Oh. Yeah. Years ago I put a 48x30x24" vinyl, (soft, dense, floppy stuff), box between the speakers as a quick and dirty experiment to explore what you're talking about. It worked. I understand the effect from an empirical and theoretical point of view.
So, what you're talking about makes badly designed, badly built speakers sound worse. Right?
With respect to loudspeaker dispersion alone, I guess it depends on how you actually hear. For you perhaps, yes. 🙂
For most people, probably not.
People generally like quite a lot of inter-channel "cross-talk" with stereo reproduction (..it's why headphone amp manufacturers make models with "cross-feed").
Also, as far as tonal aberration from combing - that's handled in production already. It's equalized to be that way under loudspeaker stereo operation, with plenty of direct-sound cross-talk. If you were to play such a stereo track on a mono or multi-channel with center loudspeaker system - then it would be tonally wrong for most people. (..this also includes play-back on headphones, even though under production the tracks were probably "checked" with headphones.)
But - if you really want to "lay blame" somewhere, it really shouldn't be the loudspeaker design (..well, unless it's really FUBAR).
Rather any serious blame should go to the stereo process itself.
Tonally-speaking a proper multi-channel with true center recording is objectively far more accurate. Anyone can use such a recording (even with identical if mediocre loudspeakers) - and get a good tonal balance (..assuming it was not messed-up by those doing the recording).
Last edited:
Yup, the principal is similar to the "wall" device that some ambiophonic people use. (..though they like inter-channel blocking down to as low a freq. as possible). Honestly though, at some point you've just got to switch to headphones and maybe an effects "rumble" transducer (or two) - and then process to binuaral. Looks only slightly strange, unlike sticking your face against the edge of a wall. 😛
Lynn, Fostex T-900A tweeters work great for me, 1st order with a Dueland 2.2uF cap (this makes a difference and is worth considering as part of the tweeter package). The LM555s are free of a LP. I would guess that there will be a 1st order phase rotation due to the VC induction. I find the tweeters completely disappear now despite the vertical separation.
martin
martin
This is something that probably cannot be eliminated without other trade offs. I would go for Binaural recording and listen through earbuds for realistic spacial effect, but strangely such recordings seems to give a different loudness balance that the real environment. Have not figured our the real reason, but it could be the recorded blood flow sound in my ears.
Scott, I seriously doubt this theory/hypothesis 🙂
My ears are very sensitive since childhood and there is no change until now. I have learned how to not feel a physical pain and I'm quite successful. But this ears pain, I don't know how to avoid or ignore (I guess its too close to the brain).
Glad somebody does!
It just sounds like some psychoacoustic puffery to me. If you are on centerline there should be no "midrange dip" in the response for the phantom center.
"If you are on centerline"
Herein lie the problem, do we limit the great sound to a single point in space. the head in a vise solution where time and phase only work in a single specific location or do we try to design for a wider listening space? The one position in space location is fine if only one person wants to hear the music and hear that effect, the other condition is when multiple people are listening on a wide couch and you want as even a response as possible. This is my trepidation with very narrow pattern control and horns. Narrow conic section horns are the worst for this in my eyes. I will give up some of that pinpoint sound field so I can move about and still hear a nice balanced sound. Of course this means that the room reflections need to be addressed to even out the response in room. Two different approaches with very different final results.
Herein lie the problem, do we limit the great sound to a single point in space. the head in a vise solution where time and phase only work in a single specific location or do we try to design for a wider listening space? The one position in space location is fine if only one person wants to hear the music and hear that effect, the other condition is when multiple people are listening on a wide couch and you want as even a response as possible. This is my trepidation with very narrow pattern control and horns. Narrow conic section horns are the worst for this in my eyes. I will give up some of that pinpoint sound field so I can move about and still hear a nice balanced sound. Of course this means that the room reflections need to be addressed to even out the response in room. Two different approaches with very different final results.
I have one of Nick McKinney's original Unity waveguides, combined with a Lambda tdx 15 (apollo). The 15 is way overkill for a home situation, as noted earlier in the thread. Unlike Gary Dahl's system, I cross to the waveguide around 400Hz.
