Integrated Amp Reviews

On program material, though, the tone controls were a pleasure to use. Advance the bass control to a bit past 1 o'clock, and the only thing you notice is a strengthening of extreme low end; the deepest material gets louder, but nothing else is affected. Open it up to about 3 o'clock, and the whole bass range comes up, while the deep material gets stronger still. Only beyond this point does the sound, typically, start to get boomy, and if it happened to be bass-shy to begin with, the result would be a pretty good job of correction. Similarly, at moderate cut settings, only the extreme bottom is affected. Deep rumble, tubbiness due to hall resonances, and even moderate amounts of hum can be reduced without a drastic effect on the overall sound.

We found the treble controls a little less useful. The musical overtone range can be very nicely controlled, but material with inadequate or excessive brilliance could not be adjusted properly without overdoing the correction in the higher ranges. But since most problems in current program material do occur in the overtone range, there are more things the AR's treble control will do than things it won't.

Volume control tracking in our sample AR amp was exemplary. There was no more than 0.75dB of tracking error through the control's entire range, and just a hair over 0.5dB of error through the normal operating range. The control's action was smooth and even (as was the case with the tone controls, too).

Bass performance was extremely good—deep, tight and solid. We'll admit that we were a little surprised that the AR amplifier's low end sounded as detailed as it did, for some experiments we conducted a number of years ago had shown that insertion of fuses in series with loudspeaker lines caused an audible loss of amplifier damping; the low end would become somewhat heavy and flabby.

Just to satisfy our curiosity, we replaced each of the output fuses in the AR amp with an identical length of copper tubing, and listened for a change in the low end. If there was any at all, we couldn't hear it, so we put the fuses back and forgot about them, which is what we advise anyone else to do, too. The penny-in-the-fuse gambit has never been a smart idea, and it isn't here, either. Fuses are installed for a purpose, and too many people have found out the hard way what that purpose is. Evidently, they serve their purpose in the AR amplifier without upsetting anything else.

In our report on the Dynaco PAT-4 preamplifier, we noted that although the measured phono preamp equalization agreed almost perfectly with the RIAA-standard curve, disc reproduction sounded somewhat sparse at the low end, in comparison with the sound of high-quality tapes feeding high-level inputs. Well, the AR's measured equalization was very slightly down at the low end, yet its phono reproduction sounded almost exactly like the comparison tapes. We would just love to be able to offer a simple explanation for this, but again, all we can do is shrug our shoulders, etc, etc.

Specifically, here's how the AR amplifier sounded on the three loudspeaker systems we used for the bulk of our tests:

With the Dyna A-25s, the sound was rather heavy and markedly dull at the top. These speakers are normally a shade soft at the high end, but they were noticeably more so with the AR amp, although turning their tweeters up a notch or two and helping them along a bit with the AR's treble controls allowed us to get very natural, balanced sound from them.

With the Janszen Z-600s, we found somewhat the same thing: a bit heavy overall and a bit short of brilliance, but here there was also a very subtle aura of "zizz" at the extreme top. In this case, there was little that could be done to improve matters, as the speakers have no tweeter-level controls, and the amplifier's treble control emphasized the "zizz" too much before it began to have any appreciable effect on the brilliance of the sound.

Vith the KLH Nine, the "zizz" at the extreme top was quite pronounced, and proved very annoying to listeners whose hearing range went appreciably beyond 12kHz, although variations in the degree and the apparent frequency of the "zizz" from one recording to another raised some speculation that what we were hearing might be high-end resonance peaks in the recording microphones and disc cutter-heads, reproduced somewhat more prominently than usual.

If this is the case, there is really no way of knowing whether the AR amplifier was tending to exaggerate the audibility of these peaks or whether it was actually reproducing them through the Nine in their proper relationship to the rest of the program material, in which case, amplifiers that do not reproduce this "zizz" as prominently would not be feeding as accurate a signal to the Nine as the AR amplifier. It's an interesting question, but a rather academic one in terms of subjective results. Wherever lies the fault, there is no doubt but that amplifiers which yield a softer top from the Nine make it sound more musically listenable than does the AR amplifier.

