INTRODUCTION
Apos Audio graciously sent me another round of review units, this time the Topping A50s headphone amplifier being one of them. Apos made no attempt to influence my opinion, asking only that I do a fair review. What you are about to read is therefore my opinion and mine alone. The A50s is Toppingās newest entry to their measurement-focused line of headphone amps. Itās causing quite a stir. Is the stir worth it? Read on to find outā¦
TL;DR
The A50s is a well-built (at least on the outside) amp with a minimal footprint. Despite having only single-ended inputs its balanced topology eliminates ground loop noise. It also has the power to drive most headphones from its balanced output. Yet, its sound is oversmoothed, the bass rolls off below 50 Hz, the soundstaging is flat compared to its similarly priced competition, and it generally falls behind its market rivals in subjective performance for the price.
WAIT, ARENāT WE MAD AT TOPPING?
Topping has been in the audio news lately ā such as it is ā for inauspicious reasons. I wonāt go into details here but you can read about the issue further here. Apos assured me that they had assurances from Topping that the catastrophic, headphone-killing failure that afflicted the L30 is not possible on the A50s. Indeed, the A50s unit I had did not blow up any of my headphones. I also agree that Toppingās customer service has been less than stellar for those who did lose headphones and their āfixā to the L30 isnāt entirely convincing. So I write this review as objectively as I can, with full acknowledgement that Topping should have done better and still has some PR cleanup to do, so that you as a prospective buyer go into potentially buying an A50s with full informed consent.
FEATURES & BUILD
The A50s is a balanced headphone amplifier built around the same circuit technology found in Toppingās flagship A90 amp. The A50s currently lists for $219.99, putting it very close to Schiitās Magnius and Asgard 3 amps at $199 each (more on these later). The physical size of the A50s is rather shocking, actually, and not because itās big. Itās precisely the opposite:
(thatās the matching D50s DAC on the left and top)
The outer chassis is all metal and quite rugged. You could probably drive over this thing with your car (disclaimer: donāt drive over an A50s or any amp or dac!) and it would be fine. You can see I got the silver finish version and it looks quite handsome, IMO. The back panel boasts the DC power input, and 2 pairs of RCA stereo analog connections for signal input and preamp output. Let me comment on further on the power supply, the input connections, and the preamp in the next paragraphs, because I think there are important points for each.
The included power cord is not great. Itās clearly intended as an internationally universal power cord because the plug portion is interchangeable. Itās also a bare-bones, fairly low-quality power supply that doesnāt match the build quality of the amp itself:
Flimsy power supplies are not new on amps in this price range and lower. Indeed, even the Monolith Liquid Platinumās power supply, on an originally $799 unit, is on the chintzy side. The plug adapter that goes on the power cord here feels cheaper than most, though, and seems to me to be an easy fail point, not to mention a high risk of getting lost during a move or if you just drop the power cord accidentally and the adapter piece rolls under the fridge ā who among us hasnāt lost something under the fridge?
The unbalanced RCA analog input initially puzzled me. At first it seemed like providing only unbalanced connections completely defeats the purpose of having a balanced amp, at least insofar as eliminating things like ground loop noise. And at this price point, eliminating ground loop noise is about the only advantage to having a balanced amp or DAC. However, credit where credit is due, the A50s still eliminates ground loop noise. My desktop computer has a rather annoying ground loop with USB DACs using unbalanced connections. I intentionally confirmed that a DAC was sending the ground loop noise out of is unbalanced output and ran it into the A50s. The A50s took it out. So, it does appear the A50s is in fact a fully balanced amp, with the lone true benefit of being balanced at this price, that uses single ended connectors for its inputs.
The preamp output and the headphone output switch automatically when a headphone is plugged in/out. I have grown to appreciate the feature of switching the headphone amp and preamp outputs ā whether by a user-controlled toggle switch or an automatic switch ā after using the Cayin HA-1AMK2 [link] tube amp that drove both the headphone amp and preamp at the same time, which could cause distortion. Thatās not a worry here. However, the volume knob on the A50s is small, both in diameter and in how far it protrudes from the front faceplate and has a fair amount of resistance on it. I found it cumbersome to change the volume at times, and my fingers are smaller than average.
Back to the overall build qualityā¦the front of the A50s sports the potentiometer knob, a 4.4mm pentacon balanced headphone output, and a 6.3mm (1/4 inch) unbalanced headphone output. There is also a button for power and gain level toggle. To turn the unit on and off, push and hold this button for about 2 seconds. To toggle between low gain (0 dB) and high gain (+6 dB), quick press this button. There are 2 white LEDs to indicate which gain level is selected. When the unit is powered on, one of these two LEDs will be lit. One more comment on the potentiometer, there is a low-level channel imbalance between about 6:30 and 8:00. This is another common problem for amplifiers with cheap pots, but for this price I think Topping could have used a higher quality pot.
Letās talk about feet! There are 4 square rubber feet on the bottom of the A50s and they are some kind of sticky! The A50s isnāt heavy, but itās also not light for its size. However, itās light enough and small enough that a signal wire that hangs off a desk could slide it around but for these feet. Wow these things have some grip! I set up the A50s, its matching DAC (D50s), and a Schiit stack on a TV tray I commandeered to use as an end table next to the couch in my basement:
When I sat down the A50s was just outside of comfortable armās length to adjust the volume. So, I leaned over and grabbed the little Topping stack to slide it across the tabletop toward me. Nope! The whole darn table tipped over because those little feet are so darn sticky! These sticky feet were a smart addition, IMO. Because even bigger heavier amps like the Asgard 3 will slide around on a desktop when you try to push a headphone plug into it. I did not have that issue with the A50s.
Finally, power output. The A50s, from its balanced output, should have plenty of power for the vast majority of price-appropriate headphones. Itās rated at 3400mW per channel at 32ā¦. And of course, it comes with all of the vanishing distortion numbers and sky-high SNR measurements for which Topping has earned a reputation and a following.
SOUND
The first thing I noticed about the A50sā sound is what I didnāt hear. It has that utterly black sonic background that has become the expectation for the plethora of measurement-focused amps currently on the market. Once it starts amplifying an audio signal its sound is clean and smooth. These are the strengths of its sound: a virtually absent sonic background, and a clean, smooth sound.
Unfortunately, the A50s suffers from some of the same sonic issues that its cheaper sibling, L30, also suffered. The sound is too smooth ā itās oversmoothed to the point of sounding sluggish with poor separation between sounds for the price. The magnitude of that perceived sluggishness is not as much as the L30, but itās present. That sluggishness also flattens the soundstage, dulls the imaging, and decreases detail retrieval. I have been listening to a lot of amps lately, including numerous amps under $200 [link to roundup here], and Iāve heard much better detail retrieval, spatial recreation, and general sense of life-like-ness than the A50s was giving me. The A50s also rolls off the bass below about 50 Hz, much like the L30 does. The magnitude of this roll-off is not as large the L30ās, but there is not as much deep-bass extension, punch, slam, quickness, or even bass quantity with A50s as with a number of even cheaper amps.
HEADPHONE PAIRINGS
The A50s appears to be most comfortable driving planar magnetic headphones. I got the best results with my Audeze LCD-2 prefazor. Some of that might be due to the fact that the LCD-2 has a magical ability to be relaxed and forgiving while also being detailed. My HiFiMan Edition X V2 sounded ok, but itās a livelier headphone than the LCD-2 and that gap in liveliness wasnāt showing up much with the A50s. Ditto the last with the Massdrop + Sennheiser HD6XX. Especially at the price point the 6XX needs an amp with a little bit of sparkle and itās not here. Strangely, the lack of sparkle and life on the 6XX made me think the Beyerdynamic DT880 600ā¦ - with that treble that some call āMt. Beyerā - would be a good match, but alas, no. The DT880 was shrill and sharp on this amp ā and that was using the balanced headphone amp on the A50s (my DT880 is balanced modded). When I first tried the A50s + DT880 pairing it was with the D50s DAC in the signal chain. I thought maybe most of that sharpness was due to the ESS chip implementation of the D50s. Switching to the Schiit Modius DAC helped, but did not come close to eliminating that sharp treble. Going all-Schiit with Modius and either Magnius or Asgard as amps brought that treble under much more control. The DT880 is also the spatial-performance reference point for <$200 headphones, IMO, and with the A50s the staging was flatter and the imaging more blended together than other budget amps can pull off.
