The APL HiFi DSD-SR MK2, I’m told it’s priced between the flagship DSD-MR MK2 ($40,000) and the new entry level DSD-AR ($7,000). That’s seriously what I know right now. I’m sure if I look through some old emails there will be a price in there, but as it stands now, I have no clue. In certain circles, all reviews should be done this way, while in other circles it makes the desired value judgement and class comparison impossible. That said, I’m charging forward with my assessment of this terrific digital to analog converter, without cost in mind. Given all the time I’ve spent with the DSD-SR MK2, one would think I’d know the price. However, this isn’t the case right now.
APL HiFi is a name that should be on everyone’s list. What list? The list of manufacturers one seeks out at audio shows, the list of manufacturers one combs through when narrowing down new component purchases, the list of manufacturers who are also really good people. After being on the business side of this hobby for over twelve years now, I can guarantee every reader that the last point on the list, the one about the really good people, is absolutely critical. If something goes wrong with a component, if there’s a support question, or even if nothing ever goes wrong, I want to do business with the good guys. Alex Peychev of APL HiFi is one of the good guys. His products are also some of the best bespoke components available today.
DSD-SR MK2 In Detail
The DSD-SR MK2 DAC is loaded with features that should make every audiophile drool, created for the sole purpose of increasing audio quality. Items such as Class-A, transformer-coupled output stage featuring special Lundahl audio transformers with amorphous core and OFC windings, paralleled Flagship DAC devices working in Mono DSD-only mode and special Class-A operation, no Op Amps, mechanical relays, or switches on the signal path, and native DSD256 support at both 11.289 MHz and 12.288 MHz. Those are just the highlights, not the fine details.
In true APL fashion the DSD-SR MK2 sounds like an Alex Peychev designed audio component. Ask anyone who is familiar with his designs or the APL product line, and they’ll immediately know what I’m writing about. The DSD-SR MK2 is the product of a master craftsman who understands both analog and digital audio in a way that not every designer comprehends.
The best way to show the Audiophile Style readers how authentic Alex is and to provide more factual details about the DSD-SR MK2, is to ask the man himself. I sent a list of eight questions to Alex. Here is that short interview.
Q1. Hi Alex, after meeting you in Munich and talking extensively about your products, your background, your approach to all things audio, and life in general, I found your authenticity very refreshing. Can you describe for the Audiophile Style community a little about your background, where you started, what you’ve done prior to starting APL Hifi?
A1. Hello Chris, it was very nice meeting you in Munich as well. It was a great pleasure exchanging notes about audio. I started designing audio equipment in high school . . . that was more than 35 years ago. I designed power amplifiers using germanium transistors because silicon devices were very hard to find (at that time, Bulgaria was controlled by a communist regime). Then I designed speakers using 10-inch drivers and ribbon tweeters (all with AlNiCo magnets). I also designed, 24-band equalizers and even tape decks with dedicated playback and recording drive mechanisms, as well as Dolby B, Dolby C and High-Com noise reduction systems. For several years I worked for a prominent company in West Berlin (I was there when Checkpoint Charlie came down), I also worked for Sony in Bulgaria. But in 1994 I decided to try my luck in the USA. I worked in a car audio repair shop at first and then moved to a much larger audio and video service center in Paulo Alto, CA. But I always dreamed of working for Sony Inc. So, I submitted my resume and was hired in 1996. At first, I worked in the Sony Consumer Electronics Group, but after six months I was recruited by the Sony Broadcast and Professional Group where I became a lead engineer responsible for eleven engineers. We mostly supported and debugged a wide range of products including JumboTrons, digital video and audio recorders, digital multi-effects, professional camcorders, microphones and everything else Sony offered to the pro industry at that time. We worked on Sony professional video and audio products costing as much as $3,500,000.
