Thoughts on a bunch of DACs (and why delta-sigma kinda sucks, just to get you to think about stuff)

[kugino]

 
I have the 10.32.  It's a really great unit and checked all the boxes for me as I really like having a good "all-in-one" to travel with.   I can really only compare with any accuracy to my EAD DSP7000 and my Theta Pro Basic 3 and Progeny as I am most familiar with those.  If really splitting hairs compared to the EAD with the PCM 63 chip the 10.32 is a little more "syrupy" sounding and can have an ever so slight etch to the upper mids/treble.  But honestly, it's a great unit, with a great amp section, exceptionally smooth and transparent sound, and never seems too thick, and I never feel like I'm missing anything from my "vintage" dac's.  There's great "weight" to the sound of the 10.32 as well, there seems to be a little less "bloom" to the sound than my Theta and EAD Dac's. 
 
I will say the DAC section of the 10.32 does have many similarities to the Theta Pro Basic IIIa that I have.
 
 My favorite DAC remains the EAD DSP7000, it just has the right balance for me, it really seems to accentuate the "right" parts of music, and is never harsh, and has an amazing balance or space, weight, speed and smoothness that makes it a great all arounder.  I don't know if this is because of the PCM 63 chip, but I'm sure that has something to do with it.
 
Regardless I would recommend the 10.32, as I've never been disappointed with any of the Audio-GD gear I've owned.  No it may not be as fluid/effortless as the Yggy's I've heard, but it's fairly close and the Yggy's don't have a 6 watt amp section, so it really depends on what you need.

cool. thanks for the impressions. i might look into buying a used unit. again. maybe. d'oh...

 

[guilders0]

Thanks for sharing zach.
Could you test its preamp part?
I am interested into having it connected to a power amp and Wonder how good its volume section is?

 

[ciphercomplete]

  Has anybody heard Monarchy NM24?

I owned one for 5~6 years.  A great dac but I modded mine by replacing the big red WIMA caps on the dac side with Mundorf Supremes.  Soundstage depth and midrange detail increased.  The NM24 holds a special place in my heart because it was my first relatively expensive dac but you would probably do just as well by buying the new Audio GD Dac19.  I owned the old version a while ago and aside from the loss of tube rolling I think they are about equal.  

 

[lukeap69]

I owned one for 5~6 years.  A great dac but I modded mine by replacing the big red WIMA caps on the dac side with Mundorf Supremes.  Soundstage depth and midrange detail increased.  The NM24 holds a special place in my heart because it was my first relatively expensive dac but you would probably do just as well by buying the new Audio GD Dac19.  I owned the old version a while ago and aside from the loss of tube rolling I think they are about equal.  

You probably have saved me lots of bucks! Thanks.

 

[prot]

The most interesting new DAC
http://www.digitalaudioreview.net/2015/05/mergings-ethernet-nadac-impresses-at-munich-high-end-2015/
Direct ethernet DAC, no bloody USB. And if you follow the link at the end of the article, they do provide some spectacular measurements.

And btw, this being the R2R lalala thread, I wonder why isnt anyone talking about the (only?) portable r2r dac/headamp http://www.head-fi.org/products/myst-1866/reviews/11660

 

[Sonic Defender]

The most interesting new DAC
http://www.digitalaudioreview.net/2015/05/mergings-ethernet-nadac-impresses-at-munich-high-end-2015/
Direct ethernet DAC, no bloody USB. And if you follow the link at the end of the article, they do provide some spectacular measurements.

And btw, this being the R2R lalala thread, I wonder why isnt anyone talking about the (only?) portable r2r dac/headamp http://www.head-fi.org/products/myst-1866/reviews/11660

Are these comparable price ranges? I doubt it. Meaning compared to the Schiit Yggy.

 

[drez]

Not sure ethernet is the answer - it's still a packetised transmission just like USB, and can generate a lot of noise.  There is also need for a computer inside the DAC which these days also probably needs to run some pretty looking mobile device UI app for remote control.  I'm not saying it's necessarily better or worse, just not automatically a free lunch.
 
http://www.akm.com/akm/en/product/featured/velvetsound/
 
Interesting new chip form AKM, looks like a bit of a challenger for Sabre type of chip.  Like usual AKM are really trying to get the noise as low as possible - I hope this can beat Sabre and have more benign sound in most implementations.

 

[prot]

Not sure ethernet is the answer - it's still a packetised transmission just like USB, and can generate a lot of noise.  There is also need for a computer inside the DAC which these days also probably needs to run some pretty looking mobile device UI app for remote control.  I'm not saying it's necessarily better or worse, just not automatically a free lunch.

http://www.akm.com/akm/en/product/featured/velvetsound/

Interesting new chip form AKM, looks like a bit of a challenger for Sabre type of chip.  Like usual AKM are really trying to get the noise as low as possible - I hope this can beat Sabre and have more benign sound in most implementations.

There is of course no free lunch. But network is, at least in theory, much better than USB ... pretty much the best we have today for data transmission. And no you dont need a computer inside the DAC, just a simple network endpoint. Posted a msg about that stuff a few weeks ago, you can look it up a few pages back. Some details also here http://nadac.merging.com/networking

The pro world is already moving to network but I did not expect such a consumer device so soon. Quite encouraging.

Sonic Defender
Just wanted to discuss/hear about an interesting new DAC .. pretty sure there are enough other threads about old schiit.

 

[drez]

A matter of syntax I guess IIRC still requires microprocessor, memory, network adapter, linux operating system to run the server?
 

At the other end these is a NAS, which is pretty much a simple server computer.  Sure it gets rid of the USB transmission, but IME the network endpoint needs to do even more heavy lifting than a USB receiver.

