DiyMobileAudio.com Car Stereo Forum banner

1 - 20 of 285 Posts

·
Registered
Joined
·
1,310 Posts
Discussion Starter #1
Ok, so this may seem like a "newb" question, but it's one I can't seem to find a good answer to...WHERE DOES WIDTH COME FROM?

I'm going to "clif note" some other points for now just to quantify where I'm coming from. Height cues come from around 3khz+ and it's why you can get good height by just putting tweeters up high (I understand shallow slopes to midbass/midrange down low can "rainbow" height, but I'm trying to keep this basic for now). Depth cues are generally from a lack of early reflections and hearing some late reflections, in my experience (which is why MS8 rear-fill helps with depth).* A good center image comes from solid balance between left and right and is primarily affected by midbass/midrange (ITD). I understand that head-shadowing in tweeters can also affect balance (ILD), but it's not the primary focus (i.e. it's easier to get a "center" with good MB spacing and poor tweeter placement than the other way around).

So, WHERE DOES WIDTH COME FROM? I ask because it seems there are two arguements commonly seen as I search.

-If you put your tweeters at the widest point, you will have good width (most people suggest sails).
-If you put MB/MR as wide as possible, you will have good width (most people say doors, but moving to kicks is worth the small width "sacrifice" for improved imaging).

I wouldn't think that width could be fixed by wide tweeters OR wide speakers, so I'm curious which actually DOES affect it. Is there a specific frequency that width comes from? Does phyical placement affect "acoustic" width? Do we need to be as 'wide as possible'?

I would think that width, like depth, is mostly a function of reflections. Thinking that, I would argue that since it is nearly impossible to avoid reflections anywhere below 3khz or so because of wavelegnth AND because crosstalk is hard to also avoid in this range...treble frequencies are the answer to good width. I believe that the reason people have good luck with sail panels and width ISN'T because they are physically wider, but IS because it's generally the best place for avoiding early reflections.

For the sake of making this short enough that people don't "give up" on this thread before even starting, I'll stop there and open up discussion. I'll make one last comment on my reflection theory first though: if you put a speaker in an anechoic chamber where no reflections where present to give you a "physical boundry"...how wide would the image be?



*Intensity differences can affect depth too (imagine your eyes closed and a speaker being moved closer towards you--increased intensity CAN lead to preceived depth or shallowness).
 

·
Moderator
Joined
·
18,619 Posts
Physical boundaries and physical location. Like you've said, reflections due to the environment seem to be the driving factor. this effects the arrival time, yes, but IMO moreso the amplitude which helps to give a sense of breadth in the system.
If we base it around time then ITD (below 1khz or so) would take precedent. If we say it's amplitude driven then ILD takes over and we focus on above 2khz.
I'm going to go with ILD here but based in reflection. The further the reflection, the less amplitude impact it imparts. At least... That's how I'm seeing it.

/weak reply from iPhone.
 

·
Registered
Joined
·
1,310 Posts
Discussion Starter #3 (Edited)
Physical boundaries and physical location. Like you've said, reflections due to the environment seem to be the driving factor. this effects the arrival time, yes, but IMO moreso the amplitude which helps to give a sense of breadth in the system.
If we base it around time then ITD (below 1khz or so) would take precedent. If we say it's amplitude driven then ILD takes over and we focus on above 2khz.
I'm going to go with ILD here but based in reflection. The further the reflection, the less amplitude impact it imparts. At least... That's how I'm seeing it.

/weak reply from iPhone.
Nice ninja edit. ;) That reply isn't as "weak" as your first one.

What gets me about physical boundaries is that some people can get a system to have apparent width outside the vehicle. If I'm reading your reply correctly, it also seems you mostly agree with my logic. If so, it still makes since why tweeters off-axis in the sails work so well for width. It isn't so much they are literally wide, but that they minimize (or delay and reduce the amplitude) reflections off the dash, windshield, and side glass.
 

