It can't of course, but sometimes it sounds like it can.
Depth is the hardest illusion for me to achieve in a small room. The early reflections just tell me that there isn't a lot of space. Width seems easier. With good room treatment and good speakers, some depth can be achieved in a small room. Many people found my Altec setup gave great depth, I rarely did. Sometimes, on some recordings..
..In any case, a large sound IS difficult in a small room. We can make the room sound larger with a lot of work, but it doesn't come easily.
This isn't that difficult, nor does it take that much imagination.
When it comes to small rooms AND STEREO REPRODUCTION it's generally NOT about reflections - it's about bounding conditions (..same "coin" different "side").
You can get good depth from as little as 3 feet (speaker distance to wall) - which is about 400 Hz.
It can substantially improve if the wall behind the loudspeakers is 7.5+ feet in distance (below 150 Hz).
...
From there the next "stop" is the loudspeaker itself - its "low-level" detail and how little mechanical resistance there is at extremely low excursions. This is CRUCIAL. If your speaker isn't designed for this then depth will always be truncated (..and even then it's a "continuum" of success/failure).
...
Then look to positioning (loudspeakers relative to the listener and each other).
"Depth" is a monophonic emphasis (person dependent) of stereo reproduction, with just the right amount of cross-feed (left speaker to right ear and right speaker to left ear) - AND it's dependent on freq..
There also seems to be a "limit" with respect to the distance between loudspeakers (speaker "spread") for most recordings. At about 7 feet and using most loudspeakers (..though diffraction dependent), that you don't get any significant enhancement in depth when the loudspeakers are closer together than that (..unless you are doing some sort of post-processing effect). Maybe it's a condition of the way most music is engineered, I don't know. But it's something I've noted before (..along with many others).
Of course as you move loudspeakers closer together you are generating lower pressure to the sides (approaching +/- 90 degrees) of your head. This reduces the perception of width and that sense of "envelopment" (reproduced envelopment) - usually negating any sounds to the sides and rear of your head (recording dependent). Even if everything else is near-perfect - you'll still get that "standing at the threshold/doorway" presentation. Still, it's almost exclusively about intensity relative to direct sound at higher freq.s. Realistically however as you increase the near reflections (left speaker to left wall and right speaker to right wall), the result is NOT a "hall of mirrors" effect that so many presume - but rather an increase in intensity at a wider angle (relative to the listener's head) which aids in expanding your level of reproduced envelopment. (..this is pretty much THE conclusion for Olive and Toole as well as other researchers. Note however that what's not fully covered in the research is how the intensity is achieved relative to those near walls.)
Unfortunately there is also a usual "price" to be paid with a wider speaker spread relative to the listener - and that's loudspeaker diffraction lowering that "pin-point" "image" positioning. Depending on the polar response of the loudspeaker, "toe-in" the loudspeakers (pointed toward the listener) can "fix" this to some degree but it also lowers the intensity at higher freq.s because it is lowering the intensity of the reflections.
...
And at the "end" of it all is the recording and the performance and venue (real or simulated).
Just about everything has some amount of reverb - and that alone gives some impression of "space".
Synthetically you can take it to an extreme like some of the stuff from Yello (..though they also have many very "dry" tracks as well):
https://www.youtube.com/watch?v=JBFywXzs9F4
Just imagine what "scrubbing-out" the soundstage would be like with this recording and seemingly replacing it with what your own pitiful room dimensions would allow. Like @ss, and in no real respect "accurate".
Last edited:
Here is a setup well designed for depth, and reasonably well designed for width (..though the loudspeakers are to close to their respective side walls, and so the owner is using absorption and/or diffusion) - scroll down for "third visit":
adnans.htm
Note the particular response by the author:
"For me the most striking point was the exceptional depth of the sound stage."
..room provided, something similar could have been done for less than what this person paid for loudspeaker cables.
adnans.htm
Note the particular response by the author:
"For me the most striking point was the exceptional depth of the sound stage."
