Benchmark has released a new headphone amplifier that doubles as a stereo pre-amp. The concept behind the volume-control portion is not dissimilar to a stepped attenuator, just done with relays: "Features features relay gain control... The gain control has 256 steps in 0.5 dB increments. Relay closures are precisely timed to deliver silky-smooth volume changes. ... Includes 4 independent 256-step attenuators: two for the L&R headphone outputs, and two for the L&R line outputs. [for] total of 64 precision relays." Which means that eventually this thing is going to crap out when the relays fail, but that should be a good long while, and in the meantime it should have better channel matching than most pots and more uniform impedance presentation to the source. But all that fancy business surrounds the volume control. What really matters to me is that everything is going through active amplification circuitry. What does the amplifier sound like is the key question. It cost $3000, so it had better sound fantastic. Before I give my views, I'll note that some SBAF members have heard this at shows at like it. @neogeosnk heard this at CanJam SoCal and writes: best amp at canjam. Very transparent yet meaty. Just ordered mine. @m17xr2b heard this at Canjam London and writes: With the HD650 I was really surprised how good the HPA4 is, sounded very similar to my daily amp the Peak/Volcano with my reference tube but not as extended in the bass. Great detail and glorious midrange, I wouldn't have said it's solid state since natural and clean is rarely found there. The HPA4 does a beautiful job on the HD800S, high end is tamed nicely, similar detail to my modded HD800, low end had lean but very well controlled impact, typical well driven HD800. Meze Empyrean...with Benchmark HPA4 and dac3 hgc, unfamiliar but quality recordings these sounded great, bass stands out a controlled extended and good weight @SPAZ told me via email: I've been going back and forth between the mjolnir2 and hpa4. The differences are not that discernible. As of right now I have only done testing with Utopia, which isn't power hungry. OK - those sound positive. So, I purchased this thing and demo'ed it. I also gave it to @porkfriedpork, who took it home to demo on his own system. Here are some impressions. Physical It has a nice footprint—compact and fits on top of the DAC3 if you have one. But it's kind of an ugly thing. The bright TFT display is totally charmless. It's boxy and sort of lame. It's compatible with the Benchmark remotes, which I happen to really like—although their LED is not that powerful so not great if you're trying to use the remote at great distance. Not a big deal, but it has a thing where it mutes the output of headphone jacks when they are unplugged, so when you reconnect a new pair, it takes a second or so for the sound to come on. This makes the amp hard to use for A/B testing. Sound With HEK: I did side-by-side comparisons with Benchmark HPA-2 (the amp that is built into the Benchmark DAC3 HGC), with Leptoni DT-Alpha (a $4K tube amplifier), with my refurbished Pioneer M22, and my Pass Labs XA30.8. These were all done using Benchmark DAC3 as the source. HPA4 versus Benchmark HPA2: Both of these amplifiers are quite competent solid state devices. HPA4 is a noticeable upgrade. It has a very slightly thicker sound with less of the HPA2's edgy quality. The edgy quality of the HPA2 is really annoying, and is why I initially transitioned to tube amps—it was driving me up the wall. If I had the HPA4, I might not have made that transition. The HPA4 also had more detail. Not gobs more, but a touch more. That's an achievement, because third-order distortion often creates the impression of detail, but at the cost of fatiguing edginess—this indicated to me that HPA4 really was a better (lower distortion) amplifier than HPA2—it wasn't just aural trickery. That advantage comes with a cost, however: the presentation seems more distant. Verdict: HPA4 > HPA2, but .... is it worth the extra $3,000? In my opinion, no. HPA4 versus Leptoni DT-Alpha: These are not really comparable amps, one is solid-state, the other tube. The Leptoni (or the loaner, which has a very carefully selected group of tubes) is a very colored amp with loads of warmth. I found it was able to drive my HEK very well. It has nice bass and good high-end. However, it did have an unclean quality to the bass. Leptoni's sound signature (or at least that of the tubes with the Loaner) causes the chest voice of male vocals to disappear behind their headvoice—which is consistent of other reports of carved-out mids. By contrast, HPA4's response is totally flat. Needless to say, HPA4 is also more detailed than Leptoni. So, HPA4 is a technically more capable amp than Leptoni. But, which would I choose given the chance to have one for free? I totally take the Leptoni. Much more fun. Verdict: Draw. Technically HPA4 is better, but I prefer the Leptoni. HPA4 versus Pioneer M22 & Pass Labs XA30.8: No contest. HPA4 loses to both on all fronts when tested with HEK. With HE-6: HPA4 cannot power HE-6. It sounds like the HE-6 was killing the thing. It produced a terrible, withered sound. It was so incredibly bad that I just unplugged it and put it back in the box after 10 seconds. Benchmark suggested I remove 10dB attenuators and try again, but I wasn't interested. Getting terrible to sound acceptable is not my idea of a $3000 amplifier. FWIW, this test was done using Soekris 1541. As a pre-amp: @porkfriedpork reports that the HPA4 just killed the stage. This seems to be something shared by all Benchmark products—because their DACs, while excellent in every other respect, also seem to have no stage. Final Verdict This is a competent solid-stage headphone amplifier. Super clean, low distortion, flat FR. Despite its size, it is not all that powerful, and that limitation makes itself evident with power-hungry planars. It is not worth $3000 in my opinion. There are other totally-dry options out there for less. Really a fail at this price point for me. Bear in mind that several others seem to like this head-amp, even though I did not. This may come down to a planar vs. dynamics thing, but the stage-killing pre-amp performance makes me think there are other limitations lurking under the hood.