Beolab
Member of the Trade (Reseller)
- Joined
- Apr 16, 2015
- Posts
- 1,279
- Likes
- 213
I fully agree that nothing sounds like live unamplified music and while the gap can never be fully closed, it can be approached much further than it is now. In the end, whether it be live unamplified music or a facsimile of it, as you've stated, it is that emotional connection that I believe we each seek.
Thank you, Fredrik. I was hoping someone would bring this up. I have repeatedly alluded to Chord's upcoming digital amps but I believe it will be the combination of these amps and DAVINA that will more fully show us what the DAVE is capable of. The onus will now be on Rob to deliver and I have full confidence he will but I believe these two components will erase all doubts about where the DAVE stands compared to the rest.
Until recently, I, too, have wondered what DAVE's ultra-low noise floor, lack of noise floor modulation, small signal linearity, almost unmeasurable distortion levels, etc. all means if other DACs can approach the DAVE's performance with their supposedly flawed designs but over the past couple of months and again this past weekend, I have had three very revelatory experiences that have changed my perspective on everything.
(1) As some of you know, a couple of months ago, I compared a variety of interconnects that differed in terms of metallurgy using my DAVE and my Abyss/HE-1000. It ranged from UP-OCC grade copper (the purest copper that is commercially available in the world today and is 99.9999% pure) to UP-OCC grade silver to an alloy consisting of OCC Silver/Gold and eventually to one of High Fidelity Cable's entry level interconnects (CT-1E). As I moved from copper to silver to the HFC interconnect, it was very clear that resolution (meaning detail retrieval) improved. The only thing that the silver/gold alloy seemed to add was greater tonal body and a more natural sounding timbre but resolution-wise, it appeared equivalent to UP-OCC silver in terms of resolution. Short of the HFC interconnect, I would go for the OCC Silver/Gold alloy. Given the choice, however, I would go with the HFC interconnect every time. The point is that whatever you use to connect your DAVE to your amp will have a significant influence on resolution and thereby transparency but it shouldn't require any stretch of the imagination to understand that even the best interconnect in the world can never be completely transparent. Just like no preamp will always be more transparent than any preamp, the same thing goes for interconnects.
(2) Last month, many of you recall that I compared a very high-quality class D amplifier against a First Watt J2 class A JFET amplifier by Nelson Pass against the DAVE directly driving my ALNICO monitors. As part of this comparison, I also introduced a tube buffer with a variable bias that allowed me to vary the impact of the tube from 0% (complete bypass) to 100%. With the tube in full effect, there was this nice warm colorful bloom, a sound I have been familiar with for years that was indeed very pleasant and for certain poorly-recorded tracks, I actually preferred it. In the absence of any comparison, it actually sounded very transparent as if the artists were in the room as they say. As I gradually reduced the effect of the tube, it was very interesting to see how this colorful bloom was replaced by clarity and depth and I realized that what I thought was transparent wasn't so transparent after all. As I went from the class D amp to the First Watt J2, it was very evident that Nelson Pass knows how to build amplifiers as the improved refinement was nothing short of breathtaking. Given the switching properties of class D amps, moving up to the class A First Watt J2 also resulted in a very noticeable improvement in transparency. As I fully bypassed the tube buffer (ie preamp) and switched from the class D amp to the First Watt J2, the improvement in transparency was so significant that I was thinking there was no way I could top this and that this was as transparent as things could get for my system. This is what I meant in a previous post when I said that we all think our systems are transparent until we hear that next "something" that is even more transparent and I believe there will always be that next "something" until you actually find yourself seated in the concert hall. In my case, that next "something" happened to be DAVE driving my speakers directly. This change has been so significant that paired with my upper-end High Fidelity Cables speaker cables, I have yet to hear a more transparent sounding headphone or speaker setup. Imagine a class A speaker amp that requires no transparency-robbing interconnects, a noise floor of -180dB with no noise floor modulation, almost unmeasurable distortion levels and the full speed, linearity, dynamic range and bandwidth of the original DAC signal and you begin to understand why this might fully change the landscape of 2-channel speaker listening. My comparison of my DAVE against the dCS Vivaldi and Nagra HD took place weeks ago and nothing I heard that night or at RMAF this past weekend has matched the detail retrieval I am getting with my DAVE directly driving my monitors. If you think you have a transparent speaker system now, just wait until next year. I believe you will be forced to recalibrate in your brain what transparency from an audio system means.
