Speakers are the weakest link in any audio chain. As such, i'd like to see someone come out with an omnidirectional point source driver that covers the full frequency range and is highly efficient with a reasonable impedance curve.
Yes, i want to have my cake and eat it too.. : ) Sean
A third ear, mounted in the center of my head
Software with as much attention to recording quality and perhaps MORE important, a format that delivers that quality with as little compromise in efforts as we all put into our systems.
A really perfect piece of software would amaze you as to how much it "equalizes" the differences in equipment. Not to say that the system is not equally important, but with (say) master tapes, first generation, you would be so deep in music that you would forget about the gear for years........perhaps forever.
That's interesting, Albert. Now I like thinking that is why I have switched to records. To me, analog is superior :) (I don't want to start a debate here, though :).
I think Sean is right, but also Albert from the other end of the chain. I've never had the chance to hear a first generation tape, but I've struggled with speakers all my audiophile life and I ain't happy to this day. So it will remain a work in progress. Cheers,
I'd like to be able to plug my turntable directly into Sean's perfect speakers
Shhhh Albert! Don't drive the price of master dubs up!
Sean, any good resources for single point source driver loudspeakers? If you have any please list, thanks. Melhuish.org is very good if you do not already use it.
I agree with Albert. The recording industry has got to get their act together. I can't believe how good the music quality is on some recordings. My equipment completely disappears. We can always buy better hardware and this is a never ending quest, but I have found that the LP or CD is definatly the piece that is lacking most often.
A new playback format designed for Audiophiles! Yeah speakers need lots 'o work in the execution dept., too. And quickly while I'm on the discussion speakers, I would like to see a ESL that has the bass slam of a dynamic speaker using cones. Flat response from 100kHz to 8Hz, or so, would be nice.
Back on the subject of the playback format. I would like to see a new format that is an overengineered as the equipment we buy. not 16/44.1 no 24/192 just because it *good enough*. Blah. Let sit down NOW, and decide on a GREAT format, going forward. How many bit, samplerate, scheme, etc? 64bit sampled 4 billion times a sec? fine with me. PCM, DSD, TLA? I don't care. Pick one, hopefully the best one, and let's stick with it! And the D/A designs should be part of the design, and NOT an afterthought!
But I suppose none of that will mean a lick of salt if the recording engineer (is that the guy?) gives us a compromised "mix." Compressed dynamics, no low freq extension, etc. mediocre microphones, mediocre A/D converter, etc.
Another vote for Albert. Audiophiles often go on and on about the importance of the source, but the true source is sat at a mixing desk twiddling knobs and adding effects to make the music sound good in the car, on a boombox, on a walkman .... but rarely on a high end hifi, probably listening through yamaha monitors, too.
AMAN Albert and Sean! I have spent a fortune on these toys for thirty Years. Software is the problem, Garbage in Garbage out! Ken
It has to be the recording phase,like many has responded. Why every thing is fine (almost) when I play exceptionally well recorded CDs (e.g Mapleshade, Reference, Chesky) and falls apart when I play ' Commercial ' titles?
The recording chain (usually dismal in quality) is the weakest link, as several folks have pointed out!
Secondly, hi-level crossovers in speakers cause losses that would be avoided with line-level crossovers designed for a particular speaker. Because we in this individualistic country hate the system approach, we somehow are not demanding our speakers to be matched with dedicated amps and getting those lossy caps, resistors out of the high level signal. We prefer to mix and match our amps and speakers, and the manufacturers are glad to keep on feeding that frenzy.
I believe this point holds true whether you're driving the speaker with digital or analog amplification, that is, the crossover should still be implemented at the line level (or before, at the digital level) where loss of efficiency is lowered.
I do not know where in the chain Meridian speakers, for instance, locate the crossover components. I think Bryston's new powered monitor uses a line level crossover, whcih might account for its good sound. In any case, at least the amps are matched to the drivers.
However, in view of poor general recording quality and the listener fatigue inherent in 16/44 digital, it also makes sense that we prefer our electronics + speakers less revealing! Therefore one hand dirties the other and around and around we go in our audiophilia nervosa, always trying to make up for the shortcomings of our recordings with different matching of components, and all the tweaks. Actually the individuality of the speaker is usually the biggest decision, because one is buying the designer's voicing after all is said and done. It ends up being a very personal decision and one highly dependent on the room one has.