In 2003, I quit Sony and opened my own company. APL Hi-Fi was established in Bulgaria and then moved to California. The first product for which I became known was an extensively re-designed version of Philips’ SACD1000 player. I designed its tube output stage and upgraded its power supply and digital and D/A conversion processing. My success with that design allowed me to exhibit at CES in Las Vegas in 2004. I also had success converting a Denon cd platform. Later, one of my Swiss customers who became a great friend and supporter of APL asked if I could upgrade anything better than the Philips and the Denon. I believed the Esoteric UX-1 or X-01 players would be very nice platforms for such a project. That is how the Esoteric-based NWO players were born. I especially thank the entire Esoteric group of Japan for their help and support through the years, even to this day. To grow APL, I decided to move back to Bulgaria. Based on all the knowledge and experience I acquired, and with the help of some very bright people here, I was able to create the entire APL Hi-Fi product line as you know it today. It was a great challenge, especially when trying to get digital to sound like analogue, but I know we are on the right path.
Q2. Can you describe your approach to component design in general?
A2. This is an interesting question indeed. As we discussed in Munich, the actual circuit design is meaningless without knowing which components to use in the final product. For example, I can draw a power amp schematic on a napkin at the restaurant while we have dinner. But this power amp can be built in hundreds of different ways using different power supplies and active and passive devices, resulting in hundreds of different sonic results.
Another example is in both our DSD-SR MK2 and DSD-AR (our recently introduced and most affordable D/A converter), we are using Cirrus Logic’s D/A conversion devices, which were introduced 17 years ago. These devices feature a non-magnetic design. I have tried just about every DAC device available but was not able to equal the natural sound I am getting with the Cirrus part. I could only better that sound in our top-of-the-line DSD-MR MK2 where the DSD to analog conversion is processed using the discrete DSD filter I’ve designed. I could go on but will conclude by saying that designing audio equipment can be very difficult when it comes to arriving at the desired result. Every single passive or active component is very carefully chosen out of hundreds in order to arrive at what I consider to be a very realistic and extremely natural sound.
Editor’s Note: I asked Alex in a follow up question about the DAC chips used in the DSD-SR MK2. Understandably he wanted to keep some information to himself, but offered the following.
“While most manufacturers use a single stereo DAC chip, the DSD-SR MK2 uses four (4) stereo chips in mono mode. This means there are two (2) stereo chips in paralleled mono mode per channel.
Such an arrangement contributes to much better channel separation figures, increased drive capability of the DAC devices, as well as a reduced noise floor.”
Q3. Can you describe your goal for the DSD-SR MK2?
A3. This is also an interesting question. I was quite happy with the DSD-SR’s performance until the smaller and even less expensive DSD-AR was recently developed. As it turned out, the DSD-AR handily outperformed the twice as expensive DSD-SR. This called for a very serious redesign of the DSD-SR which became the DSD-SR MK2 version. As with all digital products we offer, the DSD-SR MK2 version was created to even more closely approach analog sound and to regain its place as second best in our digital product line, behind our truly remarkable reference DSD-MR MK2.
Q4. What are the differences between the MK1 and MK2 and, dare I ask, the flagship DSD-MR MK2?
A4. Because so much of our technology is proprietary, I’ll offer the information I can. The DSD-SR MK2 uses a revised power supply, improved digital signal processing including DSD256 output capability and a brand-new clocking circuit. We’ve also completely revised our analog output stage which features a non-negative feedback design operating in Class-A mode as well as output transformers wound with OFC wire. The MK2 upgrade includes changes to all the circuit boards including the main digital processing and analog output boards. While an outstanding product built to obtain the best possible sound from an all-solid-state analog output stage design, the DSD-SR MK2 remains the second best of our digital product offerings. Although the USB, RCA, AES/EBU, Optical and DTR digital inputs are the same on both the DSD-SR MK2 and flagship DSD-MR MK2, the latter uses our proprietary DSD to analog conversion algorithm and also a very fast, clean and dynamic differential tube output stage that is transformer-coupled at the output. After your experience with the previous version of the DSD-MR, Chris, and now having auditioned the DSD-SR MK2, I truly hope you’ll consider listening to the new flagship DSD-MR MK2. I would be very happy to get your opinion about the differences between the first DSD-MR, the DSD-SR MK2 and the new DSD-MR MK2, especially in Zero digital filtering mode allowing for DSD512 modulator speed.
