Article: dCS Bartok Review

[austinpop]

View full article

[austinpop]

2 minutes ago, taipan254 said:

Great review, Rajiv. Actually reading this at the Odd Duck on S. Lamar. I dig your city! 

 

Cool! That's a great spot.

 

2 minutes ago, taipan254 said:

I will admit that the thought of a DAC that can't be improved with upstream tweaks scares me a bit... I guess I'd have to learn how to build speakers! 

 

Well... not quite "can't," just not as much. Remember - I said I was tempted to stop tinkering... not that I would actually stop!

[austinpop]

8 hours ago, The Computer Audiophile said:

I absolutely love this review!

 

As I told Rajiv / @austinpop, the saying, "I'm not worthy" from Wayne's World comes to mind when I read through this review. Such a great effort, I feel honored to publish such a fine piece of work. 

 

Waynes World, Waynes World
It's Party Time, It's Excellent

 

[austinpop]

5 hours ago, keeper said:

I got a shock when I looked at the price of the headphones........

 

It's a brave new world. These are not your father's Grado's!

[austinpop]

4 hours ago, ted_b said:

I am only saddened to know that you got to hear the Rossini.    Oh well, bucket list stuff. 

 

4 hours ago, wdw said:

must admit to be saddened when you brought in the Rossini...had always assumed, as you did, that the improvements would be apparent but modest...oh, well!   As mentioned above, bucket list stuff!

 

So true. I'm like a moth drawn to flame. But I couldn't resist!

[austinpop]

3 hours ago, tmarshl said:

Dear Rajiv,

 

Thank you for a detailed and objective review.  This review is refreshing, considering so many "fanboy" reviews that are published.  I have owned the Bartok since it first became available, and I consider it to be a perfect piece of gear for me.  I feed it with a Roon Nucleus via Ethernet.  I discover and listen to a lot of music via Tidal Masters and Qobuz.  I have been re-listening to my entire music collection on the Bartok, and have been amazed at what I have been missing (I upgraded from the Chord Dave).  The organic realism of the sound from the Bartok into my Focal Utopia headphones (also with Transparent Ultra balanced cable) and Raal Requisite SR1a "earfield speakers",  without warning will jolt me out of my listening reverie with sound so authentic that I could swear I am in the same room with the musicians. 

 

Thanks. I really need to get my hands on the RAAL. Time to go to a CanJam!

[austinpop]

3 hours ago, rossb said:

Excellent review, Rajiv. Thanks in particular for the comparison to the Ayre QX5. As a QX5 owner I was wondering whether the Bartok would be a worthwhile upgrade, and my wallet is saddened to hear that this is the case.

 

These things are relative. The QX-5 Twenty is, and will continue to be, a fantastic-sounding component. Like I said in the review, it's only when you put components side-by-side, that "high-end comparisons are cruel."

[austinpop]

4 hours ago, Le Concombre Masqué said:

Rajiv,

 

I wish I could share the general enthusiasm for your reviews. I can certainly praise your dedication but I find them pretty moot for they don't include the leveraging of computer based processing.

 

I can see references to usb usb cables ethernet stuff etc that certainly pertains to computer world but I don't see* leveraging of computer based processing.

 

I'b be interested in comparisons between DACs fed by, say, HQP delivering to the max of the DAC capabilities.

 

You're using a computer anyway, right ? so since you do mention diminishing-returns region and  price-performance curve, what $300 of software can save us in $K of hardware is of interest, IMO.

 

*Roon does not seemed to be used for any heavy processing and you even specify "library management and output to the DAC all in one"

 

I'm not sure exactly what you mean by "leveraging of computer based processing." You seem to be implying that running SW upsampling via HQPlayer is always superior to a DAC's internal upsampling and filters. That is certainly not my experience. It is very DAC-dependent, and as it happens, most of the DACs I have had in my system have sounded best when delivered the recording at its native sample rate. I don't doubt that some DACs in some cases benefit greatly via this approach.

 

Given the context of this thread, let's keep this discussion focused on the Bartok. This article was long enough already that I can't describe all the experiments I tried. Early on in the evaluation, I did a quick test with Roon to see if using Roon upsampling to PCM 352.8/384 or DSD128 (the max sample rates the Bartok will accept) yielded any improvements. Had the results been profound, I would have been motivated to try HQPlayer. But in fact, Roon upsampling sounded no better than no upsampling - in fact, I thought it degraded the sound.

[austinpop]

3 hours ago, bobflood said:

This is one of the best reviews that I have ever read in my 50+ years in this hobby.

 

43 minutes ago, matthias said:

Thank you,

this is a review how all reviews should be.

 

Matt

 

Thanks for the kind words.

[austinpop]

38 minutes ago, Ran said:

Any issues with the unit performing as a DLNA renderer? Gapless playback? Did you try BubbleUPnP client as well? Kudos for using MinimServer!

