This is said lightly ... 🙂markus76 said:
... and this is easier said than done. 🙄
It needs a veeery structured mind to archive and link every bit of intelligence one has sunk in forums while discussing something with other members. Nowadays I even have difficulties to navigate my own personal "loudspeaker" archive - let alone those internet forums I attend.
If you know any efficient method to really hold track of your online discussions ... let me know, please.
couldn't resist
Not really... it's just tiresome to hear the same ole story about alternative techniques... the "NIH" syndrome is alive and well...😀
markus76 said:
Not really... it's just tiresome to hear the same ole story about alternative techniques... the "NIH" syndrome is alive and well...😀
Rudolf, I was not talking about each and every discussion but fundamental findings. Just like you did with dipolplus.de
Markus,
"The Nathan is not flat on axis which would be a requirement for using it as a center speaker."
Unless you use an AT (acoustically transparent) screen, I'd not expect your ears to be on axis with the C.
Hmm, perhaps you meant 3-ch music usage, not video, but either way you wouldn't have to tilt the C forward or back very much to get the hole from being at ear height.
"The Nathan is not flat on axis which would be a requirement for using it as a center speaker."
Unless you use an AT (acoustically transparent) screen, I'd not expect your ears to be on axis with the C.
Hmm, perhaps you meant 3-ch music usage, not video, but either way you wouldn't have to tilt the C forward or back very much to get the hole from being at ear height.
I meant music and video. I wouldn't use a center if it can't be brought up to the same vertical position as the left and right front speaker.
I can look for my own postings on this topic and see what I have said, but would reiterate here if need be.
Ok "research" looking at the spectrum of the surround channels has shown that movies and music typically have very limited bandwidths and dynamic ranges. I have performed some of the spectrum analysis myself, but I've also seen this posted elsewhere. While I contend that, to a point, this is changing some with current surround formats (i.e. DTS-HD, DD+), but still remain limited in both. While surround formats have been full range and had the ability to have the same dynamic range as the fronts since the onset of dolby digital and DTS, because of the bit misery these formats use, they have not done so. In other words, while they could be full range, they aren't, so what could be is really unimportant.
While I'm not an expert on psycho-acoustics, Dr. Geddes has mentioned, and I have read before that psycho-acoustic research has shown that part of what tells the ear that something is far away is a gradually rolled off or shelfed top end. This begins to imply that surround speakers will give a better sence of space (with regard to distance) if they aren't full range.
Surround speakers should also have fairly directed dispersion, in a way. Dr. Geddes has talked before about how non-directional highs would be a bad thing as this would cause the surrounds to call more attention to themselves. I have seen this now in my own surrounds, and have taken measures to correct this. So far this has been successful, and has in fact reduced the surrounds ability to disappear.
So to quickly recap what should be inferred from those points, surrounds should not be full range, they should roll off in the highs gradually, and should have a controlled dispersion pattern so as not to call attention to themselves.
Now we have the issue of directivity with regard to direct, dipole, or bipole. Well there seems to be no argument that dipole and bipole can both enhance the sense of envelopment, which, at least with movies, is a good thing.
I have to run to a meeting, but will post more on this later. Especially with regard to my opinions on the dual duty of surrounds for music vs movies.
Ok "research" looking at the spectrum of the surround channels has shown that movies and music typically have very limited bandwidths and dynamic ranges. I have performed some of the spectrum analysis myself, but I've also seen this posted elsewhere. While I contend that, to a point, this is changing some with current surround formats (i.e. DTS-HD, DD+), but still remain limited in both. While surround formats have been full range and had the ability to have the same dynamic range as the fronts since the onset of dolby digital and DTS, because of the bit misery these formats use, they have not done so. In other words, while they could be full range, they aren't, so what could be is really unimportant.
While I'm not an expert on psycho-acoustics, Dr. Geddes has mentioned, and I have read before that psycho-acoustic research has shown that part of what tells the ear that something is far away is a gradually rolled off or shelfed top end. This begins to imply that surround speakers will give a better sence of space (with regard to distance) if they aren't full range.