Now some will say that DAVE directly driving my monitors shouldn't be that different than DAVE directly driving headphones. I even recall Fredrik suggesting that his Simaudio amp is maybe only 5% less transparent than DAVE directly driving his Abyss but what I'm hearing is well beyond that. As I mentioned in a previous post, with the appropriate recording, depth is conservatively 20% better than the First Watt J2 and detail retrieval with these ALNICO monitors is superior to any headphone I currently own. I believe the shortcoming of headphone listening is the headphones themselves and their relatively infant technology but I do believe things will continue to improve. First of all, as has been mentioned recently, headphones are limited in their abilities to fully reveal the performance of a DAC and I agree with this. While headphones can be better with regards to resolving detail, they can never fully compete with speakers with regards to resolving time. Not just depth from the standpoint of knowing that the brass section is seated 5 meters behind the woodwinds but even the reverberant sound field of the strumming of a guitar is better appreciated with speakers. With headphones, there's just too short of a space for certain sounds to fully unfold. The second problem is with headphone technology and the paucity of truly high-end headphones. Of course, high-end headphones haven't been around anywhere as long as high-end speakers and so it is to be expected headphone technology needs time to catch up and that we would have fewer choices when it comes to state of the art headphones but it sure would be great if there was a headphone that had the detail resolution of a Utopia, the air and musicality of an HE-1000, the clarity of an SR-009, the midrange of an LCD-4, the bass of the TH-900, the soundstage of an HD800 and the imaging of the Abyss. While I love the headphones that I have, oh how I wish that they could move me as much as my ALNICOs.
(3) This past weekend, while at RMAF, I paid the MQA rooms a visit. I have seen these demonstrations before and have had mixed emotions. This time was different. While at the Aurender room, they were demonstrating their new A20 music server with built-in DAC. This latest Aurender has an MQA decoder built-in and I got to hear a track by Adele with both MQA decoding turned off then on. My impression? Ho hum. Maybe a little improvement but I had to struggle to hear it. On to the Mytek Brooklyn DAC and while it was better, at no time did I feel like ditching my DAVE and buying a Mytek.
Then I went to this room:
That's right, all MSB DACs are now MQA-capable including their Select II. During a conversation with an MQA rep at CES back in January, I was told that MQA works by taking advantage of the oversampling capabilities of a DAC to remove pre- and post-echo. For those not aware of what this results in, it is supposed to lead to time smearing and so its elimination is supposed to improve clarity, timing fidelity, and depth and consequently result in less listening fatigue. Back in January, I recall being told that DACs that could oversample to very high levels had the potential to perform better than DACs that couldn't oversample much at all and so this technology was not well suited for R2R DACs. I'm not sure if things changed or else I misunderstood but this is no longer how MQA apparently works. All a DAC needs now is an MQA decoder to feed the DAC a decoded signal that has been optimized for the performance characteristics of that DAC but that oversampling had nothing to do with it and I was told by Bob Stuart that this technology should be compatible with all DACs. Fortunately, Rob himself has suggested that this should be possible with Chord DACs but the question is whether it really results in a worthwhile improvement. As I stated, with the Aurender A10 and the Mytek Brooklyn, for me, it was a noticeable improvement but it wasn't anywhere close to what I was already hearing at home with my DAVE. With the MSB Select II, however, it was quite a different story. With a certain piano track, I was simply dumbfounded by what I heard and I believe the entire room was as well. With MQA-enabled, the dimensionality of the sound, especially the depth rose to another level. Details were clearer with noticeably less smearing. With MQA off, this DAC sounded very pedestrian by comparison and I have never found the Select II to sound pedestrian in the past. The point of this is that it is clear there is much to be gained by overcoming the limitations of current ADCs. While MQA is more of a patch that addresses some shortcomings, I see DAVINA as the bigger potential game changer as it is supposed to have the full performance characteristics of the DAVE with regards to noise floor, lack of noise floor modulation, DR, etc. If the promise of DAVINA proves to be real and in theory, it should easily improve upon what MQA provides, while all will benefit, those with a DAVE should benefit the most.
This is why I believe 2017 will be such a landmark year for audiophiles but especially speaker audiophiles. In addition to whatever else Rob might have up his sleeve, if my experience with my DAVE directly driving my speakers is any indication of what is to come, I can only imagine what impact both DAVINA and Chord's digital amps will provide and I believe only then will we know what DAVE can really do.
Thanks Roy for your great post as always!
I wish that the DAVE where miles better than dcs vivaldi , MSB Select II , PS Audio , Nagra, but it is more down on subtle characteristics of the presentation, and i am still not satisfied with our answers why it does not leave the rest of the pack in a dusty cloud behind, so in my book the Delta Sigma Puls Array must also got some flaws / bottlenecks in the design, factors that we not yet have discovered or are not aware of how much it affect the sound to this date.
Maybe something Rob will discover now in the DAVINA project, or in a couple of years time, when the mesauring devices gets even better.
(For example, Rob discovered that if he set the noise shaper to -350 db you did get a positive result, then there is most likely many more hidden values that we are not aware of to day, that also affect the SQ performance.)
Or else i blame the recordings are to poor for DAVE, so it just show us 70% of what it is capable of, that could also be a factor. It going to scale when the recordings gets even better.
(2)
One thing that i dont know how many have thaught of about the comparings with MQA On and Off switch on the Mytech or MSB Select II, it is the same file you are listening from.
The file is a sort of a dubble head MQA / Flac file with an average size of like 40-60 MB , so when you deactivate the MQA switch the DAC reads the file as an normal compressed 16/44.1 PCM file, so that is why you are sort of tricked by the demonstration. For a real to real comparision the comparision should be that you fist listen to the best possible uncompressed DXD / DSD / PCM version , and then listen to the MQA file, for a honest comparison, because when you pull the switch it reads the file as if it was a slight compressed copy of a read book.
))) great news for DAVEs owners.
