My Quest for a New DAC, Part 5: Chord Electronics Hugo M Scaler & Hugo TT 2

My Quest for a New DAC, Part 5: Chord Electronics Hugo M Scaler & Hugo TT 2

• Introduction to the Series
• Part 1: Mytek Brooklyn DAC+ with Uptone JS-2 Power Supply
• Part 2: Ayre QX-8
• Part 3: Denafrips Terminator
• Part 4: iFi Pro iDSD & friends

Introduction
 
In this next installment of my DAC series, I evaluate not just a DAC, but a unique combo of DAC, amp, and scaler -- the Chord Electronics Hugo TT 2 DAC/amp and Hugo M Scaler, abbreviated for the rest of this article as TT2 and HMS. I have been intrigued by Chord Electronics’ stack ever since I heard a demo of the Blu MkII scaler and DAVE DAC at RMAF in 2017. Even in a show setting, the improvement wrought by the Blu MkII over the already-excellent DAVE was startling. However, the price points of the Blu MkII and DAVE were just not at a level I could afford. 
 
Fast forward less than a year, and Chord Electronics announced the Hugo TT 2 ($5795 MSRP) at High End Munich in May, and then the Hugo M Scaler ($4995 MSRP) at CanJam in London in July. While still well above my budget, this combo represented a price point very similar to the Ayre QX-5 Twenty that has been my reference DAC for some time.
 
Serendipitously, I attended that CanJam, so was present when Rob Watts (Chord Electronics’ Digital Design Consultant) announced and unveiled the Hugo M Scaler. I got a chance to chat with Watts at some length, both about the new creation and his design philosophy in general. While the technology is fascinating in its own right, what impressed me most was how passionate he is about music, and how every small step in his design process is validated by listening.
 
I was fortunate to get about an hour of listening to the HMS/TT2 combo in a quiet space away from the show floor. Even with my humble Macbook Pro driving them, what I heard convinced me that this was a combo I would have to seriously listen to in my own system.
 
So the request for review units was made, and then it was time for a long wait in the queue, as the demand for review units has been sky-high. Fortunately, an Audiophile Style member and friend who owns these units kindly loaned them to me while he was gone on an overseas trip. Not only that, he also sent along some cables and accessories to try along with the units. While he shall remain anonymous, this review is dedicated to him for his generosity.
 
 

Hugo TT 2

The TT2  is more than just a DAC. It is also a preamp and a headphone amp, with an astonishingly high power rating (for a source component). As readers of this series already know, I place great stock in DACs that pay attention to the fundamental aspects of DAC design: the analog electronics, the power supply, and the clock design, in addition, of course, to filtering and D/A conversion.
 
Let’s talk about the latter first. The TT2, like its cousins in the Chord Electronics lineup, eschews off-the-shelf DAC chips and implements all its digital logic in an FPGA (Field Programmable Gate Array). But there the similarity with other FPGA DACs ends. Watts hews to his own design philosophy, with the prime guiding principle being to recreate the original analog waveform with the least amount of timing transient errors. His proprietary WTA (Watts Transient Aligned) filter algorithm is a method to approach the performance of an ideal, infinite tap length FIR (fixed impulse response) sinc filter, using a large but finite number of “taps,” or coefficients, that can be accommodated within the storage and computational capacity of the FPGA. I lack the expertise to go into further detail, but Watts has kindly provided us with his slide decks (linked below) where he delves into his technology. You can also find videos of his presentations by searching online. Finally, the Chord Electronics M Scaler product page links to an informative paper that describes this design approach.
 
In the context of Chord Electronics DACs, the tap length has become a metric of evolution of the WTA filter. In the latest lineup, the Hugo 2 and Qutest DACs implement a tap length of 49,152, the TT2 implements 98,304 taps, the DAVE 164,000 taps, and the Blu MkII CD Player and Hugo M Scaler implement just over 1 million (1,015,808) taps. The actual D/A conversion in the TT2 is done with a 10-element pulse array. As a point of reference, the flagship DAVE uses  a 20-element  pulse array.
 
What about clocks? As Watts explained to me, 

“No I don't talk about clocks, as it's way way too complex. Jitter is much more complex than simple jitter figures, as different DAC architectures have profoundly different sensitivity to clock jitter. Then there is the issue of source jitter, which I have eliminated through my DPLL. Pulse array is insensitive to clock jitter, plus the active switching elements (something nobody ever talks about) is extremely low jitter. Turning off the noise shaper OP, and having DC on the pulse array outputs, only changes noise by 0.5dB - so jitter simply isn't an issue.”
 

Moving on to power supplies and the analog output, the TT2 uses a supplied 15V/4A SMPS that energizes its internal power supply built around “6 super-capacitors on the + and – 15 V power rails with 30 Farad capacity.” This allows the TT2 to deliver an astonishingly high 7.3W RMS into 8Ω via the single-ended RCA and headphone outputs, and 18W RMS into 8Ω via the rear XLR outputs. This is right up there with my Cavalli Liquid Gold.
 
