Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › What exactly makes dac <a> better than <b> if both are based on the same chip? (and related questions)
New Posts  All Forums:Forum Nav:

What exactly makes dac <a> better than <b> if both are based on the same chip? (and related... - Page 3

post #31 of 37
2/17/11 at 6:53pm
Quote:
Originally Posted by 00940 View Post



 

 

If two DACs using the same IC are properly designed with regards to PCB  layout and if their power supplies are at roughly the same level of technical achievement (often a matter of cost), they should indeed sound quite similar... if they have a similar output stage. The analog output stage can be designed along very different lines (with opamps, with no feedback discrete modules, with more or less agressive filtering, etc.) and that could make quite a difference in sound. A difference often exploited by designers to "voice" their products.

 

I think the TwistedPear folk allow for a bit of selection (at least sharp and slow roll-off) of the digital filters, with sharp roll-off being the default.  Audio-GD will probably favor slow roll-off without any ability to tweak.  That in itself, IMO, has a noticeable impact on the sound.

 

That said, I am very curious about ACSS in Audio-GD's amps.  It sounds like it's just a transconductance amplifier, which is intriguing in it's own right.  How the Audio-GD ACSS amp differs from the B22 is anyone's guess other than marketing blurbs.

 

I haven't found any actual schematics on the Audio-GD stuff, though it's very unlikely one will.

Reply
post #32 of 37
2/18/11 at 2:41am

@holland: yep, the ESS chips are indeed quite flexible wrt digital filters.

 

Actually, I had a look at both the Audio-GD and Twisted Pear PCB. What strikes me is that Audio-GD seems to have a marketing driven approach (massive use of visually appealing red wima mkt caps for decoupling of digital IC, a lot of boutique electrolytics, no ceramic caps, discrete regulators but far away from the ICs) while Twisted Pear has more of an engineer's approach (they use fewer electrolytics, no mkt and a lot of ceramic caps, integrated regs but very close to the chips). I would be surprised if all that had no influence on sound.

Reply
post #33 of 37
2/19/11 at 10:03am

you just said what i couldnt, thanks. if you want to see what an RF defense contractor does check ackos which is slightly different again. very few traces with the smd components themselves making up the cct, teflon pcb, most traces in hi speed areas are curved etc, separated ground planes for different purposes converging at the ic. ALL filter settings available via mcu, heavily decoupled with np0 and bulk polymer caps

Reply
post #34 of 37
2/20/11 at 9:22am
Quote:
Originally Posted by sphinxvc View PostMost of the more experienced members seem to be saying that the sum of all these parts in the circuit create a decisively superior dac.  But this one quote below has me stumped (doubly so because 'Pars' quoted and "agreed" with it.)

 

Truth is, most contemporary DACs using identical chips should sound very similar...

In general I would not agree with that quote. From the perspective that many commercial manufacturers may tend to use the same datasheet opamp output stage (NE5532s?), I would agree.

Reply
post #35 of 37
2/20/11 at 9:33am
^^^
As you've found out recently, circuit layout and component placement can have a large impact on the analog output. In addition to component selection(discrete or integrated), I would also add the quality and method of power supply regulation, as well as use of feedback and both digital and analog filtering. No single change may make a huge difference, but many things combined can and will.
Reply
post #36 of 37
2/21/11 at 6:32am
Quote:
Originally Posted by createorg View Post

A zhalou with a transformer output stage. 

In reference to what?
 

Reply
post #37 of 37
2/21/11 at 9:24am

I think he was talking about mine:

 

e37ee013_DSC_0004.JPG



744fd1d1_DSC_0003.JPG

 

I have since changed to 2.4K:600 transformers with a simple RC low-pass filter on the primary to scrub off anything above 20K. The transformers resonate at about 40Khz.

Reply
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › What exactly makes dac <a> better than <b> if both are based on the same chip? (and related questions)