·
Registered
Joined
·
12,054 Posts
I liked the first reply better :D
 

·
Registered
Joined
·
197 Posts
I've also wondered about this, and i am going to be rebuilding my kicks soon so it will be interesting to see the responses.

I have always thought having the speakers on axis helps, but unsure wether this is because of lack of reflections or a better FR. I plan on getting some metal baffles made up and tilting the edges of the midbass into the a frame, to get them more on axis AND wider at the same time.

Gav
 

·
Registered
Joined
·
1,205 Posts
Primary width cues are in the midrange. Primary stage width will be dictated by placement of whichever drivers are responsible for midrange frequencies. Ambient width can be augmented with treble placement. I've heard this called 'stage boundary' width. It is possible* to have primary width at or outside the pillars by placing the midranges far outboard and the tweeters in a central location designed to reduce early reflection. This can lead to a split image and other problems, however. 'Phantom' ambient width can be created with late reflection in the treble creating a heightened sense of LEV (Listener envelopment), and although it may be considered more pleasant it is also less accurate. Generating primary width outside the confines of the vehicle requires both physical width of the midrange-carrier and reflection control. I have not yet found a 'magic ratio' of direct to reflected sound in the midrange that guarantees that effect, but I feel it probable that one exists.

I'll go back to my references and try to pick out quotes (from Toole's Sound Reproduction) that reinforce the statements above. Until then, they are my opinions based on what I've determined experimentally in my car combined with information gleaned from hundreds of hours of reading. That should not be considered in any way conclusive.

Excellent subject , and I hope to hear people with experience chime in! I also look forward to the backing arguments as well. This is one of those subjects that deals with many interactions that I think challenge the installer to really understand what's happening. It's also an area that I think could use a bit of demystification. I'm not trying to say I have the knowledge to do that concisely, but I look forward to seeing other's thoughts on the matter!
 

·
Registered
Joined
·
1,310 Posts
Discussion Starter #7
I want to add another caveat. I think keeping directivity the same at crossover points is also important in helping width. If one speaker is beaming and the other has wide dispersion, you can localize the one that's beaming. I would think the key to width is not actually knowing where the soundfield ends. If you can localize by reflections at the physical boundry or by a speaker beaming, the effect is ruined and your stage stops WHEREVER that localization is.

That means either using a wideband setup, 3-way setup, or waveguides (like the JBL 660GTi) to avoid problem #2.
 

·
Moderator
Joined
·
18,619 Posts
Primary width cues are in the midrange. Primary stage width will be dictated by placement of whichever drivers are responsible for midrange frequencies. Ambient width can be augmented with treble placement. I've heard this called 'stage boundary' width. It is possible* to have primary width at or outside the pillars by placing the midranges far outboard and the tweeters in a central location designed to reduce early reflection. This can lead to a split image and other problems, however. 'Phantom' ambient width can be created with late reflection in the treble creating a heightened sense of LEV (Listener envelopment), and although it may be considered more pleasant it is also less accurate. Generating primary width outside the confines of the vehicle requires both physical width of the midrange-carrier and reflection control. I have not yet found a 'magic ratio' of direct to reflected sound in the midrange that guarantees that effect, but I feel it probable that one exists.

I'll go back to my references and try to pick out quotes (from Toole's Sound Reproduction) that reinforce the statements above. Until then, they are my opinions based on what I've determined experimentally in my car combined with information gleaned from hundreds of hours of reading. That should not be considered in any way conclusive.

Excellent subject , and I hope to hear people with experience chime in! I also look forward to the backing arguments as well. This is one of those subjects that deals with many interactions that I think challenge the installer to really understand what's happening. It's also an area that I think could use a bit of demystification. I'm not trying to say I have the knowledge to do that concisely, but I look forward to seeing other's thoughts on the matter!
I think you and I are pretty much in agreeable. Hopefully I can get some computer time later and provide some sources as well.
 

·
Premium Member
Joined
·
3,855 Posts
IMO you cannot get width to exceed the the boundaries of the speaker locations.
As Todd mentioned and We've had some great discussions on this-width cues are primarily in the midrange. There are some ambient cues above 4khz.

