a thoughtful web.
Good ideas and conversation. No ads, no tracking.   Login or Take a Tour!
comment by Vozka

I'm not as uneducated about speakers as you think I am. Even though early reflections are undesirable, constant directivity over most of the spectrum (down to at least 700 Hz) is more important for how natural the speakers sound than the angle at which they are designed to radiate (if the directivity varies), although narrow constant directivity speakers obviously behave better in small/untreated rooms. Any speakers with beaming treble will sound sound like shit unless your room is acoustically dead, in which case anything you put in there will sound like shit. You always get reflections, the point is to a) reduce the very early ones without killing the later ones b) make the reflections have similar frequency response as the direct sound, because in that case your brain will identify them as natural sources of sound in your room, making the stereo image sound real. This is the point of modern speakers with controlled directivity, speakers with beaming treble have been here for decades. Omnidirectional speakers do b) naturally and can get reasonably close to a) if you have a big room with decent treatment.





kleinbl00  ·  2849 days ago  ·  link  ·  

You realize that your statement, effectively, is that omnidirectional speakers can sound as good as directional speakers if you negate all the effects of their omnidirectionality?

Yeah - backscatter and such is no problem if you have adequate absorption. It's also no problem if you have no backscatter. And even with two perfect radiators in freespace, the cancellation between two identical sources is bad for fidelity. From an acoustical standpoint, everything beyond closed-basket headphones is a compromise; early reflections of any kind have a negative impact on fidelity and clarity but sometimes you luck out.

So. You can buy a conventional loudspeaker and stand in front of it and enjoy it... or you can buy an omnidirectional loudspeaker and hope that it doesn't sound too bad. Ultimately it will never reproduce the sound to your ears as well as a conventional loudspeaker. Wanna hear it with more reverb? More comb filtering? Less definition? Sure. Bounce that shit off the wall.

This is why, for a brief shining moment, the industry was trying to convince everyone into dipole rear surrounds: for 5.1, splashing the surround channels around make them sound more "enveloping" because they make things harder to localize. Of course, you were supposed to turn the dipole off when listening to surround music because of all the detail you lost. But then that extra two channels in the bitstream finally annoyed everyone enough into doing 7.1 so now 7.1 systems are even more common than 5.1 and if you want diffuse, you just send the same signal (ish) to both surrounds.

It's not that you're uneducated. It's that you've been lied to.

Vozka  ·  2848 days ago  ·  link  ·  

You realize that your statement, effectively, is that omnidirectional speakers can sound as good as directional speakers if you negate all the effects of their omnidirectionality?

But I didn't say that. I actually said that in an acoustically dead room everything sounds shit. Just like in a completely empty room everything sounds like shit. You need to place differently radiating speakers differently and use different room treatment (not just absorbtion, but also difusion) for the best sound, how isn't that obvious? Yes, narrower constant directivity speakers will almost always sound better, but omnis will in many cases sound better than any traditional non-CD speakers.

  
From an acoustical standpoint, everything beyond closed-basket headphones is a compromise

Well this is just silly. A closed acoustical system of such small size affected by the shape and size of everyones head and ear is almost impossible to do right, and the imaging of large headphones has many compromises, only different ones than speakers do. Maybe in-ear monitors playing binaural recordings could be closer, at least in the second aspect. But you seem to be saying that something like putting speakers in an anechoic chamber to completely remove any interactions would be the ideal system, and that's be wrong.

  
It's not that you're uneducated. It's that you've been lied to.

So on one end we have your experience and on the other end we have my experience and research of people like Siegfried Linkwitz or Floyd Toole, one of the few sane people in the industry. What do you expect me to say?

kleinbl00  ·  2848 days ago  ·  link  ·  

Have you ever listened to decent monitors in an acoustically dead room? The monitors sound great. The room sounds like ass but you're not listening to the room, you're listening to music. Music, when mixed, is mixed in a room with minimal acoustic coloration. Blackbird studio C minimal? No. But minimal.

