There is a riddle wrapped in an enigma: what to do with the feedback control? It’s not quite as simple as most reviewers, or audio gurus, say it is. “Blah blah feedback is the Devil’s work, blah blah”, said with the same confidence as the guy from Western Union who said: “The telephone has too many shortcomings to be seriously considered as a means of communication. The device is inherently of no value to us”. The feedback control on the 3.1 can be used as a tuning tool to tame unruly recordings, bright sources, difficult loads, and aging tubes.
In practice, I preferred less feedback on analog or very dynamic recordings. I preferred more feedback on loud, compressed music and digital. On good analog, less feedback allowed the high frequencies to come through with better definition. On bass-heavy popular music, or electronica, higher feedback sounded better.
Instead of viewing the feedback control as an evil, view it like the traction control setting in a car, or a four-wheel-drive system. It would be better if you weren’t driving on ice, but you can’t make the ice go away. The feedback control, like a car’s traction control, is a valuable tool for less-then-perfect situations, and many recordings have serious issues. If you have a difficult speaker load for the 3.1, the extra feedback will definitely help bass definition.
Sound and Fury!
After sorting out the tube issues, I got down to serious listening. Depending on how the feedback level was set, and which tubes were chosen, the sound can be changed quite radically. Therefore, I tried a combination of tubes and feedback settings.
With low feedback, and on very dynamic tracks, vinyl or digital, the 3.1 proved to be very dynamic, especially for an integrated amp. The dynamics were very fast, too. In contrast, most solid-state amps seem to be sluggish. Whether because it uses the 300B, which is a dynamic tube under most circumstances, or because it is push-pull, or because of the 6sn7 driver, or a combination of all these, it could produce dramatic impact from the bass all the way to inaudibility. For many, including me, good dynamics are a must. The biggest difference between live and recorded music has always been dynamic range. Given moderately efficient speakers, the 21 watts feels more like 50 on dynamic music. With higher feedback engaged, the sound lost impact, though it was still better than most tube amps.
The tonal balance, with feedback set to low, is slightly forward, but not noticeably bright. It wasn’t objectionable at all. “Sparkly and clean” showed up in my notes. The tone was “silvery” on a few recordings, using the Telefunkens. I preferred the tonal balance of the Audio Space to that of the Luxman SQ38u, which has a very classic tube sound. Mullard 12ax7s, some good used ones I had, were the cure that gave balanced tone. The Telefunken pushed things a little brighter. RCA and Sylvania were even-handed though they had less midrange beauty than the Mullards. I did not try every tube I have, and some of the examples were old used tubes, not NOS. Please don’t take my tube comments as gospel because your choice will be influenced by speakers, room, and associated equipment. As I have already said, just about any good tube made in the ’50 through ‘70s, by reputable manufactures, will sound excellent in the 3.1.
The tonal balance with feedback set “HIGH” was slightly darker, with more “classic” tube sound, but with much better resolution and dynamics than any classic gear I’ve used. Where tube rolling produced big differences when using the low feedback setting on the 3.1, it had much more consistent tone with higher feedback. I could still hear differences, but not as obvious.
Image width and depth were excellent. I heard remarkably wide sound stages, with images well outside the speakers. The feedback control and tube rolling affected imaging some, but not to the same degree as dynamics or tonal balance. This amp specializes in detail, and the result is first class imaging. It wasn’t artificial-sounding either. The images were organic and palpable. I’ve heard some better, but not in an integrated amp.
An area where the 3.1 shines (no pun intended) is detail retrieval. It is a simple circuit, with nothing extra to destroy information. It’s largely because of this, that micro-dynamics were unusually good, as was imaging. It was easier to understand lyrics with the 3.1 than the average amplifier. Across the board, I didn’t feel like information was traded in for romanticism.
A massive criticism I have about high-end components is how they eliminate musical choices: they make your record collection smaller. The biggest compliment I can pay to the 3.1 is that I never said to myself “this amp doesn’t have enough power,” or “this amp doesn’t have enough deep bass,” or “this amp can’t handle complex music,” or “this amp exacerbates weaknesses of recordings.” With careful tube rolling, and through use of the feedback control, you can tune your system to be as critical, or forgiving, as you want. This wasn’t with horn speakers in a small room; it was while driving large line sources with a 6 ohm load, nominally 93dB efficient, into a large room. This amp operates like an elite team of Special Forces: it gets in, gets the job done, and gets back out before anyone knows what happened.
I didn’t spend that much time with the phono stage. It is very good, and similar sounding to other units I’ve auditioned that used the 12ax7. It’s very good, but not reference class. Unfortunately for our wallets, phono stages almost always sound better when they have their own purpose built enclosure and power supply. But, this is a good phono stage for a full-featured integrated amp, and better than most budget phono stages.
Nits to Pick
I wish the combination BIAS/SIGNAL level meter could be deactivated. Your options are tubes 1, 2, 3 or 4, left channel level and right channel level. It’s my understanding that you shouldn’t leave a bias meter in the circuit while playing music because it affects the stability of the bias. I might be wrong about that. Either way, it seems preferable to leave it set as a signal indicator, not in bias-checking mode. It’s a welcome feature for setting bias, but as a signal level device, it’s incredibly slow. It sorta moves around with the flow of the music, but always a split second slower than whatever the music is doing, kind of like a dork who’s at a concert pretending to know all the tunes. If it were my amp, I’d figure out how to wire the meter such that it’s off when you weren’t checking bias. All that being said, it’s a welcome feature for setting bias, and preferable to using a digital multimeter.
Another obvious issue is the lack of remote volume control. The amp is packed full of features and parts, so a motorized volume control wasn’t possible. Still, it would be nice to have one.
Overall, I find the Reference 3.1 to be a well-thought out and well-built unit, packed with all the features you would need to enjoy your music collection. It has ample power to drive moderately difficult loads to adequately loud levels. Frequency response is much better than single-ended units of similar cost and size. Distortion is vanishingly low on most music, only becoming noticeable at the absolute limits of its output power, something you are unlikely to encounter on a regular basis. Dimensionality is excellent, as good as or better than most integrated amps I’ve tested. And, transparency is what you would expect of a 300B amp: excellent!
If you like triode amps, but object to the power compression and coloration like I do, this amp deserves consideration. It’s an easy recommendation.
- ← Previous page
- (Page 2 of 2)