature as animals, despite our as single individual very low
probability of surveillance, tells that we further must have developed
our sense of hearing. Remember that half the time of your existence is
in the dark and further in sleep.
When you listen, the brain sorts in the signals, building up an
understandable picture of the event, based on recognition of sound and
reflections built up from early childhood if it can. The result of this
selection is what you seem to hear.
Listening to reproduced sound and the recorded overtones and
reflections are mixed by distortion or disturbed in phase, the brain
can't detect them correctly and therefore they will be interpreted as
sound formed around the instruments, whereby the sound stage
becomes flat.
If the amount of low level information is lesser disturbed, then the
brain will detect it as filling the room between the artists, the walls and
the ceiling as reverberation and hopefully some of it as overtones
attached to the single instrument. You can now detect a room, but still
it is attached to the sound of the instruments, as if the artists turn their
back to you. The instruments become like reflected sounds supplied
with some sort of distortion, hard for the brain to interpret, why it puts
that on the instruments, and they at some notes sound a bit distorted.
NB! Sometimes it helps turning the absolute phase.
First when all information are reproduced sufficiently correct, the
brain can do its job, to separate the instruments from the sounds from
the surroundings. Listen for the silence between the sound and its
echoes.
The needed information for a good perception is normally present in
many recordings, but can be very troublesome to dig out.
That production of discs and records vary so much in quality, is an
other story. But let us communicate to find good labels and discs, easy
for the brain to understand.
The brain tries to make sense in the tiny sound of noise and
wrongdoing, and will try to interpret them as parts of it all. If they
can't be translated as overtones or reverberation, what they often will
be, we'll hear them as distortion. So even if you think, you have a big
sound of reverberation, it doesn't mean that it is in order. To find out if
it is, you should listen for the silence between the primary sound (the
artist) and the secondary sound (the reverberation), clearly heard on
recording of classical music or recordings from a church.
A phenomenon, you further have to take in consideration, which the
brain can detect, is the absolute phase. It can be heard as distortion or
as an unsettled picture of sound.
This phasing must be correct, else you will not be able to judge the correctness of the reproduced signal at all.
The absolute phase differs from disc to disc and can furthermore differ
in one take from instrument to instrument, believe it or not.
All needed by the recording should according to Richard Heyser be a
single clap of a pair of hands. In that clap all necessary information for
later improvements of the recording are present. As it is now, you are
the judge. There are no help to find anywhere, than in your brains
capacity to distinguish between over- and under-pressure. It would be
wonderful if someone could develop a device that could tell us this
absolute phase from the signal itself. It should be possible, as
transients have a tendency to generate a low frequency unbalance,
which could be used for that purpose.
Try to change the absolute phase, playing a recording made in a
church. If you can't hear the difference, something is very wrong
somewhere in your equipment.
This absolute phase has also to be correct, all the way from the main
outlet through cables and components to the loudspeakers.
This goes for power source, the one you can feel with your fingertips
on the cabinets, and equally important, cable direction must be correct.
Why? I don't know for sure.
A bet would be that it is a question of treatment of distortion slightly
different for the positive and negative half caused by net polar diodeeffect
between the crystals and the facts that high and low level aren't
treated equally. To make it even more complicated the direction is
further dependent of the frequency the wire is carrying. When it is
used for digital transfer, you can't be sure that the direction is the same
as for analogue transfer.
Should you be the owner of a single-end amplifier you even have an
absolute phasing between that and your loudspeaker to complicate it
all.
The brain, the near future, and some thoughts.
In this chapter I will try to give an explanation of how I think the brain
works with sound. It must be understood that our hearing is the latest
of our senses to be developed and that it is the most important one for
our surveillance, as it out of our 5 senses is the only one always turned
on. Even when you are unconscious it still works. You can't react on
its information, but they will be stored no matter what.
Our nerve system has a reaction time at about one tenth of a second.
In this little time the brain interprets the information received, before
they are presented for your conscious mind.
From the information it somehow builds expectations of what to come,
to verify rhythm and melody in music or concentrate on speech,
whereby it suppresses disturbing sounds. It so to say reduces and sorts
in the amount of information received from the ear nerves to
concentrate on, what it expects to come – to listen for.
But there is also a short cut, always open for transients and some
unexpected silent sounds. These serve as a signal of danger, to zoom
into and especially listen for in the sounds treated by the brain. These
specific sounds serve in the same time as a trigger signal for production of adrenaline - surely a reminiscence from our wild life.
The silent sounds are of great importance. Just remember, how scaring
tiny sounds in silence could be from childhood in the dark.
These silent sounds are even stranger - how can we distinguish them
from other more noisy sounds? Simply because they are not expected.
They are out of order so to say.
If you are a trained listener, you often feel these signals more than you
hear them, you get warm or irritated - the adrenaline productions is
raised. A fact you are not consciously aware of.