The argument you're trying to make is "different speakers need different treatments" but with omnidirectional speakers, the treatment is tantamount while with directional speakers, the treatment is bonus. After all, they don't splash crap all over the walls on purpose. All of your arguments come down to that - "in many cases sound better" is happy hand-wavey subjective observation that has nothing to do with acoustics.

Headphones close-couple. Everything with a volume under a cubic inch is close-coupled below 8,000 Hz. Size and shape has fuckall to do with it; anything below mouse farts energizes the entire volume.

I am saying that from an acoustic standpoint - as in, from a repeatable, testable, empirical standpoint - speakers in an anechoic chamber will reproduce audio better than speakers not in an anechoic chamber. That's why speakers are tested in anechoic chambers. That's why speakers are measured in anechoic chambers. You're right - speakers are not enjoyed in anechoic chambers but the fact remains: a speaker that deliberately sprays the back wall should specify how far in both directions it should be from the walls, what those walls should be made out of, and where you should be standing (in two dimensions) in order to properly enjoy the speakers.

A directional loudspeaker you need only listen to.

Amar Bose was hella more famous than Linkwitz or Toole. He made a lot more money. He was also a fucking idiot who built everything out of 2 1/4" paper cones. That doesn't make paper cones genius. Go ahead and appeal to authority - I'll appeal to every other designer that didn't make trashcans.

Look - there was a time I patched my stereo system through a Quadraverb II. It made things nice'n'roomy. Right now I've got an Eclipse DSP in my car that allows me to play bullshit stadium/club/church/whatever games with my music. But I'm not fooling myself about what I'm doing - I'm coloring the shit out of the sound. Works sometimes. Not so much other times. Pretty goddamn geeky.

Also, a substantially more controlled, more purist, more repeatable and more defensible method of adding "early reflections" in order to improve the sound.

user-inactivated  ·  2848 days ago  ·  link  ·  

    "in many cases sound better" is happy hand-wavey subjective observation that has nothing to do with acoustics.

Isn't psychoacoustics a thing? Wikipedia says yes. I'll believe you if you say "no", of course, but I'd be a little surprised to learn "in many cases sound better" wasn't a thing that had been studied empirically.

Vozka  ·  2848 days ago  ·  link  ·  

To be honest, I don't know about any research directly comparing omnis, constant directivity speakers with 2pi/imited dispersion and non-CD speakers. But yes, modern research in psychoacoustics says that constant directivity is one of the most important factors in the sound being percieved as real and natural, and some of the (well-known and rational) researchers mention omnis as one of the solutions, although they obviously mention that they're not very suitable for small untreated rooms. But to claim that omnis go against everything we know about sound reproduction is uneducated and together with assuming various types of bias just because I don't agree with him borderline delusional.

kleinbl00  ·  2848 days ago  ·  link  ·  

Psychoacoustics is very much a thing, it's just not this thing.

- turning the delay fill down 10dB so that you don't "hear" it at all but instead perceive the main cluster as louder? Psychoacoustics.

- Hearing under 12ms difference from ear-to-ear as stereo localization, 12-18ms as comb filtering and 18-up ms as delay? Psychoacoustics.

- Perceiving louder volume as better sound? Psychoacoustics.

- Hearing Sheperd tones as an endlessly ascending scale? Psychoacoustics.

Psychoacoustics is the ear-brain transfer function. It is important. Hearing is not only integrative, it's lossy. There's a lot of processing involved in hearing. The cochlea is actually a digital vibration array; encoding discrete integers as a continuous spectrum is all psychoacoustics. Psychoacoustics is also cultural and gender-based; the Japanese tend towards "bright" speakers because theirs is a language of vowels so they don't hear consonants as well, thus they emphasize the high end more than Americans or Europeans, for example.

But this isn't psychoacoustics. This is a combination of:

- The Ikea Effect, whereby consumers place a higher value on objects they partially created

- In-group Bias, whereby members of the audiophile community favor products created within the audiophile community

- Choice-supportive bias, wherein those thousand dollar cables must sound good, after all you paid for them

- Anchoring, whereby those thousand dollar cables are a bargain because you could have paid $10,000

- cognitive dissonance.