Q5. What are the DTR inputs and what’s the significance of them?
A5. The DTR inputs are based on an I2S-like data and clock transmission protocol for both PCM and DSD data between our DNP-SR network streamer, our DTR-MR CD/SACD transport and our DSD-SR MK2 and DSD-MR MK2 dacs. However, unlike the widely known I2S connections, the DTR transmission protocol is balanced and does not use a Master Clock line. This also applies to the USB, S/PDIF, AES/EBU and Optical inputs which are processed in the same manner. Both DSD-SR MK2 and DSD-MR MK2 use a proprietary PLL circuit which locks the local Master Clock to the incoming data from any of the available digital inputs. Because the Master Clock does not have to operate over long cables, the resulting jitter figure is significantly reduced. To conclude, the DTR connection providing direct data and bit clock lines is likely the purest digital transmission method available.
Q6. Can you talk a bit about the inclusion of a volume control in the DSD-SR, your feelings about it versus using an external preamp or other means of attenuation?
A6. In the initial DSD-SR version, I used a hybrid attenuator, but then I realized that any attenuation in the digital domain (even partial) cannot compare to the attenuation done in the analog domain. Also, 99% of our customers use our digital products with a preamp or an integrated amp which use analog volume control. This said, in my opinion and experience, the attenuation done in the analog domain is superior to any attenuation in the digital domain, regardless of whether it is claimed to be lossless. It is because all the noise present in the line to the power amplifiers will remain at its maximum level. It is quite simple if you think about it. Any D/A converter has a certain dynamic range and noise floor figure. Any attenuation done in the digital domain is always done prior to the actual D/A conversion. This means you will inevitably lose signal to noise ratio and dynamic range with digital attenuators. Imagine a noisy tube preamplifier. Where will you prefer to have the attenuator? Before the noisy tube or after it? If you place the attenuator before the noisy tube, you will have all the noise at the output. But if you place the attenuator after the noisy tube, the sound will improve dramatically.
Q7. You offer a selection of digital filters, PCM to DSD conversion, and upsampling, where many designers don’t give the consumer this choice, do you have a favorite and why?
A7. We have tried many types of digital filters prior to the DSD modulator. Most digital filters do not sound natural to me, although some will sound more forgiving overall. After many experiments, we prefer digital filters which have an equal amount of pre and post-ringing while preserving the linear phase response. All of our D/A converters come with Normal (sharp roll-off) and Slow (slow roll-off) digital filters. Both provide a linear phase response. And instead of employing commonly used 8x oversampling, after many experiments, we prefer 4x oversampling. While uncommon, we also offer a completely non-oversampling mode prior to the DSD modulator that we call a Zero filter. It’s fascinating that the Zero filter mode is always preferred by audiophiles who appreciate analog sound from vinyl or reel-to-reel machines. Particularly with the reference DSD-MR MK2, you need to hear the Zero digital filter and DSD512 modulator speed. The result is remarkable and my favorite.
Q8. The DSD-SR is based on a modular design, enabling future upgrades. Can you describe this upgradability and limitations if any?
A8. The DSD-SR MK2 and the reference DSD-MR MK2 have separate boards for the power supply, digital inputs, main digital processing and output stage boards. This means that
all major parts are built on separate PCBs, allowing for a cost-effective, future-friendly upgrade. For example, if we have a new output stage design, only the analog output board needs to be replaced. Similarly, if there is an upgrade to the power supply or digital signal processing boards, only those would need to be replaced. Are there any limitations? Sure, the limitation is called Alex Peychev, because he cannot usually settle on upgrading just one of the boards. Yet, regardless of the upgrade, the customer’s upgrade cost is far lower than the cost of selling his current model and buying a new one.
