Lynn Olson said:
Pretty simple IMHO: You compare a mass-market industry with well-established and recognized set of metrics, parameters and technologies with a "niche" industry that is ultimately (still) based on subjective impressions. Even a cursory look of this > 2k posts thread reveals that even those mastering the fine aspects of speaker measurements / acoustics disagree on different aspects and ultimately bring the final argument: "In my experience", "my ears", "my priorities", etc etc etc.
So, in short: apples and oranges, IMHO.
Rgrds,
Florian
Around these parts at least, I don't think anyone takes them seriously at all. I look at the Stereophile site occaisionally and find the measurements section of the reviews to be the only parts of them with value.Lynn Olson said:
Anyhow Lynn, I have a question re the latest version of your design you posted in post 2120. In the diagram you mention using cotton waste in pillows placed on the right and left sides of the rear of the B and MB drivers. Earlier in the thread you mentioned Gary P enclosing the rear of the drivers completely in a similar manner IIRC. Is the aim to control the backwave in a sort of 'soft, leaky box' (for want of a better term)?
Hi
Earl, I can't comment on the quality of codecs as I wasn't so interested in dumping the cake yet.
But what I can comment is about Dolby device sound and quality of make in terms if sound output at the end of the pipe.
I know their flaws down to the very bottom, having heavily modified some older units to a grade of even reconstructing the whole optical input, with stunning results both measurable and audible.
Believe me, they really don't have a clue – or don't care – about the sonic impact of power supply parts selection layout and grounding techniques. All they care for is specs that can be measured and documented on a Neutrik A1 – something " good to have " but not sufficient for the task of having flowers in the sound triggering butterflies in your stomach.
You say it....
------------------
BTW this sound experiences as teenager in cinema is highly educational in an unconscious way – leading to a lack of taste as a grown up having the money to by your " dream set "
Greetings
Michael
gedlee said:
Earl, I can't comment on the quality of codecs as I wasn't so interested in dumping the cake yet.
But what I can comment is about Dolby device sound and quality of make in terms if sound output at the end of the pipe.
I know their flaws down to the very bottom, having heavily modified some older units to a grade of even reconstructing the whole optical input, with stunning results both measurable and audible.
Believe me, they really don't have a clue – or don't care – about the sonic impact of power supply parts selection layout and grounding techniques. All they care for is specs that can be measured and documented on a Neutrik A1 – something " good to have " but not sufficient for the task of having flowers in the sound triggering butterflies in your stomach.
gedlee said:
You say it....
------------------
BTW this sound experiences as teenager in cinema is highly educational in an unconscious way – leading to a lack of taste as a grown up having the money to by your " dream set "
Greetings
Michael
I'm calling the Gary Pimm approach a quasi-cardioid, although it sure is awkward to spell several times in a row. I've heard his system at length, along with Siegried Linkwitz's speakers, which of course are true dipoles.
They sound different - not in terms of timbres, the usual audiophile concern, but in terms of - hmm, how shall I put this - spatial and dynamic qualities. Gary listens in a very small room, and didn't care for the excess liveliness of full dipoles (his starting point). But I found the mids a little closed-in sounding for my tastes - remember, I'm the electrostat enthusiast. But some of the credit or blame goes to the Eminence midbass drivers, which I didn't much care for.
But I have to admit the bass region was a real treat, as good as anything I've heard to this point. It certainly made me think about the sonics of dipole vs quasi-cardioid bass, which do sound different. I guess, for lack of a better word, that quasi-cardioid bass has more sense of weight, of solidity, while avoiding the box colorations of monopoles. And I'd be the first to admit that electrostats aren't at their best when it comes to bass quality.
So in purely subjective terms, my favorites are quasi-cardioid and horn bass, with each having a distinctive quality of their own. That doesn't say a whole lot, I know, and there's certainly enormous room to screw up either.
I do see the fulcrum point of the spectrum as being somewhere around 300~500 Hz, and for several reasons. Perceptually, the ear switches between two different mechanisms, as a result of firing-rate limitations of the neurons, acoustically, as a result of the wavelengths becoming large enough so that the direct-arrival wave starts to merge with the overall room energy, and in spectral-energy terms, where the greatest average spectral energy from the orchestra is centered around 300~500 Hz.
So I have no problem treating the below-500 Hz region differently than the region above it. I think the technical and esthetic requirements are different, and if quasi-cardioid sounds better below 500 Hz and dipole sounds better above, that's OK with me. Much of the reason to build this speaker - some time this fall or winter, I guess - is to experiment with the directivity patterns and see what sounds best.
