Yes, as I recall, the 800 Hz ones. Pretty sure at one point I took them out of the box, set them on top of the cabinets. Wish I got to try the OBs like the one you've shown, but that was circa 1985 and I hadn't heard of such a thing.
So those old preamps from the '60 with their "Stereo Blend" (to mono) controls were not so off, then?
Haha Yes, not so off! I still like the old loudness tone controls too.
Audio took a serious downturn when all the purist crap about no-eq and such came out...
Hello Holger. Welcome to the thread. Thank you for posting information about your grand system.
This thread is about the tonal balance of the center phantom image being different from the tonal balance of either the left or the right speaker alone. Most people do not hear this effect because their speakers or room will mask it. But with good acoustics and a tight stereo system, the effect can be heard as the phantom center sounding "darker" than either the left or right speaker alone.
Is this something that you have noticed with your system when you are playing in 2 channel stereo, without the center speaker?
Below I have attached two zip files, inside of which are FLAC files. The Basic file is spoken voice on both channels, then Left, Right and both channels again. So the voice should be Center, Left, Right, Center. Do you hear a difference in tonality between any of those positions? The other file has been phase shuffled and should sound different when both channels are playing. Do you hear a difference between Center, Left and Right with the shuffled file?
It would also be very interesting know what happens when you play the files with your center speaker active. What does that do? There are some more test files in post number 30 of this thread. You might also try them. Cheers!
Yeah, I didn't know anything about open baffle in 1985 either - well maybe I did. I loved Magnepan/Magneplanar speakers and the Quad ESLs. But I didn't think of those as open baffle or dipole, just as "panels."
Re: tonal balance of the C wrt L & R. One system I have F15s on OBs. I'm not about to buy a 3rd F15 to try some mono in the C - inevitably placed in front of the flat screen. Placed in a stand underneath the screen? Wont be in the same plane as the OBs...
Another system I have 4 MA Pluvia 7 HDs, two per 20" X 20" panel, which throw a nice deep 'stage behind the panel plane, which are about 5' apart, center to center. I cant imagine plonking down another 20" X 20" panel between those two, in the same plane - where will my nice, deep SS go with that plate covering it?
I only bring this up as real-world examples of how one would fit a C in their existing arbitrary setup. Could you have a C speaker with L & R delayed to match C physically pushed back somewhat from the L/R plane? Like C in a Cornu hanging against the back wall?
Another system I have 4 MA Pluvia 7 HDs, two per 20" X 20" panel, which throw a nice deep 'stage behind the panel plane, which are about 5' apart, center to center. I cant imagine plonking down another 20" X 20" panel between those two, in the same plane - where will my nice, deep SS go with that plate covering it?
I only bring this up as real-world examples of how one would fit a C in their existing arbitrary setup. Could you have a C speaker with L & R delayed to match C physically pushed back somewhat from the L/R plane? Like C in a Cornu hanging against the back wall?
If you measure sound with a microphone, even "late" arriving sounds will mix and interfere with earlier ones, and this can create comb filtering in the measured frequency response (e.g. try to measure a loudspeaker in an indoor space without applying any gating to the time record). But for humans, the complete hearing process does not work like a microphone except for early arrivals, so this is in general a rather poor model of what a human would perceive if late arrivals are also included in the data gathered with the mic. As others have pointed out, comb filter can be perceived as tonal shifts, etc. but this is primarily with/between early arrival sounds.
I'd have to agree with Pano. What anyone prefers is up to them, but I prefer to have a lower level of early reflections, even while it does make this comb pattern more obvious in the listening spot. The added clarity was what made me prefer it. That's why I did research for alternatives, like the shuffler and mid/side EQ (among other things like cross talk cancellation). As long as the reflections are still within the Haas limit, it can still work/help out to soften this combing effect. Avoiding/absorbing (very) early reflections (0 to 20 ms) has my preference in listening tests. With additional Haas kicker it gives me a more life like sound that still is engaging and enveloping while keeping the sound clear, less muddied.
Real reflections tend to add "a single taste" in my small room and come in too early... they widen every song that's played back. I tried that, didn't care for it. What I have now hides the real properties of the room and adjusts to the music being played. Big when it needs to be, small and intimate if that's in the recording.
I still use mid/side EQ, combined with an ambience Haas kicker. I tend to (fully) relax in the listening spot and let the music absorb me. I can move my head without any obvious tonal differences. I'm not walking around etc... I just let the music take over and enjoy that. Not a head in vice need, but I don't move that much either. We tend to move a little to observe the room, that's learned behavior. As the tonal balance now stays even, it isn't a problem any longer.
It was hindering before I added these tweaks. I'd notice the balance change with head movement on a well balanced pair of speakers. That would require head in vice to avoid it. It started my quest. Measurements at listening spot with a foam head model showed the differences at that spot, they should be somewhere in my huge thread.
Though I don't do exactly what is suggested here, reflections can give a similar blur effect. So what I do is make sure I get well timed (within Haas limits) reflections that specifically blur this part of the frequency spectrum that suffers from the combing (phantom center). I use my laterally arriving ambience speakers for this. The mains are well balanced left and right. The ambience part is split in left/right and center. The real room queues are reduced by room treatment. Floor and ceiling reflections are negligible with my tweaked arrays.
This didn't happen in one day, I tried many things and tested a lot of different options to arrive at what I liked best. My compromise, My-Fi.
Real reflections tend to add "a single taste" in my small room and come in too early... they widen every song that's played back. I tried that, didn't care for it. What I have now hides the real properties of the room and adjusts to the music being played. Big when it needs to be, small and intimate if that's in the recording.
