Audio-gd Digital Interface
Jul 16, 2010 at 1:04 AM Post #121 of 4,156


Quote:
How about asking if they can replace the RCA OUTPUT to an optical input jack? If this can be done without too much trouble or sound degradation it may be preferable because you can just make a BNC>RCA cable if you need to connect it to a dac with RCA in and not BNC in, the only drawback would be just one output. I asked them earlier and they told me to ask when they finished testing, but I plan to get a ref 3 so won't be bothering them about it.


Ok I've just asked him. This is a pretty good idea haloxt.
Kingwa amazes me on how good he is with customer service.
 
Jul 16, 2010 at 2:26 AM Post #122 of 4,156
[size=x-small]Price:[/size][size=x-small]  Version A          USD190 [/size][size=x-small](With USB and coaxial input)[/size]
[size=10pt]              [/size] [size=x-small]Version B[/size][size=x-small]          USD140  [/size][size=x-small](Only USB input)
           
[/size] [size=small](The Digital Interface can work well by USB power supply ,The external power supply only for upgrade, not necessary .)[/size]
[size=x-small]            Class A [/size] [size=x-small]e[/size][size=x-small]xternal power supply    USD75  [/size] [size=x-small](Because the Digital Interface built in high performance class A PSU, so the external power supply has not much improve.)[/size]
 
http://www.audio-gd.com/Pro/dac/USBface/Digital1EN.htm
 
Seem like the external power supply cost 75 usd but offer not much improvement?
 
I guess i will wait for reviews from the first 10 testers before buying
 
Jul 16, 2010 at 2:48 AM Post #124 of 4,156


 
Quote:
 I'll see what sounds better between USB and TOSlink on the Digital Interface.
I hope that I won't have to invest in pricey USB cables to get the best out of this unit.
I'm not familiar at all with the I2s standard but I read somewhere on the Audio-GD website that it would be I2s compatible. Does anyone know what are the benefits of I2s over S/PDIF and if the RE1 can recognize it?


 
[size=10pt][size=10pt][size=10pt]My experience is that TOSlink is not a good audio interface. I think the issue is that the necessary chips out there in the marketplace are not that good. Anyway, I have always preferred the sound from S/PDIF. [/size][/size][/size]
[size=10pt][size=10pt] [/size][/size]
[size=10pt][size=10pt][size=10pt]Benefits of I2S:[/size][/size][/size]
[size=10pt][size=10pt] [/size][/size]
[size=10pt][size=10pt][size=10pt]I2S is the means by which the DAC chip communicates with the outside world. In other words, it is the DAC chip’s own “native” language. Somewhere in the digital playback chain music data is converted from one format e.g. raw data (if being read directly from CD) to I2S format to then be processed by the DAC chip. In my set-up, I store music on a HDD in wav file format. The wav is then transferred by the music player software (cPlay or whatever) to a digital interface device (like the Musiland) by a USB cable. The interface device will attempt to recover the data and re-clock it using its own internal clock so that the wave form can be properly reconstituted by the DAC chip at a later time. The digital interface device, however, typically has to convert the data (once again) to some format in order to talk to the DAC chip. If, however, the interface device is able to convert to I2S and then transmit this to the DAC chip then one can see that some conversions are eliminated. This kind of architecture also lends itself to the idea that the DAC chip is effectively being slaved to the (precision) clock in the digital device. [/size][/size][/size]
[size=10pt][size=10pt] [/size][/size]
[size=10pt][size=10pt][size=10pt]What is a further development is for the digital interface device to control what the computer is doing to some degree. This is what Musiland and M2tech try to do. The ideal being that all three elements (DAC, interface device, and PC) are more or less tied together and running off the one precision clock somewhere in the chain (usually the interface device). However, what Audio-GD (I think) has done in the DSP-3 is avoid the issue of PC control preferring to focus on other areas. I can understand this if only because I can see that the murky waters of PC software development are expensive and deep. Of course Audio-GD could do what Empirical Audio has done and license the software from M2tech. But that could potentially add another US$300 to the cost if one were to use Empirical Audio as a yard stick. But you never know, there will likely be a DSP-4 one day..[/size][/size][/size]
[size=10pt][size=10pt] [/size][/size]
[size=10pt][size=10pt][size=10pt]RE1:[/size][/size][/size]
[size=10pt][size=10pt] [/size][/size]
[size=10pt][size=10pt][size=10pt]I don’t know if the RE1 can accept I2S from the DSP-3. According to Kingwa, the DAC-19DSP can. Perhaps you could ask him.[/size][/size][/size]
[size=10pt][size=10pt] [/size][/size]
[size=10pt][size=10pt][size=10pt]A further note: [/size][/size][/size]
[size=10pt][size=10pt] [/size][/size]
[size=10pt][size=10pt][size=10pt]It is said that I2S was never designed for “distance travelling”. What this means is that it was never meant to connect separate devices together and that if you do so you are likely to corrupt the I2S signal. I don't have the experience to make a comment. Nevertheless, I am keen to give I2S a go.[/size][/size][/size]
 
