[Yuri Korzunov]

So, what do you recon that one means, when they are speaking about:

-Sibilance
-Grainy
-Veiled
-Black
-Micro/Macrodynamics

And for a bonus: In how many part do you split your tonal specrtum, and how do you call them? (Bass-mids-highs...)

1. Me seems these term almost clear in common case, but there may be different understanding anyway.
Sound engeneer can try achieve desirable effect by playing with EQ, reverberation, compression, volume, panorama, band manipulations, etc.

2. These parts of audio spectrum I can't remember. Because, I don't use it in work. There is Hz, kHz, MHz only.

 

[pinnahertz]

If he means that in speakersystems, there could be issues in crossover area, wich could make transition harsh and somewhat spikey on that frequencyrange, i do understand what he is after. But in headphones, where there is no crossovers, only what is left to cause such similar kind of broblems, could be drivermembranes or chassis inner resonance.

I think I'll just wait for him to say what he means.

 

[ProtegeManiac]

This is the second time that I've seen you mention that sibilance is the result of time alignment issues in speakers. Would you mind posting a reference that substantiates that?

I'm on slow arse mobile tethering right now and Head-Fi is the only site that works, but what I'm saying is that if you're not hearing every driver in sync, you'll mess it all up. In any case, it will not be referred to as "sibilance" in actual engineering journals, because the problem is that in car audio forums, people hear that rought "SSSSS" or "TZZZZZZT!!!!" and call it sibilance, thinking it's the same thing as having a spike in the upper midrange and treble, and then they end up wondering why EQ barely works (and if it does, it screws up a lot of something else), precisely because the response isn't the problem. Some cars can measure relatively smoothly with a test tone and then screw up actual music - the thing is, with a test tone, you hear a constant sound, so you don't notice anything out of sync.

One easy way to test this is to go to an IASCA or EMMA event (depending on where you are), and check out the cars that aren't designed to have the subwoofers blow the glass off the car. Listen to them with the Time Alignment DSP disabled, then enabled, then disable it again, in the middle of your test tracks.

If you can't get to these, you can simulate this problem at home. Get some speakers and set them up nearfield at a corner table with no acoustic treatments, and no angled mounts nor angled front baffles on the speakers, and point them at each other like in most cars, and sit closer to one speaker.

And the second time the McLaren F1 is referenced in relation to time alignment. Now that's one I doubt many will understand or identify with. Care to elaborate? On either?

OK...pics can't upload. But imagine these.

Home audio set up, speakers : Chair and speakers form an isosceles or equilateral triangle with the chair at the apex; tweeters and midwoofers are right on top of each other in each speaker cabinet.
Home audio set up, headphones : Each headphone driver is equidistant to each ear as they sit on your head (unless you have a lopsided hydrocephalic head that is)

Car audio set-ups
British - listener's chair (driver) on right; driver side tweeter at shoulder level to his right; driver side midwoofer at ankle level to his right; passenger side tweeter waaaaaay off to his left, shoulder level; passenger side midwoofer waaaaaaaay off to his left, ankle level
US - listener's chair (driver) on left; driver side tweeter at shoulder level to his left; driver side midwoofer at ankle level to his right; passenger side tweeter waaaaaay off to his right, shoulder level; passenger side midwoofer waaaaaaaay off to his right, ankle level


Maclaren F1 : Driver at center, equidistant to where you can mount tweeters and midwoofers (almost like being at home!), but most units had no stereo on account of the very expensive BMW "sound system" located behind the driver loudly and proudly going, "VRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR_tgsssshhhh!!_WHHRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR_tgssshhhhh!!!_WHHRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR_urrrrrrm-urrrrrrm-urrrrrrmmmmmm_VRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR_tghshhhhhh!!!_WHHRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR_tgshhhh!!!_WHHHRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR_urrrrrrrmmmmmmm-urrrrrmrmmmmmmm_WWHHHHHHHRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR...."

In a nutshell - you don't typically get time alignment-related sibiance in anything but a car or a badly set up home audio system because you are sitting equidistant to each speaker. At worst, you'd have a desktop speaker that isn't angled but set low on the desk, or you sit too close to 3-way towers, which can be made worse by having reflective walls too close, but still not as bad as in a car.

You can have a Mclaren F1 because that one has a center seat for the driver, but then again, why buy a car like that to put a sound system in when BMW already provided a V12 bested only by the 5.4L Ferrari V12 and Lamborghini 5.5L V12 when it comes to smoothness at low speed while still providing dramatic, Le Mans kind of sound at full tilt? Not to mention the F1 barely has any sound deadening compared to a Camry, much less a Ferrari 456GT, so even if you put in an audio system, most of the time the V12 will just drown that out (unless you mostly drive the F1 in traffic).

