Small is better

July 25, 2017
 by Paul McGowan

Yeah, I know, the joke opportunities abound. Humor aside, I remember the very first time Arnie Nudell demonstrated the massive Infinity IRS III to me. His choice of music was a surprise.

Standing in front of these gargantuan 7.5′ tall behemoths, the first thing you’re thinking is the 1812 Overture, Pink Floyd, or Hans Zimmer’s Gladiator.

Instead, Arnie played for me a single acoustic guitar. Instead of serving fireworks, I got an aperitif.

I was immediately stunned at just how real the instrument sounded: perfectly formed, the right size, just as if the artist was in the room.

“The real test of a system is not its largess,” said Arnie, “but its ability to render singular voices and instruments in proper perspective, as if the performer was in the room.”

Once we had established the credentials of the system on the small, we moved to the big and were not disappointed.

Few systems can reproduce the power and majesty of a full orchestra.

Even fewer can manage perfection of the small.

Subscribe to Paul's Posts

32 comments on “Small is better”

  1. That’s true for sure! Would be interesting if you mean that such speakers put singular voices or instruments in proper perspective in terms of imaging only or also tonality. I think it’s both, also tonality.

    Especially the richness of let’s say piano attacks in mid/higher ranges, string sounds, brass. All around the mids, where so many are missing some flesh on the bones in their systems is delivered with the right authority with such large speakers (I don’t mean deep bass). In fact I never heard of someone who complained about a too fat sound (exempt from bass), but constantly read here and there about the opposite.

    The better components get, the more accurate and revealing they sound. Sometimes it seems the better they get, the more they are made for such speakers and the more frustration they can (not have to) create within smaller systems.

    I’m sure, with such IRS, you have much less such problems than others (but without room treatment possibly different ones ;-).

  2. Having learned from acuvox that our ears are most sensitive phase detectors I wonder how our ears combine the signal of the tweeter in the middle of the IRS line source with the time delayed signal from both the tweeter on top and at the bottom of the tweeter column. It might be that the distortions are small simply due to the fact that membrane displacement of a multiple tweeter arrangement is much smaller for the same sound pressure level than that of a single tweeter. But the ear certainly has a lot of work compensating for the time delays, isn’t it?

  3. … which is why Steve Riech, Iannis Xenakis and J S Bach are perhaps the finest composers of test signals that ever lived, not that I haven’t said that before.

  4. Small simple music sounds good on more systems and always on a good one. Going to well recorded complex even just a 4 players makes things more difficult
    As I roll tubes it’s obvious you can have a simple song play well
    Only to play more complex and be unhappy.
    Back then it must have been a analog source right ?
    Piano is one of the most difficult instruments to reproduce in digital
    Few SS dacs can do it well. My tube lampi does this very well with the right tubes.

    1. My system is entirely solid state, there are no tubes in chrome pizza boxes, I’ve had 3 DACs all solid state, and I listen to a lot of piano music. The main thing with piano is it a huge instrument that, live, can make a massive or very small sound, goes below the frequency range of most stereo systems and is also highly percussive. So it is difficult for any audio system to come close and recordings are hugely dependent on microphone positions and the amount of room effect included in the recording. That said, good speakers with sufficient amplification allows me to listen happily to piano recordings without having vacuum tubes involved. I have also been very suspicious of the concept that the sound of an audio system should be dependent on brand of vacuum tubes used. That doesn’t sound very high fidelity to me. If the tubes are changing the sound, are they acting as some sort of filter?

      1. What dacs have you used. If I may ask
        Tubes add body that seems to many to add a musicality factor.
        My Msb dac does not do nearly as well as my lampi. And tubes vary in how they effect piano.
        Having said that I am well aware that my lampi in measuremts would pale to my Msb Dac. But what is real to me is the tubes lampi

        1. A PSA PWD Mk2, then Audiolab MDAC+, now Devialet. dCS Vivaldi and Puccini DACs are far and away the best source I have ever heard.
          It does seem from your description that tubes are not high fidelity, but produce a sound that pleases you, which is perhaps more important.