Jul 16, 2010 at 6:43 AM Post #125 of 4,156
Although I2S doesn't have the problems of SPDIF as regards the clock being embedded with the data & it's recovery, it has as many issues as SPDIF (If not more) with regard to cables i.e. impedance issues & length of cable, etc. SO it's no panacea AFAIK.
 
Jul 16, 2010 at 7:03 AM Post #126 of 4,156
I was advised by Kingwa that the PSU was not necessary as the Interface provided several transistor PSU's which would easily clean the power.
i ordered the B version because I don't see myself ever using the interface with a BNC input.
 
Jul 16, 2010 at 9:52 AM Post #127 of 4,156

 
Quote:
[size=10pt][size=10pt][size=10pt]My experience is that TOSlink is not a good audio interface. I think the issue is that the necessary chips out there in the marketplace are not that good. Anyway, I have always preferred the sound from S/PDIF. [/size][/size][/size]
[size=10pt][size=10pt] [/size][/size]
[size=10pt][size=10pt][size=10pt]RE1:[/size][/size][/size]
[size=10pt][size=10pt] [/size][/size]
[size=10pt][size=10pt][size=10pt]I don’t know if the RE1 can accept I2S from the DSP-3. According to Kingwa, the DAC-19DSP can. Perhaps you could ask him.[/size][/size][/size]
[size=10pt][size=10pt] [/size][/size]
[size=10pt][size=10pt][size=10pt]A further note: [/size][/size][/size]
[size=10pt][size=10pt] [/size][/size]
[size=10pt][size=10pt][size=10pt]It is said that I2S was never designed for “distance travelling”. What this means is that it was never meant to connect separate devices together and that if you do so you are likely to corrupt the I2S signal. I don't have the experience to make a comment. Nevertheless, I am keen to give I2S a go.[/size][/size][/size]

 
Kingwa says that adding a third input would require a switch and that would degrade the SQ of both coaxial and TOSlink and regarding the SQ he prefers coaxial input over TOSlink on the Digital Interface. I guess I'll just scrap the TOSlink input and stick with the stock version.
Also, Kingwa confirmed to me that the RE1 can accept I2s.
According to him, I2s has no ground insulation between 2 gears and it is better to keep it on short distances.

 
 
Jul 16, 2010 at 3:20 PM Post #128 of 4,156
Here's a crazy idea: TCP/IP!
 
I've never heard of jitter, impedance, reflection, or re-clocking issues with my net connection. And it handles bit-perfect audio streaming over long distances just fine. As a bonus, we'd be able to use regular ethernet hubs/switches with our DACs and transports too. Not to mention wi-fi. Ethernet interfaces can be pretty inexpensive as well.
 
Something tells me I can't have been the first one to think of this. I wonder why no one does this.
 