Unusually, the TT2 can drive balanced headphones directly from its rear XLR outputs. This is not something you want to do with a standard DAC, as these outputs usually have output impedances that are too high to drive headphones efficiently. For example, my reference Ayre QX-5 Twenty DAC has an output impedance on the XLR output of 305Ω. In contrast, the TT2 has an output impedance of 0.042Ω, which can easily handle the typical range of headphone impedances.
 
Rounding out the feature set of the TT2:

• USB input: PCM supported up to 24/768, DSD up to 256 (DoP) and 512 (native)
• Dual BNC input: PCM input up to 24/768 from the HMS
• DAC (fixed output), AMP (variable output), and HP (headphone) modes
• Low- and high-gain settings
• 4x cross-feed settings that I actually found useful - I usually never use this feature
• 4x Filter settings from incisive (FIL1) to warm (FIL4)
• Single-ended and balanced XLR analog outputs
• Headphone outputs via 1x 3.5mm and 2x 6.3mm jacks 
• Balanced headphone cables can be driven from the rear XLR outputs in AMP mode

Hugo M Scaler

While a DAC performs a vitally necessary function in an audio system - after all, without a DAC, there would be no sound! - the M Scaler is an optional component. At least in functional terms. I may be giving away the plot, but once you hear it, you may take issue with the word optional!
 
In broad terms, the HMS is an upsampler - Chord Electronics labels it a “Standalone 1M-tap digital upscaling device” - that takes in inputs of varying sample rates, and sends the (selected) upsampled rate to the DAC. In the context of Watts’ architecture, where DACs have two WTA stages, the HMS can be considered an external upgrade to the WTA1 stage, that implements a tap length of 1,015,808 taps. Input streams are routed to the HMS, and output is delivered to the TT2 over the dual BNC (DBNC) input. Effectively, the TT2 bypasses its internal WTA1 stage, and routes the scaled input stream to its WTA2 stage.
 
Why an outboard scaler? Well, the FPGA that enables this level of computation, the Xilinx XC7A200T, has 740 DSP cores and requires current peaks of upto 10A, which increases the demands on the power supply, and can generate noise on the ground plane. A separate unit, with an independent power supply and chassis to isolate this noise, makes sense. This also enables modularity, allowing the use of HMS with a variety of Chord Electronics DACs starting with the Hugo 2/Qutest all the way up to the DAVE.
 
The feature set of the HMS is as follows:

• 2xBNC and 1xToslink S/PDIF inputs up to 24/192
• Galvanically isolated USB input accepting upto 24/768 PCM and DSD256 over DoP
• Upsampled outputs delivered via BNC up to 24/384 and dual BNC up to 24/768
• Output rate selectable by OP SR selector button
  • Exact operation varies by input sample rate, detailed in a table in the manual.

Finally, the HMS also ships with a 15V/4A SMPS which is identical to the unit shipped with the TT2, although the HMS’s current demands are lower. In fact, Watts personally uses a 12V portable battery on his Hugo M Scaler with the Hugo 2 when traveling. Just the fact that a designer lugs along a table top unit in his carry-on luggage because he can’t bear to be without it is quite intriguing!
 

Form and Function
 
If you’re a seasoned Chord Electronics user, you’re likely very familiar with the round buttons and color scheme. Sample rates are displayed by color, with a scheme that is consistent across Chord Electronics products:

Similarly, volume levels have a color palette:

The HMS and TT2 are eminently stackable, having identical width and depth, with the HMS a few cm slimmer in height. Placing the TT2 on top with a slight offset to the back enables easy access to the HMS control button on the front and top of the unit.
 
Once you’ve read the manual and familiarized yourself with the color scheme, operation of both units is fairly intuitive. I’ll be honest: I’m not a huge fan of the whole colored jelly-bean design aesthetic here, but since both units can be dimmed, my focus quickly turned to, and stayed on, the sound quality I was hearing. Also, while the units I borrowed were silver, I like the look of the black units better. But this is all just personal preference.
 
Operational notes on the HMS:

• The OP SR button selects the output sample rate, toggling through:
  • Red: pass-through
  • Green: ⅛ million taps
  • Blue: ¼ million taps
  • White: ½ million taps (single BNC), 1 million taps (dual BNC).
• During playback, the DX OP button color displays the input sample rate.

Methodology
 
I want to be very clear about my evaluation methodology. This is unambiguously a subjective evaluation, based on my listening impressions. The findings reported here are my own personal preferences with these components in my system. If you are looking for a measurements-based objective evaluation, then you should probably stop reading, as you will not find that here.
 
