The Garage 1217 Ember II arrived the Monday before last in a nondescript shipping carton packed in a bunch of bubble wrap and a giant ziplock bag, with an RCA 12AU7 Clear Top tube already installed in its single socket. I had ordered the amp on eBay the prior Wednesday after having had my eye on it for a while and being drawn over the line by that day’s 20% off eBay promotion. The very next day, Jeremy from Garage 1217 contacted me to collect my preferences for knob and LED color (black and red please!) and that Saturday I received confirmation that the amp had shipped.
Physical Factors
The Ember II is my first tube headphone amp, which I purchased primarily to use with my high impedance dynamics (DT 1990 and HD 600) but I wanted something general purpose enough to drive my planar magnetics too. It’s a hybrid design, but it has a selectable output impedance up to 120 ohm. I like the look of the well-machined and unique open enclosure, but expect to make good use of my canned air keeping it clean. Coming from a Magni 3, it’s a much more premium feeling product and the tube does look cool. More importantly, the volume control is much better and I’m a sucker for analog volume control. I can’t notice any channel imbalance even when turning it way down, the knurled knob allows for a secure grip and it turns extremely smoothly with no hint of scratch. The Ember II is also surprisingly small, about the same width as the Magni 3, twice as deep and about the same height without the tube, but it has a nice weight to it and I don’t feel that the unit is in any danger of being pulled around by the weight of a cable. The 1/4 inch headphone jack feels a bit tight, but it’s better to start tight than loose. Lastly, even after running it for hours, the tube doesn’t get particularly hot which as a father of small children I appreciate.
Source
I currently feed the Ember II a diet of CD-quality FLACs served up by Roon via a Hifiberry, the same diet I feed my Magni 3.
Music
I listen to a lot of stuff, but especially classical, jazz, progressive metal and singer-songwriter folk-rock type stuff. For this review like most of my reviews, representative albums would be Yuja Wang’s performance of Rachmaninov’s 3rd Piano Concerto, Dream Theater’s “Images and Words”, Phronesis’ “Alive”, and Danny Schmidt’s “Parables and Primes”.
Sound
Having heard that tube amps pair well with the HD 600, I was curious what the Ember II would do for these headphones which in most of my prior listening had suffered from congestion, notable grain and a small bit of sharpness in the lower treble. Throwing on some classical as well as modern vocal tracks seemed to reveal slightly better instrument separation, a reduction in grain and tamed treble, but the HD 600 still sounded like the HD 600, not something that could dethrone my DT 1990 from their top spot in my rotation. The softened treble did bode well for pairing with the DT 1990 which themselves suffer from an elevated treble that can quickly become bothersome on the wrong material. In all my previous listening, I’ve felt the need to EQ the DT 1990 treble downward, especially at around 8 kHz. I had also noticed this being a bit source dependent, as the treble seemed least tolerable with my LG V20 and most tolerable with my Topping NX4 (which incidentally uses a Burr Brown like the Hifiberry
So, I plugged in the DT 1990 and started listening. Whoa! The tonal response seemed as I remembered, but the treble didn’t immediately fatigue me the way I’d grown accustomed. Listening to some well mastered Phronesis albums with their punchy percussion, the cymbals and high hats were bright but tolerable. Only once I put on some Dream Theater did the treble finally reach fatiguing levels and I once again had to resort to EQ. It does sound to me as though the 120 ohm output impedance softens the sound a bit more relative to the 35 ohm impedance, but it’s difficult for me to say because switching the impedance affects overall levels somewhat. After listening over the past week I’ve landed on the 35 ohm output impedance for critical listening and 120 ohms for background listening, and ended up with a tweaked version of my typical EQ settings that follows the usual curve but with adjustments only about half as many dB in level as previously. I feel like I get plenty of detail, clarity and punch to satisfy my needs.
As for power, there’s enough for me. The Magni 3 seems like it has more gain on tap as the volume knobs find similar positions with low gain on the Magni 3 and high gain on the Ember II. The Ember II has enough gain to allow me to feed it from my LG V20 as well. I should note that I tend to listen at around 68-75 db© Z, which is to say pretty quietly. One thing that seemed to stand out to me about the Ember II was that I could turn up the volume very high without noticing any change in the character of the sound or any additional distortion. It’s possible that the Magni 3 behaves similarly but I don’t notice because of the distraction of the scratching noises from turning the pot.
Measurements
Just for fun, I also took some measurements. There’s a truism that well implemented amps will measure similarly. Here I didn’t measure the amps, but rather the amps’ impact on the headphones, but it seems to hold true in this case as well. Especially at the 120 ohm output impedance I would have expected to see some frequency response changes, but the measurements don’t show anything I would expect to be audible. I might also have expected lower 3rd order harmonic distortion as tubes are known for concentrating distortion in even order harmonics, but the measured 3rd order distortion for all amps is so low (mostly below the noise floor actually) that I can’t conclude anything meaningful other than they’re all very clean. The only measurement that seems to show some differences to the Magni 3 is the C80 clarity measurement, but when you look at clarity measurements from individual runs there’s enough variation there that this is likely within the margin of error. Looking at the impulse response, it looks like the overall rate of decay is very similar between all amps, so it’s reasonable that clarity would be similar.
Lastly, I did also try a little level-matched ABX testing on some of my usual test songs (Rachmaninov 3rd Piano Concerto, Daft Punk’s “Giorgio by Moroder” and Dream Theater’s “Metropolis Pt. 1”) using my DIY ABX tester. The test run did help me find a grounding error in my wiring, but I wasn’t able to reliably pick out which amp was which. I never have been able to distinguish sources with this method, which could either mean that the sources sound very alike, that I’m a terrible ABX tester, that my DIY switch masks the subtle differences that do exist, or that these differences require prolonged listening to discover. In any case, I’ve come to find ABX’ing tedious and will likely stop doing it and just enjoy listening to music again!
Does my inability to ABX or find measurable differences mean that tube vs solid state doesn’t matter? Not necessarily, as I’m just measuring a simple frequency sweep and not real music, besides which the Ember II is a hybrid rather than an OTL tube amp so isn’t necessarily representative of the most “tubey” of amps. To my mind’s ears, the Ember II does indeed sound a bit smoother and less fatiguing than the Magni 3. It could be placebo, it could be real, but either way I’m enjoying it. If anything, I think the similarity in measurements and ABX testing just highlights the fact that whatever differences there may be are subtle, not dramatic.
Conclusion
When all is said and done, the Ember II is a very nicely made piece of equipment with which I enjoy listening to music. To boot, it’s made by a company with some genuinely enthusiastic and friendly people behind it. I’m happy to have it in my home.
Now, please no one encourage me to get into tube rolling!
I’ switch EQ profiles the way I imagine you might switch headphones.
. Nice picture.