I'm not aware of any recordings that have an accurate representation of depth captured on them. Some recordings do give a good presentation of the acoustic size of the venue, but it's hard to differentiate between width, height, distance behind the listener, distance from the listener to the source of music and the distance from them to the wall behind them. I don't believe this is actually captured on any recordings. My dipolar ribbon speakers seemed to present more depth than my soffit mounted monopole planar magnetics and Unity horns. I think the dipoles are faking this presentation of depth and it isn't actually present on the recording. With the soffit Unities some close mic'd vocals are right in your face, easily within arms reach. You expect to feel their breath when your eyes are closed. This is almost spooky at times. The sensation is not nearly as strong with dipoles.
The presentation from the soffit mounted Unity horns is almost headphone dry. If a recording is dry the system will reproduce it as dry. If there is natural acoustic reverb on the recording you'll hear it. If the reverb is electronic, you can usually pick that too.
The presentation from the soffit mounted Unity horns is almost headphone dry. If a recording is dry the system will reproduce it as dry. If there is natural acoustic reverb on the recording you'll hear it. If the reverb is electronic, you can usually pick that too.
I had wide speakers with rounded sides, about 60cm/23" wide and 20cm of depth. There was an illusion of depth when they stayed away of the walls. However, when I placed them so that they're touching the front wall their backs, the illusion completely disappeared. It was like the sound is coming from the wall itself.
Last edited:
This is really about psychoacoustics--and not so much sound reproduction accuracy in stereo (as you imply). I believe there is a good resource on this subject: Sound Reproduction: The Acoustics and Psychoacoustics of Loudspeakers and Rooms by Floyd Toole, who compiled an excellent but a bit lengthy tutorial on the subject for the hi-fi enthusiasts. I assume that you have this book. Have you read it recently? Suppression of early reflections and very good directivity control is necessary--and attention to a wall-to-wall sound field image, very unlike most hi-fi setups that support the head-in-a-vise small sweet spot for hi-fi.
One of the things that well-done multiple entry horns (MEHs) do--that you alluded to above--is that they can do this job very well. [I also recommend you also consider in-room loudspeaker directivity well below 500 Hz to help accomplish this sense of space in the lower midrange, which you didn't really address, above.] I've personally found that means big sound directors, or conversely a lot very broadband in-room damping, the latter of which I really don't like--like being in an anechoic chamber, which is terrible for the equilibrium, I've found. This means bringing the early decay times (EDT curve) down very quickly relative to the T20 or T30 curves.
But the discussion above is basically only talking about loudspeakers, and really not the whole enchilada (the room and loudspeaker array and recording schemas). Stereo has flaws (as Toole advises), and in my experience they're insurmountable relative to other sound field reproduction schemas. If you constrain the answers to stereo-only, you're probably right in your opening sentence: I haven't really heard any of those either--in stereo.
Are the recordings that do create a relatively convincing sense of space accurate? Well, I'd first respond by asking you a question: when was the last time you went to listen to a symphony orchestra performance in a very good performing hall (acoustically) and listened carefully to sound localization or the lack thereof, and the sense of depth. This is usually a different type of sense of depth than home hi-fi enthusiasts are trying to create, I've found. I also found that my recollections of sense of sound space were actually distorted by most sound recordings, and had to recalibrate my ears...and not so much my sound reproduction setup.
There are recordings that get very close indeed but you must leave stereo behind. There are several such music-only recordings of non-amplified instrumentation that do present a realistic image when you shift from the stereo-only layer to multichannel during playback (it's breath-taking). But in order to really hear this, I've found that your listening room and loudspeakers must be in good shape psychoacoustically to achieve it.
Chris
Last edited:
I have; this one. Still one of the best couple of systems I've heard. I never heard the original boxes so I can't compare to the final system.
Edit; posted this before I read the entire thread.
Thanks Brett 
That system has gone through five iterations over the years I've had it. V1 and V2 were in the floor standing boxes. The second of which sounded pretty good and is probably similar to what most Lambda Unity horn buyers were getting out of them once their crossover bugs were fixed. You heard V3 which used the DCX2496 and a pair of Aussie Monitor 1K2 amplifiers. I've gone through some refinements since then including an upgrade to a PLX3602 on bass. My target is no longer flat, but slopes down gradually between 100Hz and 10KHz by about 5dB. CD horns sound better that way. This matches the power response tilt in a well designed conventional loudspeaker and makes the system more forgiving of poorly recorded material.
The current iteration uses Dirac Live to finesse the phase and frequency response across the listening position. Turning this on and off is pretty subtle but the measurements look superb. Amplification is now Hypex NC502 and NC122 on bass and the horn respectively. This amplifier change didn't effect the sound of the system in any way but has dropped the idle dissipation of the system to 10% of its previous value. We noticed this on our power bill!