          1. Very well said and the DCS Vivaldi is one of a few SS dacs that can reproduce piano well. It’s a short list lol.
            Your correct in that I love the sound that is most likely not correct.
            Have you ever had a lampi in your setup a GG or big 7
            You may scratch your head a bit in wonder.
            The new Msb sel 2 is also a very good dac too

            1. The Lampizator are popular in the UK. I am cured of component systems at the moment, maybe not if dCS do a DAC-only preamp with world clock and I win the lottery !
              Domestic audio is merely a conjuring trick to imitate live sound, and I agree with Paul generally to go with what you like even if it doesn’t measure well. We do this for enjoyment, after all.

        2. That was my approach in yesterday’s discussion:

          Sometimes and in some setups something like tubes in a DAC’s output stage are necessary to simulate the real thing (although they just add distortion to the pure concept).

          In an IRS III setup and several others they’re probably too much and a pure digital concept is superior even in tonality.

          There’s always something to compensate. The more perfect the whole setup is and the bigger the speakers, the less (probably).
          And there are many ways to get happy, one should just not transfer his solution to a general rule.

          Manufacturers certainly are mostly convinced that their concept is the truth, as different as they are…and those with the purest concepts are not always the best sounding for a majority of customers.

          Other than in a vinyl based system with many options from cartridge to phono amp, there’s nothing to tweak in a digital front end setup…well it could be with DSP, which no one wants

          1. My speakers are Paul’s
            Redone as his are as much as I love my setup to say it’s accurate or real
            Is just words we use to define our commnets. Real is many variations as well even when it is right there in front you playing
            In piano I like the solid notes not the reverb ones. But do get why those pedals are needed.
            There are many audio types in our group.
            I like the quality then the music some times. And even less the quality of the performance as being well played. Best is a great performance that is vibrant rich in over tones.
            Like Antonio farcioni quartet live.
            It’s striking in dynamics and very sharply pointed.
            Solid state done well is great but some tubes in there ESP in the output makes for a rich sound that is colored and distorted like a pa Ina small club in some ways. The most vibrant tubes I have found so far is the 2A3 mesh by eml.
            They are a bit hard but really convey dynamics in music. Softer tubes do good for simply music or calming music.

      1. That’s the beauty of this test. See if you can get the small sounds right then see if the large ones work. More speakers get the small right but fail miserably when the big stuff comes to bear. The IRS is one of only a handful to get both big and small correct.

        1. I have some experience with the Scaena speakers years ago. Since you’ve used the latest ones at shows, do they do a good job of both small and large?

          1. It’s hard to tell. I wasn’t happy with their performance at RMAF but once broken in at the Chicago show, they sounded damned good. Unfortunately, I had only a day at the show and had to return to work so never really got to spend any time with them. They certainly had/have the potential for both – and HP loved them for both too.

            1. He loved an earlier version, of course. I was there once with George Bischoff when he was using the Scaenas and they did big, the volcano on the Lost World CD that HP loved to use, spectacularly. And the Scaenas are really an evolution of the Pipedreams that I thought did both big and small well. Interesting that these speakers and the IRS are all large line sources.

      2. “The less you ask any system to do, the better it will sound. Its limitations show up when it has to do more.”

        At zero output it works perfectly. That’s why John Cage’s piece 4′-33″ is so valuable. Any sound system can reproduce it without any faults. You don’t even have to plug it in.

  5. The notion of proper perspective and “size” is lost on many rooms at audio shows and many high end dealers. Somehow, 40 foot tall women singers does not seem real to me, but that is the image that is frequently presented.

  6. I heard the big Infinity IRS system back in the 1990’s at an audio show in California. It was in a very large room and they were playing some big orchestral piece. And it sounded like the orchestra was spread out in front of us.

    But then they played a small jazz band and the bass player sounded like he was 8 feet wide.

    I recently heard a top of the line Martin Logan system, at the home of a local recording engineer, in a room large enough to play basketball in. As with the big IRS system, large orchestral works sounded terrific, but smaller ensembles sounded much larger than life. But his recording of the Alexander String quartet, done with a one point, AKG mike with elements configured for Blumlein recording, gave the impression of the string quartet being 20 feet wide.

    The size of the room and loudspeaker spacing have a lot to do with perceptions. My own listening room is much more what many of us listen to in, about 14′ x 17′. Small bands and solo instruments do quite well in it, but the big orchestral works just cannot sound like an orchestra that is 50 feet wide. Not possible.

