Infinity Reference Standard 1B loudspeaker Anthony H. Cordesman Returns
No speaker ever totally disguises the compromises it must make with the laws of physics. Infinity's two largest monitor speakers, however come as close to hiding theirs as any full-range speaker available. The Infinity IRS has long ranked as one of the top two or three speakers in the world. The RS-1B has slowly emerged as one of the top two or three speakers for the ordinary home.
I now use the RS-1B as my home reference speaker, and a nearby IRS-III at Excalibur Audio as a listening "control" when I'm unsure about how a given piece of equipment would sound on a true line-source speaker or with full response in the lowest octave (footnote 1). With the proper drive electronics and adjustments, both speaker systems are more than reference equipment: they provide a musically convincing and enjoyable a sound source as any equipment around.
At the same time, both speaker systems have their weaknesses, and these tradeoffs provide insights into what can and cannot be accomplished with today's speaker designs.
The tradeoffs are most obvious in the RS-1B. First, prolonged listening makes it clearer and clearer, that the treble and midrange panel outperforms the woofer column. Infinity has steadily improved the match between midrange and woofer, but the woofer is distinctly less dynamic, less extended, and more distorted than the rest of the system. The woofer also seems distinctly "slower" than the midrange, particularly when dealing with brief bass transients rather than sustained organ notes.
There are several practical ways to minimize this problem. One is to realize that the treble and midrange panel is an extremely difficult load. I do not fully trust my measuring equipment, but it is clear that the treble/midrange panel dips to 2 ohms, and can suck up incredible amounts of current. This has led me to revise my comments that they work best with tube amps. They do work superbly with the Conrad-Johnson Premier Four or Fives, and very well with the Audio Research D250-II. (The Premier Fives clearly outperform the D250-II for the application.)
The panels do not, however, work well with low-powered tube amps, and even the Conrad Johnson Premier Four an be a bit marginal at high rock 1evels. The RS-1Bs are definitely not the proper load for the New York Audio Lab or Counterpoint OTL amps. Any use of lower-powered or load-sensitive tube amplifiers will alter the timbre of the RS-1Bs, restrict dynamics, and usually create a lower midrange/upper bass suck-out that makes the limitations of the woofer column far more apparent, tending to make the entire speaker system sound a bit lean and hard.
Most transistor amplifiers have the same effect. For some reason, amplifiers as good as the Krells and Mark Levinsons do not perform well with the RS-1Bs. However, the Electrocompaniet Ampliwire 100, PS Audio 200C and even the Adcom GFA-555 provide a balanced timbre and smooth transition from panel to woofer. The Electrocompaniet and PS Audio outperform the Premier Four, Premier Fives, and D250-II in the upper bass region, and possibly in the top octave as well, although the tube gear gives superior handling of midrange transients and dynamics, and provides more low-level harmonic information. (The PS Audio furnishes more detail and focus than the Electrocompaniet.)
You also should carefully consider matching your panel and woofer amplifiers with the RS-1B. Manufacturers are curiously unwilling to send two stereo or four top-of-the-line mono tube amplifiers (footnote 2), but tests with a pair of Adcoms and PS Audio 200Cs showed that the panel-woofer blend was smoother and more convincing in timbre with matching amplifiers than with any combination of top-quality tube amplifiers on the panels and transistors on the bottom. Granted this is a tradeoff, since the tube amp may well provide a sweeter and more transparent sound; the Adcom 555 definitely lacks the focus and detail of amplifiers costing four to ten times as much.
Even with the right amps, you'll find that the Infinity RS-1B needs an "infinite" amount of fiddling. Fortunately, the crossover controls have been improved since our last review. The latest models add a third control for the upper midrange, which can help reduce the glare or leanness that results from using the wrong amplifier. The crossover also has been changed to allow more swing in the lower tweeter control, helping to rectify an area in which the speaker can sound a bit lean. These changes greatly increase the probability that only minor tweaking will be needed.
You'll need to spend a great deal of time spacing and angling the panels to get the ideal mix of depth, width, and imaging detail. I have never heard the panels image properly in any real-world listening rooms without a slight toe-in; this has been true of every other dipole design I've tried, with the possible exception of the Acoustat 1+1s.
At the end of it all you'll still "hear the mechanism creak." The RS-1Bs have good, deep bass, but hardly very good deep bass. The larger VMPS cone speakers (which are much cheaper) provide far superior bass power, control, and transients. So do the Entec subwoofers. No matter what you do, the RS-1B's bass will either be a bit slow and unconvincing, or will have to be set slightly lower than the panel to allow the lower midrange/upper bass output of the midrange EMIMs to dominate the sound.
