[Pharmaboy]

Did some testing and so far I came to the following results: (all testing done using WASAPI exclusive mode)

Amanero 1.067 drivers fails (8K bytes packages bulk transmission)
Device locks up completely after any amount of time (sometimes after only 10 minutes, sometimes after 1.5 hours)
Testing with ASIO mode give the same results

Amanero 1.062 drivers from Audio-GD website works (8K bytes packages bulk transmission)
I played a shuffled playlist continuously for 14 hours without any lock ups or errors, at this point I think I can call that "working".

Amanero Microsoft Windows provided drivers works (3K bytes packages asynchronous transmission)
I played a shuffled playlist continuously for 12 hours without any lock ups or errors, at this point like the others drivers I can call this "working"
As said before Microsoft doesn't provide an ASIO driver for the Amanero interface so that's not available. Perhaps it would work if you keep the Microsoft drivers and install the ASIO driver package from Amanero separately.. have not tested this and I'm not going to, I don't care for ASIO.

Microsoft Windows provided drivers for the Realtek ALC1220 codec optical out into the optical in of the Audio-GD NFB-11.38 works
Didn't expect any troubles with this option and non showed up, device ran without lock ups or errors for hours.
Pro: electrically isolated from the rest of the system.
Con: limited bandwidth, max 24/96 (which is more than enough for for the 2 channel PCM what I use it for, all my FLAC files are 16/44.1)
Con: Not sure if it's just my unit but the connector of the optical cable sits super loose in the socket on the Audio-GD, it's takes like no force at all to unplug the connector from the socket.


No idea what the difference(s) is/are between the generic 1.067 and the Audio-GD 1.062 drivers but there must be something as my unit at least is not happy with the 1.067 drivers at all.

I'm thinking about what interface to keep using, USB or optical. Both have something going for them.
At the moment I'm leaning towards the Amanero USB interface with the 1.062 drivers. Audio-GD unit comes with an "expensive" Amanero interface module so might as well use it. Also I have more confidence in the Audio-GD unit taking care of the clocks then the Realtek codec when using optical.


Thanks for the information
My unit seems to be working fine with the Audio-GD 1.062 drivers and the drivers provided by Windows.
So at this point I'm not so comfortable flashing firmware as it is working now.. don't try to fix something that ain't broken

You obviously know more about USB input & drivers than I ever will--but I offer this observation regardless:

• When I got my first DAC (a then praised delta/sigma design from Korea), I found its sound off-putting: bright, edgy, harmonically thin, all the digital nasties people talk about. I read that synchronous USB inputs were partly to blame, so picked up a SPDIF converter (Musical Fidelity V-Link96) + pricey toslink cable to use instead of USB. That made a real sonic improvement, so I stayed with it.
• Then I switched to a better sounding d/s DAC, the humble Peachtree Audio DAC iTx (which I still sometimes use). The iTx could decode 24 bit/192 kHz files, so I upgraded the converter to Musical Fidelity's V-Link192. I compared USB vs optical w/the iTx and found there was still a very slight advantage to optical. But I could easily hear that the asynchronous USB input board of the iTx (described as "galvanically isolated") was a clear upgrade over my 1st DAC.
• ~5 years ago I got my first NOS DAC, the Audio GD NOS 19. Again optical sounded better than USB with this DAC. I decided to fool around with coax and started out w/an inexpensive 1.5M Billy Jeans coax cable. That sounded better than optical and way better than digital.
• Eventually I sprung for a more expensive silver coax cable some praised: Oyaide R-510 (1.3M). Well, damn! With this new coax cable and the NOS 19, I suddenly had by far the best digital I'd ever heard IMS. Much of the note dimensionality, natural musical flow, and weight in lower registers of excellent high-end audio was now present.
• 2 DACs later (Audio GD DAC-19; now MHDT Labs Orchid), I still find the sonics of the SPDIF converter + silver coax cable superior to USB.

I admit that the way Audio GD handles drivers is very off-putting. I had a hard time getting drivers that worked with both AGD DACs (the drivers AGD linked to didn't work at all), though once I got the DACs' USB working, they worked well.

I also found the whole firmware-update thing very intimidating w/Audio GD gear. I've read enough about this tricky process to know I'll never do it. BTW, I remain an Audio GD fanboy despite these issues--I like the "house sound" a lot (also have/had 2 AGD headphone amps).

One of the best things about coax is that it bypasses all the tricky which-DAC-driver-works-best issue you outline in your post. I just don't think about DAC drivers now.

 

[sajunky]

One of the best things about coax is that it bypasses all the tricky which-DAC-driver-works-best issue you outline in your post. I just don't think about DAC drivers now.