But the easiest way for me to explain this would be, its very difficult to be wider than the center of the speakers location and proximity to the outer boundary of the vehicle (apillar)
So if you have a speaker mounted somewhere and the center is 3" inside the pillar then your width will be primarily about that far inside.
There may be some ambient cues that pull the width out at certain freqs, but the primary stage width will be narrower.

also, note that some of this will have to do with how the media is mixed. Where the recording engineer "placed" images in the final mix etc...

also, when evaluating any aspect, either use a track especially designed to determine that aspect like the IASCA,EMMA disc or even the Chesky Disc with the tracks used for judging width.
Or
use multiple tracks, with your eyes closed or in the dark with an Open Mind.
Its amazing what people can make themselves hear.

also, practice and hear other reference systems.
Different people cue in on different frequencies. Its very interesting to note how people hear differently.
So knowing what you are using also helps.

Most inexperienced listeners tend to pick up on ambient cues 1st, so to them the car may seem wider than it really is. some pay attention to the "base" of the sound. some are able to process the entire spectrum together to determine a location. etc.. etc...
 

·
Registered
Joined
·
1,310 Posts
Discussion Starter #11
IMO you cannot get width to exceed the the boundaries of the speaker locations.
As Todd mentioned and We've had some great discussions on this-width cues are primarily in the midrange. There are some ambient cues above 4khz.

But the easiest way for me to explain this would be, its very difficult to be wider than the center of the speakers location and proximity to the outer boundary of the vehicle (apillar)
So if you have a speaker mounted somewhere and the center is 3" inside the pillar then your width will be primarily about that far inside.
There may be some ambient cues that pull the width out at certain freqs, but the primary stage width will be narrower.

also, note that some of this will have to do with how the media is mixed. Where the recording engineer "placed" images in the final mix etc...

also, when evaluating any aspect, either use a track especially designed to determine that aspect like the IASCA,EMMA disc or even the Chesky Disc with the tracks used for judging width.
Or
use multiple tracks, with your eyes closed or in the dark with an Open Mind.
Its amazing what people can make themselves hear.

also, practice and hear other reference systems.
Different people cue in on different frequencies. Its very interesting to note how people hear differently.
So knowing what you are using also helps.

Most inexperienced listeners tend to pick up on ambient cues 1st, so to them the car may seem wider than it really is. some pay attention to the "base" of the sound. some are able to process the entire spectrum together to determine a location. etc.. etc...
Ok, I like this point. If you are correct, then shouldn't we aim for our width to be located however far outward we can get our midrange? Anything beyond that would not be actual "width" if that's the case but actually inaccurate ambient affects.

I do also agree about inexperienced listeners to a degree. When I first started, I liked reflections from the tweeter because it made things sound "fuller". Now I pretty much hate them because you realize how fatiguing they can be and how that "fuller" adjective is actually "diffuse center image". :(

What are your thought's on nailing down an actually frequency for "midrange". Is that 350-3khz? 80-1.5khz? Can you have a midbass in your door playing 60-200hz and "fix" the width of a midrange in the kicks or pillars (if not, didn't you just kill PLD's to try and help "width" when it really doesn't change anything)?

This is the type of info I think we can get out there to help people. It is already there for height and depth IMHO, so I think we can at least work on some general ideas for getting good width...or at least something better than "XXX worked for me" type subjectivity.
 

·
Registered
Joined
·
17,149 Posts
In recording width can and does come from L-R information....

Sleep on that one for a bit :D
 

·
Registered
Joined
·
6,376 Posts
I've had rears with no processing that with the right FR and right fading the car nearly sounded like headphones. I shoot for that with most of my personal installs, but sometimes don't have the time to get all the way there via trial and error.

I think spikes in the response destroy the illusion, this is worse when integrating rears. Note most rears are similar distance with less treble to make it work best. Not wanting to discuss 'rears' here, just saying it can work to pull the left out to the left, thus give a wide stage similar to headphones. I can't say exactly what makes it work or not, I often swap drivers in the rear and change the install until it works. Install meaning depth/angle of the driver, attenuating the tweeter mechanically or with install, etc. T/A gives a similar effect with the fronts alone but its not the same, at least in this car.
 