If I may ramble on, somebody posted on Audio Asylum about their experience listening to a system using Halcro amps (super low distortion solid state) and Pipedreams speakers, and several people including him just could not stand the sound! I am betting the source was just not up to the resolution/cleanness of the rest of the chain.
I have also had experience listening to Spectral electronics on Avalon speakers (all super low distortion designs) and would not put such a system in my home, even if I could afford it, simply because on most recordings it was unlistenably fatiguing. Same goes for Wilsons, Levinson, etc.
Good observations by all. My own listing, in no particular order:
Tacs. There is a lot of truth in what you say.
Why such direct things are not done is a wonder to me. Seems you like to think about these things. Answer this for me. Amp makers always measure distortion as voltage amplitude distortion , almost exclusively. It's almost all they care about - using negative feedback ect... But a driver-transducer is current controlled. Most amp makers never worry about this. It's important because the current in the driver's coil is going to be the output voltage divided by the speaker's impedance right(ohm's law)? This means that as the speaker's impedence fluctuates all over the place (with frequency and voice coil temperature) the current does too. Current will only be linearly related to output voltage if the speaker presents a purely resistive load which it never does. Anyway, the impedence fluctuations translate directly into distortion of the current driving the coil. To me this is why you get so many amps that sound crappy with certain speakers. Do you get what I mean? I like to make speakers as a hobby and i have always wondered on this point. It seems the amp people are so focused on one thing they fail to see what is important to the speaker-current. I do not see it even discussed. I believe you would have to raise the amp's output impedence to deliver more consistent current. Tell me what I'm missing.. as I said I just like to mess with speakers and am not an EE or anything.
Sincerely, I remain
I agree with you.
Many of the amp vendors do actually talk about current. At the low-end of hi-fi (where I and my budget reside!) talking about "high-current" designs and providing at least sketchy numbers (very sketchy) to partially back it up is pretty standard practice.
They do this to distinguish themselves from the JVCs of the world that are doing 5x100WPC home theatre receivers with well under 1% THD, for under $300. All in a box that weighs about 10 pounds.
I don't think they (the low-end of high-end) feel the need to provide more detailed current numbers than they do (mainly just amp ratings, with no associated information), because very few people ask about it.
I'm thinking of Rotel, NAD Silver Series, and the like. May also be true at the higher-end.
Ehart. I'm not talking just about the amp's ability to throw out current in my post but I agree with you totally. Power (in watts) is voltage times current and you need both. Current is needed when the load goes low. I agree with you as far as you go.
But I am talking about another issue in addition. Providing STEADY current into a varying load (the speaker) which gets very little attention as far as I can tell. Amp folks talk a lot about steady voltage output and the speaker coil is a current issue which should be kept constant as possible. This is true even with high end stuff. When you do hear about current it's always this artifical discussion about an amp running into a resistive load. I may just not understand something. That is why i'm throwing it out for discussion. Some of the smaller set amps drive really effecient speakers with inherent damping and avoid the current issue talked about in the post above .. at least to greater degree than the high powered amps.. Maybe one of the reasons they sound better?
Sincerely, I remain
The amount of money that some folks are able and willing to expend on this hobby these days always makes me wonder why the high-end industry stops short of trying to exceed the present paradigm of reproduced sound. It seems to me that instead of wringing hands and waiting to see what new digital format or channel configuration will prevail in the battles of the big boys - who aren't selling their products based on audiophile criteria - the high-end could form an industry consortium of their own to originate a perfectionist approach to recording, software, and system parameters that could be adopted by audiophiles and the high-end industry alone. Yes, such an effort would be very expensive, as would the resulting hardware and recordings. It would also have to be backward-compatible in some way with existing formats and software. But given the amounts of cash that some of us are apparently ready to throw at mere refinements of old technology and tweaks that don't significantly advance the illusion of recreating the original performance, and the prices that are already paid for vintage records of audiophile interest, I can imagine such an all-out assault actually reinvigorating the business. In general, the new paradigm would focus on completely integrated systems (forget the separate components, mix-'n'-match approach), be multi-channel and all-digital in nature, and would require purpose-built dedicated listening environments (the answer to the thrust of the original question is the listening room, by the way - on par with software and above speakers). The recording process would be an integral part of this system, utilizing a purist audiophile-grade chain and agreed-upon microphone techniques to enable predictable multi-channel reproduction. The digital recording and storage media and all signal proccessing would have to be accomplished using a data density exceeding theoretical distortion and detection limits of electronics and the human ear. The number of discrete audio channels would be determined solely by what is acoustically required for transparent soundfield immersion (my own guess is eight, minimum). Maybe the system would even run off of a dedicated, high-power utility line incompatible with other household devices. Whatever it takes. And whatever it may be - whether a complete room/system cost a million dollars and recordings cost hundreds each - I bet that buyers would come forth and ante up. Such is the power of music (and money). And the rest of us mortals would still have what we've got today. (But I, for one, would not feel much the poorer for it, since I primarily listen to vintage records and reissues from the golden ages of rock and roll, DIY punk, studio pop, jazz, rhythm and blues, etc., where the signature [read: distorted] sound of the recording [analog and often mono] and the medium [often less than pristine] is as important to the atmosphere of the record as the performance itself!)
Clueless (who isn't) smacks the nail right on the head. The electrical interaction between the amplifier and the speaker is wayyyy complex and cannot be deduced from reading the spec sheets of both. You have to do it with your ears. That's why amps with very similar measured characteristics sound so strikingly different when connected to the same speakers and why speakers sound so very different when connected to the same amp.
I wouldn't say that at some point in the future these matters couldn't be measured and quantified but at this point we've only got one way to do it....with the ears.
Very well put Bishopwill. Good common sense point.
Further on Bishop's & Clueless' points on amp-speakers interaction: one of the reasons high nominal power amps have a market IMO, is the current reserve the manufacturer has to include in order to achieve specs such as (fortuitous example) 250W class A, measured @ 1kHz signal/8 OHM constant resistance. This power capability allows these machines to perform under normal listening conditions at reasonable db levels and deal with the varying speaker load (+ the crap from the c/overs).
Similarly, with tubes: allegedly, many of the power supplies provided for an advertsied output of 2-digit watts the power supplies are overkill -- according to traditional EE theory. This comes from hearsay, I'm not an EE...
On the other hand, for me, the original question still stands -- bar active speakers and our zero control over recording and s/w production phases: where's the ultimate?
Lack of standards throughout the chain, from microphones to speakers. Two channel sound. Overproduced recordings. Producers and engineers by law should only be alloweed three microphones when they record. Too much editing of recordings. Too many variables in assembling systems: manufacturers should offer more all inclusive systems with everything hard wired from a to z (would audiophiles buy them: buying, assembling, disassembling and selling is the heart of the hobby, no?). At least more active speaker systems to avoid mismatches. And last but not least, improved speaker systems, not that good ones now are bad, simply that they are the hardest part of the system to design and build to the highest level.
If you ask me, the recording quality. Some of my favorite CDs sound like crap unfortunately. My Mcintosh setup does the best job of all the ones I have tried to prevent a constant hiss from coming through but the recording still sucks. Perhaps when the cat fight of SACD and DVDA is over, we can settle on spending time and money on IMPROVING a common standard (as was mentioned) that should include recording company equipment quality as well.
As far as the steady current thread above between speaker and amp, it IS totally quantifiable and I have calculated it for my setup but let me tell you, it is supremely complicated and is dependant on all parts of the system. You change one piece and everything has to be done over again. Flux linkages from the cables, output circuits, and speaker coils are all intertwined and consequenly affect both voltage and current continously and are the main reason for differences in sound and fluctuations in output. Mcintosh realized this years ago and with the inductor-coupled output (autoformers) they use on solid state amps, they can reduce the effect by dominating it with a known variable - a fantastic concept. Electromagnetics are very complicated and so you might as well forget about calculating, since even that is an estimate at best, and just listen for yourself since the end result is what your ears like.
Re Bishop, Clueless, Greg, excellent points: Amp ESL speaker interaction is even more complex, as I painflully learnt through the years, because ESLs don't show a resistive load to the amp, have crazy impedance curves and since watt specs don't necesarily tell you the current capability of an amp, you can make costly mistakes, thinkling you've bought a powerhouse, which then conks out on you on really high notes. Hence also the fable, that ESLs have no real high end. You need current, lots of it and then a violin will have its overtones, a Steinway will sound like one also in the top octaves and a Coloratura in full steam will make your hair stand properly on end. Cheers,
Detlof - ESLs. You are so right, raise completely different needs in amps and wire too. The ESL acts more like a capacitor in the circuit, and with a transformer there too, raises all sorts of unique inductance/capacitance resonance issues. Special concerns about the inductance parameters of your wire as far as I understand -which isn't too far. Helps so much to understand what can be measured and what is understandable -but in the end the reference is one's ears. Detlof-have you ever heard differences in highs, some unwanted peaks, due to wire alone?