 

[pinnahertz]

I'm on slow arse mobile tethering right now and Head-Fi is the only site that works, but what I'm saying is that if you're not hearing every driver in sync, you'll mess it all up. In any case, it will not be referred to as "sibilance" in actual engineering journals, because the problem is that in car audio forums, people hear that rought "SSSSS" or "TZZZZZZT!!!!" and call it sibilance, thinking it's the same thing as having a spike in the upper midrange and treble, and then they end up wondering why EQ barely works (and if it does, it screws up a lot of something else), precisely because the response isn't the problem. Some cars can measure relatively smoothly with a test tone and then screw up actual music - the thing is, with a test tone, you hear a constant sound, so you don't notice anything out of sync.

Ah, ok. The traditional application of time-alignment has been within a single speaker with multiple drivers. Time misalignment in that case causes the crossover region to have a different, sometimes radically different, dispersion pattern, technically "lobing". Time alignment provides the opportunity for a straight forward lobe in the crossover region. Response smoothness is another issue.

What little I've read about car audio indicates that there is a sibilance localization problem, a big one, caused by essentially really bad speaker placement (not that there's a choice).

One easy way to test this is to go to an IASCA or EMMA event (depending on where you are), and check out the cars that aren't designed to have the subwoofers blow the glass off the car. Listen to them with the Time Alignment DSP disabled, then enabled, then disable it again, in the middle of your test tracks.

Not as easy as one might hope. But if an event drops into my virtual backyard, I'll have a look.

If you can't get to these, you can simulate this problem at home. Get some speakers and set them up nearfield at a corner table with no acoustic treatments, and no angled mounts nor angled front baffles on the speakers, and point them at each other like in most cars, and sit closer to one speaker.

Thanks, I understand what that does.

Maclaren F1 : Driver at center, equidistant to where you can mount tweeters and midwoofers (almost like being at home!), but most units had no stereo on account of the very expensive BMW "sound system" located behind the driver loudly and proudly going, "VRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR_tgsssshhhh!!_WHHRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR_tgssshhhhh!!!_WHHRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR_urrrrrrm-urrrrrrm-urrrrrrmmmmmm_VRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR_tghshhhhhh!!!_WHHRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR_tgshhhh!!!_WHHHRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR_urrrrrrrmmmmmmm-urrrrrmrmmmmmmm_WWHHHHHHHRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR...."

Great. I'll keep an eye out for an F1 on Cragslist for my next daily driver.

In a nutshell - you don't typically get time alignment-related sibiance in anything but a car or a badly set up home audio system because you are sitting equidistant to each speaker. At worst, you'd have a desktop speaker that isn't angled but set low on the desk, or you sit too close to 3-way towers, which can be made worse by having reflective walls too close, but still not as bad as in a car.

Pretty much clears up the mystery for me. Your useage of "time-alignment" is non-traditional, probably specific to the automotive sound market. Here's what it means to me.

You can have a Mclaren F1 because that one has a center seat for the driver, but then again, why buy a car like that to put a sound system in when BMW already provided a V12 bested only by the 5.4L Ferrari V12 and Lamborghini 5.5L V12 when it comes to smoothness at low speed while still providing dramatic, Le Mans kind of sound at full tilt? Not to mention the F1 barely has any sound deadening compared to a Camry, much less a Ferrari 456GT, so even if you put in an audio system, most of the time the V12 will just drown that out (unless you mostly drive the F1 in traffic).

Well, I just checked my garage...sorry to report, I don't have any of those in there at the moment. But hang on, let me check.....nope...none on the driveway either. As a result, I don't listen much in my ride. I'll admit to wearing noise-cancelling headphones for long late-night drives, though. Time-aligned just fine in the US driver's seat of an F-150.

 

[ProtegeManiac]

Ah, ok. The traditional application of time-alignment has been within a single speaker with multiple drivers. Time misalignment in that case causes the crossover region to have a different, sometimes radically different, dispersion pattern, technically "lobing". Time alignment provides the opportunity for a straight forward lobe in the crossover region. Response smoothness is another issue.
---
Pretty much clears up the mystery for me. Your useage of "time-alignment" is non-traditional, probably specific to the automotive sound market. Here's what it means to me.

Uh, that's essentially the same thing. The only difference is that in home audio you can use the crossover as well as angled front baffles, and then if that's not enough, move the seat farther back, just so every driver will be equidistant to the listener's head.

In car audio the only difference is that time alignment DSP deliberately delays the output to each driver at a customized profile so that all the sound will reach the driver's head in sync with the farthest transducer, which is usually the subwoofer (without a sub, or a sub mounted on the center console like in some Nissan and Infinity cars, that will be the passenger side midwoofer/midbass). The end goal is still to simulate having each of the transducers seem equidistant to the driver's head, even if they're not mounted smack next to each other on the front baffle of a speaker cabinet on each side.