By use of these signals, the brain can, so to say, look into the future
(1/10 of a second or more) and simultaneously use them in the sort of
the sound received. It if necessary even can clear the working area
from which the conscious mind is fed, prepared with all capacity to
recognise the echoes of these trigger signatures. Some information is
thereby left untreated – masked - and so to say not heard. This is
strange but true. Further it can listen for these recognisable echoes
deep into the noise around us. (Up to -20 dB below the level of noise.
Experienced in space communication).
The brain does more than that. Based on music or sound received, it
somehow builds expectations for further development.
When in a piece of music, unknown to you, a wrong key is struck, or
your loudspeaker colours one tone, you react. Why? You know neither
the piece nor the specific instrument.
Should it be a Steinway grand, its resonant character doesn't bother.
Modern music, where the development can be hard to predict, is of
most music lovers heard as noise.
A tone from a clarinet, sampled and used for the rest of a keyboard,
sounds wrong except from the sampled one - again how can we know?
I'm sure that our perception of sound is heavily based on predictions.
Are they right you feel good, and starts singing along. Are they too
often very wrong you get irritated.
Pianists, to get the music more tense, earlier used a playing technique
where the rhythm was changed just a little bit, called rubato. -
Disturbance of expectation.
Before we continue, I must emphasise, that sound happens in the run
of time - that you can't freeze it, as you can with a picture.
All that really matters, are the brain and its tremendous work in the
dimension of time, with that enormous amount of time-distorted
information.
I really get more and more impressed of its capacity. That our hearing
never rests even if you are unconscious, and that it is the last of our
senses to be developed, tells the importance of that sense in particular.
It is well known that closing your eyes and open your mouth will
improve your hearing capacity - you look foolish but what ever.
Anything that helps you understand the event better, should be
judged as good no matter, what measurements say.
It is e.g.. well known, that distortion distributed in the right manner
makes sounds more realistic, than with no distortion. Does air distort?
Is that distortion part of our expectations?
It is also well known that a loudspeaker with linear frequency response sounds wrong, compared to one with mild decaying level
towards the upper end.
But beware! There are traps of simplicity and emotional taste for the
brain within to rest.
To understand, what I mean, think on pictures, painted contra
photographs, or photos with low contra high resolution, graphics to
pictures with a myriad of grey tones. What do you prefer?
You should of course prefer that with a myriad of grey tones following
logic, and none of the others. But all the different ways of reproduction
can be used, for you to see the subject. The principal question remains,
if this analogy can be used for our hearing - which does the brain use?
I would guess the last two in combination. Graphics for instant
recognition and gradually within parts of a second adding more and
more detail much like a painting is started from raw sketch to the end
result. In this work many hear it as right that the reproduction is
marked with a multitude of resonance. Much the same as the intensity
of colour on the TV is chosen too high. No matter how pleasant it may
seem – it is wrong.
Our brain only needs few seconds of sound, to manipulate with the
signals building a kind of basic understanding of the sound received,
and expectations of, what to come. This ability creates by itself traps at
listening, as the brain will try to glamorise it all, it's an active part,
especially experienced with musicians, who as critical listeners often
are of no use.
If you are a trained listener, you possibly can listen to a whole piece of
music and detect some information of the reproduction. But if you
really want to listen for anomalies, you must cut the time span down to
2 - 5 seconds with pauses on minutes. In this way you will learn to feel
the amount of work, your brain must perform especially in the first
second. It is like physical work - you get warm, if the work is hard,
and it shouldn't be.
To make it all even more complicated, our brain is a multi-way
listening device, where all senses plays their part simultaneously.
Our methods of measuring are normally done in one dimension. But
the 'Melissa' way, using a kind of step response, from which all single
parameters as well as their time dependency is calculated, could serve
as an analogy on a possible behaviour, done likewise by our brain. Our
hearing capacity is influenced by the resonance of the auditory canal,
which is coincident with the most sensible part of the Fletcher-Munson
curve.
There must be either an enormous part of memorised calculations or a
chain of comparisons done again and again for the decaying echoes, or
a third possibility as the bit rate is known to be rather low, but with a
great number of parallel connections.
The way the brain works must explain, why faults always are clearly
heard in treble as well as in bass, especially when the low sensibility of
hearing in these parts is taken in consideration.
It is well known that the brain can add deep bass to a sound, if the 2nd
and 3rd overtone from the deep bass note are present and the deep bass
not. I'm sure that the same procedure is valid for the treble. It is much more a question of phase than of level.
No matter how, our brain receives impulses from our ears constantly,
and finds somehow connections in time, interpreted to music, speech
and reverberation.
It is like watching a movie where every picture is doubled with its
negative. Then there should be no picture at all. But if you now
displace the film of negatives one picture, there still should be no clear
picture. But now there are disturbances reflecting the differences
between the two to show the movements in the dimension of time.
A waterfall plot illustrates the reflected sound and amount of
resonance. The immediate plot being moved into the paper with
constant speed should show how the sound itself develops. This
movement will create the time in which the sound exists, and the
reverberation following it in a further time span to form that mess from
which your brain must extract the right information – not a simple task
at all.
As you probably know, e.g. bats scream and build their orientation of
the surroundings from echoes of this sound. I believe that we
unwittingly use the same technique, but with parts of the sound itself.
It is the drift of these parts in the time domain, shown on the waterfall
plot, that is the cause of the strange differences, between what you can
hear, but can't measure by one-dimensional measurements.
Sound is time dependent by nature, therefore all measurements
concerning reproduction of sound should likewise be done in the
time domain.
Our hearing capacity is more than 120 dB in the most sensible area
around 3 kHz, therefore we must look for connections from top to
bottom of this vast depth of level somehow centred around this
frequency, where also you'll find the resonance for the auditory canal.
Experiments with the digital medium has shown the importance of
these low level signals, present in the signal or induced by the
vibrating fields around the active parts of the hardware.
This medium has shown to be trustworthy as a storage medium, where
film, tape and LP's have failed. CD is further the only medium to hold
up to 100 dB of dynamic, it is still not enough, but it will come soon.
I have recently heard the DAD solution on 24 bit 96 kHz. - Surely a
quantum leap, even if the postulated resolution isn't totally true yet.
By playing you shouldn't expect more than 20-bit resolution for now,
but that seems to be close enough. To reach 24 bit of resolution would
demand so low level of noise that it will correspond to the noise from a
47-ohm resistor, why HDCD-technique probably will be needed, but
less will do – of course.
Inspired of this I bought a Denon DVD 3000. This machine is
particularly developed for DVD-video, where faults are visible. I
shouldn't wonder, if exactly that explains the neutrality of this
machine used as CD player alone. Our eyes are much more sensitive
for change in colour than our ears are for change in sound, probably
caused its resonant character. The bigger model is developed for the
sound marked with HDCD and DTS – all sails are set – but neutral it
isn't.
The "3000" was really a positive surprise. Not only have I got a new
machine but also a new collection of CDs. It really tries to get the
sound right. I can't help it, but also this machine has undergone some
modifications resulting in a sound I didn't believe possible from a
digital source.
The capacity of more information, especially of level, seems to cause
problems. The silent parts aren't treated satisfactory well yet with 16
bit 44.1 kHz nor by the speakers, therefore parts of the silent sounds
will be unconnected with the musical event, and be heard as distortion.
This goes especially for overtones, which with the new standard, are
treated far better. But at last we can detect these weak sounds, with an
increased demand on the loudspeakers and the rest of the chain. That
of course only if you want more than a soundpicture postcard.
This is for now my understanding of parts of the brain's work, and I
am more and more convinced, that in the brain's complicated work,
these trigger signals received the quick way play a much more
important role than thought of. Isn't it possible that these guide the
brain's ability to concentrate on specific sounds?
A slightly change of these sounds and their drift in frequency and time
gives a great change in the results the brain present to you, as the
sound received one tenth of a second ago. This could be a possible
cause of the spoiled sound from wire insulated with plastic. See under
"Cables"
Those of you, who have worked with loudspeakers, have probably met
this strange phenomenon. A change in the network for the treble can
very well have consequences in the bass response, and the other way
around the treble can start to lisp, so the whole frequency band and the
whole hearing capacity is strangely connected. In other words - every
single loudspeaker in a multi-way construction must follow the
mathematically calculated slope of level to the level of noise. That is
one goal to work for.
A multi-way loudspeaker normally doesn't act as a minimum phase
unit as a whole, which is the problem. But with my solution they all
together behave as one and form an allpass function, which also like
minimum phase systems where variation in phase and level is to
calculate. With speakers attached to it, it forms a new bandpassfunction
with fare broader frequency-band and a turn of phase
determined by the order and the Q of the crossover. But there is no
room for mistakes or rules of thump..
The turn of phase, the frequency response and step response is allimportant.
These should be mathematically connected as a whole, but
are normally not. The more of the loudspeaker's reproduction for these
to be in order, the better it is understood by the brain. I shouldn't
wonder if the step response would show to be the most important one.
It shows all three at once if we can present it in an understandable
three-dimensional way.
Let the enlarged middle-band be treated by one loudspeaker and add
treble and bass with care - you'll never go wrong. But if you want the
outmost from your speaker, you must go deeper. It can be done. The
first construction in "Loudspeaker in practise" is an example on that when normal/nearly normal speakers are used.
How the brain works in detail is still a mystery, but great efforts are
done in that area in order to reduce the amount of information to be
stored on disc. Also in the very interesting work on further
development of Q-sound, hopefully ending in full surround sound
created by two, yes! Two loudspeakers.
That I'm sure will bee the future - may these two be eminent.
We still are at the starting point of cleaning up in sound reproduction,
to prepare the whole system to handle these very increased amounts of
information now possible to store.
There is so much to do, as the hi-fi manufacturer still rummage about
in the fifties to the seventies - there is refinements yes! But no real
break-through. Hopefully this paper can serve to point out the
direction, in which we have to move and develop.
Quote ends here.
Happy New Year
Gerner