Coax connection is also a preferable for the upper range DACs by many, but those have a poweful FPGA chip with PLL/FIFO. R2R-11/NFB11 performance depend on a standard S/PDIF receiver chip WM8805, there is no further jitter removal. On the other side USB can sound right, when ground loops are not entering inside.

 

[gimmeheadroom]

I'm starting to prefer AES/EBU and moving towards that. I like the locking connectors and it's relatively easy to make the cables.

 

[Telin]

You obviously know more about USB input & drivers than I ever will--but I offer this observation regardless:

• When I got my first DAC (a then praised delta/sigma design from Korea), I found its sound off-putting: bright, edgy, harmonically thin, all the digital nasties people talk about. I read that synchronous USB inputs were partly to blame, so picked up a SPDIF converter (Musical Fidelity V-Link96) + pricey toslink cable to use instead of USB. That made a real sonic improvement, so I stayed with it.
• Then I switched to a better sounding d/s DAC, the humble Peachtree Audio DAC iTx (which I still sometimes use). The iTx could decode 24 bit/192 kHz files, so I upgraded the converter to Musical Fidelity's V-Link192. I compared USB vs optical w/the iTx and found there was still a very slight advantage to optical. But I could easily hear that the asynchronous USB input board of the iTx (described as "galvanically isolated") was a clear upgrade over my 1st DAC.
• ~5 years ago I got my first NOS DAC, the Audio GD NOS 19. Again optical sounded better than USB with this DAC. I decided to fool around with coax and started out w/an inexpensive 1.5M Billy Jeans coax cable. That sounded better than optical and way better than digital.
• Eventually I sprung for a more expensive silver coax cable some praised: Oyaide R-510 (1.3M). Well, damn! With this new coax cable and the NOS 19, I suddenly had by far the best digital I'd ever heard IMS. Much of the note dimensionality, natural musical flow, and weight in lower registers of excellent high-end audio was now present.
• 2 DACs later (Audio GD DAC-19; now MHDT Labs Orchid), I still find the sonics of the SPDIF converter + silver coax cable superior to USB.

I admit that the way Audio GD handles drivers is very off-putting. I had a hard time getting drivers that worked with both AGD DACs (the drivers AGD linked to didn't work at all), though once I got the DACs' USB working, they worked well.

I also found the whole firmware-update thing very intimidating w/Audio GD gear. I've read enough about this tricky process to know I'll never do it. BTW, I remain an Audio GD fanboy despite these issues--I like the "house sound" a lot (also have/had 2 AGD headphone amps).

One of the best things about coax is that it bypasses all the tricky which-DAC-driver-works-best issue you outline in your post. I just don't think about DAC drivers now.

I'm not going into the rabbit hole of different interfaces and if one sounds better then the other. I will say that during my testing I could not hear any difference between USB input and optical toslink, I would not be able to tell the two apart in a blind test.
It appears that the Audio-GD has only the most basic optical input interface without any PLL so I guess the optical output of the ALC1220 is clean enough... or I'm just deaf

It is expected a new firmware (not beta) and a drivers package soon. Hopefully it will be a successful release.

We'll see.
New drivers or more for features then sonic improvement it seems.
At least during my testing I never heard a sonic difference between different driver versions of both XMOS or Amanero driver packages.

You noticed a lose optical connector. It is because a socket is recessed 2mm behind a thick wall and your plug moulding is little thicker than a cut in a wall, not going inside the socket properly deep enough. Check it out (some plugs are going in some other do not), but if you are going to use USB, it doesn't matter.

Now that you mentioned it I see it.
Bit of a weird design choice imho. I tried 3 cables I have at home and none fit properly in the socket of the Audio-GD.

 

[sajunky]

New drivers or more for features then sonic improvement it seems.
At least during my testing I never heard a sonic difference between different driver versions of both XMOS or Amanero driver packages.

Honestly, I don't expect any sonic improvements. Amanero is already working in the optimum mode, completely different from XMOS, but in both cases the on-board clock is used for speed synchronisation (Not PC clock as a master, it would be not good). A fact that you don't hear a difference is a proof. With XMOS there will be no feedback about dropped frames when error is detected, with Amanero you may see errors reported back (set a filter in a sniffer program to report only such cases) and even requests for retransmissions if timing allows. Be aware that transmission errors are very rare. On the other side if you load USB hub heavily, Amanero may start dropping frames, while XMOS has a reserved a portion of USB bandwith for all duration of a transfer. This the only difference between these two modes, not in SQ.