When this first appeared on the Internet, hundreds of people at the show commented on the video that it sounded awesome, there must be something wrong with the recording, what the fuck, etc. Fact of the matter is, if you're there jamming to Eddie and Crew for however many hundreds of dollars you paid, you aren't going to wrap your head around the fact that it sounds like a Merzbow cover.

Just remember. That's not distortion. It's the speaker flattening.

Vozka  ·  2848 days ago  ·  link  ·  

Yes, I have, and I found the sound unnatural like I said. I actually do most of my listening on in a near-field ambiophonics configuration because it makes sense in my acoustically bad room and it sounds great, I'm planning to build a pair of Synergy horns for the same reason, but I recognize that it is a compromise.

  
Like you said, measuring and listening are two different things. Acoustically dead rooms may make sense in a studio, but the overwhelming majority of people do not prefer them at home, this has been repeatedly proven (but you reminded me of this, which was afaik written by Toole, and may apply to you ). The quality of stereo reproduction is not just about hearing the most details but also about the illusion of space and that is better achieved with controlled room interaction.

  
What you're saying about headphones, again, is silly, and simplifies the reality extremely. I have built a few headphone prototypes and getting them sound right is much more difficult than building decent speakers, which, to a certain point, has been solved.

  
I have no idea what you're trying to say about me mentioning Linkwitz and Toole. Are you arguing that agreeing with you, a completely unknown person on the internet, whose experience is completely different than mine, makes more sense than agreeing with people whose life-long research (that everyone can read about) says something that is mostly consistent with my own experience? I don't see how Bose is relevant here, are you saying that what Toole and Linkwitz designed sounds like shit as well, or do you think they're a part of the audiophile woo community?

  
(also, I'd argue that Bose is kinda genious because he was obviously in it to make money, not to push audiophile standards, and boy did he succeed. And there's nothing inherently bad about using paper as a base material for cones, there's a reason it's still used in most pro drivers.)
kleinbl00  ·  2848 days ago  ·  link  ·  

Here's our fundamental disagreement:

You're arguing, objectively, that I am objectively wrong... based on subjective information.

I'm counter-arguing (you started this) that you are objectively wrong... based on objective information.

When I present you with objective arguments, you respond with subjective points.

When I point out that your arguments are subjective, you double down on the subjectivity as if it were objective.

Look - I will never convince you that spraying the back wall in the interest of "better sound" is a bad idea. You have subjectively decided that it's brilliant, and are immune to objective argument. What I'm saying about headphones isn't silly, it's fluid mechanics. Yes - measuring and listening are two different things. Measuring is objective. Listening is subjective. I NEVER said people like listening in acoustically dead rooms - I said that the music sounds great in acoustically dead rooms. The rooms sound fucking spooky. Then you throw some link from diyaudio.com at me after I already threw Linkwitz back in your face. I'll repeat: the preponderance of speakers ever made are directional. The ones that aren't are weird-ass audiophile bullshit. That's not in my opinion. That's physics.

But that's an objective statement against a subjective opinion. I will never convince you that MBLs sound like shit. I don't want to. You are welcome to believe they sound like angels farting and it's no skin off my nose.

SUBJECTIVELY?

They sounded about as good as EVID 6's which, in my subjective opinion, are about the worst outdoor speaker I've ever had the misfortune of listening to.

And I put about 90 Toa H-4s in omnidirectional arrays in the Seattle Tacoma International Airport.

Vozka  ·  2848 days ago  ·  link  ·  

You can't seriously think that "it sounds like shit" and "it splashes high frequency all over the back wall" are objective arguments about sound quality while discounting anything I say as subjective without actually responding, that's bizzare. Judging by your reasoning about room interactions and some hypotetical closed headphones that don't exist, you obviously know nothing about much research in psychoacoustics (have you at least read about constant directivity?), or the reality of home speaker design.

  
(and I'm saying that as someone who despises audiophile bullshit, doesn't own omni speakers or any expensive hifi really and is not planning to buy any)
kleinbl00  ·  2848 days ago  ·  link  ·  

I said "subjective" twice, yo. Once in all caps italics.

At least do me the courtesy of reading the keyboard-smashing stuff.