I have different spatial expectations than most audiophiles, partly conditioned by my early work in quadraphonics. Back in the Shadow Vector days, I spent a lot of time aiming for a spatially coherent reverberant field with no "detenting" or gaps in the coverage. This field isn't part of the actual instruments themselves, it's the space they perform in. Any reasonably normal stereo recording, made with spaced mikes, Blumlein, ORTF, or EMT stereo reverb plates, would play back and fill the room with a uniform spatial field, with instruments localized in front.
I later designed loudspeakers that throw a spatial field of about 120 degrees across the frontal arc, with the instruments typically localized between the speakers (placed 50 to 60 degrees apart). SQ, QS, or EV-4 recordings would localize rear-channel encoded instruments within the larger 120-degree arc, maintaining an accurate correspondence between Shadow Vector or QS decoding, but over a smaller arc.
I still design speakers this way - for accurate spatial projection over a large arc, and with a minimum of phasiness within that arc. Traditional multidriver omni systems, or as they are called now, "bipoles", project a wide spatial arc, but at the expense of phasiness and diffuse localization. Modern home theater speakers typically never throw an image outside the confines of the L/R pair, relying on 5/7 mixdowns to do that for them.
From my perspective, though, HT speakers do not do a realistic job of depth perspective, with flat and paper-thin images, and little or no depth. It doesn't help that multichannel systems using high ratios of lossy digital compression also compress the depth out of the signal - since low-level ambient information looks like noise to the compressor, it gets the same treatment, and the low-level bits end up discarded on the recording-room floor.
Because of my quadraphonic background, I know that speakers that accurately localize 2-channel sound are also the best choice for multichannel mixdowns, with the restriction that a spatially confusing mixdown will also sound that way on spatially realistic loudspeakers. The pasted-together acoustic is readily audible as a constantly-shifting overlay of several different spatial environments.
One simple example - on a spatially realistic loudspeaker, you can easily hear the gain-riding of individual soloist tracks in a symphony, with the bizarre result that the spotlit instrument or singer zooms towards you, does their thing, then zooms back. You can just about see the engineer's hand moving the fader up and down. This zooming is masked on speakers with poor depth rendition (typical studio monitors do not excel at depth reproduction), so it ends up on the final recording.
They sound different - not in terms of timbres, the usual audiophile concern, but in terms of - hmm, how shall I put this - spatial and dynamic qualities. Gary listens in a very small room, and didn't care for the excess liveliness of full dipoles (his starting point). But I found the mids a little closed-in sounding for my tastes - remember, I'm the electrostat enthusiast. But some of the credit or blame goes to the Eminence midbass drivers, which I didn't much care for.
But I have to admit the bass region was a real treat, as good as anything I've heard to this point. It certainly made me think about the sonics of dipole vs quasi-cardioid bass, which do sound different. I guess, for lack of a better word, that quasi-cardioid bass has more sense of weight, of solidity, while avoiding the box colorations of monopoles. And I'd be the first to admit that electrostats aren't at their best when it comes to bass quality.
So in purely subjective terms, my favorites are quasi-cardioid and horn bass, with each having a distinctive quality of their own. That doesn't say a whole lot, I know, and there's certainly enormous room to screw up either.
I do see the fulcrum point of the spectrum as being somewhere around 300~500 Hz, and for several reasons. Perceptually, the ear switches between two different mechanisms, as a result of firing-rate limitations of the neurons, acoustically, as a result of the wavelengths becoming large enough so that the direct-arrival wave starts to merge with the overall room energy, and in spectral-energy terms, where the greatest average spectral energy from the orchestra is centered around 300~500 Hz.
So I have no problem treating the below-500 Hz region differently than the region above it. I think the technical and esthetic requirements are different, and if quasi-cardioid sounds better below 500 Hz and dipole sounds better above, that's OK with me. Much of the reason to build this speaker - some time this fall or winter, I guess - is to experiment with the directivity patterns and see what sounds best.
I have different spatial expectations than most audiophiles, partly conditioned by my early work in quadraphonics. Back in the Shadow Vector days, I spent a lot of time aiming for a spatially coherent reverberant field with no "detenting" or gaps in the coverage. This field isn't part of the actual instruments themselves, it's the space they perform in. Any reasonably normal stereo recording, made with spaced mikes, Blumlein, ORTF, or EMT stereo reverb plates, would play back and fill the room with a uniform spatial field, with instruments localized in front.
