Audiowise Opto-USB, SRC-DX and early experiments with the Chord Qutest before moving to TT2

[Atriya]

2 hours ago, Gavin1977 said:

A laptop has lots of extra noise in it, e.g screen.  You can isolate noise, but you can’t recover from its original effect.  RaspberryPi is simpler, less to degrade the signal.

 

So, noise in a source (e.g. laptop, RPi4) can degrade the digital signal itself in some way other than jitter? (Given that jitter can't be the factor, since I'm using a Chord DAC.)

 

What kind of degradation might this be? I thought the signal from any source, including a laptop, is bit-perfect when using and source files and software (in my case lossless files & HQPlayer WASAPI exclusive mode for laptop and NAA for Rpi4)?

[dericchan1]

2 hours ago, Atriya said:

 

So, noise in a source (e.g. laptop, RPi4) can degrade the digital signal itself in some way other than jitter? (Given that jitter can't be the factor, since I'm using a Chord DAC.)

 

What kind of degradation might this be? I thought the signal from any source, including a laptop, is bit-perfect when using and source files and software (in my case lossless files & HQPlayer WASAPI exclusive mode for laptop and NAA for Rpi4)?

You are correct that electrical noise does not “alter” the bit perfect digital information being fed to the dac from the source device. The fact that you hear significant differences between rpi and your laptop can mean only one thing……

[Atriya]

20 minutes ago, dericchan1 said:

You are correct that electrical noise does not “alter” the bit perfect digital information being fed to the dac from the source device. The fact that you hear significant differences between rpi and your laptop can mean only one thing……

 

Which is? That I'm imagining the differences?

[davide256]

 

20 hours ago, Atriya said:

 

So, noise in a source (e.g. laptop, RPi4) can degrade the digital signal itself in some way other than jitter? (Given that jitter can't be the factor, since I'm using a Chord DAC.)

 

What kind of degradation might this be? I thought the signal from any source, including a laptop, is bit-perfect when using and source files and software (in my case lossless files & HQPlayer WASAPI exclusive mode for laptop and NAA for Rpi4)?

Devices with good hardware integration, low electrical noise do better for asynch USB out.

Using network file transmission {NAA or other protocol endpoint} allows you the benefits of a high powered CPU at HQP server without the challenges

of a complex system for the USB endpoint, file stays a file in traversing between server and endpoint, endpoint handles aysnch USB out.

 

Why is all speculation. Be nice if we had the logic probes and timing capabilities to see whats happening inside a PC

[Fourlegs]

On 10/2/2022 at 10:52 PM, Atriya said:

 

So, noise in a source (e.g. laptop, RPi4) can degrade the digital signal itself in some way other than jitter? (Given that jitter can't be the factor, since I'm using a Chord DAC.)

 

What kind of degradation might this be? I thought the signal from any source, including a laptop, is bit-perfect when using and source files and software (in my case lossless files & HQPlayer WASAPI exclusive mode for laptop and NAA for Rpi4)?

The usual mechanism for noise overlaid on a digital signal to affect sound quality is for it to cause issues in the analogue stage of the DAC. The resulting IMD is easily heard even at low levels. Note that this is not background noise..

The fact that the signal is bit perfect does not protect it from this noise causing issues. The noise is mostly out of audio band noise and sometimes can be as high as RF.

 

Whilst the Opto-USB provides an optical break, the receiving end to convert the signal back to USB requires a 5V power source and I wonder if it is possible that will allow noise to flow in the ground plane and bypass the optical break.

[Atriya]

1 hour ago, Fourlegs said:

The usual mechanism for noise overlaid on a digital signal to affect sound quality is for it to cause issues in the analogue stage of the DAC. The resulting IMD is easily heard even at low levels. Note that this is not background noise..

The fact that the signal is bit perfect does not protect it from this noise causing issues. The noise is mostly out of audio band noise and sometimes can be as high as RF.