 

[prot]

A matter of syntax I guess IIRC still requires microprocessor, memory, network adapter, linux operating system to run the server?



At the other end these is a NAS, which is pretty much a simple server computer.  Sure it gets rid of the USB transmission, but IME the network endpoint needs to do even more heavy lifting than a USB receiver.

Bit of syntax true ... but the Dac chip itself is a computer too if you wanna go that way... also the part that does filtering or any kind of DSP.

OTOH, something as simple as a modem or hub is a network endpoint. And that's all you need inside the dac, an ethernet-to-i2s "mo-dem"...actually even less cause it only has to covert/transmit one way. The Nas at the other end does pretty much all else: stores the music, runs the music player software, etc.

It's a much simpler setup and you get rid of a second PC, USB protocol & cables... and simpler usually sounds better too. But even if it doesnt sound better, what could be wrong with cheaper & easier ?!

 

[Wildcatsare1]

Anybody have any experience with the DCS "Ring DACs"? Totally forgot I have an Arcam FMJ CD23 sitting in my audio closet (I know, a mind and good audio are a terrible thing to waste).

 

[drez]

I didn't know that the server software runs from the NAS, guess that makes sense. Endpoint would then just receive bitstream then and not much else? I guess that shouldn't be too different from usb receiver apart from converted niece of remote app and NAS. I'm more from the wildly impractical for marginal sound quality gain though so my viewpoint is warped lol

 

[prot]

I didn't know that the server software runs from the NAS, guess that makes sense. Endpoint would then just receive bitstream then and not much else? I guess that shouldn't be too different from usb receiver apart from converted niece of remote app and NAS. I'm more from the wildly impractical for marginal sound quality gain though so my viewpoint is warped lol

That's pretty much it, a small network receiver board that converts ethernet-to-i2s. Doesnt have to be more complex than the Usb-to-i2s receiver boards inside most current Dacs. And the only new software needed is an ASIO-over-ethernet driver instead of the current Asio-over-usb.
Much more practical (less components & cables) and at least in theory it should also sound better.

 

[BassDigger]

  Not sure ethernet is the answer - it's still a packetised transmission just like USB, and can generate a lot of noise.  There is also need for a computer inside the DAC which these days also probably needs to run some pretty looking mobile device UI app for remote control.  I'm not saying it's necessarily better or worse, just not automatically a free lunch.
 
http://www.akm.com/akm/en/product/featured/velvetsound/
...

 

There is of course no free lunch. But network is, at least in theory, much better than USB ... pretty much the best we have today for data transmission. And no you dont need a computer inside the DAC, just a simple network endpoint. Posted a msg about that stuff a few weeks ago, you can look it up a few pages back. Some details also here http://nadac.merging.com/networking

The pro world is already moving to network but I did not expect such a consumer device so soon. Quite encouraging.

 
So as I understand it, the Nadac is used in the same manner as a usb dac; the music is played on a computer, but then the music stream is sent via ethernet, rather than usb.
 
There is something that indicates that the nadac is halfway to being in control of the playback; they say that it (rather than the source machine) dictates the data transmission rate.
It's a little confusing. Sure, ethernet is potentially waaay better than usb; it can include the timing info, for starters. (Displayport would be even better.) But they seem to be advertising that this device can be used a long way away from the controlling computer; so why is it still necessary to have the computer in control of the playback? Surely, the best solution would be for it to work as a stand-alone music player, playing music files that are located in the computer or server. The data transmission technology would be virtually irrelevant then*.
 
(*The most important thing is: Where does the data become a stream? This is usually at the device that's actually playing the file. It matters because once it's a data stream, the timing becomes important, and jitter becomes a problem. The more tenuous the connection, between player and dac, the bigger the problems! Before it's 'played', data is just data; as long as the rate is high enough, the timing is unimportant/non-existant.)
 
This strikes me as something that seems like a good idea (for a recording studio), but the solution is unnecessarily flawed. Is it just another case of a company selling what people think they want or need, because that's what they can see advertised?

 

[murrays]

   
 
So as I understand it, the Nadac is used in the same manner as a usb dac; the music is played on a computer, but then the music stream is sent via ethernet, rather than usb.
 
There is something that indicates that the nadac is halfway to being in control of the playback; they say that it (rather than the source machine) dictates the data transmission rate.
It's a little confusing. Sure, ethernet is potentially waaay better than usb; it can include the timing info, for starters. (Displayport would be even better.) But they seem to be advertising that this device can be used a long way away from the controlling computer; so why is it still necessary to have the computer in control of the playback? Surely, the best solution would be for it to work as a stand-alone music player, playing music files that are located in the computer or server. The data transmission technology would be virtually irrelevant then*.
 
(*The most important thing is: Where does the data become a stream? This is usually at the device that's actually playing the file. It matters because once it's a data stream, the timing becomes important, and jitter becomes a problem. The more tenuous the connection, between player and dac, the bigger the problems! Before it's 'played', data is just data; as long as the rate is high enough, the timing is unimportant/non-existant.)
 
This strikes me as something that seems like a good idea (for a recording studio), but the solution is unnecessarily flawed. Is it just another case of a company selling what people think they want or need, because that's what they can see advertised?

 
Ethernet is inherently asynchronous.  Data is sent in packets (e.g. TCP/IP).  As the receiver consumes them it sends acknowledgements which signal to the transmitter that it can send more data.  The receiver will maintain a buffer of received data from which it will send audio samples (16 or 24 bit) to the digital filter and DAC chips at its own stable clock rate.  There is no need for the data stream (over ethernet) to carry accurate audio timing information.