Objective vs. subjective

[philthien]

How much of this hobby is objective, and how much is subjective?

I'm always amazed by how much disagreement one finds in the opinions of highly experienced listeners when it comes to commercial loudspeaker offerings. What accounts for these differences?

And, how much will our own "hearing curves" affect our perceptions? A younger person that can hear high frequencies better may think a design is overly bright. An older person may think the high frequencies are attenuated, while a middle-aged person may think it "just right," no?

I question most those shooting for flat (anechoic) designs. It seems to me that such a design will only benefit people with perfect listening environments and perfectly flat ears.

Is it possible that, once we started getting measuring gear that would allow us to design speakers with nearly flat responses that we all started designing speakers that sound very similar, and we assume they're best, but maybe they're not?

I assume most "purists" would scoff at midrange and tweeter controls that were (are) found on many speakers like vintage JBL "monitors." Are any audible "impurities" imparted by the sound of such devices really so troubling that we wouldn't want the flexibility to adjust the speaker to its surroundings, and our own preferences?

Sorry for all the questions. As I start getting into the hobby again, I'm beginning to question many of my older beliefs about how to go about things.

[mzisserson]

Re: Objective vs. subjective

How much of this hobby is objective, and how much is subjective?

I'm always amazed by how much disagreement one finds in the opinions of highly experienced listeners when it comes to commercial loudspeaker offerings. What accounts for these differences?

And, how much will our own "hearing curves" affect our perceptions? A younger person that can hear high frequencies better may think a design is overly bright. An older person may think the high frequencies are attenuated, while a middle-aged person may think it "just right," no?

I question most those shooting for flat (anechoic) designs. It seems to me that such a design will only benefit people with perfect listening environments and perfectly flat ears.

Is it possible that, once we started getting measuring gear that would allow us to design speakers with nearly flat responses that we all started designing speakers that sound very similar, and we assume they're best, but maybe they're not?

I assume most "purists" would scoff at midrange and tweeter controls that were (are) found on many speakers like vintage JBL "monitors." Are any audible "impurities" imparted by the sound of such devices really so troubling that we wouldn't want the flexibility to adjust the speaker to its surroundings, and our own preferences?

Sorry for all the questions. As I start getting into the hobby again, I'm beginning to question many of my older beliefs about how to go about things.

There is no way measurements can provide us accurate feedback on if a speaker sounds good. They can only be a design tool to get us ther.

There is no way hearing alone can design a good speaker. It can only tell us if it sounds good to you or not.

Somewhere, there is a blending of the two. When room inteaction, diffraction, step, size, lobing, preference to types of tonaily, ect come together as a design goal, then you will find what comprimises are willing to be made. There is no such thing as a no compromise speaker.

There is also no perfect speaker, and noone's hearing is perfect.

Most can live witout the extra 20% a very high end speaker gives, and that sounds good! I have heard spakers that sell for 10K/pr that are not as good as some DIY designs.

For those who want the extra 20% its an expodential curve in learning, "training", cost, time, ect. weather in DIY or Retail.

So really, it is up to you to formulate what sounds "good", "good enough", or "bad". And the value of such lies in your hands as well. Value is defined by the end user.

This forum is failry objective. And objectivity will speak in terms everyone understands. So it is good to stick there, no matter what your listening habits and belifs are. You will see debates about cables, caps, active vs. passive, and they will rage on long after you and I. It is because of different beliefs. Its why some own Hondas, and others Toyotas.

Floyds book about loudspeakers and rooms (name is slipping at the moment) is a very good work that touches about the acoustic/psycoacoustic match-up.

At the end of the day, Have Fun... That's truly the only universal rule in this, or any hobby.

[bangleiii]

Re: Objective vs. subjective

And then there's the fletcher munson curve where the music is only flat when played at the same level that it was recorded and "music" is not sine waves.

[Jeff B.]

Re: Objective vs. subjective

There is no way measurements can provide us accurate feedback on if a speaker sounds good. They can only be a design tool to get us ther.

There is no way hearing alone can design a good speaker. It can only tell us if it sounds good to you or not.

Somewhere, there is a blending of the two. When room inteaction, diffraction, step, size, lobing, preference to types of tonaily, ect come together as a design goal, then you will find what comprimises are willing to be made. There is no such thing as a no compromise speaker.