 

I didn't play with different UPnP servers or clients. As it happens, dCS recommends minimServer, and I have been a long-time user of minimServer, so... serendipity.

 

No issue with gapless. The Mosaic client is so nice, I never considered anything else like Bubble.

[austinpop]

1 hour ago, AudioDoctor said:

More for curiosity’s sake than I might buy either of them, but I’m curious to know how this stacks up against a Warwick Acoustics Aperio and headphone combo. 

Not sure where the Aperio is. Definitely interesting.

 

[austinpop]

1 hour ago, Le Concombre Masqué said:

Maybe there's no point at all, then, in sharing each other's opinions and evaluations...

 

Here it's a 13K$ point. I'd like to see rankings on scales ; ie : if, for free once you have HQP or if you're testing it, you can set your perceived evaluation of differences between,  say, mqa mp and long lp filters to 10, how much is the perceived difference between 2K and 25 K Dacs (3, 8, 9 ; need to set a new range where that difference is 20 while the difference between filters become 7 or what have you...)

 

dCS can send me a Rossini to change my mind ; I'm not convinced that 20 K put in a DAC brings much more than a change of filters

 

Like it or not, a DAC is so much more than "just" its filtering and D/A conversion design. Indeed, if my experience has taught me anything, it's that modern DACs live and die by how well they get the good old-fashioned basics right:

1. analog output stage design
2. power supply design
3. clock architecture
4. design or choice of chipsets used for interfaces like USB, ethernet
5. Electrical isolation strategy - both at input, as well as internally
6. Mechanical isolation via careful chassis design
7. Parts quality, board layout, and quality of connectors
8. And last, but certainly not least - filtering and D/A conversion design.

To focus on #8 alone is to miss all the other factors that distinguish DACs from each other.

[austinpop]

7 hours ago, OpSpl said:

you have to factor in also the cables cost !

 

Yes indeed!

[austinpop]

5 hours ago, Darryl R said:

Clearly Rajiv has an affinity for this kind of work.  I was particularly interested in his comparisons of combinations of upstream components with USB and Ethernet input.  We all seem to be looking for that magical optimization.

 

But when I saw he was listening to Sylvia McNair (my favorite female voice), I knew this review was serious stuff

 

Well done.

 

Hi Darry,

 

Thanks for your positive comments.

 

When I first started writing reviews, I was a bit apprehensive about revealing my musical tastes and choices, which are so uniquely personal. I'm glad I overcame that, because the response here - to my identifying the music I used and my sharing of playlists - has been overwhelmingly positive. Enjoying the music that moves us is ultimately what this shared passion of ours is all about.

[austinpop]

3 hours ago, Darryl R said:

 

Honestly, we must attribute a good portion of your talent to the fact that you live in the greatest state in the union

 

PS - If you liked Sylvia McNair in the Laudamus Te, you must hear her in the Slatkin Carmina Burana, In trutina and Dulcissime, where she soars to the heavens.  PM me if you're interested.  I also have a lot of Queyras' recordings.

Thanks for the pointer to Sylvia’s Carmina with Slatkin/St Louis. It’s lovely. But dammit I was peacefully working here, and this made me go and listen to a bunch of Carmina’s and now ... totus floreo, with no domicella around! 🤓

 

For the record, the Ozawa/BPO version, with Thomas Hampson and Edita Gruberova, is still my favorite Carmina Burana. 
http://open.qobuz.com/album/fm0kymmr3f05a

[austinpop]

11 hours ago, donem said:

Given your comment above about Roon, I must ask about Alan Taffel's review of the Rossini in the September 2018 issue of TAS. He reports that there is a significant difference with Roon bettering dCS significantly. He "blames" it on the renderer. According to him, Roon's renderer and dCS' renderer (for UPnP) run side by side on the same network card. In comparing Roon to dCS' own proprietary renderer Taffel reports "It took no time at all to determine that Roon constitutes a significant sonic upgrade over dCS' renderer." After citing a few examples he goes on to say "The improvements Roon reaps are not subtle; rather, they're immediately and definitively obvious." These quotes coming on page 161 although he broaches the subject on page 154. Is he for real?

 

I can't believe that this sat very well with dCS or that they would leave this issue unaddressed. While the Rossini is beyond my means, I have been considering the Bartok for my system and am concerned that this disparity of sonic performance would be part of the Bartok as well. I was glad to see your brief comment above but am still looking for some kind of response from dCS about Taffel's assessment.

 

My biggest challenge is to get just a DAC. Don't need a player; don't need a preamp, don't need streaming, don't need Roon, don't want MQA, don't need headphone amps, all of which are price included in the Bartok. My current digital front ends consist of an ARC CD Reference 7 with CD 8 power supply and an OPPO 105D. I have not been able to assess the SQ of either DAC relative to what is available today in the $10k to $15k price range, including used gear. I was directed to your article and this site from an Audiogon Discussion Forum and am glad to have found it. Thanks for your review.

 

Hi @donem

 

Welcome to Audiophile Style.

 

I  can't speak to the TAS review or the state of the dCS renderer in 2018, since my experiences with both the Bartók and Rossini were more recent. All I can say is that my listening tests did not show any significant Roon advantage. My detailed findings are in the review.

 

As you are probably aware, the Bartók does come in a flavor that does not include the headphone amp, priced at $13,500.

[austinpop]

1 hour ago, mrkwr said:

Coming late to this, but I must say like others that it’s a really great review: I particularly value the comparisons to other DACs and between config options. Most reviews just say the unit sounds great which – as you say – is just table stakes at the price point, so gives you precisely zero new information. 

 

FWIW, I spent 2–3 weeks doing an extended home trial and direct comparison of the Bartok with my Benchmark DAC3 HGC earlier this year (mid-March). Music sources were streaming from Qobuz and from local files on a basic WD NAS, some hi-res inc. 96/192 and DSD, mostly 16/44. For the Benchmark I was using the bottom-of-the-range Auralic Mini to handle the streaming, so roughly a £2750 combo compared to £10,000 for the Bartok (ignoring the headphone amp, though the review sample did have this). Downstream I was listening to Harbeth SL5+ Anniversary Editions via a Benchmark AHB2. I also listened on headphones, which while a little bit more revealing is something I rarely do as I much prefer the experience of listening via speakers. (The more so since developing mild tinnitus.)

 

I listen primarily to classical music (I  play piano and violin to a fairly high amateur standard, and go to a lot of classical concerts, so have a good idea of what I want these instruments to sound like), and for me the priority is perceived realism/naturalism of acoustic instruments and groups from solo piano/violin up to full symphony and chorus. I also enjoy a much contemporary jazz and listened to a fair bit of this during the trial. 

 

My findings:

 

* It wasn’t always easy to tell the two apart: often you had to listen very closely to pick up minor details. Probably half the tracks I listened to fell into this category: I simply couldn’t have distinguished which DAC was playing without really focussing hard (and sometimes not even then).

* Other times, there was a clearly perceivable advantage to the dCS: this was most frequently in terms of either (a) greater perceived naturalism/realism (especialy on human voice or violin) or (b) more presence/body to the bass (e.g. double bass in jazz ensemble, bass section in orchestra). In other cases the dCS advantage was one of better separating musical voices, e.g. allowing easier following of separate lines/polyphony and clarification of complex textures, or more vivid presentation (dynamics?). 

 

Overall, though, my personal conclusion was that although the DCS offered real advantages, and that the differences all played directly to my priorities, the benefits of the dCS were not sufficiently large to warrant the additional cost – the “problem” was that the Benchmark is just too good, even if not *quite* as good as the dCS. 

 

I ended up sending back the dCS rather reluctanly, as I’d seen myself as the sort of listener that would enjoy being a dCS owner, but was forced to admit the Benchmark came too close for me to spend the extra cash. I also bore in mind that this comparison was done with a very modest streaming unit upstream of the Benchmark (the Aries Mini) which could potentially be upgraded for significantly less than the cost of upgrading to the dCS.

 

Just one listener’s opinion, of course …

 

(PS: I ended up spending the cash I’d earmarked on renovating my Steinway Model O, so it’s still gone in the service of music (re)production!)

 

What a wonderful post! Thanks for your comments. This is exactly the approach I advocate:

• audition it in your own system, and
• apply your own judgement to whether performance gains are worth the cost.

Enjoy the music!

[austinpop]

Hi Jacques,

 

While I can’t speak for the ER, as I have never heard it with the Bartok, my expectation, based on my experience with the SOtM switch with the Bartok, is that it should make a positive impact. 
 

The good thing is the ER comes with a 30 day guarantee, so you can try it for yourself. Ethernet enhancements seem to be very system-dependent.

[austinpop]

Wow thank you - that is high praise indeed! I am glad you enjoyed the article and found it useful.

[austinpop]

2 hours ago, Martin Herløv Andersen said:

I am SUPER impressed with your review.

The depth of the review and technical skills you are posses are at another level.

Keep them coming 🙂

 

Thanks, I appreciate your kind words.

 

2 hours ago, Martin Herløv Andersen said:

One question, how much did you listen to the dac in a stereo setup?

 

I'm assuming by "stereo setup" you're referring to a speaker-based setup?

 

It would be fair to say that the bulk of the listening I did with the Bartók during the time I had the review sample was with headphones, as that is my primary listening system at home. That said, I did enough listening on the speaker setup described in my review to assess it thoroughly. According to my notes, it was 4 distinct listening sessions of 3-4 hours each.