HT receivers have a similar feature but instead of being configured to time align Left Tweeter, Left Midwoofer, Subwoofer, Right Midwoofer, and Right Tweeter, it aligns Front L, Rear L, C, Rear R, Front R, and Sub. Also it's more common for this to be done manually in car audio where you input either distance (the nose while in driving position is used normally; for competitions, you move that back a little bit, since being parked can mean a more relaxed seating position) in cm and have the DSP unit compute, or directly input microseconds of delay. In HT it's more common to just hook up the Audyssey mic and have it tune automatically (this is also available on car DSP units).

What little I've read about car audio indicates that there is a sibilance localization problem, a big one, caused by essentially really bad speaker placement (not that there's a choice).

The problem again is that the driver is not equidistant to each transducer - arranged closest to farthest - driver side tweeter, driver side midwoofer, passenger side tweeter, passenger side midwoofer, and subwoofer.

You can have a similar issue at home by sitting a lot higher than the speakers (ex using non-angled front baffle speakers on desktop, and no stands that rectify that), or sitting too close to multiple driver towers. However in a car that's exacerbated by windshield reflections, so unless you can get the same reflections in your house, the second scenario is more likely to just image the notes easily localizable at least by driver height.

 

[pinnahertz]

Uh, that's essentially the same thing. The only difference is that in home audio you can use the crossover as well as angled front baffles, and then if that's not enough, move the seat farther back, just so every driver will be equidistant to the listener's head.

In car audio the only difference is that time alignment DSP deliberately delays the output to each driver at a customized profile so that all the sound will reach the driver's head in sync with the farthest transducer, which is usually the subwoofer (without a sub, or a sub mounted on the center console like in some Nissan and Infinity cars, that will be the passenger side midwoofer/midbass). The end goal is still to simulate having each of the transducers seem equidistant to the driver's head, even if they're not mounted smack next to each other on the front baffle of a speaker cabinet on each side.

HT receivers have a similar feature but instead of being configured to time align Left Tweeter, Left Midwoofer, Subwoofer, Right Midwoofer, and Right Tweeter, it aligns Front L, Rear L, C, Rear R, Front R, and Sub. Also it's more common for this to be done manually in car audio where you input either distance (the nose while in driving position is used normally; for competitions, you move that back a little bit, since being parked can mean a more relaxed seating position) in cm and have the DSP unit compute, or directly input microseconds of delay. In HT it's more common to just hook up the Audyssey mic and have it tune automatically (this is also available on car DSP units).

The problem again is that the driver is not equidistant to each transducer - arranged closest to farthest - driver side tweeter, driver side midwoofer, passenger side tweeter, passenger side midwoofer, and subwoofer.

So this is a problem confined to car audio since that's where the transducers are widely spaced, each one in a different location relative to the driver. That doesn't happen in HT. Your calibration problem is much worse.

I think I'm getting that your comment re: sibilance relating to time-alignment is 1. confined to this kind of car system and 2. the result of some rather difficult off-axis lobing in the crossover regions causing the intensity of reflections to be higher than normal, resulting in peaks in the sibilance band.

You can have a similar issue at home by sitting a lot higher than the speakers (ex using non-angled front baffle speakers on desktop, and no stands that rectify that), or sitting too close to multiple driver towers. However in a car that's exacerbated by windshield reflections, so unless you can get the same reflections in your house, the second scenario is more likely to just image the notes easily localizable at least by driver height.

I can see where you are trying to compare a home situation, but they really don't come close to the car situation at all. You've got quite a challenge there.

I think it is safe to say that sibilance is generally unrelated to speaker time-alignement, with the notable exception of automotive audio.

 

[ProtegeManiac]

So this is a problem confined to car audio since that's where the transducers are widely spaced, each one in a different location relative to the driver. That doesn't happen in HT. Your calibration problem is much worse.

I think I'm getting that your comment re: sibilance relating to time-alignment is 1. confined to this kind of car system and 2. the result of some rather difficult off-axis lobing in the crossover regions causing the intensity of reflections to be higher than normal, resulting in peaks in the sibilance band.

I can see where you are trying to compare a home situation, but they really don't come close to the car situation at all. You've got quite a challenge there.

I think it is safe to say that sibilance is generally unrelated to speaker time-alignement, with the notable exception of automotive audio.

Tell that to my Wharfedale Pi10 in my room with concrete exterior walls on one side and the attic sloping one side. If they didn't have the swivel mounts on the tweeters the only choice I have is "too much toe in" and "really sharp treble."