    Sorry, but this issue just doesn’t keep me up at night.

    1. Sorry you didn’t hear the IRS setup properly. ‘Tis a shame. Everyone should have a chance.

      Sometimes when an instrument or voice is too wide it’s a volume related issue, but this to me sounds like setup.

  7. Not small VS large, but a surprise in the audition process.

    Many years ago I was considering adding subwoofers to my BC-1 speakers. They had a wonderful midrange but lacked much low frequency extension.

    So I went to a highly acclaimed dealer in Orange Co., CA, Middleton, Kemp, & White who carried those speakers. One of the owners set up a system with their recommended subs and a crossover which rolled off the Spendors. Then he played a solo violin recording (it could have been cello, this was decades ago), while I was expecting an orchestral block-buster or heavy rock demo. Switching back and forth it was easy to hear the improvement in the upper bass and lower midrange with the low end removed from the BC-1s, the added clarity and definition was quite apparent. He was making the point that proper subs could not only extend the low end of the range but could improve higher frequency resolution when the low notes were removed from the main speakers and amp.

  8. The laws of physics are inviolate wishful thinking aside. A large speaker is quite capable of producing a single small instrument with very little effort and can produce large scale performances too with ease. This is just not possible with a small speaker and by that I mean the scale , the majesty the volume, the impact of an large orchestra or group. At best it is a miniaturized version of the real thing and the brain has to fill in the huge gap left. Regards.

  9. You are talking about two entirely different problems here. In one problem you are trying to duplicate the sound of one or more musical instruments as they would be heard in your own room. This is the easier of the two problems. In effect the speakers are a surrogate for the musical instruments. They should produce the same sound field as the instrument does and the room acoustics will affect the sound the same way between the time it is launched and the time it reaches your ears. Acuvox has solved this problem….almost. The only discrepancy between his solution and an ideal solution is that sound intensity propagation at a given frequency is a function of direction. In his solution sound propagated in all directions has the same spectral content. This could easily be adjusted although the implementation would be more complex. The difference might not be particularly significant. The limitation is that he needs different speakers for different instruments as they each propagate with their own directional characteristics. He also needs special recordings. In my own case I’d considered his solution and rejected it as impractical for listening to commercial recordings. Instead I use a somewhat different solution that is also flawed. One thing most musical instruments have in common is that they propagate sound in many directions resulting in a lot of reflections. By arranging the speakers so that both the directly heard sound is flat and a large number of reflections from different directions reach the listener and are also flat a similar but not identical solution is arrived at.

    A guitar is somewhat more directional than many other instruments. It propagates most of its sound in the forward hemisphere although not all of it. Conventional loudspeakers can come closer and dipole speakers or speakers that propagate at least some of their sound indirectly like IRS especially at high frequencies can come closer than unidirectional speakers, especially box speakers. A piano is an entirely different story. This is the primary real reason why Atkinson’s live versus recorded side by side demo with a Steinway D piano versus Vivid speakers failed. Their directions of sound propagation are entirely different and the lack of early reflections from many direction from the vivid speakers made them sound like point sources. This is a common failing of virtually all commercially sold speakers. Dipoles are somewhat better but not nearly good enough.

    The problem of reproducing the sound of a large group such as a symphony orchestra, a choral group, or a very large instrument like a pipe organ is much harder. You cannot fit these sounds into your listening room. Even if you could shrink them down to size to fit but their sound output was the same it would be deafening in a small space. These groups are meant to be heard and sound best in much larger spaces. What’s more most of what you hear at live performances comes from reflections inside the space itself.