I can see why the GPA driver might have a timbre which would give a more realistic sounding presentation. My question is: Would the difference between the two 15" drivers be as apparent at the lower crossover frequency (yeah, I know it's guesswork, but I'll take an informed guess).
Other question is: What is the volume of your PR sub cabinets Gary?
Sheldon
I can see why the GPA driver might have a timbre which would give a more realistic sounding presentation. My question is: Would the difference between the two 15" drivers be as apparent at the lower crossover frequency (yeah, I know it's guesswork, but I'll take an informed guess).
Other question is: What is the volume of your PR sub cabinets Gary?
Sheldon
Are you saying Toole's data in "Sound Reproduction" on the dip (-10dB around 2khz in the book) due to the interference is incorrect? He goes on to say wide dispersion speakers can "fill" the dip to a point with early reflections. NOT saying I agree with that approach, just sayin'.
Nate - I am well aware of Toole's claims, you are quite correct about what he says, but I am not sure that what is being discussed here is the same thing, nor have I ever found this effect to be significant. I will have to review the Toole claim, because on the centerline there is no comb filtering so what causes a dip? Off the centerline there is going to be some, but I am not sure that it is that significant. There is some timbre changes in my system from the centerline to off axis, but I can't say that this is a severe effect, but most importantly what the cause is.
I am having trouble understanding what the problem is with a driver going too low? I understand cone breakup for the high , but what is going on for the low?
Thank you
Thank you
Horns' HF Response, Room Size, Crossovers and More Mix & Match Mania
HF Extension Via Pro Audio Software:
It’s only been several weeks ago when I was surprised to learn that horns generally have limited HF response, sometimes 10kHz, or less. No wonder Gary and Pierre added (active?) EQ to extend the HF response, though wouldn’t a pair of these work better http://www.fostexspeaker.de/tweeter/horn/t900a.pdf , albeit their less than stellar off-axis response?
I was reading about the invention and widespread use ever since of phase plugs https://en.wikipedia.org/wiki/Phase_plug for both compression drivers and woofers to extend HF response. With the obvious question being why aren’t all compression drivers equipped with built in phase plugs (assuming they can’t be added without problems), what are the tradeoffs to using them? For example, how would Gary’s Radian 745NeoBe drivers perform if they had phase plugs?
Assuming phase plugs cannot be added to a compression driver (LOL?!?) to Gary’s Radians-and I certainly wouldn’t want to use anything other than the 425 horns-since my sources are all digital, would professional quality software-based EQ be sufficiently transparent to achieve hi-fi sound quality? I typically rip my favorite CD tracks to uncompressed wav files. And because I like to sometimes apply special effects (i.e. small, lossless changes like pitch reduction and/or time expansion) during playbacks, I use Samplitude Pro X-2 Suite
Editing > Functions > Samplitude > Audio Production > MAGIX Pro
Sadly, throughout my life, most of my music listening has been via (Sennheiser) headphones, so I rarely found the need to use EQ. But as shown here Mastering > Functions > Samplitude > Audio Production > MAGIX Pro , Samplitude has impressive EQ functionality http://savedbytechnology.com/catalog/news_photos/107-magix_S_S_11_05.jpg
I know there are many different types of equalizers; some are best for one mixing and/or mastering and some are better for room and speaker EQ. However, would you care to confirm if the one (s) that come with Samplitude would work well to extend the HF range of the 425 horns/Radian drivers beyond 10kHz, without penalties?
On the other hand, there are countless very high quality software EQ plug-ins that are fully compatible with Samplitude, if Samplitude’s own EQ wasn’t best suited for this application.
I believe that you can download Samplitude for a fully functional 30 day trial MAGIX Download If you have some well recorded and digitized music on CD or download and a good sounding standalone DAC, why not rip the CD to an uncompressed wav file and play it via Samplitude through your DAC into your amp(s) while tweaking its EQ? Samplitude might even allow real time EQ tweaks if your CPU and memory are up to it.
If yes, were you able to extend the highs of your horns without coloration or loss of spatial integrity?
IF Samplitude’s EQ did the trick for your drivers, then shouldn’t I be able to get a beautiful tweeter-less extended HF response from the Radian drivers-and reap the superior off axis response of horns?
Crossovers:
My knowledge of crossover design is now hardly substantial. Presently, I don’t know a whole lot more than that a coil in series with the woofer blocks HF from reach it and that a cap does the same with LF for a tweeter. But my First Watt B4 two-way active crossover (which I hope was not an expensive purchasing mistake) can generate all degrees of L-R and S-K filters. And that the steeper a two-way crossover’s slope the more dissimilar will be the high pass and low pass phase response. Did Gary use a 12db slope and the Fc was 900Hz, but not a L-R type (?). I do know that his Altec 416s start playing at 700Hz.
If yes, then would an L-R 12db passive crossover @ 900Hz between the HA425 horns/Radian drivers and the Altec 416s sound very good? And active bi-amping using the same crossover sound as good or better?
Room Size:
This is very possibly a serious issue that I now regret not giving much though to, probably because I was more concerned with planning for the future rather than the present. Within the next two years, perhaps sooner, I will be moving to a two bedroom co-op or condo. I expect my living room to be uncluttered but only about 18 ft x 14. Would that room will be big enough for the 425 horns, the Altec 416s (in Gary’s 3 cu ft sealed cabinets) and a pair of 12” sealed Rythmik subs?
However, the present room I have available for them is a cluttered 17 ft x 9 room.
Am I expecting the impossible for that system to properly work in that room?
Furthermore, even if the smaller room wouldn’t sound too bad, will later moving it to somewhat larger room require crossover changes?
Maybe I just don’t have the listening room space for horns now and/or in the future
I guess the best question to ask might be how small a room would the 425s work in without acoustical and/or crossover problems?
Speaker System Alternatives:
If you think an 18 ft x 14 and/or a somewhat cluttered 17 ft x 9 room are too small for the AH425s, then perhaps the best alternative to horns-and what might be the best answer to their HF deficiencies at the same time-are AMTs. That is, to use a particular model AMT with a direct radiator midrange driver and the Altec 416 midwoofers below that. Here Beyma TPL 150 AMT - the ultimate driver whats the catch? Tuyen is very impressed with everything about them except the vertical dispersion, which doesn’t match that of the compression tweeters he uses (Goto?).
As can be seen in the graph it’s a pretty sad story after 1kHz.
http://www.beyma.com/getpdf.php?pid=TPL-150/H ; all other Beyma AMTs had worse vertical dispersion.
I would also like to know what Tuyen may think of Mundorf’s AMTs.
Someone at that same hifiwigwam thread said that Mundorf spend $200K on R & D for AMTs, and the impressive distortion specs on most of their models seem to bare this out http://mundorf.com/PDF/Mundorf_hifiAMT_Catalog.pdf
However, vertical and/or horizontal dispersion looks no better than the Beyma 150/h, or perhaps that Fostex T900 tweeter
http://www.fostexspeaker.de/tweeter/horn/t900a.pdf How do they expect us to be able to get up and dance with our girlfriends with vertical responses like these?
But might AMTs’ off-axis HF response be extended without penalty via EQ software while playing digital source material?
And for a driver between the AMT and the Altec 416, has anyone any listening experience with this Fostex alnico midrange
https://www.madisoundspeakerstore.c...fostex-f120a-5-full-range-with-alnico-magnet/ ? Or please explain why another midrange driver might be much better for above the Altec 416s.
I thought of bi-amping the F120’s with Gary’s Altec 416 midwoofers. Do you think that a first order 6db L-R filter would work well if the Altecs start playing at 700Hz? Or how much lower to make a first order crossover possible? Or would a second order L-R be necessary? Please advise.
Then I would passively cross the F120s with that Beyma 150/h or a Mundorf AMT.
