Down-scaling - as opposed to dumbing down - has a patchy record in the annals of luxury goods. The VW/Porsche 914 didn't satisfy 911 wannabees, few would-be Leica M-Series owners cherished the economical CL and you'd have to hypnotise an Italian before he'd accept a Tudor wristwatch as a Rolex.
Read more Audio Research Tube Preamp Reviews from HomeTheaterReview.com
In hi-fi, though, such problems rarely materialise, because hi-fi companies have never shied away from using the same styling and the same series model numbering or nomenclature from least to most. Hence, no Linn dealer will lose sales of the separates just because the company issued the Classik, and a Krell KAV300i integrated amplifier can be demonstrated in the same room as the pukka FPB separates. But what do we make of Audio Research 'popularising' the Reference 1?
Over two years ago, the circa £9000 Reference 1 became, for my money, the pre-amp by which all others should be judged. Aside from the whinging of the handful of morons who couldn't understand that the controls weren't rotary devices per se, and that they worked in twist-twist fashion, I know of no criticisms worth dignifying...and any criticisms I heard levelled at the Ref 1 invariably came from manufacturers of solid-state rivals. Alas, the price alone ensured that this pre-amplifier would remain the sort of device only accessed by the very well-heeled, and the rest of us could only drool in envy.
While I would hesitate to call the near-halving of the price a 'bargain', I must regard the scaled-down, £4999 LS25 something of a gift for those of us who crave but cannot afford Ref 1. In one fell swoop, the LS25 doubles our chances of acquiring the ARC flagship. Why? Because I'm damned if I can decipher a sacrifice.
Referring back to the review of the Ref 1, you'll recall that it is a no-compromise line-level-only preamplifier with microprocessor control of volume, source selection and balance, all of this taking place outside of the signal path. The microprocessor also provides memory so the source selector remembers whether you're using balanced or single-ended inputs, necessary because its eight inputs carried duplicated RCAs and XLRs, to be used in either/or fashion.
It was the Reference 1 which introduced the spring-loaded, self-centring intermittent controls like those on the Copland CD player and the Acurus ACT1 processor/pre-amp. They were notable for their tiny operational arcs which you followed left or right to raise or lower the volume, alter the balance or select sources. Holding them scrolled through the entire arc, indicated by green LEDs. Then again, the Ref 1 came with remote control, which provided every single function bar the choice between balanced and unbalanced input selection.
Across the front, the Ref 1 carried gain, balance, record out and source selector rotaries, while the trademark channel below contained press-press toggles with the same feel as the knobs. They self-centred, so you pushed them up or down to choose on/off, balanced or single-ended input, normal or inverted polarity and mute/operate, with more green LEDs telling you what was chosen.
Across the back were the phono and XLR inputs; in addition to those eight sets were three pairs of outputs, one for record and two for main, also in XLR and phono form. Lastly, there were easily accessible fuses, and a captive mains leads.
As if to remind you why you paid the price of a VW Polo for a pre-amp, the Ref 1 weighed 30lb, featured a 3/8in thick aluminium front panel and employed eight Sovtek 6922 dual triodes. Tube regulation was solid-state for maximum silence, via a combination of MOSFETs and JFETs. Part of the weight was due to the inclusion of three separate toroidal power supplies, one each for DC regulation, high voltages and the digital section. Discrete components were used throughout.
21 LEDs surrounded the volume control, with steps in-between providing in excess of 156 gradations at 0.3dB per step. Switch-on was accompanied by a 45 second delay, to ensure that no nasties reached one's speakers. Specs were impressive, too: frequency response is +/-0.5dB from 1Hz to 200kHz, with -3dB points at 0.3Hz and better than 400kHz, distortion less than 0.01 percent at 2V RMS output. Gain was 12.3dB balanced, 6.3dB unbalanced, tape output is 0dB, and the input impedances 220k ohm balanced and 110k ohms unbalanced.
Now, here's what differs - or, rather, what you sacrifice for a four grand savings.
Described by ARC as 'a slightly smaller REF 1' because it enjoys similar chassis construction, it also features electronically-controlled rotary controls for user-selectable overall gain setting (6, 12, or 18 dB), volume, balance and input selection, two-way toggle switches for power on/off, stereo/mono, balanced or single-ended operation, true A-V pass-through processor loop, tape monitor and mute. Seven inputs (including tape monitor) and two sets of main outputs are fitted, with on-board memory storing all settings along with previously selected gain levels. There's a remote control, four (rather than eight) 6922 twin triodes supplying the gain, the tubes are mechanically damped by the new proprietary clear polymer rings, the massive power supplies are fully regulated and InfiniCap coupling caps are fitted at the output.
By my reckoning, we've lost half the tubes and weight is down by 12lb, but the LS25 actually on the Ref 1's flexibility by letting you choose your own AC cable through an IEC input. Specs? Compare these with the above: frequency response is +/-0.5dB from 1Hz to 100kHz, with -3dB points below 0.2Hz and better than 400kHz, distortion less than 0.01 percent at 2V RMS output. Gain is selectable, tape output is 0dB, and the input impedances 120k ohm balanced and 60k ohms unbalanced. Other differences include a lowering of the volume controls steps to 104 and the front panel layout differs slightly according to the functional changes. But from two metres away, it looks like a Ref 1, it acts like a Ref 1 and it feels like a Ref 1. Feeling a bit Clintonian, I stopped at taste and smell.
I've now heard the LS25 in something like 15 combinations - balanced and single-ended, driving tubes and trannies and with a variety of sources. And while the teensiest part of me reckons that the Ref 1 has the edge for overall dynamic capability and sheer presence, the LS25 does not go the way of the pretenders listed in the opening paragraph. Quite clearly, the LS25 is a pre-amp for which no apologies need ever be made. And, if ARC and its agents can get their shit together, it could - along with that budget beauty, the LS8, and the current run of VT power amps - herald a new era, returning ARC to tube amp supremacy. It's as if the time is right for an anti-SET backlash.
Auditioned mainly with source components including the hot-rodded Marantz CD94, Theta's DaViD DVD transport and the Theta Chroma DAC, the LS25 was used to drive amps including the Talks, Sutherland's 2000s and a couple of GRAAFs into Aliante's new Pininfarina speaker, Quad ESL63s and New Audio Frontiers' Reference One S. Cables included the new Kimbers, ART speaker cable, Steve Rochlin's digital wire and Musical Fidelity.
Maybe I haven't yet reached the final stage: even a month of constant use seemed not to put an end to the LS25 run-in period. Either that, or my hearing improved. Whatever the cause, the LS25's sound just seemed to get better and better on a daily basis. Then again, it could be my memory playing tricks...
By the last few days before deadline, I'd grown to cherish certain qualities which I thought I'd never hear in a pre-amp of lesser stature than the Ref 1. Using lots of recordings of Capitol-era-Sinatra calibre (Dino, Nat, Julie et al), as well as all of the audiophile CDs I'd reviewed this month, I was able to detect - in addition to the trademark ARC sound - a number of refinements which could make this the pre-amp to seduce solid-staters away from their trannies...if the Ref 1 hadn't already. The main gains which place this pre-amp so high on my wants list are its resolution of the finest details, its Vistavision soundstage and - above all - treble so sweet that you'll suspect high calorific content.
Read more on page 2