There's plenty of scope to make those old Unity horns sound REALLY good!
That system has gone through five iterations over the years I've had it. V1 and V2 were in the floor standing boxes. The second of which sounded pretty good and is probably similar to what most Lambda Unity horn buyers were getting out of them once their crossover bugs were fixed. You heard V3 which used the DCX2496 and a pair of Aussie Monitor 1K2 amplifiers. I've gone through some refinements since then including an upgrade to a PLX3602 on bass. My target is no longer flat, but slopes down gradually between 100Hz and 10KHz by about 5dB. CD horns sound better that way. This matches the power response tilt in a well designed conventional loudspeaker and makes the system more forgiving of poorly recorded material.
The current iteration uses Dirac Live to finesse the phase and frequency response across the listening position. Turning this on and off is pretty subtle but the measurements look superb. Amplification is now Hypex NC502 and NC122 on bass and the horn respectively. This amplifier change didn't effect the sound of the system in any way but has dropped the idle dissipation of the system to 10% of its previous value. We noticed this on our power bill!
There's plenty of scope to make those old Unity horns sound REALLY good!
IIRC it was Floyd Toole who discusses room acoustics in relation to early reflections and late reflections.
Early reflections can create the effect of a wider/deeper sound stage at the trade off of pinpoint focus.
Late reflections can typically add ambiance without causing too much negative impact on focus and imaging.
My memory's a bit hazy and I don't have the book on hand to refer to, but there has been a lot of research into this.
William's soffit mounted unity's would reduce (or eliminate?) early reflections in a room.
Gedde's goes one step further and removes the very early reflections (HoM's) from the speaker itself.
Early reflections can create the effect of a wider/deeper sound stage at the trade off of pinpoint focus.
Late reflections can typically add ambiance without causing too much negative impact on focus and imaging.
My memory's a bit hazy and I don't have the book on hand to refer to, but there has been a lot of research into this.
William's soffit mounted unity's would reduce (or eliminate?) early reflections in a room.
Gedde's goes one step further and removes the very early reflections (HoM's) from the speaker itself.
Exactly. The first reflection of any meaningful amplitude is 30mS after the direct sound from the speakers.
Direct radiators are unable to generate such a clean sound field in a room. I've designed and built a very large number of superb direct radiator, ribbon, planar magnetic, conventional and Unity/Synergy horn systems over the last 30 Years and for the ultimate in reproduction I keep coming back to the Unity/Synergy horns. They have a strong suite of abilities that a conventional loudspeaker just can't touch.
I would argue the two systems in the original post did not measure the same, and that the conventional speaker was actually a higher quality design than the horn loaded one. Had the same measure of design expertise been used on both systems, there is a good chance the horn based system would have come out the preferred system.
Last edited:
^^^This^^^
I've got my speakers on the wall and I can switch the perceived depth on and off by closing or opening my eyes while listening.
We are a predominantly visual species and the brain will ALWAYS side with the eyes. If we do not see space behind the speakers we will NOT hear any.
1) I've never understood the logic that blames the loudspeaker for being able to hear bad recordings better. That's like shooting the messenger. I expect that the better the loudspeakers/room acoustics, the more revealing that they will be. When you consider the opposite case--poor loudspeakers in a poor listening space--I find that I can't tell good from bad recordings, but everything sounds pretty lousy--sort of like Bose 901s that throw around almost all their energy into early reflections. That's what direct-radiating drivers really do (depending on bandwidth and cone diameter).
2) The SH-50 has so narrow a horizontal dispersion that there are relatively no early horizontal in-room reflections that are affecting the direct arrivals within what I call the "early Haas interval" (about 3-5 ms) which I find controls a sense of soundstage image. The SH-50 compared to K-402-based MEH and Klipsch Jubilees (both using the K-402 horn with 90x60 coverage), presents a very narrow and discrete soundstage image, while the K-402 image is huge but also revealing of poor recordings--but not nearly as markedly as the SH-50's "spotlight" image quality.
3) I really hate to say this, but I haven't been able to find any good recordings in the genres of EDM and punk rock. Candidly, I wouldn't use these genres as a reference for much other than perhaps less expensive car loudspeakers. The only recordings that have the quality to incite chills down the spine in terms of realism...are the ones that you say that you don't ever listen to: jazz, orchestral, and classical. These are the genres that have much higher average dynamic range, clarity, soundstage, and minute detail that sets great loudspeaker/room sound reproduction apart from the fray. For instance, data from a JAES paper on the subject:
4) I've found that there clearly is such a thing as "too narrow a coverage" and the SH-50 is in that camp in all but the smallest of listening rooms and without acoustic treatments.
