Audio-gd Digital Interface - Page 40

Some pages ago, I think someone mentioned about doing RMAA testing to make sure that the DI does not do do funky "equalizing" with the sound. Whatever happened to that? I thought that it's a valid concern and it will be a very interesting discussion.

Quote:

Originally Posted by JulioCat2  I ask a few days ago if the DI is galvanically isolated but no one answered because as you discover the ADUM Isolator is a must for any USB Audio Device in order to obtain the best sound out of it.

 

Quote:

Originally Posted by leeperry Because it's not...before ADUM4160 showed up, it used to cost a hell lot of money to isolate USB...like in this thing. 

"The Digital Interface applied a silver wires isolated transformer for insulate the ground noise between PC and the DAC, which can offer the black backstage and more analog sound flavors." http://www.audio-gd.com/Pro/dac/USBface/Digital1EN.htm

I thought that a silver wired isloation transformer in the DI would qualify as a high standard of galavinc isolation from the computer hash? But I know nothing!
 

Originally Posted by Nada 

 I thought that a silver wired isloation transformer in the DI would qualify as a high standard of galavinc isolation from the computer hash?

That's most likely over coax...coax galvanic isolation is pretty easy, it can be done w/ a ferrite or a pulse transformer(which would apparently be a grossly overpriced ferrite at heart). USB isolation is far more complicated, only AD have a ready-to-go chip for that: http://www.youtube.com/watch?v=2sJaQHp5oSk

This photo in http://www.audio-gd.com/Pro/dac/USBface/DI5.jpg suggests that the silver isolation transformer is on the spdif output? It looks like the idea is to take out the ground noise, as the web site states. Is that called galvanic isolation?

I wonder if that Analogue Devices ADUM4160 chip  (http://www.analog.com/en/interface/digital-isolators/adum4160/products/product.html) would be predicted to give the DI an even better quality of performance?

Well wikipedia says yes it is:     http://en.wikipedia.org/wiki/Galvanic_isolation
 

Quote:

Originally Posted by Nada 

This photo in http://www.audio-gd.com/Pro/dac/USBface/DI5.jpg suggests that the silver isolation transformer is on the spdif output? It looks like the idea is to take out the ground noise, as the web site states. Is that called galvanic isolation?

 

I wonder if that Analogue Devices ADUM4160 chip  (http://www.analog.com/en/interface/digital-isolators/adum4160/products/product.html) would be predicted to give the DI an even better quality of performance?

Indeed, the silver isolation transformer will provide isolation from the converter to the converter. Hence, no noise coming from the computer will directly affect the DAC that is connected to the DI.

The ADUM provides galvanic isolation from the computer to the sub device. In my opinion, it is beneficial in 2 ways: if you connect a non isolated DAC directly to the computer or if you are using a poorly/average constructed usb converter that is heavily affected by the noise of the computer.

In my testing, I have found that the ADUM affects more the Musiland 01 USD (poorly constructed, uses the computer power supply) than the Teralink X2 (especially when using an external power supply).

The DI seems to have a pretty decent power filtration built-in. This is apparent if you look at the pictures on the website and also at the comments of the people who actually listened to it.

The fact that there is very little (or no difference) between powering the DI from USB or from an external source prooves that Kingwa has done a pretty good job at the power filter section. From this observation, I can predict that the ADUM and other audiophile tweaks (esoteric usb cables) will have little or no effect on the DI. At least it should have less effect than on competing products.

Sometimes, there are people who get excited (positively) when they discover that their usb converter can be improved when adding an external power supply or an ADUM converter. To me it is a sign of a weakness in the design of such a converter.

The same can be said for DACs: the entry level FUN is more sensitive (and "revealing") regarding the quality of the transports than the DAC19dsp. To me it is a sign of weakness and not something to brag about.

So to sum up, the ADUM won't add a galvanic isolation from the computer to the DAC as the DI already has that feature through the use of a silver transformer (and not a cheap off the shelf component by the way). What the ADUM would do is that it would isolate the noise from the computer to the DI and will affect how well the DI will perform... but since the reports so far seem to indicate that the external PS has little effect on the actual performance of the unit, I wouldn't expect a eart-shaking difference.

Of course, all of this is only an educated guess. The only opinion that matters is that of people who have actually listened to the Digital Interface.

Originally Posted by Nada 

This photo in http://www.audio-gd.com/Pro/dac/USBface/DI5.jpg suggests that the silver isolation transformer is on the spdif output? It looks like the idea is to take out the ground noise, as the web site states. Is that called galvanic isolation?

 

I wonder if that Analogue Devices ADUM4160 chip  (http://www.analog.com/en/interface/digital-isolators/adum4160/products/product.html) would be predicted to give the DI an even better quality of performance?

The goal is not only to galvanically isolate the coax output, it's also to do it on the USB input so no grounds will be shared whatsoever. Any semi-decent S/PDIF transport has a ferrite bead/pulse transformer on its coax output...even a Monitor 01USD does for that matter.

If USB isolation were so easy, AD wouldn't have bothered making the ADUM4160 chip. And yes it's a magical chip that improves every USB audio device you'll put it on as it reclocks the USB data flow, filters useless messages and provides it w/ a clean signal. Your pic clearly shows that that ferrite bead is in the coax signal path....you can't filter USB this easily.

Quote:

Originally Posted by leeperry 

The goal is not only to galvanically isolate the coax output, it's also to do it on the USB input so no grounds will be shared whatsoever.

 

It seems that we have different goals: my goal is to have the best sound quality possible, regardless of the technology being used. If the best result can be obtained by using battery power, let it be. If the best result can be obtained by using a USB isolator, I don't care.