..room provided, something similar could have been done for less than what this person paid for loudspeaker cables.
Last edited:
The 5687's were probably already well-used when I installed them. They tested okay, so I plugged them in (after some rewiring--they replaced the original ECC99's, which had been in use for more than a decade).
I wouldn't expect anything other than typical tube life in this amp. The tubes are run significantly harder than those in a tuner.
Gary Dahl
Circle surround actually works.
Circle Surround Technology
I don't use the front steering or center channel anymore.
When it's setup right and you go back to just 2 stereo speakers it shows just how fake stereo really sounds. I don't know anybody with normal hearing that does not hear 360 degree sound. discrete and matrixed dolby also sounds fake, The best realism comes with the circle surround speakers being omni-directional and in my room a pair at the ceiling and out 2 feet from the back wall and another pair at the ceiling to the left and right at the walls beside you. When it's all balanced in tone and with the correct levels there is no localization of any of the speakers. Very natural soundfield with no limits or holes in the field.
Circle Surround Technology
I don't use the front steering or center channel anymore.
When it's setup right and you go back to just 2 stereo speakers it shows just how fake stereo really sounds. I don't know anybody with normal hearing that does not hear 360 degree sound. discrete and matrixed dolby also sounds fake, The best realism comes with the circle surround speakers being omni-directional and in my room a pair at the ceiling and out 2 feet from the back wall and another pair at the ceiling to the left and right at the walls beside you. When it's all balanced in tone and with the correct levels there is no localization of any of the speakers. Very natural soundfield with no limits or holes in the field.
You can build the speakers flush into the wall and it doesn't place any particular limit on depth perception that they can deliver.
tnargas,
I think the whole concept Scott is talking about is to have those delayed reflections to give the appearance of depth. With a flush mounted or soffit mounted speaker you wouldn't have any longer path lengths to give that illusion. How well that can work I don't know but I understand what he is attempting to do.
I think the whole concept Scott is talking about is to have those delayed reflections to give the appearance of depth. With a flush mounted or soffit mounted speaker you wouldn't have any longer path lengths to give that illusion. How well that can work I don't know but I understand what he is attempting to do.
Sure........gravity and orientation are key here. Take a composite material like Nidacore with its honeycomb structure and suspend it in large panels from a ceiling with equally spaced suspension components with properly placed motor(s). Such a large surface with even limited excursion can produce extreme amounts of LF. The large panels won't compromise room egress or aesthetics and furnishing. An installation such as would certainly take some time to install but certainly well within the realm of most patient and thorough DIYers. Where Nidacore excels is the ability to selectively control response by damping individual core cells as needed. R&D would be needed but again, certainly possible with a reliable measurement rig and some experimentation. Envision 80 square ft or so of flat panel surface area with excursion limits of 1" or so.
But agreed.....still to date you're LF solution offers the most practicality and portability.
I've found it just the opposite.
Even with some of the better in-wall designs there is depth, just not a lot of it regardless of the recording. (..and by depth, I mean a LOT of depth depending on the recording.)
In fact, it's one of the reasons why I don't mind it for Home Theater use - because really expansive depth behind the screen doesn't "translate" well to what's occurring on the screen (even for 3D), and also because with multi-channel you get that side-channel reproduction "envelopment". (..great for concerts as well because of that multi-channel goodness - assuming it has that or you have a really good processor.)
Note I've heard quite a few in-wall installations, from several studio to multiple HT installations. It's probably one of the reasons that some of the pricier studios use "audiophile" loudspeakers in addition to their in-wall loudspeakers - finer adjustment of "images" in the depth plane and environment.