HF Extension Via Pro Audio Software:
It’s only been several weeks ago when I was surprised to learn that horns generally have limited HF response, sometimes 10kHz, or less. No wonder Gary and Pierre added (active?) EQ to extend the HF response, though wouldn’t a pair of these work better http://www.fostexspeaker.de/tweeter/horn/t900a.pdf , albeit their less than stellar off-axis response?
I was reading about the invention and widespread use ever since of phase plugs https://en.wikipedia.org/wiki/Phase_plug for both compression drivers and woofers to extend HF response. With the obvious question being why aren’t all compression drivers equipped with built in phase plugs (assuming they can’t be added without problems), what are the tradeoffs to using them? For example, how would Gary’s Radian 745NeoBe drivers perform if they had phase plugs?
Assuming phase plugs cannot be added to a compression driver (LOL?!?) to Gary’s Radians-and I certainly wouldn’t want to use anything other than the 425 horns-since my sources are all digital, would professional quality software-based EQ be sufficiently transparent to achieve hi-fi sound quality? I typically rip my favorite CD tracks to uncompressed wav files. And because I like to sometimes apply special effects (i.e. small, lossless changes like pitch reduction and/or time expansion) during playbacks, I use Samplitude Pro X-2 Suite
Editing > Functions > Samplitude > Audio Production > MAGIX Pro
Sadly, throughout my life, most of my music listening has been via (Sennheiser) headphones, so I rarely found the need to use EQ. But as shown here Mastering > Functions > Samplitude > Audio Production > MAGIX Pro , Samplitude has impressive EQ functionality http://savedbytechnology.com/catalog/news_photos/107-magix_S_S_11_05.jpg
I know there are many different types of equalizers; some are best for one mixing and/or mastering and some are better for room and speaker EQ. However, would you care to confirm if the one (s) that come with Samplitude would work well to extend the HF range of the 425 horns/Radian drivers beyond 10kHz, without penalties?
On the other hand, there are countless very high quality software EQ plug-ins that are fully compatible with Samplitude, if Samplitude’s own EQ wasn’t best suited for this application.
I believe that you can download Samplitude for a fully functional 30 day trial MAGIX Download If you have some well recorded and digitized music on CD or download and a good sounding standalone DAC, why not rip the CD to an uncompressed wav file and play it via Samplitude through your DAC into your amp(s) while tweaking its EQ? Samplitude might even allow real time EQ tweaks if your CPU and memory are up to it.
If yes, were you able to extend the highs of your horns without coloration or loss of spatial integrity?
IF Samplitude’s EQ did the trick for your drivers, then shouldn’t I be able to get a beautiful tweeter-less extended HF response from the Radian drivers-and reap the superior off axis response of horns?
Crossovers:
My knowledge of crossover design is now hardly substantial. Presently, I don’t know a whole lot more than that a coil in series with the woofer blocks HF from reach it and that a cap does the same with LF for a tweeter. But my First Watt B4 two-way active crossover (which I hope was not an expensive purchasing mistake) can generate all degrees of L-R and S-K filters. And that the steeper a two-way crossover’s slope the more dissimilar will be the high pass and low pass phase response. Did Gary use a 12db slope and the Fc was 900Hz, but not a L-R type (?). I do know that his Altec 416s start playing at 700Hz.
If yes, then would an L-R 12db passive crossover @ 900Hz between the HA425 horns/Radian drivers and the Altec 416s sound very good? And active bi-amping using the same crossover sound as good or better?
Room Size:
This is very possibly a serious issue that I now regret not giving much though to, probably because I was more concerned with planning for the future rather than the present. Within the next two years, perhaps sooner, I will be moving to a two bedroom co-op or condo. I expect my living room to be uncluttered but only about 18 ft x 14. Would that room will be big enough for the 425 horns, the Altec 416s (in Gary’s 3 cu ft sealed cabinets) and a pair of 12” sealed Rythmik subs?
However, the present room I have available for them is a cluttered 17 ft x 9 room.
Am I expecting the impossible for that system to properly work in that room?
Furthermore, even if the smaller room wouldn’t sound too bad, will later moving it to somewhat larger room require crossover changes?
