Chord Electronics Qutest DAC - Official Thread

[Music Alchemist]

I might have mis-read, but the impression I got was that the DAVE can output the digital signal it creates before D/A conversion, or at least some kind of post-processed (post WTA filter and volume control) to the digital amp. What is being bypassed is the output stage of the DAVE, instead using the output stage of the digital amp which will convert the output to a level more suitable for driving speakers.

Yes, the confusion was cleared up in subsequent posts in this thread. Basically, if I'm reading some of the things Rob said later correctly, the digital power amps are all-in-one DAC/power amp/preamp units that can also accept a digital signal from the DAVE.

 

[Jimi Zine]

Currawong


My mac does use USB3.0, I'm not sure of the cause but I know it was unacceptable playback using DSD128 & 256 using Mojo on a Mac OS X High Sierra.

With regard to Native and DoP transmission on a mac, here's what Exasound said four years ago (2014) (the only DAC company I know who integrate an ASIO driver to transmit native DSD on a mac)

"The Mac sound system doesn't have a built-in support for DSD-encoded audio. The audio industry has adopted an open-source standard called DSD over PCM (DoP). DoP relies on a workaround to disguise DSD as PCM data stream. Unfortunately this workaround causes 30 to 50 percent overhead. The DoP implementation of DSD 256 requires support for PCM at 705.6kHz and 768kHz. Such sampling rates are a real challenge for both computer CPU and USB audio interface.

The new completely redesigned exaSound Core Audio DoP256 driver overcomes these limitations...."


Rob Watts has said this,

to which I replied,

I received no further reply or clarification since this...

 

[Currawong]

DSD256 has been working for me, on my Macs, with the Hugo 2 and Mojo. That is why I think it might be a problem with individual models, going by past experience with audio issues.

It could also indeed be system load. I had to double the memory on my Mac Mini (Late 2012, Core i7 2.3Ghz) to 16 GB before I could load a full 9-minute DSD256 track into memory and have it play all the way through. See the memory usage for Audirvana Plus just now during playback. Output is DoP 1.1 in settings.

 

[Jimi Zine]

DSD256 has been working for me, on my Macs, with the Hugo 2 and Mojo. That is why I think it might be a problem with individual models, going by past experience with audio issues.

It could also indeed be system load. I had to double the memory on my Mac Mini (Late 2012, Core i7 2.3Ghz) to 16 GB before I could load a full 9-minute DSD256 track into memory and have it play all the way through. See the memory usage for Audirvana Plus just now during playback. Output is DoP 1.1 in settings.

Maybe, I know it's not system load, as I have a powerful machine with 16gb memory, what are the memory/CPU app(s) you are using there?

 

[Currawong]

iStat Menus. It might help that I'm using an iUSB 3.0. If all else is good (ie: Your machine is dedicated to music playback as mine is and nothing else, software or hardware could interfere with transmission) then I'd strongly bet on a USB port issue. What exact machine do you have with what exact specs?

 

[Jimi Zine]

iStat Menus. It might help that I'm using an iUSB 3.0. If all else is good (ie: Your machine is dedicated to music playback as mine is and nothing else, software or hardware could interfere with transmission) then I'd strongly bet on a USB port issue. What exact machine do you have with what exact specs?

Macbook Pro 2015, 3.1Ghz intel core i7, 16gb memory,

 

[paul2qute]

I love watching films as well as music,I've got a naim nd5 xs Chord 2 qute Arcam 850 going through Monitor apex 40 speakers and a rel sub,I should still hear the difference changing to the Chord qutest even though I don't have state of the art speakers shouldn't I???

 

[jayz]

Currawong


My mac does use USB3.0, I'm not sure of the cause but I know it was unacceptable playback using DSD128 & 256 using Mojo on a Mac OS X High Sierra.

With regard to Native and DoP transmission on a mac, here's what Exasound said four years ago (2014) (the only DAC company I know who integrate an ASIO driver to transmit native DSD on a mac)

"The Mac sound system doesn't have a built-in support for DSD-encoded audio. The audio industry has adopted an open-source standard called DSD over PCM (DoP). DoP relies on a workaround to disguise DSD as PCM data stream. Unfortunately this workaround causes 30 to 50 percent overhead. The DoP implementation of DSD 256 requires support for PCM at 705.6kHz and 768kHz. Such sampling rates are a real challenge for both computer CPU and USB audio interface.