Surround speakers should also have fairly directed dispersion, in a way. Dr. Geddes has talked before about how non-directional highs would be a bad thing as this would cause the surrounds to call more attention to themselves. I have seen this now in my own surrounds, and have taken measures to correct this. So far this has been successful, and has in fact reduced the surrounds ability to disappear.
So to quickly recap what should be inferred from those points, surrounds should not be full range, they should roll off in the highs gradually, and should have a controlled dispersion pattern so as not to call attention to themselves.
Now we have the issue of directivity with regard to direct, dipole, or bipole. Well there seems to be no argument that dipole and bipole can both enhance the sense of envelopment, which, at least with movies, is a good thing.
I have to run to a meeting, but will post more on this later. Especially with regard to my opinions on the dual duty of surrounds for music vs movies.
"psycho-acoustic research has shown that part of what tells the ear that something is far away is a gradually rolled off or shelfed top end. This begins to imply that surround speakers will give a better sence of space (with regard to distance) if they aren't full range."
I'd think recording engineers are aware of this and would encode the suitable rolloff.
While generally true that surround channel program material is for ambience and may/should have rolled off highs, there are occasions when pinpoint sound effects have full response and should be reproduced as such.
A good example is in Das Boot when they have to submerge beyond the submarine's design limits and bolt heads start popping off and hitting the hull with very loud sharp reports. Now that I think about it, richocheting bullets aren't all that uncommon.
I'd think recording engineers are aware of this and would encode the suitable rolloff.
While generally true that surround channel program material is for ambience and may/should have rolled off highs, there are occasions when pinpoint sound effects have full response and should be reproduced as such.
A good example is in Das Boot when they have to submerge beyond the submarine's design limits and bolt heads start popping off and hitting the hull with very loud sharp reports. Now that I think about it, richocheting bullets aren't all that uncommon.
Stands?
What do you suggest these speakers sit on? Does it make much difference? Over the years, I have heard all sorts of amazing claims about stands, I wonder how much difference they actually make... I assume that the center of the waveguide should be at the listener's ear level, right? Thanks.
Isaac
What do you suggest these speakers sit on? Does it make much difference? Over the years, I have heard all sorts of amazing claims about stands, I wonder how much difference they actually make... I assume that the center of the waveguide should be at the listener's ear level, right? Thanks.
Isaac
Well I think you would be surprised what frequency range the "ping" actually falls within. I mean, when I say rolled off, and I think, when Dr. Geddes refers to rolled off, he doesn't mean that at 3000hz it begins falling off at 24db's per octave.
Well anyway, I won't talk on his behalf anymore, he asked me to simply post the links to discussions we have had. I hope this works, I've not done this before.
Most recent comment
Further Discussion
HT Side and Rear Speakers
Discussion that started it all
This last link is, as the title says, what started the whole discussion. I've changed my thinking a bit since then. I don't want to speak for Dr. Geddes, so this is my interpretation, but I think a distinction should be made between his comment about his speakers being too directional for surround use, and another, seemingly contradictory comment about tweeters being essentially wide dispersion at their low end. What I think he meant was that surrounds, in general, should have a very dispersed and non-directional response, but the highs would and should still roll off and would become more directional as you go up the register. Most multi-way surrounds often do suffer from presenting a wide frequency range over a wide area of dispersion which makes it very easy to detect their location.
Well anyway, I won't talk on his behalf anymore, he asked me to simply post the links to discussions we have had. I hope this works, I've not done this before.
Most recent comment
Further Discussion
HT Side and Rear Speakers
Discussion that started it all
This last link is, as the title says, what started the whole discussion. I've changed my thinking a bit since then. I don't want to speak for Dr. Geddes, so this is my interpretation, but I think a distinction should be made between his comment about his speakers being too directional for surround use, and another, seemingly contradictory comment about tweeters being essentially wide dispersion at their low end. What I think he meant was that surrounds, in general, should have a very dispersed and non-directional response, but the highs would and should still roll off and would become more directional as you go up the register. Most multi-way surrounds often do suffer from presenting a wide frequency range over a wide area of dispersion which makes it very easy to detect their location.
Hi pjpoes,
thanks for your post but I completely disagree with all you've said although I do understand some of your points from a historical perspective.
1) "Bandwidth not required"
The specifications require all channels to be full range. Typical middle-of-the-band recordings make use of that. And what can be done will be done. So there's no reason to make limited bandwidth a design goal.