 

Whilst the Opto-USB provides an optical break, the receiving end to convert the signal back to USB requires a 5V power source and I wonder if it is possible that will allow noise to flow in the ground plane and bypass the optical break.

@Fourlegs, I understand this already. My point is that even if the Opto-USB receiver does introduce some noise, this noise will be the same whether the upstream source is a laptop or an RPi4. So, it cannot help explain the differences heard between the two sources, even after both are isolated by Opto-USB. And neither can jitter, since I'm using a Chord DAC.

 

Additionally, that 5V to the Opto-USB receiver is coming from a linear power supply, in my case.

[Atriya]

On 10/2/2022 at 1:24 PM, Gavin1977 said:

You’ll still have noise from the optical transceiver at either end, but you have complete isolation of upstream noise.

 

Why noise from the optical transmitter though? Wouldn't I only have the noise from the optical receiver? It seems like the optical fiber should eliminate any noise from the transmitter.

[frieddr]

On 3/2/2022 at 5:20 PM, creativepart said:

I have a SRC-DX on the way and it is expected Friday or Monday. I used to run a Schiit Etr device to take USB from my SoTM SMS200 Ultra and run S/PDIF to my Chord Qutest. After about a year of use I removed it and went back to running USB between the network streamer and the DAC.

 

Now I run HQPlayer via Roon and thought I should give the approach a try one more time. I'll report back once I receive the device. I've been trying to decide if I want to get an M-Scaler or not so I'm testing HQP + SRC-DX + Dual BNC as a poor man's M-Scaler.

Hi did this work for you.  I have Qutest and sms200 ultra neo and I've read with arc-dx there's a volume setting glitch.  Did you solve that?  Was it difficult?

[creativepart]

On 11/9/2022 at 2:28 PM, frieddr said:

Hi did this work for you.  I have Qutest and sms200 ultra neo and I've read with arc-dx there's a volume setting glitch.  Did you solve that?  Was it difficult?

No. I ended up returning it on the day it arrived. The developer wanted me to contact SoTM to request Linux SSH Log in credentials and then install an audio volume module to make the output listenable. I declined taking that step. I was charged both return postage and a 10% restocking fee.

[frieddr]

On 11/11/2022 at 2:21 PM, creativepart said:

No. I ended up returning it on the day it arrived. The developer wanted me to contact SoTM to request Linux SSH Log in credentials and then install an audio volume module to make the output listenable. I declined taking that step. I was charged both return postage and a 10% restocking fee.

I wonder what version of the SoTM firmware you were running?  I just discovered today that the renderer plays much better with HQPlayer with the latest firmware update.  I hadn't updated the software for a couple years.

[creativepart]

I tend to keep upto date on firmware - I'm currently at Version Number : V0.5.2 which is latest. Not always, but it wasn't a SoTM firmware issue it was a Linux Kernel issue. The fix was to install a new audio mod in the Linux Kernel. But as I said, SoTM doesn't provide the SSH login except on special request and they are wary of installing new modules.

[frieddr]

I understand.  I am gobsmacked the impact upsampling that HQ Player has had on my sotm/qutest chain (holographic separation), but I can't exceed 192k without getting pops.  I tried a different USB cable, and I've tried various different filters, and I've fiddled with the settings in HD Player, but pops recur above 192 or 256k.  I'm operating from theory this has something to do with the USB implementation on the Qutest.  I've ordered the SRC-dx and I'm going to make a go of remoting in to change the SoTM volume settings.   I figure worst case scenario I can always go back to stock by flashing the micro SD card on the SoTM.  I'll let you know how it works.  Meanwhile if anyone has successfully used an src-dx with a chord dac please hmu.  

 

The other thing I'm noticing with HD Player is that it chokes on MQA files, at least with my settings.  I think that is solvable...