The RS-1B's bass columns simply do not seem capable of proving both deep bass and good bass articulation or transients in any installation I've heard. This evidently requires bigger enclosures, new drivers, or something more dramatic than crossover changes. I am tempted to suggest that Infinity should put its EMIT or EMIM drivers on a VMPS tower, or get Entec to design their next subwoofer, but as this would just get everyone angry, I'll not even hint at such a thing, much less mention it in print.
Keep in mind that virtually all the competition has worse problems in trying to provide fully integrated deep bass. I don't know of anyone who provides a better set of compromises in blending deep bass and midrange in a full-range system near the price of the RS-1Bs than Infinity. The unfortunate truth is that almost all designs that attempt to go much below 40Hz are partial failures, and breaking the 30Hz barrier always seems to result in bass boom, lack of clear frequency discrimination in the low bass, poor transition from bass to midrange, room interaction problems, etc.
I have only heard a handful of systems in my life that really balanced good, deep bass with the rest of the frequency spectrum to sound musically natural. All were extremely expensive, and most had extensive custom engineering. All required very large listening rooms, and all still had at least minor room resonance or standing-wave problems.
This has broad implications for any speaker buyer. If you live in a normal home or apartment, 'you may well find that no system with deep bass will ever fully meet your needs. As a result, you may wish to set the RS-1B crossover so that it does not play the low bass, or consider buying the RS-2Bs instead of the RS-1Bs. I feel the Infinity RS-2Bs are one of the High End's most-ignored "best buys." They have many of the virtues of the RS-1Bs, produce a better overall bass signal, and provide better overall timbre and dynamic integration in most listening rooms. They also are far less amplifier-sensitive than the '1Bs.
There are three further design problems in the RS-1Bs that I feel need correction. The first problem is twofold: the panels are top-heavy, and the feet do not work on padded carpets. Put TipToes under both panels and woofer columns; those on the panels need to be of different heights, to tip the panels upwards. You also may want to put several bricks on the rear, to weigh the panels down.
The second problem is more serious, and involves the woofer. The woofer column needs to be placed close to and just behind the panels for the best sound. The dust caps of the woofers, however, vibrate, acting as midrange drivers. You can try rubber cementing a softer, 2" dust cap over them, but this—and other, more drastic mods—simply should not be necessary: It doesn't take a golden ear to hear what is happening; Infinity should long ago have fixed this problem at the factory.
The third problem is quality control. Infinity still never seems to produce an entire RS-1B with all the jacks and sockets properly tightened. I've seen a number of RS-1Bs with loose banana sockets, and neither of the two versions I've owned have had properly assembled crossovers. The first did not complete the wiring to one high-pass socket, and the second had an intermittent shunt across the left channel. I would also dearly love to see the crossover wired with Tiffany jacks: this would sharply reduce the risk of connection problems, which can transform acoustic feedback into a near-meltdown of your amp.
I also recommend that you replace the standard Monster Cable wiring with the Straight Wire harness for this speaker, and bypass the high-pass section of the crossover. The Straight Wire is notably more transparent in the upper midrange than the Monster Cable, and the crossover wiring is too complex, and affects the purity of the high-pass signal.
Consult your dealer for details, but you can bypass the high-pass section of the crossover simply by inserting a top-quality capacitor of the right value in series with your amplifier's input jack. The best capacitors I've yet heard are the Conrad-Johnson capacitors, but you can try Wonders, Relcaps, etc. The capacitor in the Infinity crossover isn't all that bad, but it can be improved upon. You also need another interconnect, etc., if you use the high-pass section, and straight-in wiring seems to help. The EMITs and EMIMs are superb drivers, and benefit from evety possible improvement.
It takes no great vision to assume that Infinity recognizes these problems and will continue to make at least evolutionary improvements. Virtually every time I see the RS-1B, I find another set of refinements has been made to the design. I should also stress that this critique in no way means that the RS-1B is not the state of the art. The problem is that the state of the art simply is not perfect: The sport in hunting for the best high-end components is no more in danger of being spoiled by the appearance of the perfect speaker than it is by the appearance of the perfect phono front end.—Anthony H. Cordesman
Footnote 1: This doesn't apply to amplifier evaluation, since the IRS III has its own servo-amplifiers for the low end.—Larry Archibald
Footnote 2: Not all manufacturers: Audio Research recently showed up in Santa Fe with two D250-IIs and an SP-11. A mere $17,000 worth of tube electronics. We were impressed, and using matching amplifiers top and bottom did make a positive improvement. (We hadn't had the opportunity to try it before.)—Larry Archibald