    I have studied this problem of perceived power for a long time. This stemmed from my experience on that hot summer day sitting in Saint Christopher’s Church in downtown Manhattan listening to the pipe organists practice softly yet conveying an instrument of enormous power, clearly far more than any recording. I’ve identified five variables that seem to affect how powerful a source of sound appears to be. There may be others. One is absolute loudness. The louder the sound the more powerful it is. Two is bass content. More deep bass equals more perceived power. Third is perceived distance. This one is really important. If two sounds have the same loudness at the listener the one that seems further away will also seem more powerful. What’s more that power perception varies with the square of the distance. Another is the perceived size of the space the sound fills up. Filling up a very large room with sound will make the source seem more powerful. Finally the persistence of the sound, that is how long it takes to die out in that space, the reverberation time affects the perceived power. So a good home sound system is played just as loud as a live symphony orchestra with the same bass level. But the home sound system source is perceived to be about 10 or 15 feet away, the orchestra 40 to 60 feet away or more at a live performance. The room at home is often about 4000 cubic feet, the concert hall about 650,000 to 900,000 cubic feet. The RT of your room at mid frequencies is about 0.2 seconds. In the concert hall it is 2 seconds. By comparison the live symphony orchestra is perceived to be around 43 db more powerful. That is a factor of 20,000 times. For a large church or cathedral it can be perceived to be 100,000 times, 500,000 times, even 1 million times more powerful than a recording at home and its sound envelops you. This is why the sound of the Mormon Tabernacle Choir simulated to sound like it would in a room the size of the tabernacle was so astonishing from the very first second. It was hard to believe. Someone’s jaw dropped. Good thing he was sitting on a sofa and not a chair or he might have fallen over. Fortunately I’m insured.

  10. There is another aspect being overlooked? …. You could never fit the Boston Symphony Orchestra into your room… as it sounds in the actual hall. For? It would take the hall to do it.. And, then with speakers? They could never replace the instruments, that when playing, project in all directions.

    What we can hope for? Its a scaled down (but accurate) sound. Scaled down like some of us used to watch in wonder a small village before our eyes, when some artist created a masterful HO train layout with mountains and streams… roads and houses. To be done right, its got to be done smaller. But, to scale. The smaller the scale? The more over all picture you get to see. But, not too small. Just enough to make you feel like you are sitting rows back when listening to a recording that was made with mics placed close to the instruments.

    Maybe we need a scale rating system for sound stages with audio. Maybe speakers can be listed like OO , HO, TT scale as with model trains. For, a masterfully done model train layout, when being watched in action is awe inspiring. So much fine detail.. We need accurate, phase coherent, small speakers for the smaller layouts. Big floor standers for those who want larger model train scales

    When done right with accuracy you become like a happy giant reigning over your domain. Audio systems are to be to live music, as model trains are to actual trains. Yes.. Some isolated parts of a railroad may fit into your room. Like a switching light. And, in audio? As with an acoustic guitar performance. But, you’ll never get a full sized train rolling through your living room. If you did? You would have no china left sitting on their shelves in your house.

    If you could get Led Zeppelin playing full size in your room? You would have the police knocking on your door with a disturbance of the peace complaint from neighbors. Scale it down!

  11. One of the most difficult instruments to reproduce size wise is a full drum set. The drum set on Sheffield Labs Harry James “The King James Version” is one of the best test for correct sizing of instruments in space. When this is right, the drum set isn’t 20 feet wide, remember that a drum kit has to be within the armspan of the drummer, maybe 7 feet at most. A most telling test and a reference for me.

    1. I use percussion and plectrum instruments for testing. For drums I have Michael Carvin’s “Drum Concerto at Down”, Glen Velez’s “Rhythms of the Chakras” and Kodo’s “Heartbeat Drummers of Japan”.

      I am more partial to piano and harpsichord, since I have them in my music room for comparison.

  12. Back in the ’70s selling audio gear I often demonstrated with When I Was In My Prime, an acapella song by Jaqui McShee of the Pentangle from their album Cruel Sister. It showed box resonances in cheap speakers.

    I’d explain that without any other instruments or voices you could hear how well the one voice was reproduced. It could be reproduced clearly, I had Servo Statik-Is at home.

    On some speakers she sounded like she had a chest cold, others she sounded nasal. On better speakers she sounded more natural.

  13. I have three sets of manufactured speakers that can reproduce small sounds accurately (one retailing at $400/pr), but only massive surface area and fast, powerful motors can produce loud, complex music.

  14. i was astounded at THE Show Newport 2014 when, in the room with the LARGE Soundlabs full range ES, a recording of just a solo guitar sounded like the EXACT SIZE of a solo guitar. that alone was an impressive demo.

Leave a Reply

Stop by for a tour:
Mon-Fri, 8:30am-5pm MST

4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

Stop by for a tour:
4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram