Hugo M Scaler by Chord Electronics - The Official Thread

[Rob Watts]

Hello Rob,

Some background:

I use Roon's DSP Convolution engine for 131k tap FIR room correction filters along with the volume leveling feature for volume adjustment free listening. Before any DSP is applied to the signal, Roon expands the audio to 64bit floating point and does all DSP with 64bits of precision. Once the desired DSP has been applied, Roon will dither (TPDF) back down to the maximum word length accepted by the audio device. So in the case of Chord DACs, this would be 32bits for the USB input and 24bits for the Toslink input.

My questions:

A. Coming from the 64bit floating point audio stream, is it best to dither to 32bit and then send to the M-Scaler via USB, or should I dither to 24bit and use the optical input for full electrical isolation?

B. Does the M-Scaler output 32bit audio on dual BNC or is it internally dithered to 24bits before leaving the device? If it internally dithers to 24bit, are the advantages of having the 32bit input greater than the cost of dithering twice (64 -> 32 -> 24)?

C. You've stated numerous times that your devices perform their best when given a bit-perfect audio stream, for obvious reasons. In your best guess/subjective opinion, at what level of DSP volume attenuation would one be causing serious auditory consequences? Assuming that this attenuation was being done with 64bit precision and dithered back to 32/24bit using TPDF.


Thanks!

A. My guess is that optical at 24b would be best sounding. Try both, and see for yourself; use a test track that has good depth (I use an organ track), and the one that has the best perception of depth plus the warmest tone, is the more accurate/transparent one.

B. No - it's not dithered down to 24 bits but very aggressively 11th order noise shaped. Now noise shaping at 768k with my truncation noise shaper (the truncation noise shaper actually has a different architecture to pulse array noise shapers), allows for -301 dB accuracy (so a -301 dB signal is accurate to +/-0.001dB in amplitude and +/- 0.001 deg in phase) and better than 350dB THD and noise within the audio bandwidth. This is measurably perfect small signal performance, and it's something that, based on my listening tests, is essential for transparent sound quality (that is maintaining depth perception).

When the M scaler was first put together with the Blu 2, I spent a lot of time in listening tests to get the maximum SQ performance (and here the problem is trying to add a digital module without it reducing my perception of sound stage depth), and this is how I came up with the 11th order truncation noise shaping. Now this is something one can do at 768k; but at 44.1 or 48k you cannot do this, so when converting to n bits down to 24 bits than you have to use dither. As part of my ADC project, I wanted dither to sound as good as the noise shaping - so started with rectangular and triangular dither (TPDF). Rectangular sounded the worst, but TPDF was still much inferior to the 768k noise shaping; I then used pseudo Gaussian dither, and this got me much closer to the noise shaper performance compared to TPDF - but it was still no where near as good as the noise shaped performance. Incidentally, if you have a truncation setting for a filter, never switch dither to off, the small signal distortion then becomes huge. If you have the option use Gaussian dither - if this isn't available (it's very rare) then triangular or TPDF.

As far as the Davina project goes, I still have more work to do, in order to improve 44.1k truncation to 24 bits. Perhaps making the pseudo Gaussian more Gaussian may help close the SQ gap.

C. Any change in the digital data requires re-quantization if there is a truncated data residual (the bits that can't be transmitted). And my answer B shows this can only be done at 705/768 when using advanced noise shaping. So even using a -0.001dB change would create problems in depth perception; this is why my standard advice is to keep it source bit perfect - and that ignores the fact that any up-sampling filter is not the same as a WTA filter.

 

[Sound Eq]

well when I use hqplayer and poly sinc long lp and 756 sample rate I getting clicking noise with my hugo tt2 via usb , maybe my settings on hq player are totally messed up so can anyone please tell me what setting to dial in hq player so i can compare with mscaler which I will have access to in few days

I have an asus Zephyrus 501 laptop -- (Intel 6-Core i7-8750H, 24GB RAM

 

[Ards]

A. My guess is that optical at 24b would be best sounding. Try both, and see for yourself; use a test track that has good depth (I use an organ track), and the one that has the best perception of depth plus the warmest tone, is the more accurate/transparent one.

Rob, would you care to share what test tracks that you find particularly useful? Always on the lookout for good test material...

 

[Rob Watts]

You can find the depth test on Youtube here:

 

[Triode User]

and it is available on Qobuz here . . . . (I have been listening to this ever since RW suggested it some time ago).

https://www.qobuz.com/gb-en/album/m...nts-from-petrushka-jean-guillou/0053479011725

 

[ZappaMan]

http://www.6moons.com/audioreviews/vibrapod/cones.html

An old but good review, I’ve just ordered the pods,

https://www.analogueseduction.net/isolation/vibrapod-isolators-pack-of-4.html

Its my Christmas present

 

[dmance]

@ZappaMan
Yes, footers on my TT2 work to extract 'more' from the presentation - subtle but its clearly audible with direct-to-headphones or speakers. Courtesy of @rAd-dude 's suggestion, i tried the Level3 footers from Ingress Engineering (here) - and i'm sure any of a number of other products work as well. Any discussion with my engineering peers or friends in digital electronics R&D cannot delve too far into the 'why' these footers work because I look the fool ...as low level vibrations should have no effect. Microscopic movements of a digital PCB must exclusively be detrimental to the clock crystal stability or induce microphonics in components. (somewhere i read that even rob watts says Dave is microphonic). However, unlike the RFI isolation of DACs (of which I am developing an excellent understanding of and can correlate remedies to SQ improvements), the anti-vibration tweaks on digital components are most puzzling. We are not talking about shakey-shakey movements - but micro/nanometer scale movements: the vibrameter app on my phone shows no measurable difference with or without the footers - and this is with walking around the floor or playing music (via loudspeakers) soft or loud. So even though i can measure the extremely low levels of jitter induced in the DAC's output due to RF noise, I cannot measure any markers in a DAC's output from vibration isolation at all. Solving this mystery is top of my list for 2020...