 
Listening Experiments
 
Chord Electronics strongly encourages the use of the stock PSUs that are shipped with the units, as using anything else can invalidate the warranty. For the bulk of this review, that is exactly what I did. While I did vary the BNC cables in the Upgrades and Accessories section of this review, for most of the listening, I used the cables as pictured in the system diagram in the Associated Equipment section.
 
Since this combo comprises 2 units, one of which - the TT2 - can be considered on its own, the number of experiments was rather large.
 

Key Questions

1. How does the TT2 perform as a standalone DAC?
2. How does the TT2’s headphone amplifier compare to my reference Cavalli Liquid Gold?
3. How does the addition of the HMS improve the sound quality (SQ)?
4. How does the HMS/TT2 combo perform as a high-end DAC combo?
5. How does the quality of the source chain affect SQ of this combo?
6. How does the HMS/TT2 combo scale with upgrades and accessories?

Review Playlist
 
HMS/TT2 Review Playlist on Qobuz (US)
 
To enable you to listen to the same tracks (see caveats) I did, I created a public playlist on Qobuz USA. This playlist includes the tracks mentioned in this review, as well as others I listened to in the course of this evaluation. Please note that in some cases, the Qobuz track will only stream at 16/44.1, whereas I may have used a local hi-res version. Still, this gives you a sense for the music I listened to for evaluation.
 
 
TT2 as a DAC
 
Let’s dig in - we have a lot of ground to cover! Since these units were several months old, they were thoroughly burned in, so I could hit the ground running. I started out with just the TT2 in my chain, replacing the DAC in the system. I used the supplied SMPS wart, as that is how Watts recommends the unit be powered. He sourced these PSUs for their low noise characteristics, and this, combined with the super-capacitor design of the TT2’s power supply, is said to render expensive linear PSUs unnecessary.  Of course, being the curious sort, I would put this to the test later on!
 
The TT2 impressed from the get-go in my system. The overall impression is full-bodied and detailed, with plenty of texture. I have had a lot of great DACs in my system these last few months, and the TT2 drew me in like the best of them. One telling sign of how good a DAC is when you can’t stop listening to music! I’d start a test track for a quick listen, and find myself having finished the album an hour later.
 
Another feature of the TT2 that I found really impressive was Crossfeed. I’ve played with crossfeed before, but always found it gimmicky at best. Until now. Perhaps because of the TT2’s inherently excellent imaging depth, I actually found the crossfeed function to enhance imaging further. To my ears, the subtlest XFD-1 setting usually sounded the best, and I found myself actually using it quite often. Not always, though. With well recorded classical music, I still preferred no crossfeed.
 
As for filters, I spent some time experimenting, but quickly settled and stayed on the most incisive FIL1 setting. Transparency and incisiveness are the hallmarks and strengths of the TT2, so why mask those?
 
 
TT2 as a Headphone Amp
 
While several of the DACs I’ve reviewed - most, in fact - have nominally had headphone amps in them, they have not come close to the level of sound quality of my Cavalli Liquid Gold (LAu) headphone amp. And understandably so, as the $4000 MSRP LAu ranks among the best headphone amps ever. Yes, there are better and more expensive amps (there always are), but to date, I have never regretted my investment, as the LAu has just never been the bottleneck in my system. So I didn’t have particularly high expectations of the TT2’s headphone amp.
 
Watts claims the single-ended outputs of the TT2 (both RCA and headphone) are the most transparent, while going to the XLRs provide significantly more power, at the cost of a slight loss of transparency. This is due to the presence of another output stage in the analog chain. In contrast, on my LAu, the dual-3pin and 4-pin XLR headphone outputs of the LAu sound significantly superior to the single-ended ones. On the other hand, the Cavalli sounds equally great with single-ended and balanced inputs.
 
Given the above, this comparison had to be constructed with care. I connected the TT2’s RCA outputs to the LAu, leaving the XLR outputs free for headphones, if needed.
 
I happened to have 4 excellent headphones on hand, with which to compare the TT2 and the LAu, and I’m glad I did, as there were some interesting differences. The headphones (and their published impedance and sensitivity) were:

• HD800 (SD mod) - my reference - 300Ω, 97 db estimated (1mW/1kHz)
• Meze Empyrean - 32Ω, 100 dB (1mW/1kHz)
• Focal Utopia - 80Ω, 104 dB (1mW/1kHz)
• Abyss AB-1266 Phi CC - 42Ω, 88 dB (1mW/1kHz)

Luckily I had enough cables and adapters on hand to make either SE or balanced possible. For this study, all the headphones were equipped with Cardas Clear balanced cables (shout out to Cardas for providing these).
 

Comparison: TT2 SE vs. XLR for headphones
 
This was a test to validate Watts’ assertion of transparency loss vs. power gain when choosing between these output modes. I used my trusty HD800 (SD mod) for this test. 
 