I later designed loudspeakers that throw a spatial field of about 120 degrees across the frontal arc, with the instruments typically localized between the speakers (placed 50 to 60 degrees apart). SQ, QS, or EV-4 recordings would localize rear-channel encoded instruments within the larger 120-degree arc, maintaining an accurate correspondence between Shadow Vector or QS decoding, but over a smaller arc.
I still design speakers this way - for accurate spatial projection over a large arc, and with a minimum of phasiness within that arc. Traditional multidriver omni systems, or as they are called now, "bipoles", project a wide spatial arc, but at the expense of phasiness and diffuse localization. Modern home theater speakers typically never throw an image outside the confines of the L/R pair, relying on 5/7 mixdowns to do that for them.
From my perspective, though, HT speakers do not do a realistic job of depth perspective, with flat and paper-thin images, and little or no depth. It doesn't help that multichannel systems using high ratios of lossy digital compression also compress the depth out of the signal - since low-level ambient information looks like noise to the compressor, it gets the same treatment, and the low-level bits end up discarded on the recording-room floor.
Because of my quadraphonic background, I know that speakers that accurately localize 2-channel sound are also the best choice for multichannel mixdowns, with the restriction that a spatially confusing mixdown will also sound that way on spatially realistic loudspeakers. The pasted-together acoustic is readily audible as a constantly-shifting overlay of several different spatial environments.
One simple example - on a spatially realistic loudspeaker, you can easily hear the gain-riding of individual soloist tracks in a symphony, with the bizarre result that the spotlit instrument or singer zooms towards you, does their thing, then zooms back. You can just about see the engineer's hand moving the fader up and down. This zooming is masked on speakers with poor depth rendition (typical studio monitors do not excel at depth reproduction), so it ends up on the final recording.
You think it's bad to read reviews? Imagine writing them - that's worse, I can say from experience.
During my brief stint as a reviewer for Positive Feedback magazine (back in their print days) the only way I could tolerate it was to adopt an intentionally gonzo style, mixing in science-fiction, UFO literature (long before the X-Files TV show), verite descriptions of what the CES was really like, and when I felt like it, a little bit of subjective-speak.
Since David Robinson didn't rewrite the reviews to fit a "style guide", I got away with it - I was appalled to discover all of the "legit" magazine editors quite routinely rewrote all product reviews to fit the corporate "style guide", as horrible as that sounds. Even writing scripts for porno movies would be better - and you probably wouldn't feel as dirty afterwards.
Here's a little sample of David Robinson would let me get away with - this time pretending to discuss the Karna and Gary Pimm amplifiers. I was at a loss to describe something as dull as amplifiers, so I wrote the cinema-noir/science fiction framing device. Some nice passages about Portland, though - I was probably missing it when I wrote the article.
During my brief stint as a reviewer for Positive Feedback magazine (back in their print days) the only way I could tolerate it was to adopt an intentionally gonzo style, mixing in science-fiction, UFO literature (long before the X-Files TV show), verite descriptions of what the CES was really like, and when I felt like it, a little bit of subjective-speak.
Since David Robinson didn't rewrite the reviews to fit a "style guide", I got away with it - I was appalled to discover all of the "legit" magazine editors quite routinely rewrote all product reviews to fit the corporate "style guide", as horrible as that sounds. Even writing scripts for porno movies would be better - and you probably wouldn't feel as dirty afterwards.
Here's a little sample of David Robinson would let me get away with - this time pretending to discuss the Karna and Gary Pimm amplifiers. I was at a loss to describe something as dull as amplifiers, so I wrote the cinema-noir/science fiction framing device. Some nice passages about Portland, though - I was probably missing it when I wrote the article.
Re: There is one thing to be said for the subjective
I suspect that I'd do a few things differently were I to work that side, but I haven't and won't. However, one must recognize that there may be some ability to adjust the playback for what may be seen as deficiencies in the recording side in some situations. I simply see some playback adjustment as possibly being beneficial in the right circumstances. You obviously don't, so we simply disagree on that point. Again, it would only be useful if the majority of the playback was from recordings such as this.
I have some recordings from the 50's that benefit to a small degree from a bit of extra low end boost. This is not good across the board, but for this music in these recordings, the deficiency in the recordings can be partially corrected. Likewise with some 70's recordings that were engineered for vinyl that have a similar deficiency due to engineering, not technology of the recording side. If these were my primary sources, I'd be inclined to make some small adjustments. Right now I do, but then I'm using the Digital Filter in SoundEasy, so I can change the "design" in minutes and revert again as desired.