 

[llamaluv]

On the topic of PC-based upsampling, I gotta say, I'm very happy with where I'm at with my M-Scaler setup, especially after having added the Opto DX.

But! The recent discussion makes me want to revisit using HQ Player again for upsampling (including putting in the necessary amount of work to feel confident in having optimized such a setup sufficiently).

Because I don't know about anyone else, but if I can get as-good results from a PC-based upsampling setup, I'd drop the M-Scaler in a heartbeat (#cantlie). If not, then the M-Scaler can stay put, with extra appreciation points added.

 

[audio_1]

and it is available on Qobuz here . . . . (I have been listening to this ever since RW suggested it some time ago).

https://www.qobuz.com/gb-en/album/m...nts-from-petrushka-jean-guillou/0053479011725

Thanks for the link. Do you use the Qobuz PC app or listen to Qobuz via Roon. I can't find a way to defeat the volume control on the app. Max volume appears to be 2 or 3 db higher compared to a CD played on the Blu 2. Have you noticed this? It is hard to know if the CD and Qobuz streams are the same though. I just go by the release date. I want to get a bit perfect output from Qobuz. Thanks Neil

 

[Progisus]

Both achieve the same result as I perceive it BUT the mscaler is one beautiful piece of equipment and much more stable.

 

[Sound Eq]

Both achieve the same result as I perceive it BUT the mscaler is one beautiful piece of equipment and much more stable.

that is very interesting to read, then can someone please share their settings so I can get the best settings for HQ player software and compare later on to mscaler as the setting in HQP seem to be not giving me the best result without a drawback. I used poly sinc long lp. dither non, sample rate 756. I perceived a bit metallic or harshness in the top end, but yes the sound stage and sound in general were improved quite noticeably using HQP with hugo tt2

so are there any other settings maybe to perfect the HQP

 

[llamaluv]

Both achieve the same result as I perceive it BUT the mscaler is one beautiful piece of equipment and much more stable.

Hi Progisus, you've mentioned a few times that you find HQ Player to be less stable. Can you elaborate?

 

[Progisus]

Hi Progisus, you've mentioned a few times that you find HQ Player to be less stable. Can you elaborate?

Sure...
Firstly, I use hqp on a win 10 nuc7i3 and that is all that is running. It gets it signal from Roon running on another nuc8i7 elsewhere in the house over copper. It then does it’s magic and sends over copper to a signalyst naa image running on a pi 4. That signal is then fed through a jitter bug on usb 3 port to an audiquest cinnamon cable to my hugo tt.
I use pcm poly-sinc-long lp 384k no dither as the filter.
Once or twice a week while manually changing songs it will glitch causing a high pitched white noise. Sometimes stopping and starting the song may fix it but a reboot may be required.
Turning off the hugo tt may also cause hqp to loose the naa connection. This will require rebooting the pi and restarting hqp. This can be avoided by leaving the tt and pi running.
These are the main glitches that occur. The mscaler does not have these issues but I have rebooted it a couple times when the music seemed lifeless.
Hope this helps. A first world problem to be sure.

 

[Sound Eq]

Sure...
Firstly, I use hqp on a win 10 nuc7i3 and that is all that is running. It gets it signal from Roon running on another nuc8i7 elsewhere in the house over copper. It then does it’s magic and sends over copper to a signalyst naa image running on a pi 4. That signal is then fed through a jitter bug on usb 3 port to an audiquest cinnamon cable to my hugo tt.
I use pcm poly-sinc-long lp 384k no dither as the filter.
Once or twice a week while manually changing songs it will glitch causing a high pitched white noise. Sometimes stopping and starting the song may fix it but a reboot may be required.
Turning off the hugo tt may also cause hqp to loose the naa connection. This will require rebooting the pi and restarting hqp. This can be avoided by leaving the tt and pi running.
These are the main glitches that occur. The mscaler does not have these issues but I have rebooted it a couple times when the music seemed lifeless.
Hope this helps. A first world problem to be sure.

i miss simplicity in set ups. I miss just pressing play on my cassette player in 80s, and I swear that during those days listening to bands that I still listen today to, I used to adore the sound back then ten times more than now with all this digital tech. It was all bout music in the 80s now its all bout critical listening and analyzing and music takes a back seat

 

[llamaluv]

Sure...
Firstly, I use hqp on a win 10 nuc7i3 and that is all that is running. It gets it signal from Roon running on another nuc8i7 elsewhere in the house over copper. It then does it’s magic and sends over copper to a signalyst naa image running on a pi 4. That signal is then fed through a jitter bug on usb 3 port to an audiquest cinnamon cable to my hugo tt.
I use pcm poly-sinc-long lp 384k no dither as the filter.
Once or twice a week while manually changing songs it will glitch causing a high pitched white noise. Sometimes stopping and starting the song may fix it but a reboot may be required.
Turning off the hugo tt may also cause hqp to loose the naa connection. This will require rebooting the pi and restarting hqp. This can be avoided by leaving the tt and pi running.
These are the main glitches that occur. The mscaler does not have these issues but I have rebooted it a couple times when the music seemed lifeless.
Hope this helps. A first world problem to be sure.

Thanks for the description.

I'll post an update as I try stuff out myself.