First up was Henryk Górecki’s Symphony (No. 3) of Sorrowful Songs, David Zinman, London Sinfonietta (16/44.1). This is a gorgeous piece with slow tempi, but dense textures. In the second movement, soprano Dawn Upshaw’s soaring lament over the percussive piano and lush strings comes over just a bit  clearer on the SE outputs. I have to admit I was surprised I’d hear a difference. While the XLR output did seem to be more dynamic, and  can definitely drive the HD800’s past the threshold of pain, the SE output was more than adequate for the HD800.
 

Using Lucky Man from the 2012 remaster of Emerson Lake & Palmer (Qobuz 24/96), I again found the SE output to be just a tiny bit cleaner, with more of the fine micro-details of the drums and cymbals, for example, coming through.
 
I also experimented with the Lo- and Hi-Gain settings (the former is 9dB lower), but couldn’t hear a significant difference in quality. For all further listening, where I used the headphone outputs of the TT2, I used the SE output in Lo-Gain, except where noted.
 
 
 
 
Comparison: TT2 vs. Ayre QX-8
 
To put the attributes of the TT2 as a DAC in context, I compared it to the very similarly priced Ayre QX-8 ($5450, vs. the TT2’s $5795 US MSRP). I reviewed the Ayre a few months ago, and found it to be a formidable piece, with outstanding sound quality. As I wrote then, the QX-8's key strengths are an ability to portray an expansive soundstage, combined with a refined tonality, and ample dynamics.
 
On Veinte Anños, Buena Vista Social Club (DVD-Audio 24/96), the QX-8 conveyed the energy, detail, and ambience of the piece as well as always, but the TT2 did more. With the TT2, there was a coherence to the rhythm, an ability to distinguish the various percussion instruments, and a level of clarity and transparency that drew me more emotionally into the song. The QX-8 had a sweet and smooth  tonality to it that was very pleasing, and that has always drawn me to Ayre DACs. The TT2 was no slouch in tonality, and delivered better articulated bass, and an overall more muscular sound. Most impressive with the TT2 was the clarity and attack of leading-edge transients, like guitar plucks, and the assorted percussion instruments that are struck or shaken in this song - leaving you stirred.
 
On Brahms Symphony No. 3, II. Andante, Thomas Dausgaard, Swedish Chamber Orchestra (BIS, 24/96), the QX-8 came into its own, rendering a huge soundstage, with excellent instrument placement. While the TT2’s image wasn’t quite as expansive, there was a greater degree of clarity. Dense passages with multiple instruments were easier to tell apart and follow. Bass was fuller, and instruments like the cello, double bass, and tuba were better articulated. Here again, there was a greater sense of detail and coherence. 

Comparison: TT2’s headphone amp vs. Cavalli Liquid Gold
 
This is where having 4 different headphones with different impedances and sensitivities proved very informative. Using the tracks above, along with others, here are my notes for each headphone.
 
HD800 (SD mod): I have lived happily with the combo of HD800 with Super DuPont mod and LAu for the last 2+ years, and I now see why. These two are very well matched. With high quality sources, the LAu drives the HD800 with excellent low end extension, and without the excessive brightness or harshness for which it is often criticized. The TT2 also drove the HD800 very well, even in SE mode. While the LAu had deeper bass, and was more full-bodied, the TT2 had its strengths too. Being driven directly from the TT2’s DAC output stage certainly sounded extremely clean and detailed. Since imaging is the HD800’s strength, I found the TT2 produced an outstanding sound stage. The depth and dimensionality in particular were striking. The LAu had a bigger soundstage, but without quite the precise depth cues. All in all, the LAu won out, but by just a nose.
 
Abyss AB-1266 Phi CC: While the LAu could drive the Abyss as loud as I wanted to - and louder - the overall signature was leaner and seemed a bit lacking in excitement and dynamics. In contrast, the TT2 was more dynamic, full-bodied and meaty. In addition, especially using the SE output of the TT2, there was a marked improvement in resolution and micro-details. The Abyss was the least sensitive of the headphones, and for the  Górecki and some classical tracks, I had to use the TT2’s XLR outputs to get sufficient gain. This was the rare instance I found the use of the XLR outputs necessary. Indeed, going from SE to XLR did result in a slight loss of transparency. To my ears, this was a subtle loss, so I don’t want to overstate this effect. If your headphones require the gain, switching to the XLRs is an easy way to go, and I didn’t find the minute transparency loss a deal breaker. Certainly, I’d recommend the SE output where it provides sufficient gain. On this headphone, the TT2 won by a clear margin.
 
Meze Empyrean: These sounded equally wonderful with both configurations. The LAu was more full-bodied with deeper bass, while the TT2 more transparent and detailed. A familiar theme! Call this one a draw.
 
Focal Utopia: Another case of both configurations sounding wonderful. I found the exact same tradeoff as with the Empyrean. This one is a draw, too.
 