[blueninjasix]

1 hour ago, frieddr said:

I understand.  I am gobsmacked the impact upsampling that HQ Player has had on my sotm/qutest chain (holographic separation), but I can't exceed 192k without getting pops.  I tried a different USB cable, and I've tried various different filters, and I've fiddled with the settings in HD Player, but pops recur above 192 or 256k.  I'm operating from theory this has something to do with the USB implementation on the Qutest.  I've ordered the SRC-dx and I'm going to make a go of remoting in to change the SoTM volume settings.   I figure worst case scenario I can always go back to stock by flashing the micro SD card on the SoTM.  I'll let you know how it works.  Meanwhile if anyone has successfully used an src-dx with a chord dac please hmu.  

 

The other thing I'm noticing with HD Player is that it chokes on MQA files, at least with my settings.  I think that is solvable...

I also use qutest and have suffered with pops and crackles at times. These have always been solved by increasing the buffer setting beyond 100ms. I also get the white noise issue at 768kHz but never at 705.6kHz. This last problem made me think about src-dx but I've heard that even that does not solve that particular problem.

[frieddr]

So like 200 ms?  

[davide256]

Curious, never had a problem using PGGB pre up-sampled files at 705/768 with a Chord Mojo and SRC-DX. HQPlayer under Euphony also behaved well other than start of play delay. This was with an 11i3 NUC which only starts to falter with DSD upsampling faster than 64

[deathdisco]

@frieddr — did you end up successfully using the src-dx with your SoTM/Qutest set up? I have a SoTM 200ultra w HQ player NAA setup into a Chord Qutest also... and am extremely curious to hear about your experience if you're willing to share!

I'm on the fence about purchasing one to try it out.

With thanks,
Andrew

[creativepart]

On 3/25/2023 at 10:23 PM, deathdisco said:

@frieddr —

I deleted my response. Sorry, I didn't notice that this was addressed to another user.

[davide256]

On 3/25/2023 at 11:23 PM, deathdisco said:

@frieddr — did you end up successfully using the src-dx with your SoTM/Qutest set up? I have a SoTM 200ultra w HQ player NAA setup into a Chord Qutest also... and am extremely curious to hear about your experience if you're willing to share!

I'm on the fence about purchasing one to try it out.

With thanks,
Andrew

This is an issue with linux updates defaulting devices at startup with "adjustable" volume to -20db for user protection. I believe it affects any version of HQPe after 4.28.

For whatever reason linux thinks SRC-DX has a volume control. I would not suggest using an SRC-DX with SoTM since ir appears you don't have root access  to install

ALSA mixer and reset the default to 0 db (100% volume.

[deathdisco]

On 3/30/2023 at 5:26 AM, davide256 said:

This is an issue with linux updates defaulting devices at startup with "adjustable" volume to -20db for user protection. I believe it affects any version of HQPe after 4.28.

For whatever reason linux thinks SRC-DX has a volume control. I would not suggest using an SRC-DX with SoTM since ir appears you don't have root access  to install

ALSA mixer and reset the default to 0 db (100% volume.

Thanks for the context Davide. For the record (for anyone else following along) I've spoken with May at SoTM and they will provide root access to customers, though they don't publish this information publicly.

Might go down this path when funds allow... will post my findings in this thread if I do.

Cheers!

[davide256]

Have learned that the Mojo2 supports 768 via the SRC-DX whereas the original Mojo does not, so have placed an order, already have the appropriate dual BNC

to dual mini RCA cable from Headgear Audio. Asked Audiowise  if any other DAC 's could do 768 PCM with the SRC-DX, looks like its only Chord in the consumer space,

 

[creativepart]

SOtM has just posted in their sponsored forum about beta testing a new firmware version that looks promising for using the SRC-DX with SOtM devices. Here is a list of updates in the new firmware:

Quote

• •  

Version: V0.5.5
- Date: 11-24-2023
- Added 'CPU frequency' selector.
- Added 'Set Max volume' button.
- Updated eunhasu, smb avahi services.
-Disabled unused avahi service(nfs).
- Changed a few GUI texts.
- Fixed up 'Start with Logitech Media Server' bug.