·
Registered
Joined
·
17,149 Posts
Chad, you have been spending WAY too much time reading Lycan's posts. :p

Dropping a bomb and leaving is just wrong.
I do it on the other side for a living :D

Ever look at how old skool "dolby surround" worked?

they put that L-R somewhere.

you can get width AND depth from L-R and depending on where you place that information you can move, in reason, anything anywhere you want (within reason) it as long as you are recording anywhere outside an anechoic chamber. the 'ol M-S mic technique is a cool way to do this with few channels live.
 

·
Premium Member
Joined
·
3,855 Posts
Ok, I like this point. If you are correct, then shouldn't we aim for our width to be located however far outward we can get our midrange? Anything beyond that would not be actual "width" if that's the case but actually inaccurate ambient affects.

I do also agree about inexperienced listeners to a degree. When I first started, I liked reflections from the tweeter because it made things sound "fuller". Now I pretty much hate them because you realize how fatiguing they can be and how that "fuller" adjective is actually "diffuse center image". :(

What are your thought's on nailing down an actually frequency for "midrange". Is that 350-3khz? 80-1.5khz? Can you have a midbass in your door playing 60-200hz and "fix" the width of a midrange in the kicks or pillars (if not, didn't you just kill PLD's to try and help "width" when it really doesn't change anything)?

This is the type of info I think we can get out there to help people. It is already there for height and depth IMHO, so I think we can at least work on some general ideas for getting good width...or at least something better than "XXX worked for me" type subjectivity.
Imaging localization typically start around 150hz or so. One of the very few things I ever agree with someone like Scott Buwalda with is I prefer to have a midrange that can cleanly play up as high as possible, preferably to at least 4 or 5khz.

So YES, IMO the goal is always to get the speakers as far away and as wide as possible and away from as many direct reflecting surfaces as possible.

The ability to do that at least gets you in very good neighborhood. after that the ability to control individual phase response, eq, time etc...can come into play to move things around.
I still donot fully understand How Mark Elderidge is able to do with his phased arrays--but from what I hear its amazing.
and I know after talking to him at Finals he is working on the ability to move the left side out physically much wider than the boundaries while keeping the right side at the pillar so in the driver seat -you are truley in the acoustic center listening position.

for the rest of us stick with the basics above.
If you can use acoustic treatments, they can help alot as well when used in the right places.
 

·
Premium Member
Joined
·
3,855 Posts

·
Registered
Joined
·
1,310 Posts
Discussion Starter #18
I can't tell everybody how much I appreciate this thread staying in a positive direction. The information has been good as have the links that Mic provided. I'm looking forward to seeing what Bikin and Highly can dig up as well. ;)

That being said, the comments below are a bit frustrating. I'm not trying to be a dick, but you can't really have one or the other. I understand Mark E. has more experience, resources, and processing power than most of us, but if it can be done...it CAN be done. If Mark can manipulate the sound to appear to come from beyond the driver's physical location, it means there is a "science" to width just as their is to height and depth.

That is the purpose of this thread, to see how to get good width, what frequencies primarily control it, and if we can "width" to exceed the physical boundaries of the speakers themselves (without simply being inaccurate ambiance). I think we've got some solid stuff on #1, and a good idea of #2, so can we/how do we get to #3 in the car?

IMO you cannot get width to exceed the the boundaries of the speaker locations.
I still donot fully understand How Mark Elderidge is able to do with his phased arrays--but from what I hear its amazing.
and I know after talking to him at Finals he is working on the ability to move the left side out physically much wider than the boundaries while keeping the right side at the pillar so in the driver seat -you are truley in the acoustic center listening position.
 

·
Registered
Joined
·
17,149 Posts
It's an undeniable fact that even in a closed environment you can get the stage to extend.... Headphones do it.

Ambio rigs do it.
 
1 - 20 of 285 Posts
Top