[SodaBoy]

Can anybody comment on the quality difference between using the digital volume control of the Gustard directly into an amp vs fixed output through a preamp? Possibly even a passive.

Basically I'm newish to the scene and am moving from integrated to separates. I'm trying to decide if I should go for the Gustard as a dac/preamp or get a separate preamp. I basically only use my desktop USB (mostly 24b & DSD, some rebook) as a source so the only function I really need is volume. This is also why something simple, such as a passive is appealing. I'm on a budget so "trial & error" can get expensive!

Thanks!

Dan


Sent from my iPhone using Tapatalk

The digital volume control has nearly no channel imbalance. You would have to go very high end in the analog realm to match that. The downside is that even with internal 32-bit internal processing, there is a limit to how much attenuation you can get without reducing the audible dynamic range of the recording. In analog volume control, when you turn down the volume, both signal and noise are attenuated. With digital volume, when the volume is turned down, the signal is attenuated, but the noise floor remains at the same level, so dynamic range is reduced. The ES9018, if implemented well, has more than enough headroom in dynamic range to have useful attenuation without affecting audible dynamic range. I don't own the X20, only the X12, but I'd trust Gustard to do a good job on the implementation of the ES9018 and keep the noise floor low.

 

[Quadman]

@dcepela I use my X20 with a home made buffered preamp modeled after the  Pass B1 schematic.  Since my preamp is 15 feet away from my listening position and it does not have a motorized volume control I use the gustards digital volume control to control the loudness and is works really well and as far as I can tell it has no negative effect on the SQ (and I upsample all music to DSD512).  The Gustard has a very high output, SE it is either 2.8V or 3.3V and balanced double that.  Most amps have a much lower input sensitivity and you would easily over drive most amps with that level of input unless you set the volume on the gustard to -12 or even -18dB.  My advice is to use a preamp (active of passive) and you are golden.  BTW I usually play the gustard with volume ranges of -10dB to 0 depending on the track, most music being -6 to -8 dB.  

 

[mocenigo]

  @dcepela I use my X20 with a home made buffered preamp modeled after the  Pass B1 schematic.  Since my preamp is 15 feet away from my listening position and it does not have a motorized volume control I use the gustards digital volume control to control the loudness and is works really well and as far as I can tell it has no negative effect on the SQ (and I upsample all music to DSD512).  The Gustard has a very high output, SE it is either 2.8V or 3.3V and balanced double that.  Most amps have a much lower input sensitivity and you would easily over drive most amps with that level of input unless you set the volume on the gustard to -12 or even -18dB.  My advice is to use a preamp (active of passive) and you are golden.  BTW I usually play the gustard with volume ranges of -10dB to 0 depending on the track, most music being -6 to -8 dB.  

Not only. With the Gustard you can also add 6 or 12 DB gain, bringing the mac output to something like 22.4Vrms or 26Vrms, which can probably be used to drive anything!
 
 Roberto

 

[mocenigo]

Can anybody comment on the quality difference between using the digital volume control of the Gustard directly into an amp vs fixed output through a preamp? Possibly even a passive.

Basically I'm newish to the scene and am moving from integrated to separates. I'm trying to decide if I should go for the Gustard as a dac/preamp or get a separate preamp. I basically only use my desktop USB (mostly 24b & DSD, some rebook) as a source so the only function I really need is volume. This is also why something simple, such as a passive is appealing. I'm on a budget so "trial & error" can get expensive!

 
if we trust what the manufacturer of the ESS9018 chips states, up to about -40Db there should be no SQ degradation. This may be exaggerated, but I would have no problems going to -24Db and even a bit more.  I know that if i connect the Gustard directly to my power amp, -22Db is a normal listening volume, and the sound is very similar to the one I get through the preamp at -9Db (no preamp implies: scene marginally wider, highs a bit more fatiguing, so I prefer to have the preamp)
 
 Roberto

 

[mocenigo]

 
The digital volume control has nearly no channel imbalance. You would have to go very high end in the analog realm to match that. The downside is that even with internal 32-bit internal processing, there is a limit to how much attenuation you can get without reducing the audible dynamic range of the recording. In analog volume control, when you turn down the volume, both signal and noise are attenuated. With digital volume, when the volume is turned down, the signal is attenuated, but the noise floor remains at the same level, so dynamic range is reduced. The ES9018, if implemented well, has more than enough headroom in dynamic range to have useful attenuation without affecting audible dynamic range. I don't own the X20, only the X12, but I'd trust Gustard to do a good job on the implementation of the ES9018 and keep the noise floor low.

Since the ear can perceive at most about 20bits of resolution and only at some middle frequencies, one has to lose 12bits from those 32bits before having a reduction of the perceived dynamics.

Which means a theoretical attenuation of 72Db before the perceivable dynamic range is actually reduced. This seems too high, so the argument probably has some fault. Let us tackle in a different way. The ES9018 in mono mode (as used in the Gustard) has a dynamic range of 135Db. Orchestral music at "FFF" levels reaches 115Db close to the orchestra and inside can get painful.  So even for a recording simulating a position in the first row (which is nor realistic, as the microphones are usually a bit further, esp because they are a bit elevated) even reducing by 20Db there would be no perceivable audible difference. And I doubt you could actually be able to reproduce that DNR at home with your speakers even in a large and well treated speakers.

So, even attenuations significantly larges than 20Db, say up to -30Db, should have no effect on the SQ.
 
Regarding noise. There is noise that is intrinsic in any recording and this is often much higher than the level of noise introduced by the electronics both in analog and digitai domains. Attenuation in the digital domain reduced also that noise. And then there is the intrinsic noise of the analog parts of the DAC, both in the chip and downstream from that. So the function determining the overall noise at the end is quite complex 
 
 Roberto

 

[SodaBoy]

 
Since the ear can perceive at most about 20bits of resolution and only at some middle frequencies, one has to lose 12bits from those 32bits before having a reduction of the perceived dynamics.

Which means a theoretical attenuation of 72Db before the perceivable dynamic range is actually reduced. This seems too high, so the argument probably has some fault. Let us tackle in a different way. The ES9018 in mono mode (as used in the Gustard) has a dynamic range of 135Db. Orchestral music at "FFF" levels reaches 115Db close to the orchestra and inside can get painful.  So even for a recording simulating a position in the first row (which is nor realistic, as the microphones are usually a bit further, esp because they are a bit elevated) even reducing by 20Db there would be no perceivable audible difference. And I doubt you could actually be able to reproduce that DNR at home with your speakers even in a large and well treated speakers.

So, even attenuations significantly larges than 20Db, say up to -30Db, should have no effect on the SQ.
 
Regarding noise. There is noise that is intrinsic in any recording and this is often much higher than the level of noise introduced by the electronics both in analog and digitai domains. Attenuation in the digital domain reduced also that noise. And then there is the intrinsic noise of the analog parts of the DAC, both in the chip and downstream from that. So the function determining the overall noise at the end is quite complex 
 
 Roberto

I don't see why my post needed further elaborating, if you want to lecture, direct this to the original post. I made the distinction between audible dynamic range and actual dynamic range very clear.

 

[mocenigo]

 
I don't see why my post needed further elaborating, if you want to lecture, direct this to the original post. I made the distinction between audible dynamic range and actual dynamic range very clear.

I did not want to lecture you, from the free flow of my style it would have been clear I was essentially brainstorming openly - mostly to justify your conclusion. So, I apologise if it seemed to you I was trying to lecture you, and there is no need to be hypersensitive, it was not my intention to argue in any way against you.
 
 Roberto

 

[SodaBoy]

 
I did not want to lecture you, from the free flow of my style it would have been clear I was essentially brainstorming openly - mostly to justify your conclusion. So, I apologise if it seemed to you I was trying to lecture you, and there is no need to be hypersensitive, it was not my intention to argue in any way against you.
 
 Roberto

Well my apologies then. When I see someone quoting me, and restating variations on the points I have made without making it clear that these points were made in support, that goes down very bad with me.

 

[mocenigo]

  Well my apologies then. When I see someone quoting me, and restating variations on the points I have made without making it clear that these points were made in support, that goes down very bad with me.

Do not worry. It was my fault after all to reply in the wrong place. Let us now do not flood this thread with this and return to listen to the music we love with the equipment we love (sent also a DM)
 
 Roberto

 

[MLGrado]

 
This is an interesting 'modern' perspective on audio quality and OS'
 
http://archimago.blogspot.co.nz/2015/08/measurements-audiophile-sound-and.html
 
.. The bottom line is that I am unable to find evidence to say that OS changes will make any difference to analogue audio output from a modern asynchronous USB DAC despite having heard comments already that somehow upgrading to Windows 10 will make things sound "better". Likewise, I see no evidence that software like Fidelizer will do anything to the sound output even though I can show that it "worked" to reduce the number of processes and threads of execution in the OS. Finally, and yet again, JPLAY demonstrates no clear ability to affect the sonic output and in fact (again after 2 years since the last test) I see problems with Kernel Streaming in playing back 24/48 accurately. This is obviously worse than "no effect" in that it actually deteriorates the quality of playback. I seriously wonder how these developers actually perform testing on their software and what they're actually doing! (Beyond marketing and dreaming up terminology like "ULTRAstream" for example.)
 
 
 

 
All tested with a cheap consumer ADC.  
 
Just saying.  If you are going to argue for measurements as being the end all, if you are going to argue that you know exactly what measurements are important, (and unimportant) and if you are going to arbitrarily define how small a difference is small enough to still be 'identical', then please, at least START with something more professional than a consumer level EMU ADC.  
 
Okay.  End of rant about that guy.  

 

[denti]

There is a discrepancy between the website and the manual on what inputs this can take:
 
USB: according to the manual DoP64 and DoP128, and also DSD512 (native)-- can that be true
 
How are people getting DSD252 and DSD512 to the Gustard? I'm curious. Is there a sound card that can output these over I2S (HDMI)? I don't know of one.

 

[Quadman]

  There is a discrepancy between the website and the manual on what inputs this can take:
 
USB: according to the manual DoP64 and DoP128, and also DSD512 (native)-- can that be true
 
How are people getting DSD252 and DSD512 to the Gustard? I'm curious. Is there a sound card that can output these over I2S (HDMI)? I don't know of one.

denti, you really should go back in this thread and read some, especially my posts, it is pretty much all laid out there.  Gustard will accept, via USB, DSD (native or DoP) signals up to DSD256 (if you have a driver that allows this) the driver supplied with gustard will allow up to DSD128 via USB.  You can only get DSD512 if you tap into the I2S lines and have a USB to I2s card that will handle DSD512 (really it needs the xmos 208 or 216 chip).  DIYinHK has one such card.  As to drivers, you can get a demo driver, which beeps every 60 minutes from Xmos web site that will allow DSD256 playback.  You need to set up an account first before downloading.  the DIYinHK also has a free xmos driver download, 2.23 I believe, that they claim will handle DSD256 (DoP) I have not confirmed this.  I also understand that you can use the singxer su-1 (mentioned on this thread) with an HDMI cable and adjust dip switches so the gustard's channels play correctly, tho it is my understanding this will NOT sound as good as tapping directly into the internal I2S lines of the dac.  It is a way easier solution than tapping into the internal I2S lines.

 

[denti]

 
denti, you really should go back in this thread and read some, especially my posts, it is pretty much all laid out there.  Gustard will accept, via USB, DSD (native or DoP) signals up to DSD256 (if you have a driver that allows this) the driver supplied with gustard will allow up to DSD128 via USB.  You can only get DSD512 if you tap into the I2S lines and have a USB to I2s card that will handle DSD512 (really it needs the xmos 208 or 216 chip).  DIYinHK has one such card.  As to drivers, you can get a demo driver, which beeps every 60 minutes from Xmos web site that will allow DSD256 playback.  You need to set up an account first before downloading.  the DIYinHK also has a free xmos driver download, 2.23 I believe, that they claim will handle DSD256 (DoP) I have not confirmed this.  I also understand that you can use the singxer su-1 (mentioned on this thread) with an HDMI cable and adjust dip switches so the gustard's channels play correctly, tho it is my understanding this will NOT sound as good as tapping directly into the internal I2S lines of the dac.  It is a way easier solution than tapping into the internal I2S lines.

 
Thanks. I've been slowly making my way through the thread.
 
I don't even have a PC yet that is powerful enough to do DSD256, but I'd like to plan ahead for that. So if I don't worry about DSD512, then I can get DSD256 with the right drivers, but there's only a demo driver that beeps? Why is that driver not available as non-demo? Are there no other drivers? Might they appear from Gustard later?

 

[denti]

So, I think I've now got a better handle on things. 
 
It seems that without DIY skills there are only two good ways to get DSD256 into the X20:
 
1) SU-1
 
2) the right drivers (which drivers these are is still unclear to me -- link, anyone?)
 
To get DSD512 at this point is only possible with DIY solutions. Is that right? Ideally we'll get something like an SU-2 or an X-13 or whatever, that will take DSD512 from USB and output it through I2S. My much is that this is around the corner, and that would save a lot of us a lot of trouble.
 
For now I will plan to build a mini PC that can handle the conversion involved in J.River to send out DSD512 via USB. Until such a non-DIY device is available (or the DIYing is made super easy), I'll stick with DSD128 or (assuming I find the drivers) DSD256.
 
(Until then I'm stuck with DSD64.)

 

[MHzTweaker]

Well my waiting is finally over.
 
DHL delivered my Singxer SU-1 at 11AM + some extras
It finally shipped last Thursday after almost 2 weeks after I ordered it.
China to N.C. in 4 days over the weekend is hauling mail.
 
How did they know I needed some earbud replacement earpiece thingeys???
 
Hopefully now I can experience my Gustard X20U at DSD256