I'm checking for full details, but that "- Added 'Set Max volume' button." looks like exactly what is needed to use the SRC-DX.

[creativepart]

I have confirmed with SOtM that the new 0.5.5 firmware in Alpha testing right now does indeed have a Set Max Volume button that sets ALSAmixer to 100% volume and "Should" solve the volume issue with SRC-DX working properly with SOtM gear.

 

The following quote is from May at SOtM:

 

Quote

you are right, the set max volume button makes the volume to be max. you don't need to go to ALSAmixer settings. Just try to use this button.

 

[creativepart]

I've purchased a new SRC-DX to test out this new Set Max Volume feature in SOtM's new 0.5.5 firmware. But just to be safe, I've also tested loading and running ASLAmixer in Linux on the SMS200 Ultra Neo just in case the new feature doesn't work as expected.

 

Maybe two years ago or more now, but I bought the SRC-DX to use with my then DAC (Qutest) and was very disappointed with the volume issue. I didn't want to mess with Linux command line changes back then, and returned the SRC-DX because of that.

 

I used to know Linux CL years ago, and didn't want to get back into it. But I figured before buying the SRC-DX again to test the new Set Max Volume fix from SOtM  I'd better be prepared to intervene manually if necessary so I dusted off my CL chops before ordering.

[creativepart]

I'm back with real world testing of the SRC-DX with a SOtM SMS-200 Ultra Neo.

 

As expected, when I first installed the SRC-DX the volume was reduced almost but not quite to silence.

 

I installed the new ALPHA version of the v0.5.5 firmware for the SOtM devices and sure enough there is now a "Set Max Volume" button located far bottom left of the Configuration menu. Clicking this button immediately sets the SRC-DX to 100% volume in AlsaMixer in the Linux firmware.

 

Once set the SRC-DX works normally with the SOtM network streamer devices.

 

Here are some screen grabs:

 

[creativepart]

As I recently posted, I purchased a new SRC-DX for my streaming setup. I'm running a SOtM SMS-200 Ultra Neo streamer with a Chord Hugo TT2. Until now, I've always used USB output from the SOtM into the Chord DAC.

 

Now with the SRC-DX acting as a DDC to convert that USB output from the streamer into BNC S/PDIF outputs to feed the Hugo TT2. I've had a week or so of back and forth listening to compare the two configurations.

 

My initial impressions were that not much changed... but with more careful listening and comparisons I do find a fairly noticeable difference between the Hugo TT2 on USB and the Hugo TT2 on BNC S/PDIF. The thing I haven't been totally successful with is deciding if these differences matter to me or not.

 

Fed USB out of the streamer the TT2 sounds great. Very smooth, lots of separation between instruments and all the typical descriptions you've heard about the Chord sound. But when fed S/PDIF out of the SRC-DX that doesn't really change but there is an increase in high frequencies, most noticeable in cymbals - high hat especially. There may be a slight change in bass as well. I can't tell if its as clean as the USB fed sound but it is a bit more prominent.

 

Mind you, I'm not talking day or night differences. Very slight changes - except with High Hat cymbals - you really notice these from the TT2 with the SRC-DX.

 

Every day I switch back and forth and each day I think "OK, this is the better of the two" but that's a problem because one day that's feeding the DAC with S/PDIF and the next day with USB.

 

I need more time that this week to decide which is better or perhaps preferred is the better terminology.

 

If you're thinking the dual BNC output from the SRC-DX is going to be the best part of adding this device, think again. I'm not noticing any meaningful change between 768hz, and two BNC output and 384hz and one BNC connection. The same differences I pointed out above exist regardless of using one BNC connection or two BNC Connections. Evidently, there is a basic difference between the Chord's USB input and the DAC's S/PDIF input.