Coming Soon: Audio-GD Reference 3 USB>SPDIF converter/wonder-box

[auee]

Quote:

Originally Posted by DoYouRight /img/forum/go_quote.gif Musiland will have mac drivers very soon.

Please let us know when this occurs. Then those of us who use Macs will be very happy.

 

[sandchak]

Quote:

Originally Posted by FauDrei /img/forum/go_quote.gif Does Ref3 + DAC-9mk3 translate as Ref1 with additional functionality (Ref1 with two SPDIF/BNC inputs + optical + USB)? ...or are there some shortcomings of such separated setup (f.e. DAC-9mk3 does not take 192KHz/24bit SPDIF/BNC input)? From what I read on Audio-GD's site DAC-9mk3 is Ref1 without fancy DSP... having the missing functionality in a separated box would make such combination even more appealing. But if this is true, then Ref3 MUST HAVE at least 96KHz/24bit USB implementation to make it most appealing.

In my opinion DAC19MK3 + REF3 does not translate to REF1, although it does give you the additional functionality (as you have mentioned + it also has CAST outputs apart from usual transistor ones).

The REF1 is totally a different beast if you compare the internals with DAC19MK3, apart from the fancy DSP1, it has fancy power supply section, 4 X PCM1704U K per channel, Balanced DAC etc.

I have both the REF1 and DAC19SE (older version) and the difference between the two is quite big.

Btw, I think A-GD has released the REF3 for overseas market, although only 3 pcs for testing purposes : н¨ÍøÒ³ 1

 

[some1x]

Looks like its priced is $550

Unless the Ref 3 has an amazing USB implementation ... I'd rather get the Empirical Off Ramp for I2S and 24/96 support.

 

[scootermafia]

Yeah, $550 for a box not doing 96/24 is disappointing. This is why you crack the whip and have someone make you the implementation from scratch. I'm willing to pay $1000 if Lee's one makes a difference, and good circuit design does matter. It's a little scary the USB signal going through another massive integrated circuit, I mean come on.

 

[Pricklely Peete]

I cannot understand why USB is even being considered other than as an emergency fallback solution (Kingwa has this view as well) ....if any of you have this month's TAS mag read the comprehensive article on USB....let's put it this way USB is a major fidelity killer (no matter how good USB drivers and chip sets become, 24/96 or no).

If you have a SC that has a spare SPDIF header make your own breakout panel using a spare blank (covering a PCI slot) on the back of the comp. Use a pair of jacks, one BNC and one SPDIF RCA...or stick with one type it doesn't matter. It really isn't that hard to do. Drill a 10 mm hole in the blank and add a section of 75 ohm COAX from it to your SPDIF header on the sound card. Done...no USB tom foolery needed.

Peete.

 

[IPodPJ]

Quote:

Originally Posted by Pricklely Peete /img/forum/go_quote.gif I cannot understand why USB is even being considered other than as an emergency fallback solution (Kingwa has this view as well) ....if any of you have this month's TAS mag read the comprehensive article on USB....let's put it this way USB is a major fidelity killer (no matter how good USB drivers and chip sets become, 24/96 or no). If you have a SC that has a spare SPDIF header make your own breakout panel using a spare blank (covering a PCI slot) on the back of the comp. Use a pair of jacks, one BNC and one SPDIF RCA...or stick with one type it doesn't matter. It really isn't that hard to do. Drill a 10 mm hole in the blank and add a section of 75 ohm COAX from it to your SPDIF header on the sound card. Done...no USB tom foolery needed. Peete.

I don't know. It boggles my mind, too.