I still use mid/side EQ, combined with an ambience Haas kicker. I tend to (fully) relax in the listening spot and let the music absorb me. I can move my head without any obvious tonal differences. I'm not walking around etc... I just let the music take over and enjoy that. Not a head in vice need, but I don't move that much either. We tend to move a little to observe the room, that's learned behavior. As the tonal balance now stays even, it isn't a problem any longer.
It was hindering before I added these tweaks. I'd notice the balance change with head movement on a well balanced pair of speakers. That would require head in vice to avoid it. It started my quest. Measurements at listening spot with a foam head model showed the differences at that spot, they should be somewhere in my huge thread.
Though I don't do exactly what is suggested here, reflections can give a similar blur effect. So what I do is make sure I get well timed (within Haas limits) reflections that specifically blur this part of the frequency spectrum that suffers from the combing (phantom center). I use my laterally arriving ambience speakers for this. The mains are well balanced left and right. The ambience part is split in left/right and center. The real room queues are reduced by room treatment. Floor and ceiling reflections are negligible with my tweaked arrays.
This didn't happen in one day, I tried many things and tested a lot of different options to arrive at what I liked best. My compromise, My-Fi.
Mostly from direct sound, is what I've found. I'd have to go back and read the paper to know if the authors said anything about early reflections. I suppose if they are early enough, they would add to the comb filtering.
Basically this was started as a thread about a problem many people have never noticed, and might not even have. I does, however, seem prevalent enough that some research has gone into it for dialog channels.
After all these years and over a thousand posts, the problem has been well identified and some good fixes found for those who want them. The thread has served its original purpose.
This thread should now head off on a close tangent, the phantom image in general, not just the tonal balance. What have you done to improve or maybe even damage the center image? Use a center speaker? Toe in or out? Positioning in the room? Haas reflections or other room tweaks? Let us know!
Closely related to phantom image so I'll ask if any of you have more knowledge on this.Mostly from direct sound, is what I've found. I'd have to go back and read the paper to know if the authors said anything about early reflections. I suppose if they are early enough, they would add to the comb filtering.
Basically this was started as a thread about a problem many people have never noticed, and might not even have. I does, however, seem prevalent enough that some research has gone into it for dialog channels.
After all these years and over a thousand posts, the problem has been well identified and some good fixes found for those who want them. The thread has served its original purpose.
This thread should now head off on a close tangent, the phantom image in general, not just the tonal balance. What have you done to improve or maybe even damage the center image? Use a center speaker? Toe in or out? Positioning in the room? Haas reflections or other room tweaks? Let us know!
I got smooth directivity speakers, toed in, both channels in parallel to single amp so strong phantom image. Perception that I'd like to know more is that if I move back and forth along center normal of stereo listening triangle there is point where the phantom image becomes clear and also what I think as envelopment happens, like stepping into the sound. Take step back and now the phantom is kind of veiled a bit and the surrounding envelopment turns into something that I perceive being in front of me instead of all around. Sensation of stepping out from the good sound.
I think this could be what Griesinger calls proximity and I'd like to know how to manipulate it. I'd like the effect to extend further into room, as now the position where this into sound sensation happens is not apex of stereo triangle, but perhaps 1/3 way into it.
I should say that when it happens I'm about at center axis of each speaker (45deg toe in) but it seems toeing them out to point ears at listening position doesnt change situation much. I also tried if listening triangle position in relation to walls would change this but didnt notice.
Any pointers / tips how to extend the good phantom image and envelopment durther into room, apex of listening triangle, to practical listening spot?
I suspect its mostly directivity thing and relation of boundaries, size of room to size of listening triangle, but could be something else as well?
Closely related to phantom image so I'll ask if any of you have more knowledge on this.
I got smooth directivity speakers, toed in, both channels in parallel to single amp so strong phantom image. Perception that I'd like to know more is that if I move back and forth along center normal of stereo listening triangle there is point where the phantom image becomes clear and also what I think as envelopment happens, like stepping into the sound. Take step back and now the phantom is kind of veiled a bit and the surrounding envelopment turns into something that I perceive being in front of me instead of all around. Sensation of stepping out from the good sound.
I think this could be what Griesinger calls proximity and I'd like to know how to manipulate it. I'd like the effect to extend further into room, as now the position where this into sound sensation happens is not apex of stereo triangle, but perhaps 1/3 way into it.
I should say that when it happens I'm about at center axis of each speaker (45deg toe in) but it seems toeing them out to point ears at listening position doesnt change situation much. I also tried if listening triangle position in relation to walls would change this but didnt notice.
Any pointers / tips how to extend the good phantom image and envelopment durther into room, apex of listening triangle, to practical listening spot?
What does an IR as recorded on the (preferred) listening spot look like? And what does it look like further back? Mostly looking at the first ~20 ms.
I suspect that you're looking for the right mix of direct + indirect sound. Room treatment could manipulate this, is that an option? If you move closer and closer to the speaker, the direct sound will be dominant, moving back the indirect sound rises in SPL compared to direct.
It's more what you can learn from it, for instance looking at the levels of reflections at both positions might tell what you are looking for. Looking at RTA figures might also reveal something. Disect that IR with every tool available to you, you'll learn a lot about your space.
I went a more extreme way and absorbed most of the early reflections. That way I sort of sculpt my own space with bringing in virtual new queues (Haas kicker).
Mostly done because I am in a small room without options to be closer to the speakers.
I went a more extreme way and absorbed most of the early reflections. That way I sort of sculpt my own space with bringing in virtual new queues (Haas kicker).
Mostly done because I am in a small room without options to be closer to the speakers.