My goal is not to have a specific set of components (ADUM, ...) inside (or plugged to) a converter but to have the best sound quality out of a converter. So far, I have tested/owned more than 8 external usb converters during the past 2 years. What I have learned is that you can have adaptive sound better than async, you can have external power supply sounding worse than usb power and vice versa.

There are no amount of tech talk that is going to give you with certainty the outcome of a technology until you have tried it on your system.

Quote:

Originally Posted by leeperry 

Any semi-decent S/PDIF transport has a ferrite bead/pulse transformer on its coax output...even a Monitor 01USD does for that matter.

 

Not all output stages are created equally. There is pulse transformer and ... silver transformer. Since you mentioned the Musiland 01 USD, the pulse transformer it uses is not even suited for 192KHz. The one used in the stock Hiface is not up to standards (regal had the DIY to mod it, I didn't).

The 75 ohms driver and the silver transformer used on the DI are a objectively a step up from the cheap components used on many entry/mid level converters. You have to go something like Empirical Audio Units to get that kind of quality output stages. Whether that is audible or not should be left to the people who have actually listened to the units... and reports are positive so far.

Quote:

Originally Posted by leeperry 

If USB isolation were so easy, AD wouldn't have bothered making the ADUM4160 chip. And yes it's a magical chip that improves every USB audio device you'll put it on as it reclocks the USB data flow, filters useless messages and provides it w/ a clean signal. Your pic clearly shows that that ferrite bead is in the coax signal path....you can't filter USB this easily.

In order to put your comment into perspective, could you please answer 2 simple questions:

1. Do you have first hand experience with the Audio-gd Digital Interface? Or perhaps one of Audio-gd's DSP-1 based DACs?

2. With how many converters on the market (could you name them?) have you tried the ADUM?

Keep in mind that this a thread about the Audio-gd Digital Interface. If you answer by a personal attack or by quoting some random poster about the ADUM, I will just have to assume that the answers to the 2 questions are: 1.you have no exposure with the DI and 2. you have only tried the ADUM with your Firestone converter.

For the sake of the conversation, let's keep this civil and talk about facts (it is very important to distinguish between proven facts and personal opinions and personal speculation).

Originally Posted by Nada 

 

ADUM4160 chip

I forgot to mention ADUM5000, but it's limited to 0.1A so no dongle using both seem to be available atm(USB can need up to 0.5A)..soon enough those two dynamite brothers will show up within USB audio interfaces input stages..it's only a matter of time.

Being this is the DI thread, and Kingwa being Kingwa, he'll probably read this (or someone will email him and ask about it), get hold of said USB isolating chip and test it, if he believes it will make any useful improvement.  If it does, it will appear in a revision of the DI. That's how it usually goes, doesn't it? Remember the 75 Ohm BNC socket discussion?

Interesting topic though.

Quote:

Originally Posted by Currawong 

Being this is the DI thread, and Kingwa being Kingwa, he'll probably read this (or someone will email him and ask about it), get hold of said USB isolating chip and test it, if he believes it will make any useful improvement.  If it does, it will appear in a revision of the DI. That's how it usually goes, doesn't it? Remember the 75 Ohm BNC socket discussion?

 

Interesting topic though.

I do agree with your assesment. Kingwa has been very responsive to suggestions these past few months: the 75 BNC socket but sometimes also by bringing new products (ES9018 DAcs, ROC SA for orthodynamics,...). I wonder if he ever finds the time to sleep.

I already suggested to him to make an upper scale usb converter (maybe on the CD7 level). I am sure he has the knowledge and capability to do so ... Maybe if enough people ask him ... he will think about it :)

Originally Posted by Currawong 

Being this is the DI thread, and Kingwa being Kingwa, he'll probably read this (or someone will email him and ask about it), get hold of said USB isolating chip and test it

get him to reclock the USB input via a WM8804 chip(50ps jitter) as well(using a good masterclock such as the zero PPM PLL1707), and I'll be glad to pay for shipping to test it(together w/ the DPS) :)

Quote:

Originally Posted by leeperry 

 

If USB isolation were so easy, AD wouldn't have bothered making the ADUM4160 chip. And yes it's a magical chip that improves every USB audio device you'll put it on as it reclocks the USB data flow, filters useless messages and provides it w/ a clean signal. Your pic clearly shows that that ferrite bead is in the coax signal path....you can't filter USB this easily.

I have the diyparadise Ultravox version of the ADUM. It has a painfully obvious degrading affect on an Empirical Audio Off Ramp.

Originally Posted by Newk Yuler 

I have the diyparadise Ultravox version of the ADUM. It has a painfully obvious degrading affect on an Empirical Audio Off Ramp.

O RLY? well, this is said to be 1ps jitter gear...what's the chip being used on its USB input? most likely not PCM2702 or TE7022L ^^

As explained in the youtube I recently linked, you can isolate/reclock USB by other means, but this doesn't come cheap...like in that Peachtree Audio Nova.

But you can see that this Monica2 DAC user was sold on it, and everyone trying it has given the same kind of raving feedback(including me, but my computer PSU is as dirty as can get): http://diyparadise.com/forum/index.php?topic=1063.msg11167#msg11167

"it is funny, all the sounds are there, some details too,
it just doesn't  sound so realistic without the USB-isolator. Music Dies away. Some of the magic dies out, the room perspective, depth is not so clear.. It's not coming out of the speakers properly without the isolator... Funny!"

and another one: http://www.computeraudiophile.com/content/Big-Bear-USB-2ISO

"With the rDac it provides a greater degree of a black background and lower noise floor."

I clearly couldn't think about using my current transport w/o it, and most other TE7022L users will agree on this