Dave
gedlee said:
I suspect that I'd do a few things differently were I to work that side, but I haven't and won't. However, one must recognize that there may be some ability to adjust the playback for what may be seen as deficiencies in the recording side in some situations. I simply see some playback adjustment as possibly being beneficial in the right circumstances. You obviously don't, so we simply disagree on that point. Again, it would only be useful if the majority of the playback was from recordings such as this.
I have some recordings from the 50's that benefit to a small degree from a bit of extra low end boost. This is not good across the board, but for this music in these recordings, the deficiency in the recordings can be partially corrected. Likewise with some 70's recordings that were engineered for vinyl that have a similar deficiency due to engineering, not technology of the recording side. If these were my primary sources, I'd be inclined to make some small adjustments. Right now I do, but then I'm using the Digital Filter in SoundEasy, so I can change the "design" in minutes and revert again as desired.
Dave
Robert F said:
I had this initial impression about the Benchmark when I listened to it firstly. I.e. Dry sonics. Then by chance I got to listen to it in different studios. It was a chameleon result. Its high resolution was always there but dry vs musical results in different situations put me into thought about mixdowns it was fed and systems/rooms it was feeding. I got a demo unit to my home and tested it with several loudspeakers and headphones. I concluded that it was honest, with unusually 'real' treble dynamics, in a way that it could irritate small faults in the rest of the chain not readily apparent with other sources. I bought one, started a thread here, got to change its sole pair of coupling caps into Auricaps and its digital input caps into SCR tinfoils as the two crucial and conservative tweak upgrades of a multitude proposed, in order not to change the fundamental engineering. DAC-1 helped me locate and tame some stuff in my system that although I could see by measurement, seemed benign subjectively with many other sources. It sounds just as the data fed, it can be gloriously spacious and overtone rich or flat and pinched depending on the recording, and there are many of bad recordings. There are recordings that manage to sound flat even on my full range TB3R mini bipoles with top firing super tweeter, which extract any sound stage data -if recorded- to the fore.
The cap tweaks free it from some component distortion (its built down to a price) but they don't make it into something different. Just purer.
Hi Salas,
You clearly have much greater experience with the Benchmark than I. I have heard it in two different systems neither of which are my own. I think it has potential i.e. there is certainly clarity there and yes the high frequencies sound rather good, but otherwise it sounds like bad solid state to me. I wonder if it could have anything to do with all the cheap ic op-amps in the output stage of this device?
One interesting thing we did do was compare digital leads. There were clearly audible differences despite the manufacturer's claims for Jitter immunity.
Rob.
You clearly have much greater experience with the Benchmark than I. I have heard it in two different systems neither of which are my own. I think it has potential i.e. there is certainly clarity there and yes the high frequencies sound rather good, but otherwise it sounds like bad solid state to me. I wonder if it could have anything to do with all the cheap ic op-amps in the output stage of this device?
One interesting thing we did do was compare digital leads. There were clearly audible differences despite the manufacturer's claims for Jitter immunity.
Rob.
Its true, leads and transports can be easily picked up through it. But they don't fundamentally shake its rhythmic or tonal delivery. As far as they could measure the effect of jitter they are honest. But there must be other interface issues at play too. The 5532 op amps were the only ones not exhibiting IMD as the Benchmark engineers have emailed me, and they were not picked on cost grounds but on performance ones as they assured. Its clarity can be disturbing systems built around high end sources. Its a demanding little devil. In studios it sounds correct 9 out of 10 times. Well accepted in the pro world. I changed my view (which was exactly as yours) when I first listened to it directly feeding pro ATC 12'' actives. Hell, it was sounding towards sweet! In a Hi-Fi system I first heard it, it was relentlessly harsh! I asked some engineers that had a Weiss pro DAC in a remastering studio (they do mainly classical), and they told me, ah! its so sweet and we got shocked by its rendition because we could not initially differentiate it from the Weiss! Go figure...
Lynn Olson said:
I remember once sitting in the office of an editor at one of big European audio mags. He was asking me how I liked the magazine. "Well", I replied, "it's nice - but you guys like everything. Don't you ever write a bad review?" He just signed and said "I know, I know. But we survive on advertising. It puts us in tough spot." And I agreed it was not a spot I'd want to be in.
Sometimes you just have to read between the lines. Art usually isn't a plain statement of obvious facts, is it?
(BTW, my gripe is more with movie critics).
Lynn Olson said:
Touché. I guess that would be of primary importance, wouldn't it?
Lynn Olson said:
I'm really curious how this will turn out. I may just bite the bullet and try the 10NDA or 12NDA myself.
Lynn Olson said:
So are you looking at a series damping resistor on your bass driver, whether it ends up being a 15 or 18?
imho aiming for an idealized, almost fundamentalist goal of "accurate" sounds hopeless. Accurate to what? Most records are a few mics recorded, eq'ed, compressed, levels mucked around with and panned around to suit whatever guess the engineer is making about the range of playback equipment this music might be played on. It's common to take a mix and listen to it in the car, in a bedroom system etc. He knows there's no one exact way this mix is going to sound to everyone, he just tries to make sure many people will enjoy it.
Would you rather enjoy something or be the only one that knows you were right in listening to what you think someone else is saying you should hear?
To each his own, but would you really rather be right than happy?
I think accurate (as in faithful reproduction of input signal) is a good place to be. Otherwise I won't get maximum enjoyment out of a good recording.
Once I have accurate I can always install a "nice" button, if I am so inclined, but I can't really even do that, unless I start out with accurate reproduction because I won't even know what it is I'm attempting to modify - electronic, recording, or acoustical artifact .
poptart said:
As a recording, mixing and mastering engineer I have to make this sort of decision every day. Who or what am I mixing for - airplay? boom boxes? clock radios? earbuds? well-healed audiophiles?
I long ago decided to mix "for the future", presuming that in the future the average playback system would be more accurate, tonally at least, than is currently the case. This presumption is based on my observations over the last 40 years. I do believe that quality of playback is steadily improving, taken as a whole.
In the meantime, I think it only fair to mix for the guy who has taken the trouble to assemble an accurate playback system, as he demonstrably cares how it sounds, as do I. Moreover logic tells me that as long as speaker manufacturers (in particular) have accuracy, i.e., flat response as the target which I believe they do, apart from a little fudging to fatten up the bass in tiny boxes, then the target will be hit more often if the mix was done on genuinely accurate speakers in a quality acoustic environment.
If that means I must be unhappy with a bad recording, then so be it. I need to be!
So yes, for my work I would rather be right than happy.
On the same hand, good recordings sound fabulous, as they should.
I should add:
1) I will happily tolerate awful sonics if the performance is great
2) The more I do this the more I appreciate live un-amplified music!
Russell
JohnL said:
Ugh. No way.
As mentioned earlier, you can have low efficiency everywhere (small magnet or series resistor) or you can have high efficiency in the mid band and have about the same efficiency as the other approach close to the Fs region. Throwing away power for no good reason - especially since I'm pretty dubious about amplifier sonics - just makes no sense to me.
Thus, I'm seeing bi-amplification - with EQ near Fs, as needed - for the Bass and Midbass drivers. Since these drivers cover the region where the room is having major deviations in response, independent per-channel parametric EQ is needed anyway.
The widerange driver and tweeter share a highest-quality moderate-power amplifier (with no EQ and a passive mid/hi crossover), the midbass and bass drivers share a substantially higher-power amp with active crossovers and parametric EQ, and there's a separately powered subwoofer coming in around 60~80 Hz.
Using a parametric EQ to create higher-Q rolloff (with a mild amount of peaking) is quite a different thing with an efficient large-area pro driver than an inefficient small-area audiophile driver. Efficiency and low IM distortion really matter in this application - this is the traditional weak area of dipole systems, and it's an area I want to address. The usual comments of "thin bass" have a lot to with drivers - or electrostatic panels - that just don't have the grunt necessary for the job.
salas said:
I honestly don't know, and am reluctant to follow along and use the opinions of others - I guess I'd rather make my own mistakes. Maybe another example of that Swedish/Norski stubbornness that used to be a standing joke in our family as I grew up.
A 40~200 Hz true dipole is going to be: very large, use a W-baffle (and associated sharp lowpass crossover), or many dB of bass-boost equalization. A quasi-cardioid relaxes all of these constraints, and to my ear, sounded very, very good at Gary Pimm's place. Frankly, as good as anything I'd ever heard in that frequency range, and with Eminence drivers I'm not that fond of.
As for higher frequencies, I really don't know. The smoothness of the walkaround test is important, though, regardless of the formal designation of polar pattern. I strongly suspect if a rear tweeter is used, it does need to be 6~10 dB down relative to the front tweeter, so the backwave doesn't diffract around the front and screw up the (frontal) impulse response. It has crossed my mind that a plain old Scan-Speak in a rear-facing hemispherical baffle might be just fine for this application.