Amanero had very stable drivers, everything has changed when MS implemented (very poorly on the beginning) UAC 2.0 native drivers. They had to re-write the entire USB driver stack. MS implementation was not matching a Linux, it started to be very messy. As MS implementation is now more mature, I do expect integration of the UAC 2.0 firmware for Windows and Linux and more stability.

I see you are working hard, I was surprised with results, so I am sure you will not stop on this. You can work on SQ improvements in number of ways. The simplest way is to put ferrite clamps on the USB cable. It will increase impedance for common mode noise and redirect ground loops between devices to a power supply cable. My original cable has a molded one on the computer side, you can also put another one on the DAC side. To see effect immediately, you need to cycle a power of a DAC. It seems is doing some ladder recalibration during powering on. (EDIT: it is for R2R-11). It is reported that some more expensive double shielded cables give better SQ, but I cannot afford audiophile cables, didn't try it. Get a cable as short as your connection require.

There is also a trick with removing contact of +5V pin on the Type A connector. Amanero module on the Audio GD DAC do not use USB power, removing this connection is beneficial. You can put electric tape on the 5Vpin.

Induction based ground loops are proportional to the area between cables. You will reduce this area by placing power cables together (for PC+monitor and a DAC). As long a power outlet has a ground pin, you can use a trick with a wire.

 

[Telin]

Honestly, I don't expect any sonic improvements. Amanero is already working in the optimum mode, completely different from XMOS, but in both cases the on-board clock is used for speed synchronisation (Not PC clock as a master, it would be not good). A fact that you don't hear a difference is a proof. With XMOS there will be no feedback about dropped frames when error is detected, with Amanero you may see errors reported back (set a filter in a sniffer program to report only such cases) and even requests for retransmissions if timing allows. Be aware that transmission errors are very rare. On the other side if you load USB hub heavily, Amanero may start dropping frames, while XMOS has a reserved a portion of USB bandwith for all duration of a transfer. This the only difference between these two modes, not in SQ.

Interesting information, thanks for sharing!

I see you are working hard, I was surprised with results, so I am sure you will not stop on this. You can work on SQ improvements in number of ways. The simplest way is to put ferrite clamps on the USB cable. It will increase impedance for common mode noise and redirect ground loops between devices to a power supply cable. My original cable has a molded one on the computer side, you can also put another one on the DAC side. To see effect immediately, you need to cycle a power of a DAC. It seems is doing some ladder recalibration during powering on. (EDIT: it is for R2R-11). It is reported that some more expensive double shielded cables give better SQ, but I cannot afford audiophile cables, didn't try it. Get a cable as short as your connection require.

I will look into the ferrite clamps. Not very expensive so I'm willing to give it a try, worst that can happen is that it doesn't do anything.
I already have a double shielded USB cable, AudioQuest Forest USB cable. Didn't buy it myself, just something I got from a colleague who went the optical route and had no use for it anymore. Everyone must decide for themselves but personally I don't believe in (super) expensive USB cables. A good cable does not sound any worse than a >1000 euro one IMHO.

There is also a trick with removing contact of +5V pin on the Type A connector. Amanero module on the Audio GD DAC do not use USB power, removing this connection is beneficial. You can put electric tape on the 5Vpin.

I have a cleaner solution..
I connected my Audio-GD to the USB DAC port on my motherboard. I can completely turn off the 5V power on this port in the BIOS settings of my motherboard. I tested this with a multimeter and there is indeed no power on this USB port anymore.
Have to say honestly that I don't hear any difference in the SQ with or without the 5V enabled on the USB port.

 

[sajunky]

I have a cleaner solution..
I connected my Audio-GD to the USB DAC port on my motherboard. I can completely turn off the 5V power on this port in the BIOS settings of my motherboard. I tested this with a multimeter and there is indeed no power on this USB port anymore.
Have to say honestly that I don't hear any difference in the SQ with or without the 5V enabled on the USB port.

This is ..... probably an option for turning off power when computer off, or switching off USB_PD option. Check it out, if in doubt, post a photo or a link to the user manual.

As I wrote before, I didn't try audiophile type USB cable, but if I had such power cable, I would certainly try it.

 

[Telin]

This is ..... probably an option for turning off power when computer off, or switching off USB_PD option. Check it out, if in doubt, post a photo or a link to the user manual.

Manual

Page 98
Read for yourself. As I said it checked it with a multimeter and it does what it says it does, it cuts the 5v power from the USB port.

 

[sajunky]

Manual

Page 98
Read for yourself. As I said it checked it with a multimeter and it does what it says it does, it cuts the 5v power from the USB port.