5) So your hunch is correct in my experience: much wider directivity "lights up" the room with consistent polars and wide-enough coverage. I've found that more-or-less magic horizontal coverage level to be about 90 degrees, or about two SH-50s side-by-side.
Chris
Last edited:
I have not read all of this thread, so I'm mistaken, but I guess what he says is the extra ambience can blur the defect of the original recordings. It is probably like a Karaoke echo makes a bad singer better, and I can understand his logic that way, if I'm correct.
And I'm not against Patrick. I have had both many different horns and domes in many different rooms, and I think I do understand the strength and weakness of each topology.
There are many different perspctive, but in Patrick's context, my experience does not contradict his theory. Dome speakers are good for grabbing the whole picture of the recording, and horns tend to be more microscopic in general. This may explain why near field monitors in the recording studios are mostly dome, and large monitors are horn, and mastering engineers usually prefer dome. I personally think it's not only because of the directivity, but the transient also.
My own hypothesis about overall direct radiator preference is the most mastering engineers finalize the source balance to match "average home listening environment", which I guess they expect the direct radiator speakers in a rather wet room. So if you playback those materials with horn in a controlled environment, they should be drier than mastering engineers expected. But it does not mean horn is inferior to the direct radiator.
Well, one thing I'm against Patrick is that it seems like he wants to try to conclude one is better than the other based on one aspect. I really wish my life would be that easy... Maybe I'm just stupid.
There are many different perspctive, but in Patrick's context, my experience does not contradict his theory. Dome speakers are good for grabbing the whole picture of the recording, and horns tend to be more microscopic in general. This may explain why near field monitors in the recording studios are mostly dome, and large monitors are horn, and mastering engineers usually prefer dome. I personally think it's not only because of the directivity, but the transient also.
My own hypothesis about overall direct radiator preference is the most mastering engineers finalize the source balance to match "average home listening environment", which I guess they expect the direct radiator speakers in a rather wet room. So if you playback those materials with horn in a controlled environment, they should be drier than mastering engineers expected. But it does not mean horn is inferior to the direct radiator.
Well, one thing I'm against Patrick is that it seems like he wants to try to conclude one is better than the other based on one aspect. I really wish my life would be that easy... Maybe I'm just stupid.
Last edited:
Snipped only for brevity.
I'm in a V2.0 rebuild at the moment as for years they were in a rough unfinished mdf enclosure, a bit like your V2 dimensionally but WUW with the AE TD15S. As I'm in an apartment and can't really mount them like yours, the new design is going to be a wide and shallow design with a 30* baffle angle back from just outside the Unity's width. They will be placed hard in the corners so as close to soffit mounted as possible as the room is only 3.2m wide.
Amps will be the same as before; M.E.'s for the HF and MF and 2 Yamaha P7000S for the AEs, one amp channel/driver. This time however, the xover will not be the modified DCX but a DEQX as I scored one for an awesome price. They went to about 40Hz flat in room in the previous iteration and have enough power to EQ flat to 20Hz, but I'll be adding a miniSHARC to do the subs.
Likewise, I'd EQ'd the top end approximately the same as yours.
I also have an AE, ATC, TPL150 build in the planning stage (all drivers on hand) built like the SF Stradivari that I'll be able to roll in front of the Unitys when I want a bit of a change.
Hi William,
I'm in a V2.0 rebuild at the moment as for years they were in a rough unfinished mdf enclosure, a bit like your V2 dimensionally but WUW with the AE TD15S. As I'm in an apartment and can't really mount them like yours, the new design is going to be a wide and shallow design with a 30* baffle angle back from just outside the Unity's width. They will be placed hard in the corners so as close to soffit mounted as possible as the room is only 3.2m wide.
Amps will be the same as before; M.E.'s for the HF and MF and 2 Yamaha P7000S for the AEs, one amp channel/driver. This time however, the xover will not be the modified DCX but a DEQX as I scored one for an awesome price. They went to about 40Hz flat in room in the previous iteration and have enough power to EQ flat to 20Hz, but I'll be adding a miniSHARC to do the subs.
Likewise, I'd EQ'd the top end approximately the same as yours.
I also have an AE, ATC, TPL150 build in the planning stage (all drivers on hand) built like the SF Stradivari that I'll be able to roll in front of the Unitys when I want a bit of a change.
G'day again Brett
In an apartment I'd use a dipole bass section under the horns crossing at about 80Hz to a near field dipolar subwoofer beside or behind the LP.
I'll be in the same position in a few years time and have given the problem considerable thought.
If you're interested in exploring this idea we'll take it offline because it's OT on this thread.
In an apartment I'd use a dipole bass section under the horns crossing at about 80Hz to a near field dipolar subwoofer beside or behind the LP.
I'll be in the same position in a few years time and have given the problem considerable thought.
If you're interested in exploring this idea we'll take it offline because it's OT on this thread.