Overall, this was a very exciting finding for me. In effect, I could very happily live with the TT2 directly driving my headphones! While I don’t (yet) own any of the headphones other than the HD800, the fact that the TT2 could drive even the inefficient Abyss (88dB/mW) not just sufficiently loudly, but with excellent dynamics, means that I don’t need a bespoke headphone amp like the LAu with a TT2. This makes the TT2 a really outstanding value as a DAC/amp!
 

Enter the Hugo M Scaler
 
Since the TT2 had proven its chops as a DAC and a headphone amp, it was now time to add the “special sauce” of the M Scaler. What would this curious device, this culmination of Watts’ and Chord Electronics’ decades-long quest for “a million taps,” deliver? To find out, I placed it upstream of the TT2, connected by dual BNC cables, driven by my my chain using USB, and sat down to listen. 
 
And couldn’t stop listening. I’d start a track and listen to the entire album. I fired up The Remembering from Tales from Topographic Oceans, one of the discs in The Steven Wilson Remixes - Yes (24/96). Tales... has its moments, but tends to drown in its self-indulgence. I can usually only listen to a track at most, but here I just listened to the whole album. 
 

I had a similar experience with the newly released  Mahler: Symphony No. 2, Osmo Vänskä, Minnesota Symphony (24/96). I’m a sucker for Mahler, and this is now the eighth version of the 2nd in my library. A plan of listening to a movement quickly became a marathon listen to the whole album.
 
I’ve really had to think hard about how to describe the effect of the HMS. While it dramatically improved conventional areas of sound quality: tonality, extension, resolution of micro- and macro-details, and imaging, it also did more - almost along another axis of improvement I hadn’t registered before.
 

The best way I can describe it is temporal coherence. There is just something “right” about the sound in terms of its timing. Take dense orchestral music like the Mahler. Normally, with music like that when different instruments come to the foreground in terms of volume, one’s ear is drawn to them, and the others recede from attention. It’s akin to surfing - riding on the crest of each wave. Forgive the mangled metaphor - I don’t surf! What I found with the HMS is that as an instrument receded from its crest, it remained perfectly easy to follow. As a result, I found myself able to keep my attention on what instruments were doing off their peaks. On dense music, this is a revelation! It draws you so much deeper into the piece, both intellectually and emotionally. 

The OP SR “Magic” Button
 
My favorite “jelly bean” on the HMS’s fascia is the button labelled OP SR (output sample rate). With it, you can step through passthrough (red), ⅛ million taps (green), ¼ million taps (blue), and either ½ million taps (if only using a single BNC) or 1 million taps (white, with dual BNCs). Each step makes a noticeable improvement, but the last one from ¼ million to 1 million taps (since I was using dual BNC) is startling. Whoever had the idea of this button is a genius! It’s the perfect sales tool. I had several audiophile and non-audiophile friends visit during the time I had this unit. I assure you every single one of them said some variant of “Holy $h!t!” when stepping from blue to white. Very entertaining!
 

Effect of the source chain
 
If you glance at my system diagram, you’ll see I have optimized my digital transport chain upstream of the DAC. This is apparently a sticking point with some, who argue that this part of the digital chain shouldn’t matter to sound quality, especially with DACs that have been engineered with galvanic isolation and buffered inputs. All I can say is that in my listening experience, this part of the chain makes a profound impact on sound quality, and I have invested accordingly.
 
The HMS is also claimed to have a galvanically isolated USB input, so I put the HMS/TT2 to the same test I sometimes do with visitors to my system, who wonder how much of a difference all my upstream optimizations make. The experiment is to compare 2 digital chains, both using Roon as the music server:

1. The optimized chain shown in the diagram below
2. The upstream chain replaced by a MacBook Pro laptop running on battery, driving the HMS/TT2 directly with an Amazon USB cable.

The result? Not even close! Compared to the optimized chain, the Macbook Pro chain:

• Had a smaller soundstage, and particularly lost image depth, sounding flat and boring
• There was a loss of clarity, a blurring effect, making instruments harder to isolate
• Massed instruments, especially strings, sounded more homogenous
• Worst of all, there was a slight harshness and glaze to the sound.

Let me put this in context. Even driven directly by a laptop source, the HMS/TT2 delivered excellent sound quality. Do upstream optimizations further improve the sound quality? Emphatically yes! In my experience, such investments continue to pay rich dividends, just as they have with any of the other DACs and DAC combos I have evaluated. All of the evaluations and comparisons in this article were done with the full optimized chain in the diagram.
 

Effect of Power Supplies
 
As I’ve reported in my previous DAC evaluations, I’ve come to appreciate the importance of low noise and low impedance power supplies, and in selecting DACs for evaluation, have preferred those that enabled the use of external PSUs. The impact of a Paul Hynes SR-7 DR (dual-regulated) linear PSU on the Mytek Brooklyn+ and the iFi Pro iDSD was quite profound.
 