Mea culpa. I can't believe!

BTW, I found something interesting on your mobo: 2x USB 2.0 headers. Do you use it?
I suggest to try. I have computer with only USB 3.0 ports, all ports are very noisy. I use USB 2.0 hub with this computer, it sounds better. Find a hub with multi transaction translator, it will allow to plug other devices without a worry of dropping speed of a hub. Or use any 2.0 hub with only a DAC connected.

 

[Telin]

Mea culpa. I can't believe!

BTW, I found something interesting on your mobo: 2x USB 2.0 headers. Do you use it?
I suggest to try. I have computer with only USB 3.0 ports, all ports are very noisy. I use USB 2.0 hub with this computer, it sounds better. Find a hub with multi transaction translator, it will allow to plug other devices without a worry of dropping a speed of a hub. Or use any 2.0 hub with only a DAC connected.

I used the USB 2.0 headers to connect the front USB 2.0 ports from my case.
Might test it out later, headers are blocked by the graphics card which is connected to the other graphics card with a hard bridge.. I know SLI is pretty much dead in 2021 but can't get myself to rebuild my entire system while is still doing what I need it for.

 

[Telin]

So after about 40 minutes of playback my unit locked up again using the Audio-GD 1.062 drivers
No idea how it is possible that it works for 12+ hours one day and fails within an hour the next...

Now running the Microsoft Windows supplied Amanero drivers and see how that goes.
These drivers actually make the transfer go in smaller packages and different method so it might work.... hopefully.

If not I'm stuck with optical I guess.
Not the end of the world but the Amanero interface should be technical superior so would've preferred to use that.

I'll report back how it goes.

 

[Pharmaboy]

So after about 40 minutes of playback my unit locked up again using the Audio-GD 1.062 drivers
No idea how it is possible that it works for 12+ hours one day and fails within an hour the next...

Now running the Microsoft Windows supplied Amanero drivers and see how that goes.
These drivers actually make the transfer go in smaller packages and different method so it might work.... hopefully.

If not I'm stuck with optical I guess.
Not the end of the world but the Amanero interface should be technical superior so would've preferred to use that.

I'll report back how it goes.

The only drivers I could get to work on my 2 AGD DACs were the drivers on the Amanero website--not the drivers linked on the Audio GD website.

 

[sajunky]

Beware of having multiple port drivers! This is what Kingwa wrote to me one year ago:

I think the driver issue may your computer install multiple driver but computer may not recognize correct. You may update the USB ports driver of your computer to try.
Or update the Amanero frimware to this version: CPLD_for_1080"

This is a top position on the table. Download oemtool117 from the Amanero website, not A-GD. It will flash with all defaults, there is no need to fiddle with combo boxes.

I had blue screen issues. Removed duplicate drivers, everything started working fine. The same happened now when playing with different drivers, this time no blue screen, but port lockups.

This is what should be done on Windows 7:

1. Uninstall ASIO driver and Amanero driver in Control Panel.
2. Go to the Device Manager. Expand "Devices by connection".
3. If Amanero driver is seen healthly, you have multiple port drivers!
4 Go to the Amanero properties and uninstall driver. Select an option to delete from a disk.
4. Cycle power on a DAC. If a healthly Amanero device comes back, there is another port driver! Repeat #4.
5. When device comes up unrecognised, OS is ready reinstall Amanero drivers.

I expect Windows 10 will complicate things by installing built in UAC 2.0 drivers. I don't have updated Win10, I can't verify.

For me there is no problem with the drivers. On Win7 1.062 package from A-GD website install a version 1.0.57, the same (I think) the latest driver from Amanero website. On Win10 I have installed the latest drivers from Amanero website.

 

[Telin]

I've been extremely busy with work so not much time to spend on this issue.

Windows 10 is making it more complicated as it will (re)install the UAC drivers as soon as you uninstall the Amanero. I will also try to dig around in the device manager to remove it that way, as much as Windows 10 allows me that is.

As said have not spend a lot of time with the Audio-GD lately but the Windows UAC drivers seem to work fine..... for now at least...
Although it is kinda "itching" me that it's not running with the Amanero drivers which should technically be the best..

What would be the preferred (best) option if the Amanero driver continue not to be an option on my system?

1. USB connection using the Windows UAC 2.0 drivers
2. Optical (Toslink) connection using the Windows drivers

 

[sajunky]

1. USB connection using the Windows UAC 2.0 drivers
2. Optical (Toslink) connection using the Windows drivers

It is reported that Coax can give better SQ that USB, nut sure about optical, it is always more jittery than coax.
Test it yourself what works the best in your setup and report back.