There is also no perfect speaker, and noone's hearing is perfect.

Most can live witout the extra 20% a very high end speaker gives, and that sounds good! I have heard spakers that sell for 10K/pr that are not as good as some DIY designs.

For those who want the extra 20% its an expodential curve in learning, "training", cost, time, ect. weather in DIY or Retail.

So really, it is up to you to formulate what sounds "good", "good enough", or "bad". And the value of such lies in your hands as well. Value is defined by the end user.

This forum is failry objective. And objectivity will speak in terms everyone understands. So it is good to stick there, no matter what your listening habits and belifs are. You will see debates about cables, caps, active vs. passive, and they will rage on long after you and I. It is because of different beliefs. Its why some own Hondas, and others Toyotas.

Floyds book about loudspeakers and rooms (name is slipping at the moment) is a very good work that touches about the acoustic/psycoacoustic match-up.

At the end of the day, Have Fun... That's truly the only universal rule in this, or any hobby.

Yes, I agree. It is hard to get off the ground without understanding some theory, having good measurments, and employing good design choices - But, in the end, what we hear and prefer will always be subjective. Stereo is an imperfect playback medium, and room acoustics can change the whole playing field. Depending on our rooms, and speaker and listening placement, even objectivity is subject to its environment. As a result subjective preferences will tend to always make the final call, but hopefully only after good objective design has gotten us in the neighborhood.

[GranteedEV]

Re: Objective vs. subjective

Subjective is what we choose to listen to, and how we choose to EQ to compensate for bad mixing.

Objective is the speaker, and the room. No voicing should really be necessary if the measured design accounts for as much as we want it to - although admittedly I don't think most DIY designs really can account for everything, though.

But I do think that with enough money, subjectivity when it comes to LOUDSPEAKERS and ROOMs should be thrown out the window - these things should be based on measurements and the correlation of these measurements to all the research out there. I doubt Kieth Yates goes around saying "How warm do you like your room" - pros use a reference measurement they strive for, every time.

That said, unfortunatley if you don't start with a good fundamental (A room that measures well) then you may need to compensate for that elsewhere.

I think the approach should always be objective, and then, if your budget precludes you from doing everything possible to be objectively optimal, then you compensate. But that's all voicing is - compensation for objectively measurable phenominon.

I'm always amazed by how much disagreement one finds in the opinions of highly experienced listeners when it comes to commercial loudspeaker offerings. What accounts for these differences?

Listening environment and often not experience but an accustomation to listening to speakers.

And, how much will our own "hearing curves" affect our perceptions? A younger person that can hear high frequencies better may think a design is overly bright. An older person may think the high frequencies are attenuated, while a middle-aged person may think it "just right," no?

This is not true. so-called hearing curves do not affect our perceptions. Is the sky above you blue or green? Blue, you say. Is the sky above me blue or green. Blue, I say. So why would the sky all of a sudden look green to you if you reproduced it through an excellent photograph? Our perception is based on how we interpret that material.

Subjectivity comes from familiarity - IE x person is familiar with the effect of y bad photography software, therefore said person prefers the sky to look green, even if he looked up and the sky was blue.

I question most those shooting for flat (anechoic) designs. It seems to me that such a design will only benefit people with perfect listening environments and perfectly flat ears.

Huh?

So when you hear a live violin in your not so perfectly flat ears, and then hear that same violin through an anechoic flat speaker (and with other good measurements, mind you), your not so perfect ears do what exactly to the sound of the recorded violin compared to what you heard live?

But yes, listening environment is important. It's a PART of reproduction - right back to objectivity. But speakers with good polar response aren't NEAR as sensitive to listening environment as a lot of hi fi speakers out there with thier 1st order crossovers at 4khz and what-have-you

Is it possible that, once we started getting measuring gear that would allow us to design speakers with nearly flat responses that we all started designing speakers that sound very similar, and we assume they're best, but maybe they're not?

Speakers that truly measure well should have zero timbre of their own. Timbre from a speaker is BAD.