The HMS and TT2 both have outboard 15V/4A SMPS “bricks,” and a 5.5x2.5mm input that accepts external supplies from 9-15V. I cannot stress strongly enough the importance of ensuring any PSU you try has the regulation and quality to stay within these voltage levels. As previously stated, Chord Electronics will consider any damage caused by external PSUs to invalidate the warranty. 
 
How would the SR-7 DR improve sound quality, if at all? The SR-7 at my disposal had 2 adjustable DR rails, one from 3-13V, and the other from 9-19V, both rails conservatively rated at 6 amps. Since the friend who owns this PSU uses both rails set to 12V, I first evaluated both the HMS and TT2 at 12V. I then also tried 13V and 15V.
 
On the TT2, the effect of the SR-7 was modest at best. This is really a validation of Watts’ claim that the supplied PSUs are of very high quality. Indeed, when the SR-7 was set to 12 or 13V, the stock PSU beat it on dynamics and slam. It was only when the SR-7 was set to 15V did the balance tilt in favor of the SR-7. At 15V, the SR-7 matched the dynamics of the stock PSU, while sounding a little less harsh, and with a darker, more relaxed presentation. Was it better? Yes. Would I recommend you run out and buy an external PSU for the TT2? No.
 
The HMS was a slightly different story. Here again, I noticed the same correlation between voltage and dynamics. You really have to run this thing at 15V to match the dynamics of the stock PSU. However, on the HMS, I felt the SR-7 was a more significant step up in quality. The same reduction in harshness I noticed on the TT2 was more pronounced, accompanied by an uptick in resolution and clarity.
 
Let me be clear: the HMS and TT2 sound wonderful with the supplied PSUs. The benefit from upgrading to an end-game PSU like the SR-7 is much smaller than I have heard on other gear. Is there a benefit? Sure. Given the cost and difficulty to obtain the SR-7, I don’t think HMS/TT2 owners need to feel deprived in any way just running the stock PSUs. If, on the other hand, you are the lucky owner of a PSU of the quality of the SR-7, you can certainly elevate the performance of the HMS/TT2 combo even further. If I had to pick only one, I would apply the upgrade to the HMS, and then only if you can run at 15V.

 
Effect of different BNC cables
 
The AS’er who loaned me the HMS/TT2 units also sent along some different BNC cables for the dual BNC connection, so I was able to assess how much of an impact this upgrade makes, if any. The cables I tried (all rated at 75Ω) were:

• Stock BNC cables
• Cardas Clear digital BNC ($600/1m each)
• Habst 5N silver digital BNC (1259€/1m each)
• SOtM dCBL-BNC ($700/1m each)
• WAVE Storm Reference BNC (£705/1m each, sold in pairs)

I don’t want to suggest you have to spend this much on digital cables. These just happened to be what I had on hand. There are plenty of lower priced cables to experiment with. The idea here was to determine how much of an effect the digital cables made.
 
As with the stock power supplies, the stock BNC cables sound excellent, and for most owners, they need look no further to enjoy their units. Did the additional outlay from $1200 to approx. $2800 improve sound quality further? In absolute terms - yes. In value terms, that is up to each buyer. Were it me, I would definitely save up for an upgrade, while happily enjoying the stock cables in the interim.
 
Each of the cables had different strengths. The SOtM was the smoothest and warmest sounding, the Cardas also a bit warm, but more incisive than the SOtM, while the WAVE had the most dynamics and bass slam. I found the Habst to have the most natural and organic sound, especially for piano and voices. These were also sadly the most expensive of the bunch. 
 

Comparison: HMS/TT2 vs. Ayre QX-5 Twenty
 
If you’ve been following my series, you already know how highly I regard the QX-5 Twenty. In fact, many of you have already placed bets that it will be the winner of my quest. Didn’t your mothers warn you of the dangers of gambling? So how did the HMS/TT2 fare against the mighty QX-5?
 
First up, I listened to the beautiful Trio and Chorus In holder Anmut stehn, from Haydn: The Creation, John Elliott Gardiner, English Baroque Soloists (16/44.1). On lesser DACs, this 1995 recording can sound a bit thin and constricted. Not so on these two. The QX-5 exhibits its signature smooth tonality, and imparts a welcome physicality to the low end, with a large, textured soundstage. So far, so amazing. The HMS/TT2 in contrast isn’t quite so visceral, and has a leaner tonal balance. While the image isn’t quite as large as the QX-5, it’s made up for by the pinpoint localization of the voices. And oh, those voices! The HMS really does something special with human voices. I’ve never been as engaged in the interplay between the 3 soloists and the chorus in the piece as I was with the HMS.
 