Thank you so much, Chris, for giving me the opportunity to help you and the AudiophileStyle community to better know me, APL and our products.
While of course APL is a business, I take enormous pleasure in knowing I’ve helped so many people to hear music in a way that brings them tremendous happiness.
My Own Take
I used the APL HiFi DSD-SR MK2 in two different audio systems. My main Constellation Audio > Wilson Audio Alexia Series 2 system and my new headphone based system with the RAAL-requisite SR1a true ribbon headphones. The main part of my recent SR1a review was conducted using the DSD-SR MK2 DAC. In that review I said the SR1a was, “unequivocally the most sensational audio product I’ve ever heard.” Readers should consider that a major role in my listening sessions was played by this APL DAC. I listened to several DACs with these headphones and the APL gave me a sound unlike anything in my considerable, and possibly ridiculous, DAC stable.
One really great feature of the DSD-SR MK2 is its ability to handle different digital inputs and volume control options. What I mean by this is the DAC performs exquisitely well in several different configurations, giving consumers fantastic options to go with the flow or experiment a little on their own.
On its own the DSD-SR MK2 has built-in volume control and can be connected directly to a power amplifier. This configuration was really good and is highly recommended for those with only digital sources, who also want to save a little or a lot of money bypassing a preamp. It can be done with the DSD-SR MK2 and done well. Along similar lines, one can also connect the APL DNP-SR streamer to the DTR input of the DAC for a very seamless and excellent sounding experience. The DNP-SR is a Roon Ready endpoint featuring a full Auralic digital platform embedded inside. The combination is very good and shouldn’t be overlooked.
Much of my time with the DSD-SR MK2 was spent on the more esoteric side of HiFi, pushing the boundaries. Given that this DAC accepts native DSD256 and bypasses filtering when receiving DSD content, I decided to try Jussi Laako’s HQPlayer upsampling everything to DSD256, and at times acting as the sole volume control for the system. I also put my Constellation Audio Inspiration preamp in between the DAC and amps on occasion to hear what was possible in an analog attenuation configuration. The sonic results of these configurations were absolutely wonderful.
Which configuration was best? I really can’t answer with a high degree of certainty. I heard things in each setup that I enjoyed immensely. We mustn’t forget the other factors such as ease of use and tweeter safety, when we consider using a product like the APL DNP-SR compared to connecting the DAC straight to the amps with only software volume control. There are many roads leading to Rome with this DAC. Some are winding, some are straight, some are a bit bumpy, but all eventually arrive at the same highly satisfying destination.
I must also note that using the DSD-SR MK2 with an external upsampler or volume control is just an option that says nothing about the DAC’s native sound quality or Alex Peychev’s design prowess. Many people get mixed up thinking DACs that I combine with external upsampling or volume control, require these methods to sound good. Nothing is further from the truth. I absolutely love options and take advantage of them when given the chance. As a counterpoint to my experience, I know of one person who recently moved away from HQPlayer and other esoteric options, in favor of an APL DAC with the APL DNP-SR streamer. He loves the simplicity and above all, the sound quality in his reference level system.
Listening through the APL DSD-SR MK2 being fed by HQPlayer upsampling to DSD256 at 12.288 MHz and the ASDM7 modulator, the sound was incredibly engaging, vivid, saturated, and again unlike other DACs I have in my system. Listening to my new favorite Bruce Springsteen album, Western Stars – Songs From The Film, the DSD-SR MK2 reproduced the entire album in what I’ll call a rich tone versus some other DACs that sounded a bit flat. I don’t think this experience should be lumped into the musical versus clinical debate either. The APL sound is different from that endless debate.
Everything from Springsteen’s voice to acoustic and electric guitars and the backing of the full orchestra, sounded better to me through the APL than through my other DACs. Who’s to say what’s neutral or what Springsteen’s nearly 70 year old vocals are supposed to sound like. I can only testify to what I heard in my systems, compared to some others, and how the music from the APL DAC was just more engaging.