That said, speakers can still portray a different stereo illusion - sense of width and depth and spaciousness. It's tough to say "which" stereo illusion is correct, because stereo itself is not the live event. Only binaural type recordings, played back through ambiophonics really might be objectively "correct" if you think about it, though I've never played with them myself. So yes, stereo imaging/soundstage is subjective, only because stereo itself is objectively "wrong". I find people who listen to pop type music prefer the stereo effect to be left/center/right distinctness, and people who listen to classical prefer depth and less location distinction. "Niether is wrong, because both are wrong" ;P

And this goes as far as room preference. A room with minimal comb filtering, controlled/tunedg bass response below the shroeder frequency, i think that is measured and desirable either way. But subjectivity comes into play in terms of spaciousness - should the room add anything later to the recording? Should the recording be the one adding the spaciousness? It's tough to say, but I think in general, you want the room to do very little over the first 15-20ms, at the very least. Whether you want a dead audiophile room after that 20ms or a natural sounding, immersive room, well that's subjective.

I assume most "purists" would scoff at midrange and tweeter controls that were (are) found on many speakers like vintage JBL "monitors." Are any audible "impurities" imparted by the sound of such devices really so troubling that we wouldn't want the flexibility to adjust the speaker to its surroundings, and our own preferences?

Your own preferences? No. A speaker isn't an instrument, it's a reproduction tool. Do you prefer your drill in reverse when you're trying to drill forward? Do you prefer your keyboard behind you or in front of you as you type? Likewise, you shouldn't be playing around with a GOOD speaker's measurable behavior just to see what is funnest. It's there to reproduce, not produce.

A mixer's poorly done job? Use a 32-band graphic EQ instead, if you're really that anal, or just accept that different recordings are, well different.

Your surroundings? Yes. Finding the right MEASURED (objective) balance in a room is of course a valid goal, but again it pertains to rooms that don't measure optimally.

[philthien]

Re: Objective vs. subjective

There is no way hearing alone can design a good speaker. It can only tell us if it sounds good to you or not.

Really?

That's discouraging.

Are there no mass-market products out there that were designed/manufactured before sophisticated measurement tools were available that you'd consider pretty good?

Nobody had a good enough ear to design something that sounded good to not only himself, but others as well?

[Wolf]

Re: Objective vs. subjective

I actually disagree with Mike's statement there, in terms of xover design alone. I've done them by ear, but it was from knowing about what values might work where in the circuit. It takes some serious time, but it's doable if you have the experience.

The cabinet and tunings really can't be designed properly without some sort of measurement of the drivers' inherent T/S specifications. Without that, it's a lot of trial and error, and just personal preference like Mike said.

Before Thiele and Small developed the specifications for drivers, it was just trial and error, and stabs in the dark. I suppose with experience there may have been a few standouts back then, but I can't say for certain.

Later,
Wolf

[philthien]

Re: Objective vs. subjective

Your own preferences? No. A speaker isn't an instrument, it's a reproduction tool. Do you prefer your drill in reverse when you're trying to drill forward? Do you prefer your keyboard behind you or in front of you as you type? Likewise, you shouldn't be playing around with a GOOD speaker's measurable behavior just to see what is funnest. It's there to reproduce, not produce.

Well, in the case of the drill analogy, I think I may prefer a different speed (RPM's) than you. And in the case of the keyboard, I still like my classic PC/XT keyboards with the function keys on the left (no, I don't have one, but wouldn't mind if we returned to them).

Thanks for sharing your perspectives.

It kinda reminds me of old muscle cars. Engines with huge displacements and tons of HP, but hindered by the tire technology (which limited the ability to put that HP into the pavement).

I sort of feel like we will always have tire problems in this hobby. The tires are the listening environment. Yeah, I know there are devices that will "fix" the room. But for people that aren't taking that step, I'm unsure of the benefit of spending tons of time/money on designing a perfectly flat speaker.

[philthien]

Re: Objective vs. subjective

The cabinet and tunings really can't be designed properly without some sort of measurement of the drivers' inherent T/S specifications. Without that, it's a lot of trial and error, and just personal preference like Mike said.

Thanks. I'm assuming we're using T/S parameters. In reality, we don't need a computer to do that. We can use the manufacturer-provided #'s and a calculator.

What I'm really getting at is crossover design and driver integration, I guess.

I wonder if a crossover designed with the help of a computer and measurement system, designed to be measurably flat, is going to be perceived as superior to one designed the old-school way.

