Page 1 of 5

DACs -- a technical discussion

Posted: Sun Jun 22, 2014 3:18 pm
by jesuscheung
tony wrote:....
Personally I seem to have gone a different route lately to many here. Valves everywhere and big box Dac from matchbox.
At a recent meet primarily to compare dacs we found interconnects,isolation and dac type all influenced the final sound.
Reality is as I think most of us realize and accept that all things in the chain often affect the final result(Positively/Negatively/ and often not at all).

From hearing various systems over the last few years to get a bit of magic takes effort,investigation, willingness to trial/test and an open mind.
I went to Scalford with a closed mind to CDp's as transports for a dac and came back scratching my head at the people who used this method. Maciej's Meridian transport into a dac opened my mind to what could be achieved with it.

To cut to the chase give us some hi res versions of 3.14avx!
i think there is a misconception of DAC.
people thinks DAC is a CD player.

so manufacturers make DAC to sound better.

if D A C is neutral, it is supposed to sound much worse with people's shitty OS/software.
it is also supposed to sound much more wonderful with good OS/software.

DAC is just like a musical instrument. it is about who plays it. DAC isn't supposed to make music "better"

so i always suspect expensive DACs e.g. chord is doing some trick. i am sure many would be pissed off i say this hehe

Re: MQN

Posted: Sun Jun 22, 2014 3:48 pm
by nige2000
jesuscheung wrote:
DAC is just like a musical instrument. it is about who plays it. DAC isn't supposed to make music "better"

so i always suspect expensive DACs e.g. chord is doing some trick. i am sure many would be pissed off i say this hehe
DAC can only make signal/sound worse
Maybe more accurate to say some are worse than others
Only reclocking can try improve signal

You mentioned buffers the other day
Although they're necessary I think less will be more
In the long term

Re: MQN

Posted: Sun Jun 22, 2014 4:35 pm
by jrling
so i always suspect expensive DACs e.g. chord is doing some trick. i am sure many would be pissed off i say this hehe
DAC can only make signal/sound worse
Maybe more accurate to say some are worse than others
Only reclocking can try improve signal
JC/Nigel you are both right.
There is no trick with Chord DAC64 MKII anyhow. It buffers 4 seconds of incoming SPDIF data, reclocks it, sends it to a FPGA DAC chip programmed to their own formula and sends it out to an Opamp. They engineered it beautifully and have done a great job. More to the point AFAIK, they are the only ones doing this - particularly the FPGA-based DAC. So they are not beholden to or suffering from other manufacturer's poor DAC chip implementation. They are in control.

I think this degrades the signal less than other DACs.

They are expensive and you do pay for the lovely case, but in this instance, I do think you get something for your money rather than another ESS 9018 implementation.
For £1,000 second-hand though, I think my DAC was very good value.

Re: MQN

Posted: Sun Jun 22, 2014 5:02 pm
by tony
jesuscheung wrote:
i think there is a misconception of DAC.
people thinks DAC is a CD player.

so manufacturers make DAC to sound better.

if D A C is neutral, it is supposed to sound much worse with people's shitty OS/software.
it is also supposed to sound much more wonderful with good OS/software.

DAC is just like a musical instrument. it is about who plays it. DAC isn't supposed to make music "better"

so i always suspect expensive DACs e.g. chord is doing some trick. i am sure many would be pissed off i say this hehe
No bother when you say that though you remind me of some characters I come across on another forum! It would piss you off
if you knew who they are!!

Re: MQN

Posted: Sun Jun 22, 2014 5:42 pm
by Aleg
jesuscheung wrote: ...
so i always suspect expensive DACs e.g. chord is doing some trick. i am sure many would be pissed off i say this hehe
I consider those remarks just as spite, because it also shows you haven't a very good understanding what is happening inside a DAC and its various implementations.
This will probably piss you off ....

Re: MQN

Posted: Sun Jun 22, 2014 5:57 pm
by cvrle59
LOL...
Sure, it is some kind of trick JC, but they won't tell you what. They figured out something that nobody spotted so far, so their little portable out performs 3-4 times more expensive systems.
Too me, that is a HOME-RUN TRICK!
A lot if people are actually pissed off these days, Chord's competition the most.

Re: MQN

Posted: Sun Jun 22, 2014 6:48 pm
by DaveF
jrling wrote:
so i always suspect expensive DACs e.g. chord is doing some trick. i am sure many would be pissed off i say this hehe
DAC can only make signal/sound worse
Maybe more accurate to say some are worse than others
Only reclocking can try improve signal
JC/Nigel you are both right.
There is no trick with Chord DAC64 MKII anyhow. It buffers 4 seconds of incoming SPDIF data, reclocks it, sends it to a FPGA DAC chip programmed to their own formula and sends it out to an Opamp. They engineered it beautifully and have done a great job. More to the point AFAIK, they are the only ones doing this - particularly the FPGA-based DAC. So they are not beholden to or suffering from other manufacturer's poor DAC chip implementation. They are in control.