Last edited:
Mayehm,
Having worked on a rectangular flat panel speaker I can tell you there are many more issues than you would want to think about. It isn't as simple as you think to make a non terminated edge surface function, the bending modes are very complex. I have done plenty of work with composites and core materials and i can't imagine that a honeycomb material would truly react the way you would think. Rohacell types of foam core would be more homogenous but like I said it isn't easy to just put a motor on the back and move the panel. You actually get many sets of vibrational modes that are in anti-phase working against you. Not a simple problem and as Earl has said if it was really a viable solution you would see this. It has been tried by companies with a lot more money and effort than you can imagine, the Japanese majors spent lots of money trying to make this work. Everyone seems to come to the same conclusion in the end, dynamic drivers work much better, are higher efficiency and are well understood. The concept sounds good but the actual implementation doesn't match the expectations.
Having worked on a rectangular flat panel speaker I can tell you there are many more issues than you would want to think about. It isn't as simple as you think to make a non terminated edge surface function, the bending modes are very complex. I have done plenty of work with composites and core materials and i can't imagine that a honeycomb material would truly react the way you would think. Rohacell types of foam core would be more homogenous but like I said it isn't easy to just put a motor on the back and move the panel. You actually get many sets of vibrational modes that are in anti-phase working against you. Not a simple problem and as Earl has said if it was really a viable solution you would see this. It has been tried by companies with a lot more money and effort than you can imagine, the Japanese majors spent lots of money trying to make this work. Everyone seems to come to the same conclusion in the end, dynamic drivers work much better, are higher efficiency and are well understood. The concept sounds good but the actual implementation doesn't match the expectations.
Please show me where I said it is always possible, or stop misquoting me.
You know better, stop making stuff up.
That's just you. And since you already know it is a recording, you are hardly without bias. This report is just anecdotal, not evidence.
My car does not go 140 miles per hour, despite the markings on the speedometer. From this I should conclude that it's not possible for any car to go 140 miles an hour?
Nope.
(..but sometimes I'm "clear as mud".)The reflections just don't have the timing/delay to be relevant for most small room conditions (..though not all), nor do you usually have the sort of "discreet" reflection (and intensity near the average or above it) that would compete with direct sound - unless you have unusual hearing.
On of the things it does have going for it can be a near absence of diffraction - which can help with depth reproduction to some extent, but really helps with other aspects of reproduction - particularly lateral image placement.
Now, if the reflection is a bit more "focused" or "discreet" (greater intensity on one wall in one area), AND the direct sound is LOWER in level - then yes, at that point the reflection starts dominating. This is particularly true if the pressure gradient is physically blocked to some extent and diffracting. (..Note: this is also covered in Toole's book, though again - some listener's "buck" this trend and will focus on a lower - pass-band that doesn't have this characteristic.)
Last edited:
I should also note that this "bounding condition" shows up in several respects.
Horns/waveguides - assuming the diffraction is reasonably well handled, usually provides greater depth of field when it's NOT physically connected to the typical loudspeaker cabinet below it. Lift it up a few inches off of the cabinet and correct for diffraction effects (and other linearity problems) - almost always results in some increased depth (sometimes a lot).
Floor-standing vs. stand-mounting. Same basic thing - even after correcting for any potential floor-bounce suck-out that the stand-mounted design might have vs. the floor-standing design. That extra length down the baffle and particularly connecting to the floor does something to foreshorten depth of field.
Horns/waveguides - assuming the diffraction is reasonably well handled, usually provides greater depth of field when it's NOT physically connected to the typical loudspeaker cabinet below it. Lift it up a few inches off of the cabinet and correct for diffraction effects (and other linearity problems) - almost always results in some increased depth (sometimes a lot).
Floor-standing vs. stand-mounting. Same basic thing - even after correcting for any potential floor-bounce suck-out that the stand-mounted design might have vs. the floor-standing design. That extra length down the baffle and particularly connecting to the floor does something to foreshorten depth of field.
Last edited:
No prob.s.
The book itself is "OK", but really there is a lot of "padding", whereas most of the "substance" is located within about 3-4 chapters. (..and you can often get a fair amount of it with a google search that "turns-up" specific page numbers from various preview/"quick looks".)
