Opinions on this DAC

[ak_47_boy]

http://www.dddac.de/ma_dac31.htm

Is this DAC good? I am looking for the highest quality DIY dac. I am looking to build the 60 dac version.

 

[ak_47_boy]

Whoa 132 views and no posts. I know this is a popular dac, is there anything better for the money?

 

[FastEddy]

alternates = possibly a better DAC: http://www.wavelengthaudio.com/usbdac.html

Worth of careful comparisons for sure, for sure ... (I don't appreciate USB connected DACs in any case, being spoiled by FireWire connected DACs = what the pro recording studios use.)

... and this DAC / ADC is as good as any, period: http://www.digitalaudio.dk/ax24.htm ... it is used to make SACD & DVD-A masters ....

 

[DcibeL]

132 views and no posts may be due to the fact that the DDDac has been extensively discussed on this forum previously. Of course you would know that because you searched before posting, right?

Merry Xmas and a happy Festivus for the rest of us!

 

[hifimaker]

Personally, I'm no fan of the TDA-1543 DAC. Comparing this chip to the far superior CS4398 and TI 179x designs, the TDA1543 lacks sound stage and detail and punch. The 1543 is a DIY favorate since it is dirt cheap and easy to design with.

It's tough to take a low end DAC chip and add enough highend parts to it to sounds decent. To compansate for this chips short comings some designers are ganging together multiple 1543's to gether just like this design. I'm seen some designs with as many as 64 of the cheap chips stacked together!

-David

 

[FastEddy]

dw8083: " ... Comparing this chip to the far superior CS4398 and TI 179x designs, the TDA1543 lacks sound stage and detail and punch. ..."

http://www.docethifi.com/TDA1543_.PDF ... dual, two channel 16-bit

http://www.cirrus.com/en/products/pro/detail/P1023.html ... dual, two channel 24-bit

Duhhh ... you don't suppose there could be a connection here?

Come on guys, anyone trying to design a modern Digital to Analog converter should concentrate on 24-bit devices, exclusively ... even when playing 16-bit sources into a 24-bit DAC the results will always be superior ... and what are you going to have, quality wise, when you try to play 24-bit sources through a 16-bit converter (DVD-Audio, SACD and movie sound tracks from DVD's) ... except "lacks sound stage and detail and punch ..." ??

24-bit rules, dude ... its what the pros use for studio masters and anything less than 24-bit conversion just ain't ever gonna make it ... ( http://www.oxsemi.com/products/audio/OXFW970.html ... British made by Canadian engineers: check it out.)

BTW: Happy New Year

 

[ssmith]

http://www.dddac.de/ma_dac31.htm

Is this DAC good? I am looking for the highest quality DIY dac. I am looking to build the 60 dac version.

I have 24 chips + the USB board. I'm very, very happy with it, it has replaced my Benchmark Dac1.
a 60 chip DDDac seems a big step -- how about going for 12 or 24, use upgraded components and expand from there?

Some info on mods are here

Is it the best DIY DAC at the price? No idea - it's the first diy dac I've had.

 

[ssmith]

... even when playing 16-bit sources into a 24-bit DAC the results will always be superior ...

I prefer my 24-chip DDDac to my 24bit Benchmark Dac1. Matches it on detail, soundstage, but beats it on the fatigue stakes.
The DAC1 has better specs and features, though the DDDac, in my not-terribly expensive DIY system, is more enjoyable to listen to. I only like the DAC1 into some kind of tube pre/buffer, which "warms" it up at the cost of detail. I haven't tried modding it.
(As i've mentioned before, I was feeding my Dac1 with spdif (both glass toslink and coax) through a modded M-audio USB Transit board, whereas the DDDac is being fed USB> i2s directy -- which could explain differences).

and what are you going to have, quality wise, when you try to play 24-bit sources through a 16-bit converter (DVD-Audio, SACD and movie sound tracks from DVD's) ... except "lacks sound stage and detail and punch ..." ??

I don't have any 24 bit recordings, though understand a 16 bit DAC is not a very logical choice for 'future proofing'.

 

[FastEddy]

ssmith: Your Benchmark DAC1 = http://www.benchmarkmedia.com/dac1/

... This has fatigue questions? ... Do you have the version with XLR connects? ... connected to an XLR balanced input pre-amp/amp ??

I am amazed that the DDDac would appear to have less fatague than the (24 bit, single chip) Benchmark ... One would have thought that the daisey chain of D to A chips would have a jitter buildup problem (error accumulations) and be power hungry (reducing the gut factor, increasing fatigue) ...

 

[rfbrw]

I am amazed that the DDDac would appear to have less fatague than the (24 bit, single chip) Benchmark ... One would have thought that the daisey chain of D to A chips would have a jitter buildup problem (error accumulations) and be power hungry (reducing the gut factor, increasing fatigue) ... [/B]

If they the dacs were daisy chained, the whole thing wouldn't work.

 

[FastEddy]

rfbrw: " ... If the dacs were daisy chained, the whole thing wouldn't work. ..."