Another way I can describe the DSD-SR MK2 is lush, in the way that tube amps can only dream of, and in a way that’s not at all full of harmonic distortion. The DSDS-SR MK2 way is chock full of detail and transparency as part of this lushness. The DSD-SR MK2 is a solid state design that reproduces John Coltrane’s Standard Coltrane album in a way that figuratively forces the listener to sit through the entire album. I have several masters of this one, and listened mainly to the 2019 Kevin Gray DSD version during this review. On the opening track, Don’t Take Your Love From me, Coltrane’s tenor sax is absolutely as lush and gorgeous as sinking into a large velvet pillow with a glass of your favorite red wine. In addition, Paul Chambers’ double bass, Jimmy Cobb’s drums, and Red Garland’s piano are all superbly delineated and distinguished for listeners who love to audibly meander through the recording studio in their heads.
Switching gears quite a bit, I really got into the Dixie Chicks’ new single Gaslighter while reviewing the DSD-SR MK2. This 3:23 track is pure pop country, compressed as usual, but created by masterful musicians. While Natalie Maines’ voice isn’t as pure sounding on this track as it is on Let Him Fly, it’s backed by wonderful vocals from sisters Emily Strayer and Martie Maguire. From the opening verses, this track sucked me in when listening through the APL DSD-SR MK2. Each of the three women’s voices are clearly audible in their own right, distinct from the others, and easily identifiable because of the APL DAC’s performance. In addition to the three voices becoming one, yet remaining separate, the richness of the vocals is both real and seductive. They were so seductive that I listened to this track on repeat a half-dozen times per sitting. That’s what the APL DSD-SR MK2 does, it forces one to keep listening. This happened to me countless times during the review period.
I could say I was testing out this DAC’s dynamics with the next track, but that would be a misnomer. I love this track and the entire album. It just so happens to be a great test of dynamics. The Terumasa Hino Quintet’s Stella By Starlight, from the Live! album, has a DR (R128) score of 21 and will wake any audiophile asleep at the listening chair. Hino’s trumpet weaves through this track in short bursts, they could also be considered blasts by the most uncouth of listeners, demanding instant transient spikes from a nearly sleeping quintet. The DSD-SR MK2 nailed this track for the most part, but wasn’t quite up to my current transient champion the $30,000 EMM Labs DV2. However, without the ability to place both DACs next to each other, going through the same preamp, and A/B’ing them, most listeners wouldn’t spot the sonic difference. It’s one of those things that those of us who listen to this stuff all day every day, doing this so-called “job,” recognize only under direct comparison.
Seeking a detail and delicacy through the DSD-SR MK2, I played an old favorite of mine that I hadn’t virtually dusted off in many years. David Oistrakh’s Bruch: Scottish Fantasy, performed September 13-14, 1962 at Kingsway Hall for Decca. “Track” 1, Op. 46: Introduction. Grave. Adagio cantabile, is an absolutely gorgeous piece of music, played by arguably the best violinist the world has ever seen. Through the DSD-SR MK2, the sweetness of Oistrakh’s 1702 Conte di Fontana Stradivarius is like an addictive nectar for audiophiles like me. Starting very slow, almost whining and winding his way through this beautiful piece of music, one can’t help but feel the emotion in Oistrakh’s playing. Sure this is undeniably great music from great musicians but it doesn’t always sound as detailed, as delicate, as great through other DACs as it does through the APL DSD-SR MK2. Trust me, I’ve done the homework.
The APL DSD-SR MK2 is yet another “hidden” gem from Alex Peychev and his company APL HiFi. I say hidden because almost nobody, in the US at least, knows about APL. There is always something to be said for finding hidden gems, putting together a system with bespoke and somewhat unknown components, and enjoying the end result of such a journey. During my time with the APL DSD-SR MK2 I certainly felt like I’d found a key piece of audio gear that made me smile like Sylvester the cat, while a feather from Tweety shoots out of one side of his mouth. The DSD-SR MK2 was my ace in the hole that nobody really knew about. I was delightfully smiling while enjoying my favorite music, but knowing full well that most people were unaware of how I was getting to such a sonic destination.