COMPARISON WITH OTHER AMPS
When I listened to just the A50s for awhile, over time, my brain adjusted and I became less distracted by its deficiencies. I even tapped my toes and bobbed my head a few times. But as soon as I would switch back to another amp ā even slightly cheaper ones like the Magnius and Asgard 3 ā itās like the world of music opened up again. The A50s, unfortunately, is simply behind its competition on detail retrieval, spatial re-creation, timbre, dynamics, and I think this picture is clear enough without belaboring the point further. It is true the A50s wins the measurement race at its price point, but that does not translate to sonic performance. Magnius is just behind it in the measurement game, and Asgard ā while itās not a measurement chaser ā measures well beyond any threshold of concern. I can only recommend the A50s if 1) the buyer lives in a location where obtaining products made by Schiit, JDS Labs, Monolith, Geshelli Labs, or iFi is difficult/pricey; 2) desk space is top priority; and 3) thereās a ground loop issue in the system. And even if those things are true, the L30 amp and a $10 RCA ground-loop isolator is likely the better value, provided enough time passes that buying the L30 feels safe again.
FINAL THOUGHTS
May I level with you readers and say that this review was really difficult for me? My honest opinion on this product is at odds with many reviewers who are far more established than I. Thatās an awkward position to be in as someone hoping to launch a reviewing āside hustleā ā for lack of better term, haha. Some may say that Iām simply not used to the ultra-clean sound that a measurement-champ-amp and I dislike it because Iām unfamiliar with it. Itās true that when given a choice I gravitate toward sound signatures more often associated with Class A, Class A/B, or tube amps (and letās also be clear that there is lots of variation in sound signature within those amp types, too). But, itās also unfair to say Iām unfamiliar with squeaky-clean, measurement chasing sound; Iāve owned the JDS Labs Atom and the SMSL SP200 and put hundreds of hours on both. Iāve also reviewed the Topping L30, Geshelli Labs Archel 2 GMR, and the Schiit Magnius ā all amps that seek to minimize distortion measurements and drive up SNR and SINAD measurements. The available evidence shows Toppingās approach has been to maximize results on a certain subset of quantities that can be measured in an audio signal. For some product categories that translates to good-sounding results when music is playing. I have favorable opinions of their D10 and E30 DACs, for example. For amps though, the approach seems to be a bit too all-in on the measurement side. IMO, JDS Labs, Geshelli Labs, and Schiit have shown that chasing measurements doesnāt mean that sound quality during music playback also has to be sacrificed to the degree that my ears tell me with Topping amps. I donāt have all the answers as to why measurement excellence doesnāt always add up to excellent sound quality. However, it appears to me that there is a framing within the audiophile community that focuses more on what an amp (or DAC) does not do more than what an amp (or DAC) does do. To me, focusing on vanishing distortion and making sure SNR or SINAD measurements reach Everest-like heights is focusing on what an amp should not do ā mess up the audio signal with excess noise or alter a waveform from what comes in. What gets lost when that becomes such a dominant focus seems to be what an amp should do ā increase the amplitude of the audio waveform while making sure voices sound like voices, pianos sound like pianos, etc. and place those sounds in a sonic space that sounds convincing and real. Unfortunately, we donāt have a good way to measure timbre or soundstage and imaging beyond what we hear with our own ears. Some will still reply that our senses are imperfect and our ears and sonic memories cannot be trusted, therefore we need objective standards. I completely agree that our senses and memories are subject to all sorts of biases and flaws. I also respectfully disagree that that means they should be discounted. The science that the objective measurements being used in audio is based on owes its very existence to the careful application of human senses in observing natural phenomena. Our eyes and ears (and nerve endings and taste buds when appropriate ā do not lick a circuit board!) and qualitative descriptions of what they tell us are central to our understanding of pretty much everything; we canāt just discount them.
The Topping A50s amp has a sturdily-built chassis and a thoughtful touch on the inclusion of sticky rubber feet that make it slide around a desktop very little. It also has a form factor that some will find attractive. It has plenty of power, a virtually absent sonic background, prodigious distortion and noise measurements, and a generally clean, smooth sound. However, at $220 it does not compete with its closest competition in overall sound quality and I cannot recommend it.
Thanks for reading, all, and as always, enjoy the music! 