The new completely redesigned exaSound Core Audio DoP256 driver overcomes these limitations...."


Rob Watts has said this,


to which I replied,



I received no further reply or clarification since this...

My vote is we keep this thread limited to discussions on Qutest specific items - it is the official thread for Qutest after all. Looking forward to initial impressions, etc. Bristol Audio Show is coming up and hopefully, we will see increased activity of relevance to Qutest here soon.

 

[jayz]

But with USB it is easy to connect a portable battery powered USB charger to it - and that for sure would represent the ultimate PSU.

Does that mean we can power it through a high capacity mobile power bank? I guess one of the fast-charge capable ones will be better. Not sure of how good such a power supply might be in terms of noise. As long as Qutest powers down automatically after a period of inactivity i.e. no signal then that use-case should work.

 

[PANURUS]

Can the Qutest usable with 2go one day? With an interface module ?
Or is a streamer, with the same design, in the pipe?

 

[dmance]

I have a Qutest on pre-order ...which basically means i've told my dealer to call me when they come in.
Having auditioned the Hugo2 for several weeks, I know the sonic qualities of the technology. I am on loudspeakers so the Qutest is the more appropriate desktop device for me.
My reasons for buying it are:

• it can run on batteries (using commonplace and inexpensive +5V USB power pack). This allows me to float my entire source chain off AC. Laptop->USB conditioning->DAC without any AC mains noise or ground loops. And I expect lots of forum input into what power pack sounds best.
• USB Galvanic Isolation. So, yes, thanks for this ...but I will also continue to use my Intona+Battery+Ferrites because my ears have heard how important it is to get rid of all RF/power/ground noise from the source.
• 32-bit samples from source is as good as a volume control. I am 100% playing JRiver or Roon - which internally processes audio at 64-bits (volume, DSP, etc) and will downsample to 32-bits for Qutest. No problem here ...and the ability to change output line levels lets me select the best sweet spot for my downstream electronics.
• High resolution PCM768+DSD512. I am intrigued by how close software up-sampling can be to M-Scaler. My experiments with HQP+Hugo2 demonstrate that it's pretty darn good. Yes, I love just listening to WTA filtered 44.1k music but there is so much forum dialog on upsampling and this lets me have some fun until Chord gets its M-Scaler product and pricing strategy in order.
• Its just a DAC ...and just enough DAC for my needs. I run single-ended and USB input. More inputs or outputs are superfluous and money not well spent. Also, given what I have learned about removing sources of RF/power/ground noise from the D/A portion of the DAC, i dont want any integrated streaming or bluetooth or WiFi in the same chassis. I don't need convenience ...i need the best sound quality.
• Its reasonably priced.

Dan

 

[Rob Watts]

Does that mean we can power it through a high capacity mobile power bank? I guess one of the fast-charge capable ones will be better. Not sure of how good such a power supply might be in terms of noise. As long as Qutest powers down automatically after a period of inactivity i.e. no signal then that use-case should work.

Sure - that was one of the motivations for USB power, as battery supplies are inexpensive. Qutest consumes only 2.5W, so a 100W/H would last for 40 hours. If you can hear no change in connecting a battery, then that would prove that a "better" PSU to the one shipped is not possible. Battery operation provides absolutely the lowest possible RF and LF noise - they are ideal.

 

[Music Alchemist]

As long as Qutest powers down automatically after a period of inactivity i.e. no signal then that use-case should work.

Sure - that was one of the motivations for USB power, as battery supplies are inexpensive.

The 2Qute, powered by a wall wart, only turned on/off when you connected/disconnected the power supply.

With the included power supply, does the Qutest shut down automatically? If so, when?

32-bit samples from source is as good as a volume control. I am 100% playing JRiver or Roon - which internally processes audio at 64-bits (volume, DSP, etc) and will downsample to 32-bits for Qutest. No problem here ...and the ability to change output line levels lets me select the best sweet spot for my downstream electronics.