In all other respects, though, the AR amplifier's performance with a single pair of KLH Nine panels was superb. When the two panels of the Nine are separated, for a wider stereo "stage," their combined response often tends to be rather thin at the bottom and tipped up at the top. Careful experimentation with placement will usually remedy the situation, but the ultimate location may not be ideal in terms of stereo imaging or center fill. As it turned out, the AR amplifier's characteristic sound was almost perfectly complementary to that of a separated-panel Model Nine system, resulting in better overall balance than is usually obtained from other amplifiers.

Like all other solid-state amps that we've encountered, the AR unit is designed for loudspeakers of 4 to 8 ohms impedance, and loses almost half of its 8 ohm power when connected to a nominally 16 ohm load like a Model Nine panel. But unlike previous mismatched solid-state amplifiers we have tried with the Nines, which sounded on the verge of overload much of the time, the AR amplifier sounded quite "comfortable" with the Nines, and when brief overloads did occur, the resulting clicks were noticeably less offensive than usual. All of which makes the high-end "zizz" even more disappointing.

Listening with AR-3As
After we completed our main round of tests, we decided that it might be nice to answer what seemed to us (and would to most readers, too) a very logical question: namely, How would the AR amplifier sound with AR speakers?

For this test, we borrowed a pair of AR's best speakers, AR-3As, and tried them in several different locations in three rooms. Not surprisingly, the AR amplifier affected them in much the same way as it had affected the Dyna A-25s—with their tweeter balance controls set for what sounded flattest on most other top-notch amplifiers, the AR amplifier made them sound somewhat more bottoms-up and tops-down. We were able to get better overall correction with the AR-3A's two balance controls than with the A-25's one control, but it struck us that more bottom was one of the last things that AR-3As need.

Testing the Protection
Finally, we got around to testing the AR's fail-safe provisions. We specifically avoided doing the one thing that AR admits will blow things apart: shorting together the two Hot loudspeaker terminals. But we did try just about everything else we could think of, including shorting the outputs and then yanking the phono leads from their receptacles with the gain turned up full. We were not able to pop any transistors, although we did go through several of the "speaker-protection" fuses. We will admit, though, that we did not try to find out whether these would actually protect our standard test loudspeakers from burnout. We'll take AR's word for that.

Conclusions
Whether or not the particular sound of the AR amplifier is worth your consideration depends on the loudspeakers you use it with, the way they behave in your listening room, and your own personal taste for reproduced sound. For example, we have encountered cases where no amount of careful placement of loudspeakers could make them sound anything but thin, and it was not always the fault of the speakers, either. Tweeter balance controls are rarely able to correct a condition like this, and very few tone controls can do it either. But the peculiar behaviour of the AR amplifier did solve the problem very nicely in two such instances.

So, just how good is the AR amplifier? Better, we would estimate, than about 95% of the current competition. But beyond that, we are going to have to hedge a bit, because specific comparisons tend to become somewhat "iffy." For example, the AR amplifier is price-competitive with the Dynaco PAT-4/Stereo 120 combination, which sounds a shade sweeter at the top, has equal transparency, and more natural overall balance, except on phono reproduction, which is thin. And the PAT-4's tone control action is as lousy as the AR's is good. A Stereo 120 and PAS-3X preamp will give fine low end from phono too, but the tone controls are almost as bad as in the PAT-4, and the sound is not as clean nor lucid as that of the AR, although the overall cost is a bit less.

On the other hand, if there are any similarly-priced amp/preamp combinations that sound as good as either the Dynas or the AR, we have yet to find them. As to the choice between them, though, it should be obvious by now that there's no clearcut winner. A buyer who's considering something in this price range must decide for himself, based on whichever aspects of amplifier performance—balance, transparency, tone control action, etc.—he deems to be of most importance to him.