Maybe I just don’t have the listening room space for horns now and/or in the future
I guess the best question to ask might be how small a room would the 425s work in without acoustical and/or crossover problems?
Speaker System Alternatives:
If you think an 18 ft x 14 and/or a somewhat cluttered 17 ft x 9 room are too small for the AH425s, then perhaps the best alternative to horns-and what might be the best answer to their HF deficiencies at the same time-are AMTs. That is, to use a particular model AMT with a direct radiator midrange driver and the Altec 416 midwoofers below that. Here Beyma TPL 150 AMT - the ultimate driver whats the catch? Tuyen is very impressed with everything about them except the vertical dispersion, which doesn’t match that of the compression tweeters he uses (Goto?).
As can be seen in the graph it’s a pretty sad story after 1kHz.
http://www.beyma.com/getpdf.php?pid=TPL-150/H ; all other Beyma AMTs had worse vertical dispersion.
I would also like to know what Tuyen may think of Mundorf’s AMTs.
Someone at that same hifiwigwam thread said that Mundorf spend $200K on R & D for AMTs, and the impressive distortion specs on most of their models seem to bare this out http://mundorf.com/PDF/Mundorf_hifiAMT_Catalog.pdf
However, vertical and/or horizontal dispersion looks no better than the Beyma 150/h, or perhaps that Fostex T900 tweeter
http://www.fostexspeaker.de/tweeter/horn/t900a.pdf How do they expect us to be able to get up and dance with our girlfriends with vertical responses like these?
But might AMTs’ off-axis HF response be extended without penalty via EQ software while playing digital source material?
And for a driver between the AMT and the Altec 416, has anyone any listening experience with this Fostex alnico midrange
https://www.madisoundspeakerstore.c...fostex-f120a-5-full-range-with-alnico-magnet/ ? Or please explain why another midrange driver might be much better for above the Altec 416s.
I thought of bi-amping the F120’s with Gary’s Altec 416 midwoofers. Do you think that a first order 6db L-R filter would work well if the Altecs start playing at 700Hz? Or how much lower to make a first order crossover possible? Or would a second order L-R be necessary? Please advise.
Then I would passively cross the F120s with that Beyma 150/h or a Mundorf AMT.
Last edited:
Earl, there is evidence out there. I found plenty when I went looking on Google .
Psycho-acoustic, yes. Puffery, no.
The results I got from the phase shuffler were exactly as predicted. And yes, it is somewhat subtle. That's the point.
Psycho-acoustic, yes. Puffery, no.
The results I got from the phase shuffler were exactly as predicted. And yes, it is somewhat subtle. That's the point.
Not if you play the short wall and put them in the corners. 9ft to the outside edges isn't much, but it's OK. My Altecs were on 9 or 10 ft centers, IIRC. I sat about 12 ft back. A larger room would have been nice, but they did work OK in that space.
Clutter is usually a good thing. 🙂
I'll have to dig. It's been awhile. Somewhere I also have the shuffler file I made. I'll post it here for those who can use convolution in the playback stream.
Example please, I just don't get what all the discussion is about.
A shuffler will de-correlate the two channels and yes will reduce the lobbing, but it will also destabilize any center image in the recording. Sounds like a tradeoff - one that we should expect to be "preferred" by the implementer.
Last edited:
The implementation I used did not destabilize. That could be system dependent, I don't know.
Some Googling found the paper I used to understand and eliminate the effect. If you look at section 3.4, you'll see the impulse file I made and sent to the convolver. It works.
http://www.sfxmachine.com/docs/FixingThePhantomCenter.pdf
Some Googling found the paper I used to understand and eliminate the effect. If you look at section 3.4, you'll see the impulse file I made and sent to the convolver. It works.
http://www.sfxmachine.com/docs/FixingThePhantomCenter.pdf
Though not the same effect(s) under discussion..
Remember the problems that Markus was having with a stable center phantom with your speakers "crossed in front"? (..at least not until he was tired and a bit intoxicated.)
Best to keep in mind that not everyone "processes" the same.
..I'm pretty sure forum member Elias would have a somewhat different reaction as well.