A couple of points...

Back in my Madisound forum days, they had an A/B comparison for (I think) interconnects or speaker cables. It was double blind. Anyone attending could "take the challenge." Determining superiority wasn't the goal, just accurately identifying which cable was being used was the purpose. And (IIRC), only one person out of dozens showed any statistical ability to do so. And it wasn't a wide margin, either. He may have gotten "lucky," for all I know.

So I often wonder whether people would be able to identify different crossovers in a similar test. In an A/B test, would people be able to identify (for example) series vs. parallel crossovers? 2nd, 3rd, 4th order parallel? Crossovers with baffle step compensation vs. those with none?

Of course, every designer says they can hear these differences. The question is, can they really? Or is it some sort of placebo effect? And, if they can identify a difference, would they always "prefer" the one that is ideal, or would they sometimes prefer the sound of the design that strays from perfection?

[GranteedEV]

Re: Objective vs. subjective

Well, in the case of the drill analogy, I think I may prefer a different speed (RPM's) than you. And in the case of the keyboard, I still like my classic PC/XT keyboards with the function keys on the left (no, I don't have one, but wouldn't mind if we returned to them).

Thanks for sharing your perspectives.

It kinda reminds me of old muscle cars. Engines with huge displacements and tons of HP, but hindered by the tire technology (which limited the ability to put that HP into the pavement).

I sort of feel like we will always have tire problems in this hobby. The tires are the listening environment. Yeah, I know there are devices that will "fix" the room. But for people that aren't taking that step, I'm unsure of the benefit of spending tons of time/money on designing a perfectly flat speaker.

Actually, making a speaker "by the measurements" is going to get you that much closer. If your speaker has good polar response then it becomes a LOT less room sensitive compared to speakers which just strive for a desired on axis response. That doesn't mean building an omni, necessarily. It might mean narrow directivity with a waveguide like the JBL LSR or Geddes speakers or with a quality coax like the Pioneer drivers, or a true narrow baffle multiway dipole like the Nao Note.

Better speakers aren't just ones that measure flat on axis, but are less room sensitive too.

[andy19191]

Re: Objective vs. subjective

How much of this hobby is objective, and how much is subjective?

It depends on what the hobbyist wants from the hobby. At one end we have the cold and rational taking a similar approach to professional engineers and at the other end we have the bouncy and enthusiastic audiophiles trying whatever takes their fancy with little to no requirement for the performance to build on a model of how speakers work.

I'm always amazed by how much disagreement one finds in the opinions of highly experienced listeners when it comes to commercial loudspeaker offerings. What accounts for these differences?

If you look at the evidence (e.g. Toole's book) you will find scant support for statements like this.

And, how much will our own "hearing curves" affect our perceptions? A younger person that can hear high frequencies better may think a design is overly bright. An older person may think the high frequencies are attenuated, while a middle-aged person may think it "just right," no?

No. As older people's hearing degrades it applies to live music and recorded music equally.

I question most those shooting for flat (anechoic) designs. It seems to me that such a design will only benefit people with perfect listening environments and perfectly flat ears.

There are one or two arguments for deviating from a flat on axis response but this is not one of them for the reason given above.

Is it possible that, once we started getting measuring gear that would allow us to design speakers with nearly flat responses that we all started designing speakers that sound very similar, and we assume they're best, but maybe they're not?

To answer this you need to define what best means. Once you have defined that then you can use a wholly rational procedure to develop a design towards it.

I assume most "purists" would scoff at midrange and tweeter controls that were (are) found on many speakers like vintage JBL "monitors." Are any audible "impurities" imparted by the sound of such devices really so troubling that we wouldn't want the flexibility to adjust the speaker to its surroundings, and our own preferences?

People that scoff at the need for tone controls are making rather strong statement about themselves. As far as I am aware, almost all modern "pro" speakers have tone controls.

Sorry for all the questions. As I start getting into the hobby again, I'm beginning to question many of my older beliefs about how to go about things.

Where do you think they come from?

A fair amount of modern audiophile belief would seem to be a straightforward belief in old fashioned magic. This is not necessarily a bad thing in itself if the audiophile is having fun but it does lead to problems when interacting with conventional scientific belief.

[jinjuku]

Re: Objective vs. subjective

Well, in the case of the drill analogy, I think I may prefer a different speed (RPM's) than you.

I would say subjectively people can prefer to run their drills at what ever speed they want. Objectively you have to run the drill at speeds that the application you are putting it to calls for. You can like 300 RPM all you like but if the job calls for 900 RPM and you refuse to face the science of the bit, the torque, and the RPM that the material calls for...

What I would say is that it's ok to have a brand preference. I like my Hitachi router (subjective) because it is what I own and it gets the job done. I may even say I like it over a comparable Porter Cable. That's subjective. Now when I sit both units down, go through them, try some trial cuts I may come to the conclusion that we are talking at tenths of a percentage point Objectively.

I guess the question that I would ask: Do you want sound reproduction that is to your liking or, given all the variability of room and placement, reproduction that is as true to the source as possible with everything disappearing as to coloration?

I know there are devices that will "fix" the room. But for people that aren't taking that step, I'm unsure of the benefit of spending tons of time/money on designing a perfectly flat speaker.

I wouldn't hazard a guess that it takes any more time or money to design and flat response then to purposefully design for another curve cough*Bose*cough.

Here is are some links to John 'Zaph' Krutke. Someone I tend to prescribe to when it comes to the type of question you posted:

The worn out flat response debate (Scroll down just a bit)

Evaluation Myths Busted

[Wolf]

Re: Objective vs. subjective

Back in my Madisound forum days, they had an A/B comparison for (I think) interconnects or speaker cables. It was double blind. Anyone attending could "take the challenge." Determining superiority wasn't the goal, just accurately identifying which cable was being used was the purpose. And (IIRC), only one person out of dozens showed any statistical ability to do so. And it wasn't a wide margin, either. He may have gotten "lucky," for all I know.

So I often wonder whether people would be able to identify different crossovers in a similar test. In an A/B test, would people be able to identify (for example) series vs. parallel crossovers? 2nd, 3rd, 4th order parallel? Crossovers with baffle step compensation vs. those with none?

Of course, every designer says they can hear these differences. The question is, can they really? Or is it some sort of placebo effect? And, if they can identify a difference, would they always "prefer" the one that is ideal, or would they sometimes prefer the sound of the design that strays from perfection?

If you're referring to the wire-test of 2003 from Bob Cordell, I took that test.
It was me who got them right every time. He was me. Difference between Soundking and Kimber.

Different xovers sound different.

Series vs parallel, I dunno- maybe. If everything else was matched like transfer function and such, and the only difference was that they were series or parallel topologies, it'd be hard to say without doing it.

2nd vs 4th would probably be audible due to the drastic difference in slope, and I imagine the 3rd would also sound different, all else being equal.

BSC or no would be easy. No bass reinforcement is really easy to determine.

Yes- we really can.
In the 12dB/24dB question, a lot of people will prefer the 12dB xover to the 24dB. That's the truth.
In terms of 'perfection', there is none. There is always a set of defined tradeoffs that someone makes along the design process, and it's to achieve a certain goal. So- perfection within those constraints of the goal would be more appropriate.

For example, some people prefer a little BBC dip (as generally defined, not from actual measurements of the old BBC) as they call it, and I used to. Now, I feel the midrange lacks resolution with that presence removed.

Later,
Wolf

[dantheman]

Re: Objective vs. subjective

I actually disagree with Mike's statement there, in terms of xover design alone. I've done them by ear, but it was from knowing about what values might work where in the circuit. It takes some serious time, but it's doable if you have the experience.

.....snip....
Later,
Wolf

But Wolf, you've got more experience doing this than some professionals. Makes me wonder why you don't make a run at a commercial production? You certainly have a passion/creativity for it.... but anyway. Most people would endlessly screw this up w/o measurements--some even with measurements.

Otherwise if you(OP) are serious about this, Read Dr. Toole's book on the subject(twice). It's easily the most complete I know of on the subject. Even Dr Olive's blog will give you enough to get the wheels turning.

Dan

[billfitzmaurice]

Re: Objective vs. subjective

How much of this hobby is objective, and how much is subjective?

That varies from person to person. If you don't know how gear and acoustics work it's 100% subjective. If you do know but are deaf it's 100% objective.
For the rest of us the answer lies somewhere in between the two extremes.