Quote:
Although I2S doesn't have the problems of SPDIF as regards the clock being embedded with the data & it's recovery, it has as many issues as SPDIF (If not more) with regard to cables i.e. impedance issues & length of cable, etc. SO it's no panacea AFAIK.


 
 
Jul 16, 2010 at 3:25 PM Post #129 of 4,156
oh no!!!! You never hear about software jitter, jitter induced by network delays packet colissions etc, on cable or wifi??? Ehernet is an option but not the perfect one.
 
Jul 16, 2010 at 3:28 PM Post #130 of 4,156
There is also jitter on TCP/IP too. So, modems has their own buffers and correction algorithms. We can't notice that jitter because timing is not so important on data transfer. Also online sound streams has very low quality compared to hi-fi streams. But ethernet interface can still  a proper solution (Some professional equipment using it if I remember correctly)
 
Jul 16, 2010 at 3:32 PM Post #131 of 4,156
@Yoga Flame
 
I think that you should share your idea to Kingwa at Audio-GD [size=13.5pt]audio-gd@126.com[/size]
I'd like to know what he has to say about this.
 
Jul 16, 2010 at 3:37 PM Post #132 of 4,156


Quote:
Here's a crazy idea: TCP/IP!
 
I've never heard of jitter, impedance, reflection, or re-clocking issues with my net connection. And it handles bit-perfect audio streaming over long distances just fine. As a bonus, we'd be able to use regular ethernet hubs/switches with our DACs and transports too. Not to mention wi-fi. Ethernet interfaces can be pretty inexpensive as well.
 
Something tells me I can't have been the first one to think of this. I wonder why no one does this.
 

 


Squeezebox, anyone?
 
Jul 16, 2010 at 4:09 PM Post #133 of 4,156
 
Quote:
oh no!!!! You never hear about software jitter, jitter induced by network delays packet colissions etc, on cable or wifi??? Ehernet is an option but not the perfect one.

 

That's a good point. Net connections do have their share of problems. It's a pretty crowded network out there after all.
 
But within my home LAN I have had zero issues. I'm sure there are still dropped packets and collisions on my local network, but it's all transparent to me since TCP/IP has error correction built into the protocol. The end result for me is that data is sent and received bit-perfect.


Quote:
Squeezebox, anyone?


I guess that qualifies as an example. Not exactly what I had in mind though. Devices like Squeezebox and Airport Express function as DACs, and must replace the existing one in the typical audiophile rig. Or they might have digital outputs for a dedicated DAC, using SPDIF,  which is back to square one.
 
Perhaps if the Squeezebox type of device had an I2S output. We could sit it right next to the dedicated DAC where the distance wouldn't be an issue. Still, the ideal scenario as I see it would be for all DAC manufacturers to adopt ethernet as a standard input.
 
Jul 16, 2010 at 4:17 PM Post #134 of 4,156


Quote:
@Yoga Flame
 
I think that you should share your idea to Kingwa at Audio-GD [size=13.5pt][ ... ][/size]
I'd like to know what he has to say about this.


I will leave that to his regular correspondents here. I too am interested in his take on that, but I'm here merely out of curiosity and am not really in the market for a transport at the moment.
 
Jul 16, 2010 at 4:18 PM Post #135 of 4,156


Quote:
 
 
Kingwa says that adding a third input would require a switch and that would degrade the SQ of both coaxial and TOSlink and regarding the SQ he prefers coaxial input over TOSlink on the Digital Interface. I guess I'll just scrap the TOSlink input and stick with the stock version.
Also, Kingwa confirmed to me that the RE1 can accept I2s.
According to him, I2s has no ground insulation between 2 gears and it is better to keep it on short distances.

 

 
I guess I will check with Kingwa about the I2s connection with the DAC-19 and how it can be done as I am hoping to test the Digital Interface with the FUN and DAC-19/C-2 soon.
 

 
 

Users who are viewing this thread

Back
Top