 

[tosehee]

Quote:

Originally Posted by Pricklely Peete /img/forum/go_quote.gif I cannot understand why USB is even being considered other than as an emergency fallback solution (Kingwa has this view as well) ....if any of you have this month's TAS mag read the comprehensive article on USB....let's put it this way USB is a major fidelity killer (no matter how good USB drivers and chip sets become, 24/96 or no). If you have a SC that has a spare SPDIF header make your own breakout panel using a spare blank (covering a PCI slot) on the back of the comp. Use a pair of jacks, one BNC and one SPDIF RCA...or stick with one type it doesn't matter. It really isn't that hard to do. Drill a 10 mm hole in the blank and add a section of 75 ohm COAX from it to your SPDIF header on the sound card. Done...no USB tom foolery needed. Peete.

Is jitter non-issue with RE1 + this approach? The reason why I ask is because RE1 has DSP-1 which supposed to reclock and etc to fight the jitters. Do we really need a sophisticated transport to fight the jitters? or are we really fighting a unnecessary evil here?

 

[Pricklely Peete]

Quote:

Originally Posted by tosehee /img/forum/go_quote.gif Is jitter non-issue with RE1 + this approach? The reason why I ask is because RE1 has DSP-1 which supposed to reclock and etc to fight the jitters. Do we really need a sophisticated transport to fight the jitters? or are we really fighting a unnecessary evil here?

Well the Transport matters a whole lot......why hamstring yourself with USB ? I realize the RE1 will only accept BNC/SPDIF but that is by design. In all honesty I don;t use a comp T with my ref system it simply isn't in the same league as the CD7 or the CDM12 Pro T top spinner I used before it. Even with a decent SPDIF out from a Auzen prelude the PC/Comp T came in last place fidelity wise.

Jitter is major factor with any USB solution and should be avoided at any cost IMO. The RE1 (and RE3 I'm willing to bet) needs a strong glitch free clean signal from it's transport in order to delver the goods without digital lock "hunting" I like to call it. The DSP1 isn't a panacea even though it's made out to be, much like other DAC's attempts to eliminate jitter with their own expensive solutions...the bottom line is getting the T signal right and keeping it far far away from anything that can influence it in a negative fashion. It is the number 1 goal or at least it should be, for any competent T IMO. Part of the reason why Kingwa is creating the RE3 was to eliminate the USB option altogether (kinda ironic) ....in it's original form it did not have USB at all...it would have been out last year if USB was left out but Kingwa bowed to pressure and the RE3 morphed into a USB/BNC/SPDIF comp transport.....you still won't get the most from it unless you use BNC .....again the last resort as a temp measure even with the RE3 is USB.

Sorry if this reply sounds a little preachy ...it really isn't meant to be like that.I like getting straight to the point you see.

Peete.

 

[tosehee]

Ok. You got me confused a bit here.

T refers to "timing"?

So, in other thread, you said to avoid USB and solder directly to sound card to add the BNC/RCA to the SPDIF header.

Even that is not a good solution then since the PC as a transport would lack the timing which would kill the fidelity.

I'd like to pursue your route and have THE best source for the transport, but I am also damn lazy and using PC as a transport spoil the hell out of me. I am simply too lazy to swap the CD back and forth.

As much as I hate to use PC, it is THE ultimate, most convenience tools to play the music. With that, what is the best way to get the sound to the RE1 then?

Quote:

Originally Posted by Pricklely Peete /img/forum/go_quote.gif Well the Transport matters a whole lot......why hamstring yourself with USB ? I realize the RE1 will only accept BNC/SPDIF but that is by design. In all honesty I don;t use a comp T with my ref system it simply isn't in the same league as the CD7 or the CDM12 Pro T top spinner I used before it. Even with a decent SPDIF out from a Auzen prelude the PC/Comp T came in last place fidelity wise. Jitter is major factor with any USB solution and should be avoided at any cost IMO. The RE1 (and RE3 I'm willing to bet) needs a strong glitch free clean signal from it's transport in order to delver the goods without digital lock "hunting" I like to call it. The DSP1 isn't a panacea even though it's made out to be, much like other DAC's attempts to eliminate jitter with their own expensive solutions...the bottom line is getting the T signal right and keeping it far far away from anything that can influence it in a negative fashion. It is the number 1 goal or at least it should be, for any competent T IMO. Part of the reason why Kingwa is creating the RE3 was to eliminate the USB option altogether (kinda ironic) ....in it's original form it did not have USB at all...it would have been out last year if USB was left out but Kingwa bowed to pressure and the RE3 morphed into a USB/BNC/SPDIF comp transport.....you still won't get the most from it unless you use BNC .....again the last resort as a temp measure even with the RE3 is USB. Sorry if this reply sounds a little preachy ...it really isn't meant to be like that.I like getting straight to the point you see. Peete.