" ... Modular set up with I2S Bus structure. Easy to create multiple DAC-chip construction in multiple of 12 TDA1543 chips ..." From http://www.dddac.de/ma_dac31.htm

This implies a "stacking" of chips or a "daisy chain". I haven't studied the DDDac docs thoroughly, so I'm only guessing about this, but the block diagrams indicate a "master" DAC chip with several (12 or more?) "slave" DAC chips on parallel I2S bus ... (I did not know that the I2S bus was able to handle this parallel topography ... probably should be asking some one like John Swenson.)

 

[rfbrw]

This implies a "stacking" of chips or a "daisy chain".

A daisy chain, to me at least, implies a series or end to end arrangement. A parallel arrangement would not 'buildup jitter'.

I haven't studied the DDDac docs thoroughly, so I'm only guessing about this, but the block diagrams indicate a "master" DAC chip with several (12 or more?) "slave" DAC chips on parallel I2S bus ... (I did not know that the I2S bus was able to handle this parallel topography ... probably should be asking some one like John Swenson.)

Or you could just read the I2S Bus specification.

 

[ssmith]

ssmith: Your Benchmark DAC1 = http://www.benchmarkmedia.com/dac1/

... This has fatigue questions? ... Do you have the version with XLR connects? ... connected to an XLR balanced input pre-amp/amp ??

I am amazed that the DDDac would appear to have less fatigue than the (24 bit, single chip) Benchmark ... One would have thought that the daisy chain of D to A chips would have a jitter buildup problem (error accumulations) and be power hungry (reducing the gut factor, increasing fatigue) ...

I'm using the DAC1 single ended out, direct to power amp (which does not have balanced inputs).

Fatigue... Don't know why, it just has it. But I can guess....
- the DAC1 is more 'accurate', which given the quality of many CDs is not necessarily a good thing. It could be that the DAC1 is producing grating highs which the DDDac is not even capable of resolving???
- the stock DAC1 is let down by using cheap parts (this has been noted by modders) and could easily be smoothed out by some simple mods.
- the DAC1 really shines when fed into a quality preamp/buffer (which I don't have -- I have a DIY unity gain tube buffer but this is a bit detail robbing).
- My comparison between the two has been with the PC as source. It could be that the DAC1 interface I've been using (USB>SPDIF) is inferior to the DDDac (USB>I2S). (Note that regardless of their marketing blurb, the DAC1 is transport sensitive).
- system synergy -- perhaps the components I have work better with one and not the other???
- and so on...

I'm not saying the difference is huge -- certainly not "night and day" - it just occasionally crops up in certain passages. With the DAC1, it's 'ouch' every so often, with the DDDac not very often. With my 2-way speakers this is less apparent. With my frugel horns and their extremely accurate mids and highs, more so. At the same time I don't find the DDDac at all lacking in terms of noise floor or the details that matter. Soundstage width, height, no difference. Soundstage depth -- perhaps a very, very slight advantage to the DAC1. Instrument separation -- blindingly clear with the DAC1 but a bit clinical and cold, and more 'airy' with the DDDac.... which leaves me with fatigue and preferring, on balance and ever so slightly, the DDDac.

 

[DcibeL]

A daisy chain, to me at least, implies a series or end to end arrangement. A parallel arrangement would not 'buildup jitter'.

Daisy chain means parallel...always

 

[Dave]

This looks very promising http://jos.vaneijndhoven.net/dac2/index.html

 

[rfbrw]

Daisy chain means parallel...always

As far as I am concerned a T- off is not a daisy chain but an active loop-through is but what say you we agree to disagree ?

 

[phn]

The by digital fans highly rated Linn CD12 has paralleled DAC chips.

I'm not into digital. But I cannot see what paralleled ICs have to do with jitter. And if you think jitter is a problem, going 16 bits is probably the best way to avoid it.

 

[dddac]

Reading this post, I see what I have seen so many times before ......

One "camp", never tried the dddac, arguing with 24 bits, hypothesis on paralelling etc etc.

Than the other "camp", having built it, experimented with it, modified with better components (so real DIY) and are mostly very happy wit the results and very pleased with the Value for money........

Deja Vu, Deja Vu

 

[rfbrw]

The by digital fans highly rated Linn CD12 has paralleled DAC chips.

Ah, that must be a different Linn, a Scandinavian one perhaps. The CD12 from the Linn in Scotland has 4 dacs in a dual differential configuration.

Reading this post, I see what I have seen so many times before ......

One "camp", never tried the dddac, arguing with 24 bits, hypothesis on paralelling etc etc.

Than the other "camp", having built it, experimented with it, modified with better components (so real DIY) and are mostly very happy wit the results and very pleased with the Value for money........

Deja Vu, Deja Vu

In case any of the above is directed at me, I have tried the TDA1543 and I still say it is a piece of junk.

 

[phn]

Ah, that must be a different Linn, a Scandinavian one perhaps. The CD12 from the Linn in Scotland has 4 dacs in a dual differential configuration.

Could be. Maybe that's why it sucked--jitter, that is.