I think this degrades the signal less than other DACs.

They are expensive and you do pay for the lovely case, but in this instance, I do think you get something for your money rather than another ESS 9018 implementation.
For £1,000 second-hand though, I think my DAC was very good value.
I dont know the details of the Chord DAC64 but an FPGA sitting before the DAC is most likely to be doing some type of digital filtering or a choice of filtering as well the possibility of upsampling/interpolation etc. An FPGA is a pretty powerful device capable of doing a lot with audio. Any idea what FPGA it is? (Just to get an idea of the size, logic gates and mem internally)

dCS use Xilinx FPGA's in their DAC's. Well the older Elgar DACs anyway.

Re: MQN

Posted: Sun Jun 22, 2014 7:26 pm
by Aleg
DaveF wrote:
jrling wrote:
so i always suspect expensive DACs e.g. chord is doing some trick. i am sure many would be pissed off i say this hehe
DAC can only make signal/sound worse
Maybe more accurate to say some are worse than others
Only reclocking can try improve signal
JC/Nigel you are both right.
There is no trick with Chord DAC64 MKII anyhow. It buffers 4 seconds of incoming SPDIF data, reclocks it, sends it to a FPGA DAC chip programmed to their own formula and sends it out to an Opamp. They engineered it beautifully and have done a great job. More to the point AFAIK, they are the only ones doing this - particularly the FPGA-based DAC. So they are not beholden to or suffering from other manufacturer's poor DAC chip implementation. They are in control.

I think this degrades the signal less than other DACs.

They are expensive and you do pay for the lovely case, but in this instance, I do think you get something for your money rather than another ESS 9018 implementation.
For £1,000 second-hand though, I think my DAC was very good value.
I dont know the details of the Chord DAC64 but an FPGA sitting before the DAC is most likely to be doing some type of digital filtering or a choice of filtering as well the possibility of upsampling/interpolation etc. An FPGA is a pretty powerful device capable of doing a lot with audio. Any idea what FPGA it is? (Just to get an idea of the size, logic gates and mem internally)

dCS use Xilinx FPGA's in their DAC's. Well the older Elgar DACs anyway.

The Chord Hugo is using a Xilinx Spartan 6 with up to 147k logic cells. I'm not quite sure which model of the Spartan 6 it has. I'll see if I can find.

Re: MQN

Posted: Sun Jun 22, 2014 7:33 pm
by nige2000
DaveF wrote:
jrling wrote:
so i always suspect expensive DACs e.g. chord is doing some trick. i am sure many would be pissed off i say this hehe
DAC can only make signal/sound worse
Maybe more accurate to say some are worse than others
Only reclocking can try improve signal
JC/Nigel you are both right.
There is no trick with Chord DAC64 MKII anyhow. It buffers 4 seconds of incoming SPDIF data, reclocks it, sends it to a FPGA DAC chip programmed to their own formula and sends it out to an Opamp. They engineered it beautifully and have done a great job. More to the point AFAIK, they are the only ones doing this - particularly the FPGA-based DAC. So they are not beholden to or suffering from other manufacturer's poor DAC chip implementation. They are in control.

I think this degrades the signal less than other DACs.

They are expensive and you do pay for the lovely case, but in this instance, I do think you get something for your money rather than another ESS 9018 implementation.
For £1,000 second-hand though, I think my DAC was very good value.
I dont know the details of the Chord DAC64 but an FPGA sitting before the DAC is most likely to be doing some type of digital filtering or a choice of filtering as well the possibility of upsampling/interpolation etc. An FPGA is a pretty powerful device capable of doing a lot with audio. Any idea what FPGA it is? (Just to get an idea of the size, logic gates and mem internally)

dCS use Xilinx FPGA's in their DAC's. Well the older Elgar DACs anyway.
Would be great if some fellow audiophile, highly Knowledgeable in fpga would take tell lead on a project for a fpga source for usb dacs :) ;)

Re: MQN

Posted: Sun Jun 22, 2014 7:59 pm
by DaveF
Interesting that the Hugo uses a Spartan 6 device and I think the dCS Elgar uses one too as I've seen a unit with the lid off once. I know every inch of the internals of a Spartan 6 device as I've worked on it extensively over the past 3 years or so. Many different types of Spartan 6's though in terms of package size, IO's and available logic and memory blocks.

On the subject of an FPGA based project here, I've been asked about this a few times and regrettably I just wouldnt have the time to take on such a task. I could get a few days here and there but then would be pulled off with work for weeks on end.