2) High frequency cut-off as a cue for distance
Indeed, for known sources a frequency cut-off can be a cue for distance. But that is something even the most basic reverberation effects processors have as an editable parameter. So this is something that is taken care of in the mix.
Furthermore there is a much stronger cue for distance: the initial time delay gap (ITDG). This is the time difference between the direct sound and the first strong reflections arriving at the listening position. Big ITDG means source is near. Small ITDG means source is far away.
But the main reason for limiting the bandwidth was HF-artefacts that became audible when decoding the surround channels from the optical 2 channel soundtrack. Nothing to worry about nowadays. But still the old speakers from that era are used in cinemas. Maybe this is the bandwidth limiting factor you found in certain recordings.
3) "Fairly directed dispersion"
In the beginning of cinema multichannel there was only one surround channel. This lead to a highly correlated surround sound signal because only two speakers were used at home. The use of dipole speakers helped in making the surrounds more decorrelated. Today decorrelation is done electronically in your AVR when playing one of those old formats. When using one of the common discrete formats the mixing engineer takes care of that.
So at home the surrounds should have
- the capability to deliver a discrete full range signal
- a dispersion that covers the whole seating area
If a large listening area has to be addressed then having a constant level at every listening position would be a design goal for a speaker too. Maybe line sources can help.
Best, Markus
thanks for your post but I completely disagree with all you've said although I do understand some of your points from a historical perspective.
1) "Bandwidth not required"
The specifications require all channels to be full range. Typical middle-of-the-band recordings make use of that. And what can be done will be done. So there's no reason to make limited bandwidth a design goal.
2) High frequency cut-off as a cue for distance
Indeed, for known sources a frequency cut-off can be a cue for distance. But that is something even the most basic reverberation effects processors have as an editable parameter. So this is something that is taken care of in the mix.
Furthermore there is a much stronger cue for distance: the initial time delay gap (ITDG). This is the time difference between the direct sound and the first strong reflections arriving at the listening position. Big ITDG means source is near. Small ITDG means source is far away.
But the main reason for limiting the bandwidth was HF-artefacts that became audible when decoding the surround channels from the optical 2 channel soundtrack. Nothing to worry about nowadays. But still the old speakers from that era are used in cinemas. Maybe this is the bandwidth limiting factor you found in certain recordings.
3) "Fairly directed dispersion"
In the beginning of cinema multichannel there was only one surround channel. This lead to a highly correlated surround sound signal because only two speakers were used at home. The use of dipole speakers helped in making the surrounds more decorrelated. Today decorrelation is done electronically in your AVR when playing one of those old formats. When using one of the common discrete formats the mixing engineer takes care of that.
So at home the surrounds should have
- the capability to deliver a discrete full range signal
- a dispersion that covers the whole seating area
If a large listening area has to be addressed then having a constant level at every listening position would be a design goal for a speaker too. Maybe line sources can help.
Best, Markus
The point to consider is that in my setup the sides are dipoles and the listener is on the dipole axis. This is the situation recommended by Thomason Holman, but quite the opposite of Floyd Toole. Holman is coming from the film side and Toole the music side. I also see these as very different situations and I expect that they are mixed quite differently. You almost have to decide which you are trying to achieve.
If you have side dipoles and you put a tweeter in them, then at the tweeters lower band edge it is basically omni-directional. This places a lot of sound energy right into your ear if the surrounds are close to you as they have to be in a small room. even if there are two tweeters wired as dipoles, the dipole axis will be very narrow at these frequencies. Better, IMO, is to use a single full range larger driver that has some directional characteristics and falls off slightly at the high end - something like -6 dB/oct above maybe 2-3 kHz. This is basically what a good 8" or 6" driver does. I will put two of these at the ends of a tube of the same diameter and hang them in the corner of the side walls.
With a speaker like the Nathan as a side surround it would basically be like wearing headphones. In my room they would only be a few feet away pointed directly into you ear - ouch! The direct field would dominate any reverberant field and there would be no spacial effect from the surrounds at all. You could use Nathans as the rear speakers in a 7.1 setup, but this would be overkill IMO. I just use a close box set of the same 8" loudspeakers that I use on the sides.
