This review page is supported in part by the sponsors whose ad banners are displayed below
Very fast, extremely lucid and detailed. If we group all that under adrenaline—built for speed in booty slang—it's fair to wonder. What might some added comfort by way of curves and lusciousness accomplish? One perennial tightrope act is finding the right balance between attack mode (3rd-order sharpness, energy, PRaT, timing, impact) and bloom (2nd-order sweetness, mass, density, flow, relaxation). Too much speed and things get relentless and skeletal. Too much warmth and things turn mushy and gentrified. Everyone's idea of a good time is different. Still, chances are high that long-term satisfaction won't be an extreme but somewhere in the middle.


My tastes—here it gets purely personal—wanted a bit more weight and tone color intensity. For the Gladius speaker I also wanted more bottom-end power. I hit upon my ultimate combo with Alex Peychev's NWO-M modification of Esoteric's UX-1. Operated as a DAC fed from the iMac (outrage alert: or via a Pure digital dock and my 160GB iPod Classic) this ran into Dan Wright's LS-100 preamp. The latter's primary contribution was bass mass. This also worked towards the oft-cited blacker blacks in the color palette. A subtler secondary contribution was a mild shift from transient to bloom phase.


The 20 paralleled '32-bit' AKM 4399 DACs per channel connected "in an unusual configuration known only to us and Asahi Kasei" and terminated into a class A amorphous-core Lundahl transformer-coupled NOS E182CC triode buffer turned up the tone color temperatures. With this source plus what the amp already did on its own, I achieved triode color saturation without fuzziness or sacrificing any of the nearly crystallized clarity the SIT was so good at. Now I had a quite idealized single-ended triode sound without the usual compromises of control, noise floor, rise times, bandwidth and thick THD disguised as warmth.


SIT straight is usually good for the back. Here a deliberate relaxation via injection of mild tube fluffiness plus down-low assist from an active preamp that's proven unusually endowed in the nether regions before was even better. Usually mature systems are quite finely calibrated. Change anything too much and balance is off to require a counter adjustment. The SIT amp proved unusually tolerant of being weighed down. Think powerful sports car whose street performance doesn't notice the first passenger no matter the weight. I'm sure that the audiophile equivalent of a washing machine in the trunk would eventually turn the amp into something less exciting and brilliant. In this case however two preceding valves—E182CC in the source, half a 6SN7 in the pre—made the ride feel no slower. Just curvier.


That was unexpected. When playing in a hifi sandbox whose dimensions are dictated by budget (i.e. the designers of competing machines all suffer the same practical restraints), the variability of performance attributes tends to operate in a quite narrow window. You want mo' different usually means shopping in a different price class. The SIT amp broke that mold. Its particular advantages were more significant in scope. This reflects right back to handling two external valve stages (one buffer, one voltage gain) without blackening its own score card.


The NWO as the final iteration of progressive Esoteric universal player rebuilds is ultra refined and elegant. For my tastes it also errs a bit on the side of politeness. I'd thus usually mate it with my Bent Audio Tap-X autoformer passive to 'regain' some speed and edge. Conversely I'd usually match up my ModWright pre with a converter more incisive and sharp like the Metrum Acoustics NOS Mini DAC Octave or Antelope Audio Zodiac Gold. A small step to the right gets countered by a step to the left. One stays roughly in center aisle. With the SIT amp my scope of flavorization options expanded. That was seriously practical and fun.


It also segues right back from personal preferences to more universally relevant commentary. To put it basic, you can surround this amp with a broader variety of play mates than expected and it'll remain recognizable as described. Adding a deliberately light rather than deep triode preamp to my ears cloned the perfect single-ended triode amp playing into perfectly matched speakers - except that the speakers wouldn't have been considered perfect even had this been a really good 10-watt 300B amp. And the 300B amp would have been noisier, less extended on top, warmer, less resolved and blurrier/bloomier down low.


To be clear, to optimize such a 300B amp—I recently reviewed the CHF 28.000,- Swiss Colotube monos to have a very concrete example in mind—I'd mate it to my passive autoformer volume control or Esoteric's massive C-03 active pre set to zero voltage gain. Using the latter on the SIT amp produced a clearly cooler leaner more lit-up sound. Adding the ModWright LS-100 to the 300B monos would already get a bit thicker and softer than I'd want. On the SIT amp meanwhile it simply loosened up textures while adding bass oomph. The NWO-M player-as-DAC then intensified tone colors. This simulated the oft-cited SET amp effect of appearing to be illuminated from within. I think of it nearly as color pressurization. Timbres pop harder.


Neither of these valve machines is overt. Their thermionic contributions are very finely tailored and only carefully dosed. Whiffs rather than winds. It thus took very little to nudge the single-ended transistor SIT into single-ended triode sound - except that it was superior to what a real SET(riode) would deliver 9 out of 10 times.