And like I said, I add this to "sibilance" even though it technically is not because when such a thing happens - which, again, is inherent in cars, but that doesn't make it impossible in a home audio setting (really, try using a low desk and no stands for nearfield monitors - I have that kind of problem with my desktop system if there isn't enough toe-in) - the problem is that when it does occur outside of a car setting, you'd end up with the same non-solutions being applied. Like EQ or switching out the speakers entirely. The latter might work, but that might have more to do with one speaker's dispersion pattern than a treble spike in the other.

 

[pinnahertz]

Tell that to my Wharfedale Pi10 in my room with concrete exterior walls on one side and the attic sloping one side. If they didn't have the swivel mounts on the tweeters the only choice I have is "too much toe in" and "really sharp treble."

And like I said, I add this to "sibilance" even though it technically is not because when such a thing happens - which, again, is inherent in cars, but that doesn't make it impossible in a home audio setting (really, try using a low desk and no stands for nearfield monitors - I have that kind of problem with my desktop system if there isn't enough toe-in) - the problem is that when it does occur outside of a car setting, you'd end up with the same non-solutions being applied. Like EQ or switching out the speakers entirely. The latter might work, but that might have more to do with one speaker's dispersion pattern than a treble spike in the other.

What you describe is the interaction of a speakers dispersion pattern with a reflective surface. That has nothing to do with time alignment. Though both may be at least partially corrected using digital filters the distortion mechanism and result are entirely different.

 

[ProtegeManiac]

What you describe is the interaction of a speakers dispersion pattern with a reflective surface. That has nothing to do with time alignment. Though both may be at least partially corrected using digital filters the distortion mechanism and result are entirely different.

My point about it was that such things need to be comprehensively tackled so people will know better what they're actually dealing with, rather than the non-solutions I described in my previous posts.

 

[pinnahertz]

My point about it was that such things need to be comprehensively tackled so people will know better what they're actually dealing with, rather than the non-solutions I described in my previous posts.

My point was that your statement earlier in the thread, "In speaker systems, it can be caused by reflections and time alignment issues (ie you hear the output from each speaker on the same note out of sync)" is misleading, and frankly, just wrong. You may see something like that happen in the car with widely spaced drivers, but not elsewhere. You've also blended time-alignment with reflection issues.

 

[ProtegeManiac]

My point was that your statement earlier in the thread, "In speaker systems, it can be caused by reflections and time alignment issues (ie you hear the output from each speaker on the same note out of sync)" is misleading, and frankly, just wrong. You may see something like that happen in the car with widely spaced drivers, but not elsewhere. You've also blended time-alignment with reflection issues.

Precisely because you can have both problems exacerbate each other and then some people mistake that for sibilance, which is where we started.

I wonder how many carbs I could burn just repeating myself to this guy.

 

[gregorio]

[1] Precisely because you can have both problems exacerbate each other and then some people mistake that for sibilance, which is where we started.
[2] I wonder how many carbs I could burn just repeating myself to this guy.

1. No it's not! Where we started, was assuming headphones, not surprisingly seeing as this is a headphone forum! We progressed onto speakers when the OP mentioned auditioning hi-fi equipment at dealers. At no point though, at the start or anywhere else, has there been any mention of car audio systems, until you bought it up!
2. I can't see why you bought up car audio systems in the first place, let alone bother repeating yourself?!

G

 

[ProtegeManiac]

1. No it's not! Where we started, was assuming headphones, not surprisingly seeing as this is a headphone forum! We progressed onto speakers when the OP mentioned auditioning hi-fi equipment at dealers. At no point though, at the start or anywhere else, has there been any mention of car audio systems, until you bought it up!
2. I can't see why you bought up car audio systems in the first place, let alone bother repeating yourself?!

G

Because

1) To understand what something is you need to clarify what it is not.

2) It was the extreme case as far as time alignment-caused "sibilance" is concerned and yet the same issues can still happen at home. As much as most home systems have people sitting equidistant I've seen people posting where they plan to put a nearfield system or already have one and in some cases the speakers aren't the same distance and in some cases aren't even the same height, and they're close to the corners of a room.

But the again, you're right, why bother repeating myself. I just wasted energy typing this reply and it's not like you will comprehend what I just typed down. I could type that with a jackhammer and your skulls will not get a scratch.

 

[frodeni]

[...]
So, what do you recon that one means, when they are speaking about:

-Sibilance
-Grainy
-Veiled
-Black
-Micro/Macrodynamics

And for a bonus: In how many part do you split your tonal specrtum, and how do you call them? (Bass-mids-highs...)

I never use any of this expressions, and given that people cannot even agree upon what soundstage and imaging is, forget about these. Also, I really only use spectrum terms to describe things to people who use spectrum terms.

I do how ever, use terms related to enjoying music. Like soundstage, imaging, separation, harmony, attack, articulation, etc. Plain text as "perk sound horrible" also works wonders. Particularly if related to a particular tune.