In a movie about 40% of the sound comes from the center channel - the dialog, 20% from each left and right speaker and 20% from ALL of the surrounds. The surrounds don;t need to be the same level of capability as the front three.
If you have side dipoles and you put a tweeter in them, then at the tweeters lower band edge it is basically omni-directional. This places a lot of sound energy right into your ear if the surrounds are close to you as they have to be in a small room. even if there are two tweeters wired as dipoles, the dipole axis will be very narrow at these frequencies. Better, IMO, is to use a single full range larger driver that has some directional characteristics and falls off slightly at the high end - something like -6 dB/oct above maybe 2-3 kHz. This is basically what a good 8" or 6" driver does. I will put two of these at the ends of a tube of the same diameter and hang them in the corner of the side walls.
With a speaker like the Nathan as a side surround it would basically be like wearing headphones. In my room they would only be a few feet away pointed directly into you ear - ouch! The direct field would dominate any reverberant field and there would be no spacial effect from the surrounds at all. You could use Nathans as the rear speakers in a 7.1 setup, but this would be overkill IMO. I just use a close box set of the same 8" loudspeakers that I use on the sides.
In a movie about 40% of the sound comes from the center channel - the dialog, 20% from each left and right speaker and 20% from ALL of the surrounds. The surrounds don;t need to be the same level of capability as the front three.
I don't think that this is possible but in the end that's exactly what we are looking for: the recorded space becomes real by the speakers delivering the necessary cues.
That's an argument I never understood: I would want every sound source to be full bandwidth at the exact position as intended even if it's only required for one second in the whole movie.
Best, Markus
I just want to say that the bandwidth limiting in surround channels is true of all surround mixes I have tested, right on up to 2008 and beyond. It's done because, while the specs allow for full range, this uses needed data bits, and so an additional part of all surround encoding formats to date is a scheme for bit pooling or bit sharing between channels, to save space. It's a sort of dynamic compression of the data if you will. Since studio engineers appear to believe that surrounds don't need to be full range, and making them so would use up needed bits, they have instead chosen to limit the bandwidth and dynamic range to save bits. While I'm guessing at the exact reason, I think its a pretty good and educated guess. None the less, for all movies I have tested thus far, right on up to 2008, none have had full range surround channels and a surprising amount don't even have full range main channels. I've only recently been able to begin looking at the newer DTS-HD and other new surround formats, and so far they are a mixed bag. My guess is that many are simply porting over the soundtracks made for the regular def versions of the movies with less compression applied to the main channels, but little done to the surround channels. However, its possible that engineers still feel that high frequencies are not needed.
I would also note that, as of current, my main surrounds do use a tweeter, and use a sort of "tripole" design. They have a midbass and waveguide loaded tweeter (shallow waveguide) on the front, with two full range 3" peerless drivers on the sides. The tweeter has a small .47uf film cap in parallel to offer a slight roll-off of the top end, and smooth some ripples caused by the waveguide. The side firing full rangers have a 1uf parallel cap and a 45uf series cap to limit their bandwidth into their best operating range, and keep down any destructive interference they may have with the front midbass drivers bass. The Peerless midbass crosses over to the Peerless/MK tweeter around 3.7khz and these drivers work in phase with the full range driver facing forward, but out of phase with the full range facing toward the back of the room. The full range drivers are wired in series which keeps the load above 4 ohms, well actually around 14ohms taken alone, and act to pad down their output. They also have a switch to disconnect them, but I've never found a preference to using them that way.
I don't personally like music recorded to put the listener into the center of the mix, i.e. in the middle of the band. I feel that I never go to see music in this way, and when I play gigs with guys, this isn't what I hear either. I much prefer surrounds mixed to give the ambiance of sitting in the audience, and for this purpose, I find that dipole surrounds still offer benefits. This is of course my opinion, and given that there really is no consensus on the topic, to each their own. None the less, I would disagree that my view or even Dr. Geddes view is based on history, which to me at least, implies an antiquated idea. I think this is still current thinking, especially given my very recent surround modifications and experimentation.
I would also note that, as of current, my main surrounds do use a tweeter, and use a sort of "tripole" design. They have a midbass and waveguide loaded tweeter (shallow waveguide) on the front, with two full range 3" peerless drivers on the sides. The tweeter has a small .47uf film cap in parallel to offer a slight roll-off of the top end, and smooth some ripples caused by the waveguide. The side firing full rangers have a 1uf parallel cap and a 45uf series cap to limit their bandwidth into their best operating range, and keep down any destructive interference they may have with the front midbass drivers bass. The Peerless midbass crosses over to the Peerless/MK tweeter around 3.7khz and these drivers work in phase with the full range driver facing forward, but out of phase with the full range facing toward the back of the room. The full range drivers are wired in series which keeps the load above 4 ohms, well actually around 14ohms taken alone, and act to pad down their output. They also have a switch to disconnect them, but I've never found a preference to using them that way.
I don't personally like music recorded to put the listener into the center of the mix, i.e. in the middle of the band. I feel that I never go to see music in this way, and when I play gigs with guys, this isn't what I hear either. I much prefer surrounds mixed to give the ambiance of sitting in the audience, and for this purpose, I find that dipole surrounds still offer benefits. This is of course my opinion, and given that there really is no consensus on the topic, to each their own. None the less, I would disagree that my view or even Dr. Geddes view is based on history, which to me at least, implies an antiquated idea. I think this is still current thinking, especially given my very recent surround modifications and experimentation.
Sorry mis-spoke, there is an LCR on the full rangers, not just a cap, just not done in a very conventional way, the "full range" basically has a notch filter at 12khz, so not much beyond that point.
Surround channels more often than not had nothing above 5khz to be often, I don't think ever above 10khz, and again, I was surprised to find many not having any material above 3.5khz or so.
I do think we are starting to debate semantics here, I mean, the 8" full range would have response to 5khz+ without issue, maybe not 10khz, but Dr. Geddes did talk about using an 8" with wizzer cone, so most of those do have response past 10khz. As for my own designs, well they use tweeters and full rangers with response past that, its just all padded down and tilted down a bit in level.
I do think we are starting to debate semantics here, I mean, the 8" full range would have response to 5khz+ without issue, maybe not 10khz, but Dr. Geddes did talk about using an 8" with wizzer cone, so most of those do have response past 10khz. As for my own designs, well they use tweeters and full rangers with response past that, its just all padded down and tilted down a bit in level.
Do I understand you right that you "tested" the bandwidth of surround channels by listening to them?
Markus
There is very little concrete data about surrounds to work with. I have experimented a little, and I definately find that if I can tell the surrounds are working I don't like it. I can tell when they aren't on but if they are working right, IMO, you can hardly tell that they are on. I can say with certainty there there is no consistancy of surround mixing. Most engineers arn't very good with two channels, some grasp three, but its a rare engineer who can deal with 7.1 effectively. Perhaps this is why I keep them so low and unobtrusive, to keep me from hearing the bad mixes.
I'm guessing that you don't have much experience with surround because I don't think that youd like a pair of Nathans in your ears on the surround channels.
There is very little concrete data about surrounds to work with. I have experimented a little, and I definately find that if I can tell the surrounds are working I don't like it. I can tell when they aren't on but if they are working right, IMO, you can hardly tell that they are on. I can say with certainty there there is no consistancy of surround mixing. Most engineers arn't very good with two channels, some grasp three, but its a rare engineer who can deal with 7.1 effectively. Perhaps this is why I keep them so low and unobtrusive, to keep me from hearing the bad mixes.
I'm guessing that you don't have much experience with surround because I don't think that youd like a pair of Nathans in your ears on the surround channels.
No I tested surround channels by removing the sound track from the dvd and pulling out the surround channels, then converting to an uncompressed wav file and examined the entire 1-3 hour movies worth on spectral analysis software. I want to say that the software will give you proportions of total time that x-y range of frequencies exist, but now I can't remember if I figured out how to do that or not. Mostly it was just setting a line at 5khz, 10khz, etc and seeing what hits or lies above that line. I want to say I had lines at 3,3.5, 4.5, 5, 7, 10, 15, and 20khz. I would display examples but my laptop is still in for repair, and its what had the software and files.
- Status
- Not open for further replies.