The underlying research really isn't provided in much detail - but as long as you read those significant chapters carefully several times, the conclusion and the reason's behind them are there.
(..The surreal moment for me reading it was when they starting listing other older sources - sources I remember reading in college (..Baron, Ando, etc..).)
The book itself is "OK", but really there is a lot of "padding", whereas most of the "substance" is located within about 3-4 chapters. (..and you can often get a fair amount of it with a google search that "turns-up" specific page numbers from various preview/"quick looks".)
The underlying research really isn't provided in much detail - but as long as you read those significant chapters carefully several times, the conclusion and the reason's behind them are there.
(..The surreal moment for me reading it was when they starting listing other older sources - sources I remember reading in college (..Baron, Ando, etc..).)
Like the old-timers, I usually run tubes between 70~80% of rated plate dissipation. 100% will definitely shorten the tube life, and 50% discards a lot of headroom or runs the tube in a nonlinear low-current mode. 70~80% is a good compromise between tube life (typically 3000~5000 hours) and performance.
There aren't a lot of applications where tubes can be run at 50% or less, unless distortion is an intentional goal (like effects boxes in studios).
Within that 70~80% window, the tradeoff is between desired headroom and getting the current out of the nonlinear low-current region. I think it's bad practice to let an amplifier all clip at once, so I try to give the driver tube a peak voltage swing at least 30% greater than the onset of output-tube clipping.
If the coupling is direct or through a transformer, the driver section can then drive the power tube a short way into Class A2 (positive grid drive), as well as providing immediate recovery from overload (no RC recharge time). The entire amplifier acts as a progressive limiter and will provide 3 dB (or more) of subjective headroom above the steady-state power. One advantage of a zero-feedback amplifier is the overload region can be shaped to provide immediate recovery (no device sticking to the rails) and an overall progressive characteristic.
The measures that provide additional headroom also provide better low-level performance. If the driver has less (high-order) distortion than the output section, that reduces IM distortion that arises from distortion components in one stage being re-modulated in the following stage.
This is different than a typical Golden Age amplifier, where 20 dB of feedback is pretty much used as a cure-all for the nonlinearities of the circuit and problems in the driver section. Granted, pentodes like the EL34 and KT88 only need 35V peak grid drive versus the 70~85V peak grid drive required by a 300B, but the Golden Age amplifiers had trouble doing even that, with the Dynaco Stereo 70 as a near-worst-case example. Open loop, the ST70 only has a bandwidth of 50 Hz to 7 kHz. The published specs come from feedback linearization. Of all the Golden Age amplifiers, the Mullard and Acrosound circuits probably have the best inherent linearity.
Although the Amity is a good amplifier, the driver stage is asked to do an awful lot. The 5687/7044/7119 is close to the limit of its capabilities driving a 300B, which asks three things of a driver:
(a) voltage swing of 85V peak, with an additional 30V in reserve.
(b) better linearity than the 300B, which is a tall order, since the 300B is one of the lowest-distortion amplifying devices ever made.
(c) the ability to deliver that very linear peak voltage into a 60~80 pF capacitive load that also becomes nonlinear as the power tube approaches zero-bias condition.
Last edited:
People who follow vacuum-tube amplifiers will note similarities with the "Beyond ..." project. It re-visits concepts from the Thirties and Fifties, and looks at them in light of modern measurement capabilities and modern parts (better interstage transformers and power supplies, and beryllium diaphragms). In principle, similar to an Altec Model 19, but the LeCleac'h profile and Bjorn Kolbrek's computer simulations weren't available in the early Seventies.
+1.
In terms of Woofer parameters:
"little mechanical resistance" = low Rms with respect to the moving mass, i.e. low Rms/Mms, which equates to low mechanical damping (Fs/Qms);
"extremely low excursions" implies/requires a large radiating surface (Sd).
Marco