SITups: I considered it a true privilege to be first and report on this amp in prototype form. As a valve lover who slowly returned to transistors whilst accompanying the FirstWatt output from model to model, my hifi tastes have quite effortlessly deglommed themselves from earlier tubular excesses. I now prefer a clearer more unadorned honest and less colored view. Even so I still want certain valve virtues. The most important ones would be maximized microdynamics, intense color richness and good image density. Solid-state amps tend to do things just a bit differently. Here upstream ancillaries compensate. In other areas transistors are superior to require no compensation.

With his SIT amp—first of many—Nelson Pass has narrowed the gap between transistors and triodes. Again. When it comes to single-stage circuit simplicity, this low-power no NFB amp arguably behaves more triode than most more complex valve amps. That it still doesn't quite sound like a triode per se is somewhat of a contestable statement. No amp or other hifi component sounds like anything by itself. We can only listen to complete systems. The moment we do, unknown interactions muddy the picture on what component is doing what and by how much. What it might sound like by itself is guess work by triangulation.


Following the J2, we have two single-ended Power Jfet designs of nearly identical gain though the J2 makes 25wpc. The SIT's performance lead—and by how much it over-shadowed the now discontinued F5—proves how real meaningful advances continue despite widespread and observant cynicism that much of high-end audio is really old wine in new bottles. The Pass SIT-1 from SemiSouth is new wine in an old bottle if a single-ended single-stage circuit without feedback is old. How others will respond I can't predict. But I can say categorically that with the advent of this amp and how cleverly it harnesses the new part, owners of advanced SETs have lost more (previously still good) reasons to ever turn to power tubes again. That's a carefully considered loaded statement. Boom!


SITdowns: The amp runs very hot. It only makes 10 watts with high output impedance. While my speakers and listening habits were perfectly happy with the power and gain on hand, if I could add one item on my Nelson wish list, I'd ask for a bit more brute LF force. Likely the mono version addresses that already.


Conclusion: In final form with an official model name and published retail price, this amp deserves a big fat award that would have to include aspects which normally don't factor. Since we don't have awards for prototypes or new audio transistors, you'll have to imagine one. Its byline pointing at a grave might read "the amp that killed off my power tube jones for good?" But not only that. More importantly, this amp deserves a big fat juicy thank you to Nelson Pass who invested serious personal funds and time to gift our community with a brand-new output device. Giants like Sony and Yamaha have explored this in the past. In 2011 it's taken a far smaller non-corporate guy to shoulder the risk and continue this exploration in the spirit of real innovation and 'predatory' curiosity. I understand that the man's interests in this field have made him into a present-day VFet magnet. In a private email Nelson reported that he's been contacted by various sellers of NOS transistors to have amassed quite the inventory by now. More data points make for more insight and understanding. That informs progress and R&D. Damn, 10 years later and FirstWatt's original kitchen table operation is really cooking. Who says you gotta slow down as you get older? Here's to speed and many more SITuations, Nelson. Cheers!


Going forward: "I failed to mention that this amp provides the best argument I've experienced for absolute phase. You know the drill - we flip the speaker phase and sometimes we hear some difference and occasionally it actually sounds better one way than the other. Try it with this amp and let me know what you get. I have a nice monoblock prototype in the works with a little more power but it is more unique in that the SIT is the only semiconductor - the bias elements and such are all passive. This piece is designed to offer front-panel load-line adjustment so that you can tweak the amplifier more perfectly to mate with a given loudspeaker - or alternatively, adjust the harmonic character. It will also have a meter showing the operating point. The thing with going totally passive except the single device has a strong attraction in that it takes the sound down just to the one part as much as possible. The biggest downside is that the efficiency drops even more. The S2 you've got is about 12% efficient, the monoblock is about 6%.

"There's more of this sort of thing in the works. With the XP30 preamp Wayne catered to a couple requests - dual sets of outputs independently adjustable for biamping, much lower output impedance and much higher gain options. The latter two make it an excellent front end for an amplifier, providing all I need for various follower output stages. Now I can just cobble them together and listen apples to apples. It's very interesting and illuminating to isolate the specifics of different devices, topologies and so on and trying to correlate them to the subjective side of things. So far there have been a few surprises and I'm hoping there will be more but it's terribly complicated.

"One thing that has helped is that we are now running four sets of Tannoy 15" HPD drivers in Jensen Imperial enclosures all with Joe's special crossovers. They are extremely revealing of the phenomena of interest and this puts our listeners more on the same page so the turn-around time for subjective testing is greatly speeded up. Bigger power is definitely for the future. The Yamaha 2SK77 parts are rated at 300 watts, which would definitely get me into the 75 watt @ 8 category à la Digital Do Main but without the multiple stages, feedback and degeneration. Pricing and availability of that part remain to be seen. Alternatively I can start paralleling SIT-1s but realistically First Watt is not likely to go beyond about 25 watts. Far more probable is that you will see some of these things embedded in the next generation of Pass Labs. When they heard what you heard, my partners acquired a particular gleam in their eyes." - np

Postscript.

FirstWatt website