I guess the cat is out of the bag. My secret APL ace now has a spotlight upon it, and for good reason. The DSD-SR MK2 is a terrific DAC that sounds nothing like the competition. Vivid, saturated, analog (in the best way), and lush are solid characteristics of this DAC sonically. Not to be confused with harmonically distorted components, the DSD-SR MK2 also delivers the goods with respect to dynamics and delicacy, while maintaining its seductive sonic illusions.
At $15,000 (yes, I checked before writing this conclusion), the APL DSD-SR MK2 is certainly not inexpensive but is definitely capable of providing more enjoyment and better sound than many other DACs at double the price. Try it, put it next to the competition, in the same system, and sit back with the substantial metal remote in one hand and a favorite beverage in the other. I’m pretty sure of the results. This coming from a teetotaler, who enjoyed many pots of Organic Nepali Golden Black Tea while enjoying the distinctive sound of the APL DSD-SR MK2.
Community Star Ratings and Reviews
I encourage those who have experience with the RAAL-requisite SR1a to leave a star rating and quick review on our new Polestar platform.
Additional APL HiFi Digital Products
Where to Buy
- Source: QNAP TVS-872XT Roon Core, Aurender W20SE, CAPS 20.1
- DAC: dCS Rossini, EMM Labs DV2, Berkeley Audio Design Alpha DAC RS3, Schiit Audio Yggdrasil
- D-to-D Converter: Sonore Signature Rendu SE (optical), APL HiFi DNP-SR, EMM Labs NS1 Streamer, Berkeley Audio Design Alpha USB
- Amplifiers: Constellation Audio Mono 1.0 / Monoblock Power Amplifiers
- Preamplifier: Constellation Audio PreAmp 1.0
- Loudspeakers: Wilson Audio Alexia Series 2
- Digital Signal Processing: Accurate Sound, HQPlayer
- Remote Control Software: Roon Remote, JRemote, Aurender Conductor
- Remote Control Hardware: iPad Pro
- Playback Software: Roon, JRiver,
- Network Attached Storage (NAS): QNAP TVS-872XT
- Audio Cables: Transparent Audio Reference Interconnects (XLR & RCA), Transparent Audio Reference 110-Ohm AES/EBU Digital Link, Transparent Audio Reference Speaker Cables, Gotham GAC-4/1 ultraPro Balanced XLR Audio Cable (40′)
- USB Cables: Transparent Audio Premium USB Cable
- Power Cables: Transparent Audio Reference Power Cables
- Power Isolation: one 4kVA and one 5 kVA 512 Engineering Symmetrical Power Source
- Ethernet Cables: Transparent Audio High Performance Ethernet Cables
- Fiber optic Cables: Single Mode OS1-9/125um (LC to LC)
- Acoustic Room Treatments: Vicoustic Diffusion and Absorption, ATS Acoustics Bass Traps
- Network: Ubiquiti UniFi Switch 24, Ubiquiti UniFi Switch 8-150W x2, Ubiquiti UniFi Switch 16 XG, Ubiquiti UniFi Security Gateway Pro 4, Ubiquiti UniFi AP HD x2, Ubiquiti FC-SM-300 Fiber Optic Cable x2, UF-SM-1G-S Fiber Optic Modules x6, Commercial Grade Fiber Optic Patch Cables, Calix 716GE-I Optical Network Terminal, CenturyLink 1 Gbps download / upload
This graph shows the frequency response of my room before (top) and after (bottom) tuning by Mitch Barnett of Accurate Sound. The standard used for this curve is EBU 3276. This tuning can be used with Roon, JRiver, and other apps that accept convolution filters. When evaluating equipment I use my system with and without this tuning engaged. The signal processing takes place in the digital domain before the audio reaches the DAC, thus enabling me to evaluate the components under review without anything changing the signal further downstream.