Thiel CS2 loudspeaker Larry Archibald page 2
AHC's original description of the CS2 is accurate, as far as it goes: neutral tonal balance, very deep bass considering size and price (–6dB at 43Hz in JA's measurement), an intimate, tactile midrange, a flat and smooth high-frequency response. I would agree wholly with this assessment; the highs themselves have only been a problem when the source material or driving amplifiers have been flawed.
But I have to add that there is, in my rosewood samples, an aberration in the 1–6kHz range—where, as you probably know, our hearing is most sensitive. The most expressive way of describing this problem was coined by JGH, when visiting my old house (where the CS2s were set up): "From in here, they sound live." The speakers were playing in the living room, we were sitting in the dining room and separated from the speakers by a 16"-thick adobe wall with a door in it. The unfortunate part was that in the listening room the sound generally was too aggressive; it did not sound live. This same response has occurred in my current house, except that the dining room has a 6' opening into the living room; the degree of discomfort with the CS2s has been correspondingly greater.
I'm certain this is the problem referred to by our reviewers, but let me emphasize that this is extremely equipment-dependent, and also has a lot to do with your actual listening position. I have had many different kinds of electronics, speakers, turntables, and CD players in my houses over the years that I've listened to the CS2s. Any product with an aggravating upper-midrange or high-frequency characteristic (the highs are smooth, but they are flat; the speakers will not tolerate problems in this region) made for an immediately unfavorable match. Frequently, even products that came out neutral in other listening situations would not show well with the Thiels. The best components I've found for use with the Thiels are those listed at the beginning of this review.
Even so, I found that LP listening could be excellent, with very well recorded music, and tolerable with that which was less well recorded. CD, however—even CDs that I regard as excellent, such as the Chesky Sibelius 2—ranged from discomfiting to unbearable. This characteristic was alleviated significantly with the CAL Tempest II, but I think it's hardly fair to recommend a speaker on which CD can be auditioned only with a $2200 CD player which, even given CAL's success, can be owned by only a tiny number of prospective speaker purchasers. The Denon combination (DCD-3300/DAP-5500) is much more representative of what's out there, and it is actually superior to a large majority of the machines. On it, I simply didn't want to listen to the CS2s (and, though it's inferior to the CAL, the Denon was quite acceptable with the Magnepan MG2.5/Rs or Spica Angeluses).
What was the problem? Basically, if you take the characteristic brightness of most CDs, and act as if these older CS2s had a special sensitivity to that brightness, you've got it. It was as if the speaker "took off" at those frequencies, making even excellent CDs unpleasant if not played back on the best, most forgiving machines.
After long and careful audition, without knowing that current production was in any way different, I concluded that the Thiel CS2, in terms of its tonal balance, could be recommended, but only in quite specific circumstances: with forgiving electronics (the Levinsons are as good as any I know of in this area, rivaled by the VTL amps) playing LPs, particularly using a moving-coil cartridge with the ubiquitous upper-midrange suckout (but not the ubiquitous rising high treble). I was disappointed to come to this conclusion, given the speaker's other virtues; it is not a recommendation that could have survived weeding-out for the next "Recommended Components."
Fortunately, this problem with the CS2 must not have been unknown to the people at Thiel, though they have never mentioned to me that ongoing production changes, which occur with all manufacturers, would have been aimed at it. Nor would I say that the changes wrought with my newest samples (finished in a black plastic laminate—similar to high-quality Formica—that is a dead ringer for black lacquer) could justify a different model number or "point 2" (though I've heard that might be in the offing for the CS2, probably sometime late in 1989 or early 1990).
Rather, the change that has been made—which somehow fails to change the overall character of the speaker while at the same time making it much easier to live with—converts my judgment of the CS2 from a reluctant thumbs down to a cautious thumbs up. It is still necessary to use electronics that fall into the forgiving camp, though you can probably now spend much less money. I haven't tried it, but I suspect even a lowly (though highly recommendable) B&K ST-140 would do. I still recommend using the best CD player you can buy, and high-frequency grundge must be avoided at all costs. But now, the best CDs can sound fabulous, with no penalty on LP. The speaker's other virtues (see below) now predominate. One is aware that the upper midrange is not to be abused or trifled with—sort of like a placid lion in the wild—but it can be overlooked and the music enjoyed. I will be interested to know what was done to accomplish this subtle but very important improvement.
I should also mention listening position. Most often, just the right listening position has its primary importance with respect to soundstage and image correctness. With the CS2, the correct listening position also has a dramatic impact on spectral balance. For two reasons, I think: first, the Thiel's off-axis high-frequency response is almost as flat as the on-axis,$s1 resulting, in any reasonably live listening environment—such as mine—in a lot of reflected high frequency in the room. Being in other than the correct listening position, where direct radiation from the loudspeaker dominates, results in a notably brighter sound. Second, the Thiels, more than other speakers, depend on their whole package of virtues to make their case. The Thiel tonal balance, sans the Thiel achievements in imaging specificity and soundstage coherence, fails to convince.
I also found the grille important for ameliorating the CS2's tonal balance "problem"; it may be that part of my difficult reception for the rosewood CS2s was due to their predominant lack of grilles. As an experiment, I did some listening to the newest sample without their grilles. Sure enough, back came that extra sharpness and, to some degree, the intolerance. But only halfway—the old CS2s still had substantially more of the characteristic. With the best recordings, the lack of a grille actually improves the sound (though I'm sure the people at Thiel will not agree with this): There's less between you and the music, the speaker is more there, the sound more vivid. Considering your whole record collection, though, you will definitely want to leave the grilles on. Overall, their presence is so positive that I don't even recommend you experiment.