 

[Pricklely Peete]

T is short for transport in my parlance, sorry tosehee I should have said earlier

....I get a little lazy at times with the keyboard so I shorten words from time to time....

No the SPDIF out is much much better than USB at conveying the signal source integrity when used in a Comp environment but it still comes up short against a standalone quality transport IME.

USB has an effect on everything it passes.....which disappears for the most part when you switch to a BNC or SPDIF COAX output.

Of course this is my personal experience thus far and IMO...I can recreate the effect here and it's easily audible. The USB signal loses focus and soul, treble gets glassy, bass a little muddy, piano and strings sound artificial. Swap to SPDIF COAX and things get a whole lot better but imaging and overall top to bottom freq response still suffers a little. This is with the Comp T. Switch to a high quality spinner and things get better again...with the last step being a pure BNC--->BNC signal driven by the CD7...it's markedly superior to any other T (comp based or stand alone) that I have. I have several standalone players ranging from mid fi to top shelf...the CD7 stands alone with it's partner the RE1 by a huge margin. That threw me for a loop how much better it was.

Sandchak has experienced this very same thing and we've talked about it rather in depth at times. I think Bob has a CD7 now....and I bet he will soon chime in with some similar observations.

Peete.

 

[tosehee]

Thanks a lot, Peete, for indepth explanation.

 

[gevorg]

SPDIF is simply inferior to properly implemented USB, especially if you use an internal soundcard. Of course it doesn't mean that SPDIF sucks and if you have a good jitter "immune" DAC like the Benchmark or Ref1, it shouldn't really matter to you since the differences (if any) would probably be insignificant. Kingwa's USB implementations have been based on PCM270x chip, so obviously he found the SPDIF to be superior.

Many highend audio manufacturers see that USB audio is the way to go for computer-as-source users. Scroll down in the link below to see what Bel Canto, Empirical Audio, Wavelength Audio, Weiss, Ayre Acoustics, Red Wine Audio, and others have to say about computer-as-source.

ca intro

 

[some1x]

Quote:

Originally Posted by Pricklely Peete /img/forum/go_quote.gif The USB signal loses focus and soul, treble gets glassy, bass a little muddy, piano and strings sound artificial. Swap to SPDIF COAX and things get a whole lot better but imaging and overall top to bottom freq response still suffers a little. This is with the Comp T. Peete.

For clarification
1) what USB converter have you tried?
2) what is providing the SPDIF out of your computer, that is superior to the USB converters in 1?

 

[DoYouRight]

usb isolates the computer noise therefore jitter dead mostly compared to soundcard if its done right.

 

[tosehee]

Jitter is the glitches caused by the timing. I don't think it has anything to do with the computer noise. That would be interference/EMI or something of that nature. I doubt the PC noise travels to optical and affects the sound quality. If it is, that wouldn't be a jitter.

Just my $0.02

The usb would be better than spdif because it's got a timing information there. Therefore, it has the potential to be the best implementation. However, only TAS1380(??) comes close to this solution, and rest are pretty poor.

The Musiland is interesting beast since it's got its own driver, and tries to tackle the jitter in different manner. How good is it going is not technically tested/proved yet. It can do the bit perfect output, but when it comes to 'jitter', we do not know how it fairs other than rough reviews from one or two online reviewers.