With regard to conventional speakers, I still want to play around with the idea of an amp that is variable impedence and designed specifically for one load/speaker, which is a point Tac brought up a few posts ago.
An audio system is really one circuit and it help to think of it that way instead of as discrete, separate, little enclosed boxes. We don't buy a car from five different vendors; body from one, frame from another, motor from third, all not knowing what the others design intentions are.
Back to topic on this thread we are so far from ultimate that it is depressing if you think about it. The amount and kinds of distortion just in speakers is crazy. Harmonic, intermodulation, crossmodulation, mechanical resonance.....the speaker is playing it's own tune really and it will be this way until the driver diaphram's density is equal to that of air and has absolutely uniform acceleration over the entire surface at all frequencies. Requires a speaker with trillions of microscopic transducers and nanotechnology or something- Lynn Olson has some interesting things to say about this stuff and most of my opinions just reflect a few folk like him.
Joke: what do you call the time between the point you step on the banana peel and when you fall on your arse?
- a bananosecond
That joke has a peel.
I'm still brainstorming on a "full-range" plasma driver without the ozone problems or paying $2-$5 an hour to run pressurized gas tanks. If I could get one of those down to 100hz-step aside ribbons, electrostats, electrodynamics, and bending wave transducers. And spend all the loving time building the mother of all subwoofers to go with. Wouldn't have to worry about crossover points in the midband, time delay, phase problems, cabinet resonances, doppler distortion, puh.. gone. One beautiful pulsing point source. Who knows. I know they've been run down to 700hz :). Nelson Pass' "ion cloud" put him in the hospital, but it ran full-range I believe.
I would say the Audio press "Elite" are the biggest anchor to advancement. I just received a Sony SCD-777ES SACD Player
last night. It will be paired with my Pass Aleph P pre-amp
& Pass Aleph 2 monoblocs. All these pieces are state of the art to me. They're also all discontinued & essentially forgotten by The Absolute Sound & Stereophile.
Harry Pearson says you need a minimum 200 watts per channel. Robert Harley says the new multi channel Sony SCD-XA777ES player shows potential. I say B.S.
I'll settle for (I want) HIGH QUALITY 2-channel MUSIC. My music system components get moved to my video syetem as they are replaced by something better.
There's nothing more bogus than reading a review of a "totally neutral" tube amp or pre-amp. I've owned tube equipment. I bought it for "tube sound"! Why on earth would anyone put up with tube heat & maintenance issues to obtain "neutral" SOLID STATE sound?
In Absolute Sound isssue 133, Anna Logg says the BAT VK-75SE Amp "is a reference-level amplifier that produces, without coloration, exactly what is fed to it". In the NEXT paragraph she says the amp delivers the fabled triode midrange, paired with a solid state or tube pre-amp.
Aren't triode fanatics looking for "romantic"(colored)sound? Isn't that the point of tubes?
Is 9:00a.m. too early for a beer?
Daniel: "I would say the audio press elite are the biggest anchor to advancement." Amen Brother.
Hype and gloss and big buck wannabe materialism instead of audio. Especially, as you point out, in tubes. All of the basic circuit topologies are how old? There have been a few interesting improvements but some basic old stuff stands the test. Give me an old st70 (sells around here for $250) and a soldering iron.
Where can I meet ya for that beer?
Sincerely, I remain
Daniel and Clueless - you guys are getting back to the essential fact that one does not require "the ultimate", or even a close approximation of it, to receive enjoyment from reproduced music (I often get maximum enjoyment from my stone-stock base-model OEM cassette player in the car). Regardless, I still think the theoretical question of getting past the bogus "latest and greatest" paradigm promoted by the mags, manufacturers, and retailers alike is fascinating in its own right. Will the industry all get together to try and exceed the "same-old same-old"? Probably not. Could I benefit from it if they did? Definitely not (would be too expensive). Would this matter to my enjoyment of music played back utilizing the status quo? Of course not. As I said, our capacity for the enjoyment of reproduced music can be remarkably free from issues relating to its fidelity to the absolute sound - fortunately for us! (BTW, Daniel, I disagree that "tube sound" is always sought-after for its "colorations", or that solid-state sound should be thought of as "neutral". Both will deviate from reality, but the best of both will do so less.)