Next up was some “nouveau flamenco” courtesy of Ottmar Liebert with the song Danza Viva from the Solo Para Ti album (16/44.1). Here again, there was the contrast between muscularity and precision. The QX-5 didn’t put a step wrong and rendered a big image with richness, density and texture. The HMS/TT2 was a bit leaner, but what it added was coherence. Hand claps were exquisitely real, and here again it was possible to simultaneously follow the foreground guitar with the strings, percussion, and hand claps with clarity. 
 

Rounding out this section is another beautiful choral piece, Quia Fecit, from the incredible 2L recording Magnificat (Arnesen) (DSD64). This album was recorded in DXD, which I also own, but I used DSD here to see if DSD streams tilted the balance between these 2 DACs. The HMS handles DSD identically to the TT2 - using Watts’ DSD filter that he’s refined since the DSD+ filter in the DAVE. I’m happy to report the essential natures of these 2 DACs remained intact with DSD as well. Listening to this first on the QX-5 would make you wonder how this could possibly sound better. On the initial deep bass drone, the HMS/TT2 can’t quite match the QX-5, but once the vocals and chorus start, they are so natural and holographic. Ultimately, it’s about which one of these magnificent pieces of gear connect you more emotionally to the music, and in each case, I found the HMS/TT2 did that.
 
It’s also natural to ask: why not connect the HMS as an upstream scaler to the QX-5? How would that sound? Indeed I did this experiment, but consider this: the QX-5, like most other DACs at my disposal, can only accept up to 24/192 on its S/PDIF BNC inputs. Therefore the HMS could only deliver ¼ million taps using a single BNC cable. Only a few Chord Electronics DACs support the Dual BNC protocol, which Watts tells me is built on the AES3 standard.
 
The HMS driving the QX-5 at 24/192 sounded wonderful. However, used at only ⅛ of its potential, the sound quality improvement would not justify its price - to me. It’s also natural to wonder why Chord Electronics does not expose the M-Scale’d output via USB, since that connection can handle the 24/768 sample rate? Again, Watts tells me this would have required substantial development, and would have significantly delayed release. One can only imagine the demand for this unit if it could drive any (not just Chord Electronics) DAC capable of 24/768 on its USB input with its 1-million tap filter!

Comparison: HMS/TT2 vs. Ayre QX-5 Twenty, Part Deux with speakers
 
So far, my impressions have been on my primary setup with a variety of high-end headphones.  I also compared the HMS/TT2 and the QX-5 Twenty in a loudspeaker-based system - the alternate system detailed in the last section. Here again, the two DACs were fed from a USB-based chain, except that in this case, the output drove Magnepan 3.7i’s speakers with dual Rhythmik subs, through Audio Research and Hegel electronics. Since this is a friend’s system, and he was present for all the listening, we conducted some of these comparisons as single-blind tests. Also, we adjusted the levels from both DACs to within 0.2dB SPL for each track we compared.
 

We started with the Marvin Gaye’s classic Mercy Mercy Me (The Ecology) from The Marvin Gaye Collection (24/96). This was one track where I clearly preferred the QX-5. Being a very smooth song, it wasn’t particularly challenging tonally. The QX-5 sounded more exciting and alive.
 

Moving on to Max Richter’s On the Nature of Daylight from The Blue Notebooks (15 years Edition) (24/96), the lush orchestration and deep bass notes were very well portrayed by the QX-5. What the HMS/TT2 added was to portray the massed strings more clearly as many individual instruments. The HMS/TT2 localized the hypnotic, primary violin more clearly. While the QX-5’s soundstage was more expansive, the HMS/TT2 was more holographic - i.e. precise in instrument placement in 3-dimensional space. Here again, the analogy that came to mind was of an image snapping into clearer focus.
 

The final track was another choral piece - For unto us a child is born, from Handel: Messiah, Emmanuelle Haïm,  Le Concert d'Astrée (Qobuz 24/96). Just as it did with headphones, the HMS/TT2 excelled in its portrayal of voices - solo and in chorus.
 
At this price point, I would expect outstanding sonics from any DAC and indeed, both the QX-5 Twenty and the HMS/TT2 combo deliver. Picking between the two then becomes a matter of personal preference, and what attributes one values. With the QX-5 Twenty, one is getting an approx. $10k DAC, with the designer having decided how best to apportion the cost to various subsystems. With the HMS/TT2, you have a $5800 DAC married to a truly transformational $5000 upscaling device.  With all that said, let me be clear. In this comparison, from my listening, I would pick the HMS/TT2 combination. 

Summary
 
The Chord Electronics Hugo M Scaler and Hugo TT 2 combo set a new high-water mark for sound quality in my system. It exceeded even my long-standing, comparably-priced reference, the Ayre QX-5 Twenty, although as I’ve noted, there were certain areas where the QX-5 still dominated. While the TT2 is an outstanding DAC at its price point, it is really the HMS that I consider the game changer. The TT2 excelled in reproducing music, whereas the HMS allowed me to experience my music at a completely new and deeper level. That emotional connection defies price, and even at its $4995 price, I consider the HMS to be an absolute bargain.
 