I have many of the audiophile players out there (Bug Head Emperor / Infinity Blade, HQPlayer, JPLAY, etc.) but usually just use foobar2000.

I heard that foobar2000 has 32-bit internal processing that also applies to the volume control. Does that mean it could do bit-perfect volume control with the Qutest and without a preamp?

I still prefer having a preamp because there are too many variables otherwise...but I might want to try other solutions.

I am intrigued by how close software up-sampling can be to M-Scaler. My experiments with HQP+Hugo2 demonstrate that it's pretty darn good.

Rob's position is that software upsampling (including HQPlayer) should not be used with Chord DACs because they already do much more advanced digital processing, and messing with things on the software end will just result in lower fidelity.

Now, if you like the sound, that's fine. But if your goal is superior technical performance, I'd ask Rob about the specifics on why it's not going to work that way.

 

[dmance]

@Music Alchemist

Nobody but Mr. Watts knows what his secret sauce is. The rumor is that M-Scaler WTA is a million-tap linear phase filter with carefully calculated coefficients developed over Rob's long tenure in digital audio. its just math + time, right? Well, as they say, the devil is in the details.
My limited experience in software up-sampling is running HQPlayer's XTR filters from 44.1 to 705.6k PCM and feeding them to Hugo2 - no problem on my lowly Yoga laptop. Jussi Laako, the man behind HQPlayer, also has a secret sauce of hand-crafted filter optimization and he claims that XTR filters hundreds of thousands taps. However, when I tried it, the Hugo2 still sounded better using WTA rather than the XTR filters. And this is just with Hugo2's 49k taps let alone M-Scaler's million taps. I have not tried filtering to DSD512 or using closed-form and this is apparently much better sounding ..but I then I'll be investing in an expensive CPU+GPU system just to play music.

So, long term I think that if all we want to do is just listen to our music in its best form we're better off just buying the damn M-Scaler - whatever the cost. However, between now and then I can explore how close we can get - and maybe reach most of the way for far less. Or not.
Dan

 

[Music Alchemist]

Nobody but Mr. Watts knows what his secret sauce is. The rumor is that M-Scaler WTA is a million-tap linear phase filter with carefully calculated coefficients developed over Rob's long tenure in digital audio. its just math + time, right? Well, as they say, the devil is in the details.
My limited experience in software up-sampling is running HQPlayer's XTR filters from 44.1 to 705.6k PCM and feeding them to Hugo2 - no problem on my lowly Yoga laptop. Jussi Laako, the man behind HQPlayer, also has a secret sauce of hand-crafted filter optimization and he claims that XTR filters hundreds of thousands taps. However, when I tried it, the Hugo2 still sounded better using WTA rather than the XTR filters. And this is just with Hugo2's 49k taps let alone M-Scaler's million taps. I have not tried filtering to DSD512 or using closed-form and this is apparently much better sounding ..but I then I'll be investing in an expensive CPU+GPU system just to play music.

So, long term I think that if all we want to do is just listen to our music in its best form we're better off just buying the damn M-Scaler - whatever the cost. However, between now and then I can explore how close we can get - and maybe reach most of the way for far less. Or not.

An important thing to note is that the taps are calculated differently. Basically, even the highest number of "HQPlayer taps" is still lower than the lowest number of "Chord taps" as far as their production DACs go, if I recall the explanations I read. And there's a lot more to it than just that. Again, you'd have to ask Rob for specifics or track down his posts about it in the past. But he was very clear that no matter how good the digital processing is on the software end, it would interfere with the digital processing of his DACs and result in lower fidelity. Simply thinking about it mathematically as adding two things to get a higher number isn't how it works at all.

I'm sure HQPlayer is technically fine for most other DACs, and I do use it from time to time. (I really hate the interface, though.) I generally use the most extreme settings, but it makes my laptop (which does have good specs: quad core i7 processor, 8 GB RAM, etc.) heat up and slow down to the point that I can't do much else on the computer at the same time. If I don't load a track into memory beforehand, it stutters like crazy. If I load web pages while music is playing, it can sometimes make the program crash and make alien noise until I restart it. Storing everything in virtual RAM drives helps dramatically, and I'd recommend that to make things easier.