I'm afraid this discussion has focused overlong on only one characteristic of the Thiel's tonal balance. There are many other virtues to be heard here. The bass, though bass reflex, is well controlled and extends to satisfyingly low frequencies, only when severely strained giving a hint of looseness, say, in comparison with the Thiel CS3.5, which is non-pareil in this area. On organ, kick drum, synthesizer, and the deepest orchestral fundamentals, you will certainly notice a lack of immense power and underpinning, but this will not show up as a lack of fullness on bass guitar, male voice, tympani, or other than quite low orchestral fundamentals. The bass will play reasonably loud, with a lesser sense of strain than the competitors I surveyed. The Spica Angelus is not unsatisfying, but conveys less rollicking enthusiasm; the Magnepan '2.5 is big-sounding but does not go as deep, and compresses a bit at high levels. Still, with the CS2 you do well to pay attention if you begin to sense strain, particularly if you listen in a large room such as mine. The woofers can give out, but even more sensitive, particularly to upper-bass and lower-midrange transients, is the midrange driver, probably due to the 6dB/octave crossover slopes. I managed to do in a number of midranges (though this was partially due to a problem Thiel had with the production they received from their supplier; this problem has been solved for some time now).
More important than this individuation of tonal-balance areas—bass, midrange, upper midrange, highs—is the fact that the CS2 integrates the sounds it produces extremely well, as noted by AHC in his original review. All speaker designers face the problem that achieving excellence with one driver can frequently make the speaker sound worse overall due to dissimilarities between inherent driver sound or radiation patterns. I feel that Jim Thiel is right at the top of a small group of designers—I would include John Bau, Richard Vandersteen, Jim Winey, and Robin Marshall in this group—at getting his products, all of them, to sonically cohere. It takes hard work, perseverance, and a bit of genius.
Soundstage and Ambience: This is where the CS2 really shines. Except for a somewhat inferior capability at defining the overall size and character of the listening hall, it is every bit the equal of its larger brother, the CS3.5—which is itself as good as the best I've heard (with the exception of much, much more expensive products like the IRS Beta and WAMM, which, again, excel in their definition of the space in which the performance takes place). In the very important area of individual instrumental ambience—the "pools" of space within which each instrument performs, and which is audible on good recordings reproduced well—the CS2 excels. It may be the best I've heard in this respect, bettering its smaller brother, the CS1.2, as well as another speaker whose preservation of instrumental specificity is extraordinary: the Spica Angelus.
Some day Stereophile will have instruments other than our ears that can measure phase coherence; until then we will have to rely on those ears. To mine, the Thiel CS2 is astonishingly good in this respect. Both individual instruments and the performing group as a whole hang together with remarkable conviction. As pointed out above, other speakers better portray the hall in which the group is playing—I suspect because of deeper bass response which lacks the phase disruption near resonance of a bass reflex design—but the CS2 is damned good. There are not too many speakers at any price which surpass it.
In fact, I would say that, now that the newer CS2s have largely conquered their insensitivity to impure upper midrange phenomena in the preceding chain, one is freed to simply pay attention to the instruments and singers that are playing—almost as if they were right there in the room! (Somewhere back when I became enthusiastic about hi-fi around 1961, I remember thinking that was what this hobby was all about.)
Dynamics: The dynamics of live music is the singular area where generally little progress has been made in high end, outside of heroic—and vastly expensive—efforts like the Wilson WAMMS, the Duntech Sovereigns, and the various IRSes, all of which are able to easily deliver SPLs that, if not literally reproducing orchestral sensations, begin to overwhelm you in the same way. In fact, a convincing case could be made that the huge woofers of the 1950s and the greater prevalence of horn-loaded systems in those days made that time a better one in terms of imitating the dynamics of live music. I am certain this is why the Klipschorn maintains such loyalty, even among readers of this magazine.
What has changed is the amount of coloration that must be suffered at lesser levels of dynamic performance. The Thiel CS2 doesn't break any new barriers in this area, but is able to preserve its sound character almost unaltered within a significant dynamic range. Many speakers change significantly in clarity (usually improves up to a point) and tonal balance (gets steadily brighter with increasing SPL) in the output area of 60–95dB. Within these relatively conservative limits, live music does not. (Above 95dB, it's common for your ears to experience both compression, ringing, and a kind of splattering effect that begins to increase your level of tension with increasing SPL.) The CS2 is one of the best I've heard at maintaining its character within these limits. It will play louder than 95dB, but then you are aware of both compression and some disorder in the sound; your usual response is to turn down the volume. The range I've specified, though, makes it almost ideal for chamber groups, small numbers of voices, most jazz, and pop or rock if not played at loud party levels. It's fine on orchestra, kick drum, and loud rock, but you have to settle for sub-live music levels. This is no insignificant accomplishment; in my experience you have to go to a substantially larger speaker to do better.
Overall, my evaluation of the CS2 has changed substantially on listening to their new samples. I used to see it as a model with many attractions but flawed by the difficulty of making it sound easily natural. Now it is a speaker whose attractions—primarily a neutral and very well-integrated tonal balance and very precise retention of image size and individual ambience—can be realized quite readily, but only if you're willing to take care in setup and selection of associated components, and have the interest to really pay attention while listening.—Larry Archibald