Do We Want All Speakers To Sound The Same ?

[pma]

Two different real speakers will never never sound same. It is technically impossible. Only ideal sound sources would sound same. For this reason, I cannot see the point of this thread.

 

[Koeitje]

No, because some applications require tighter or wider directivity. I would like to see the same design criteria.

 

[Chrispy]

While I like the idea of the speaker ideal of accuracy I don't think our current tech is anywhere near it....and may never be, let alone speakers that could automatically overcome/adjust for room differences or at any spl and remain as accurate but maybe technology can solve that some day.....but I don't see it happening in my lifetime either.

 

[MattHooper]

Two different real speakers will never never sound same. It is technically impossible. Only ideal sound sources would sound same. For this reason, I cannot see the point of this thread.

I think that is taking too narrow a view of the question.

It can be approached from various angles:

1. If it were POSSIBLE to make all speakers sound the same...would we want that? Even discussing hard to reach, even impossible goals, can bring to light the logic of current trains of thought. (It's why thought experiments can often be enlightening). Someone answering "yes" to that question vs "no" could explain why, and that could indeed cast light on their general thinking on the goals for speaker design. Even if perfection is impossible, ideals provide directions for design goals, right? It explains why designs improve over time towards a goal.

On a more practical level:

2. I'm sure you wouldn't say "speakers will never sound the same" THEREFORE let's abandon any criteria for judging speakers. Clearly understanding speaker design CAN allow people to build more-similar-sounding designs if that is the goal. That is one reason why ASR can have the criteria it does: there will be certain similar features in the sound of the speakers judged as "Good Designs."

Even if you don't have perfection we are STILL talking about speakers headed more towards certain criteria than others (and sounding more similar for it), while dispensing others as "poor designs." So my question can still be viable - does one propose speaker designers OUGHT to adopt a design criteria that will make speakers converge more in their sound, and reject designs that fall outside that criteria?

 

[Andysu]

doesnt matter what we want

we could ask that all amps sound the same too, like all dacs

but the reality is we have a million companies all with different ideas on what sounds 'good'

i have one . its called real life sound and it is 100% free . matching speakers tonally isn't impossible to the small degrees of proper multi surround arrays . takes effort of getting out of the seat , walking around with microphone with REW sound generated , pink noise and watching the RTA and using some xyz coordination of getting soundly physical with the speakers , that are all stupid . spending tens of £ k speakers is mindless . only make them richer . some speakers have good wide dispersion others have narrow dispersion to each and how they are going to be applied in home cinema that can also i remind all of you , do stereo 2 channel .

what get more intimate with speaker that isn't even closely to be perfect with each speaker driver . mid range tweeters compared to ( compression driver horns ) rather have CD horn , all in one . but far from perfect . frequency is all over the place up and down . it needs to follow the same as the amplifiers frequency response . theoretically .

PEQ a lot of PEQ is needed as each frequency on the REW sine wave say start at maybe 8.000KHz then what does it sound/look like at 8.001KHz and continue upwards and onwards till you reach 20.000KHz a lot of PEQ is needed to dial in each PEQ filter to address each single speaker driver that is so cockeyed out of balance . then the amp becomes cockeyed so now everyone has some cockeyed audio gear and its all cockeyed measurable lol . only thing to worry about is the amp provides power of the PEQ to
provide enough PEQ for that speaker to speaker to hopefully when all manual PEQ has been addressed after month of manual PEQ . don't expect auto-EQ to even be perfect it is far from perfect .

manual PEQ if want it done right . run the frequency sweeps each time . to really do this would take few months . as each CD horn needs to be addressed . each speaker used for surround as to be addressed and up close on table/bench with microphone close to the each speaker drivers and each with their own DSP crossover/amp . yes it would be extremely and costly expensive . but that is the way to do it .

so can make some small trade off compromises of around a cutting few corners . they i personally go with the above ^ . rather than chasing my tail around . i rather get it all done .

passive crossovers are they all the same ? has anyone cared to connect them to computer and run REW sweeps graphs and compare each passive crossover ?

i be thinking does that bass driver measure same in the other same box ? i would test the impossible just for sceptical curiosity sake lol .

 

[Ken Tajalli]

. . . . others suggesting it ought to be the goal of any speaker to be accurate/neutral (which if followed through implies a best case scenario where speakers sound essentially the same).

Well I was one of those others, but the goal may not be exactly the same for all manufacturers.
A panel speaker (Electrostatic or Maggies etc.) manufacturer may not cater for Rock music lovers, but aim for Classical music lovers. They still have the same goal of low distortion and as even an FR as possible, but accept that some qualities of the panels would/may trade off for its weaknesses.
Then the end user, upon deciding his usage, can decide.
BTW, the listening room would always be in the equation. I may love a pair of Maggie's, but not have the right listening room for them.
Yes I would like a pair of panel speakers, that would fit and work in my humble living room, and has the sub-bass of a transmission line - but I can't get one.
So I decide, what mostly would fit my bill and choose.
Not all speakers are made for everyone, It's horses for courses.
What we want, is everything!
what we can get is something else.

 

[-Matt-]

@MattHooper: I hope you will take this in the light hearted spirit it is intended... You are possibly the most polite and engaging troll on ASR!

No, seriously, thank you for your thought provoking comments.

Even if meant in a light hearted fashion, I have to come to Matt Hooper's defence.

I think understanding of the world troll has changed over time, it used to mean someone who would say things purely to get an angry response. On the contrary, I think Matt likes to ask interesting/difficult questions because he is inquisitive, and honest in seeking knowledge. He is not looking just to rile people up, even if that is occasionally the result. Those that get riled up at the polite asking of probing questions, rather than resorting to the word troll, might want to turn that finger inwards once in a while.

Anyone who is particularly inquisitive is at risk of being called a troll, just through their persistence in pursuing an idea where others give up/fear to tread. I don't think that kind of person is actually a troll, although more and more people resort to using that term to attempt to terminate, oftentimes interesting, discussion.

If Galileo had access to the internet in his time, perhaps he would have been considered the biggest troll of them all.

Just briefly I'd like to clarify that this was intended as a joke (I thought I had made that abundently clear).

The crux of the 'joke' was that: to anyone who has read a few of @MattHooper's posts it is obvious that he is inquisitive and thoughtful and not a troll.

It seems to have missed the mark, so please accept my appologies for any unintended offence caused. I hope this can be the end of the matter.

 

[Tangband]

Do We Want All Speakers To Sound The Same?

The thread title may seem a tad hyperbolic, and clearly there will be plenty of nuance involved in replying to such a question.

The question arises first of all because I've seen some criticize Amir and ASR along these lines "That place is boring, they want all speakers to sound the same!"

That strikes me as a caricature. After all, I know members have owned all sorts of different speakers over their audiophile career, and it seems there is some nice variety in member's current speaker set ups.

On the other hand...might there be, in some sense, some element of truth in the proposal "we want all speakers to sound the same?"

After all, any effort to evaluate something like speakers, based on an engineering (and sometimes science) heavy approach will tend to arrive at some sort of "best practices" for speaker design, upon which speakers will be evaluated. It would seem that the general characteristics arrived at from research from Floyd Toole and others have provided such standards for this forum - so ones that are neutral (with the proper off axis response) are selected as "good" and those departing as "poor" to one degree or another.

And since an underlying goal for many ASR members seems to be "accuracy" the logical extension of this would seem to be that the more speakers tend to meet that goal, the more alike they will sound. Which at least implies that if all speaker manufacturers adopted these same goals "ideally" speakers (for any given frequency response) would sound closer to indistinguishable.

It's my impression that some (many?) on ASR would in a sense prefer the speaker to "sound like nothing" in the same sense that a good solid state amp would "sound like nothing." No character of it's own, just neutral, so one isn't 'listening to the system' or thinking about "how the system/speaker sounds" but is simply listening to "the recording."

If much of that does indeed capture some people's goal here, it would imply that..yeah, in some sense, "Ideally, all speakers would sound the same."

I'm not writing any of that to IMPOSE this view on anyone here, only as some talking points to get off the ground. This forum isn't a monolith, it's made of individuals with varying views, so I'm interested in YOUR response to any of these questions:

Would it be THE ideal, or your ideal, that all speakers eventually sound the same, if you could wave your magic wand and send things in that direction? If so why. If not, why not? Should they sound roughly the same, like most should sound close to neutral but you are good with variation in X, Y parameters? Or are you happy with the essentially "Wild West" approach as it has been - some manufacturers striving for neutrality/Toole-approved performance, many heading off in different directions? Do you see the general approach by Amir's approach to evaluating speakers as too narrowly defined and limiting in terms of vetting "bad" from "good" - or does it match your own ideals for performance?

The stereo system and all the flaws with it dont get much attention in this forum .
The spatial quality and the illusion of the recorded event ( yes, its only an illusion ) can be perceived very different by the listener when using two loudspeakers in a stereo setup. The room also contribute in very different ways, so do the setup of the two stereo loudspeakers. 20 cm different placement of the speakers changes the way we perceive the stereo illusion. The variables of the perceived sound is much bigger with two stereo loudspeakers than with one in mono.

Further - some speakers are constructed to be close to the frontwall, ie having much less baffle step correction than constructions made for freestanding use. The Klippel test system always favours loudspeakers made for freestanding use.
Dipol loudspeakers are very different sounding than ordinary loudspeakers.
With those facts, its clear that different constructed loudspeakers never gonna sound alike, even If our goal was that.

In recent time, much attention has been taken regarding the importance of good directivity. In the next 20 years, Im sure that even better loudspeakers can be made with 4 or 5 way dsp speakers,- this will lower IM distortion in a dramatical way. They still gonna sound slightly different though, and never as good as the real concert event .

Good directivity and a flat frequency response is very important and a goal to have, to avoid ” the circle of confusion” from studiomonitors to home speakers. They still gonna sound slightly different anyway, because there are two of them in stereo and the setup variables are huge.

 

[westyjeff]

I think sound is dependent on age, as I have gotten older and the cost of standing next to performance monitors has caught up to me I like the highs to be more pronounced. Bass should be there but does not need to overwhelm the room, mids need to be present and not subdued.

To be more on topic, as age has caught up to me speaker changes have been helpful to meet my hearing loss.

 

[Ken Tajalli]

. . . . . They still gonna sound slightly different though, and never as good as the real concert event .

Actually, I have been to so many Classical concert events at some famous London venues over the years.
Yes they were good, some were excellent.
But a finished recording, mastered and equalized expertly, using multiple microphones, can sound really wonderful - sometimes even better than the live event!
The speakers alone may not match the venue, but together with the right recording material, they may even surpass it.

 

[Tangband]

Actually, I have been to so many Classical concert events at some famous London venues over the years.
Yes they were good, some were excellent.
But a finished recording, mastered and equalized expertly, using multiple microphones, can sound really wonderful - sometimes even better than the live event!
The speakers alone may not match the venue, but together with the right recording material, they may even surpass it.

Yes, the illusion can be very good.

 

[YesChickenNuggets]

Is it possible given how much we don't control like the room? Even when we have a high accuracy television, and most are reasonably accurate these days, the background color of your wall can dramatically impact color perception. If your wall is a high saturation color, that color will be visually discounted from your TV. That's why color grading studios have a specialized neutral grey wall paint and light conditions which must fit a certain spectral distribution.

 

[Tangband]

Is it possible given how much we don't control like the room? Even when we have a high accuracy television, and most are reasonably accurate these days, the background color of your wall can dramatically impact color perception. If your wall is a high saturation color, that color will be visually discounted from your TV. That's why color grading studios have a specialized neutral grey wall paint and light conditions which must fit a certain spectral distribution.

True.
I would say that the flaws in the stereo system with only two loudspeakers are much bigger than your example with the background color. Sound travels trough air, and bounces on the walls in the listeningroom, coloring the sound in a big way.

 

[MattHooper]

Actually, I have been to so many Classical concert events at some famous London venues over the years.
Yes they were good, some were excellent.
But a finished recording, mastered and equalized expertly, using multiple microphones, can sound really wonderful - sometimes even better than the live event!
The speakers alone may not match the venue, but together with the right recording material, they may even surpass it.

This invokes questions of what one is trying to get out of a sound reproduction system. I've included what I seek in my answer to the question I posed in this thread. Though I don't want to make "live vs reproduced" the subject of this thread...some musing...

Many eschew the idea of seeking more realistic sound from a speaker system, or using real life sound as a criteria. (" We only have the recording, not the real sound, so we should just stick to evaluating how accurately that recorded signal is produced, through looking at distortion measurements etc").

On the other hand, relevant to this thread, the criteria Amir and many are using to vet speaker designs is generally based on the results of research, including blind testing, in which patterns were identified in what type of sound people prefer.

Which leads to the question: WHY did people prefer X type of sound?

It's been a while since I revisited the particulars of the research, but as I recall a number of "well recorded" tracks, ones taken to be of "good sound quality" were used (e.g. Fast Car – Tracy Chapman or some such). That makes sense: if you are trying to study what is possible in speaker design, selecting "poor sound quality recordings" which would limit the sound quality wouldn't make sense. But then...what makes for "better sound quality" when selecting the tracks?

We are particularly familiar with the human voice, so it makes sense "well recorded vocals" would be included in the tracks, and that tends to mean more "natural sounding" recordings of vocals. Something that at least has the potential of more life-like.

So it seems reasonable to assume that part of what people were responding to under blinded conditions as "sounding better" where voices sounding "more natural, more real" where colorations deviating from the normal human voice were deemed less pleasing.

Someone more knowledgeable about the details of those tests could correct anything there. But if so, to some degree it would suggest that "comparison to the real thing" IS part of the criteria deemed as "good sounding" and which guide us to seek certain speakers. So to think that "we are just seeking accuracy to the recorded signal" is the goal in of itself, and comparing sound to real life is a non-starter, would be missing that piece of the puzzle.

But...anyway...that's a bit of a side step. Back to the main issue...

 

[Killingbeans]

So an important part of my question had to do with whether "we want all speakers to sound the same" in that sense - not just "what I want" but "what speaker designers OUGHT to do."

They definitely shouldn't be forced to do anything.

One of the marketing blurb staples that always makes me throw up a bit in my mouth is: "Music, the way it was meant to be heard."

I think there should be a certification for recording studios (if there isn't already one?) that can only be obtained if the playback setup in said studio adhere to a strict set of criteria for flat/neutral sound. Then the consumer can choose to aim for a home setup that also qualify for that certification if they truly want to "hear what the producer did" within a reasonable margin of error.

Besides that, speaker designer should, IMO, be allowed to do whatever the f##c they want, as long as they are clear about their intentions.

 

[MattHooper]

Is it possible given how much we don't control like the room?

In the "should all speakers sound the same" room correction can be incorporated in to that question. (So if speaker design combined with room correction could make all speakers 'sound the same/very similar' is that the goal people should adopt?)

Even when we have a high accuracy television, and most are reasonably accurate these days, the background color of your wall can dramatically impact color perception. If your wall is a high saturation color, that color will be visually discounted from your TV. That's why color grading studios have a specialized neutral grey wall paint and light conditions which must fit a certain spectral distribution.

Indeedy! As a long time home theater obsessive I paid tons of attention to the effect of the room on the perception of the image. For instance, I gradually went through the process of covering almost every visible area, speakers included, in black velvet so that it all disappears with the lights down. It was always fascinating what taking away the next visual distraction would do for the image.

In fact: I find that this also affects my perception of sound as well. I listen to music in my home theater room, which means I'm staring past my stereo speakers to a projection screen behind them. I have colored lights aimed at that screen, and I find my perception of the sound can be slightly influenced by whether the colored lights are on, or the regular lights instead, or if I'm listening in the dark, or if I'm closing my eyes.

 

[MattHooper]

They definitely shouldn't be forced to do anything.

One of the marketing blurb staples that always makes me throw up a bit in my mouth is: "Music, the way it was meant to be heard."

oh gawd, same here! I have a similar reaction to the audiophile trope (x component) "Sounds Like Music!" One of the most useless (and annoying) descriptors I can think of.

 

[MaxwellsEq]

Thanks @MattHooper for a stimulating set of questions.

I've had a thought - we manage to drive down distortion in amplifiers through the use of negative feedback (and before anyone declares a dislike for feedback, I can recommend Bruno Putzeys' "The F Word".

Can we do something similar with loudspeakers? Everyone here using PEQ and REW etc to control some room nodes or to crossover to a sub is doing "manual negative feedback". But it requires a lot of work and is quite coarse. And when someone walks in the room, the settings should be tweaked, or if the curtains are opened or closed, we need to change it again.

But why can't we do this in real-time? Why can't we create a setup that is always correcting for speaker errors, driving down distortion (by subtracting the speaker distortion from the source) and handling errors in real time?

 

[lashto]

Do We Want All Speakers To Sound The Same?

Of course. And I also want all food to taste the same, all cars to look the same and all jeans to have the exact same size & color. That would be so much fun

Sarcasm aside, my ideal speaker will sound like no speaker (or like an invisible speaker if you wish).

Someone made a very good point that all-the-same will not be an issue because you can still use EQ/tone controls/etc to make them sound however you want. Sounds good to me.
And an all-the-same scenario would make audio shopping a giant lot easier... kind of like that proposition too.

And since we are at it, please build them at 140+dB SINAD. Sometime during the next ten years would be just great!

 

[Killingbeans]

But why can't we do this in real-time? Why can't we create a setup that is always correcting for speaker errors, driving down distortion (by subtracting the speaker distortion from the source) and handling errors in real time?

The loop delay would be way too massive?

I think it's more feasible to do feedback internally in the speakers, like you see in servo subwoofers?

Speaking of Bruno, I think I saw some mentioning of the Kii Three using current feedback to hold the drivers by the balls?