Yes, the HMS/TT2 combo is expensive, clocking in at just over $10k USD. But I can also say I experienced my music in a way, and at a level of engagement, that I’ve never done before. If your budget allows, you owe it to yourself to audition this combo. 
 

In Conclusion: My Quest is Over
 
I’ve decided to buy the HMS/TT2 combo. It is significantly over my original approx. $5k budget, so I have some saving up to do. Yes I know: an audiophile who exceeded his budget - what a surprise. Hey - let he (audiophile) who is without sin (cost overrun) cast the first (Shakti) stone!
 
What clinched it for me was:

1. the transformational sound quality of the M Scaler,
2. the fact that the TT2 as a headphone amp competes head to head with my Cavalli Liquid Gold, which means that the cost of the HMS/TT2 combo could partially be funded by the sale of my Cavalli,
3. and the modularity, which would enable me to upgrade the scaler and the DAC incrementally.

 
So does this mean I won’t be doing more DAC reviews? Not at all, but every quest must come to an end, and I’ve strung you readers along long enough. You deserve closure!
 

Product Information:

• Product:  Chord Electronics Hugo M Scaler ($4995)
• Documentation:  Hugo M Scaler Manual
• Product:  Chord Electronics Hugo TT 2 ($5795)
• Documentation:  Hugo TT 2 Manual
• Rob Watts’ Hugo TT 2 Powerpoint deck (link 1.6MB)
• Rob Watts’ Hugo M Scaler Powerpoint deck (link 2.4MB)

Associated Equipment
 
Primary System: Here is a diagram of how the Chord Hugo M Scaler and Hugo TT 2 integrated into my system:
 

 
 
 
 
Roon Core:                     Intel NUC7i7DNBE running Audiolinux (in RAM)
Streamer:                       The Linear Solution DS-1 OCXO running Audiolinux (in RAM)
Headphone Amplifier:  Cavalli Liquid Gold
Headphones:                 Sennheiser HD800 (Super DuPont Mod), Focal Utopia, 
                                         Abyss AB-1266 Phi CC, Meze Empyrean
USB Regenerator:          SOtM tX-USBultra
Ethernet Switch:            The Linear Solution OCXO switch 
Reference Clock:           Mutec Ref 10 10MHz clock driving the tX-USBultra
Power supplies:             Utpone LPS-1.2 for switch & tX-USBultra, Paul Hynes SR-4 for DS-1
Power Details:               Dedicated 30A 6 AWG AC circuit, PS Audio P5 PerfectWave Regenerator
Power Cables:               PS Audio AC-12 (wall to P5), Cardas Clear Beyond (Cavalli Amp), 
                                         Cardas Clear (Mutec Ref-10), Cardas Clear to all DACs under test
                                         Cardas Clear (SR-4, SR-7), Pangea AC-14SE MkII to all PSUs
USB cables:                   Phasure Lush & Lush^2 USB
AES/EBU cables:          Cardas Clear
Clock cables:                 Habst 5N Cryo Pure Silver
BNC cables:                   Habst 5N Cryo Pure Silver
Ethernet cables:            SOtM dCBL-Cat7, Supra Cat 8
DC cables:                      Audio Sensibility Signature Silver (LPS-1.2)
                                         Paul Hynes fine silver (SR-4, SR-7)
Interconnects:               Cardas Clear XLR balanced (DAC to Amp)
Headphone cables:       Cardas Clear balanced and SE cables for all headphones
Accessories:                  Synergistic Research Tranquility Base XL UEF with Galileo MPC
 

Alternate system for speaker-based evaluation
 
Roon Player:                  Innuos ZENith MkII SE
Preamp:                         Audio Research Reference 6
Power Amp:                   Hegel H30
Speakers:                       Magnepan 3.7i
Subwoofers:                  2x Rhythmik 12g
USB Regenerator:         SOtM tX-USBultra
Reference Clock:          Mutec Ref 10 10MHz clock driving the tX-USBultra
Power supplies:            Paul Hynes SR-7 for  tX-USBultra
USB cables:                   Phasure Lush & Lush^2 USB
BNC cables:                   Habst 5N Cryo Pure Silver
Ethernet cables:            SOtM dCBL-Cat7
DC cables:                     Paul Hynes fine silver (SR-4, SR-7)
Interconnects:              Cardas Clear RCA and XLR (DAC to preamp)
Accessories:                 Synergistic Research Tranquility Base UEF
 
 
Acknowledgments
 
Many thanks to Cardas Audio for providing a loom of Cardas Clear cables to aid evaluation.