Latest Thread Images

Thoughts on a bunch of DACs (and why delta-sigma kinda sucks, just to get you to think about stuff)
Mar 8, 2015 at 11:55 PM Post #2,941 of 6,500
bill-p said:
 
If that's the case, you can just grab one of the Geek Out's... and you can also forget about the amp. Bargain, I'd say.
 
No, honestly... 
beerchug.gif

 
I'm using a Geek Out 450 as my DAC. Cheap, very functional, and does a fairly decent job. Far better than some of the more expensive offerings I have heard.
Click to expand...

Thanks for your advice, but I would rather to find a AC powered one instead of a USB powered one.
 
I am using Schiit Fulla now and not really satisfied with it...
 
Mar 9, 2015 at 12:09 AM Post #2,942 of 6,500
sapientiam said:
   
As a part-time (armchair?) designer of DIY multibit DACs myself, which measurements are the ones that matter to you?
 
I agree that the retrieval of ambient detail whilst having a higher overall noise floor is something of a puzzle, but its been something I've perceived too in my DAC experiments. How I look at it is that there's dynamic noise from a DAC which broadly tracks the program content (noise modulation in other words) and because this noise is strongly correlated with the signal, our brain interprets it as signal. Thus it doesn't end up being perceived as ambience when the signals are decoded in our brain.
 
On S-D artefacts they tend to be rather impulsive so relying on an FFT to gauge their level is misleading because the FFT time window is normally measured in seconds nowadays (64k and up FFTs are common) whereas for our ear/brain to register something it can be much shorter duration. I suspect we'll need wavelet analysis to get to grips with what happens in S-D modulators on short time-scales.
Click to expand...


I think this might explain the paradox.  If the noise is perceived as part of the music rather than something foreign.  I guess it also depends what the noise or distortion does to the sound - which I don't think I would be able to understand this from specs without hearing the product.  Were the measurements for none PSS noise on the ESS presentation from RMAF FFT plots also?
 
I'm trying to figure out what the offending delta sigma noise is and what level it is.
 
To be honest it could be at a very low level and still matter.  I can't imagine the noise from my computer hard disk, passing through storage controller, computer, galvanic isolation in my transport, finding it's way to the DAC, would be of a substantial level either, but I could hear that.
 
Mar 9, 2015 at 12:11 AM Post #2,943 of 6,500
The D-S hash stuff, I wondering if it's something out of band that can be correlated to the experience. It would have to be measured pre analog filter with good equipment.
 
Another interesting thing is that I can write a program to simulate R2R error according to a generated INL plot to use as a transfer function. However this cannot be done on D-S DACs because the error is going to change depending upon what data preceded it. In other words, R2R error is predictable for each bitcode assuming stable temperature among other things.
 
I would have no idea how to write an algorithm to degrade a WAV file for D-S DACs. But the point is this error tends to be random in the context of each bitcode. Maybe this contributes to the D-S sound?
 
Mar 9, 2015 at 12:18 AM Post #2,944 of 6,500
drez said:
 
I think this might explain the paradox.  If the noise is perceived as part of the music rather than something foreign. 
Click to expand...

 
RIght - that's more of less where my hypothesis comes from. Noise (tape hiss on an analog original recording which often have best ambience) doesn't get decoded into the soundscape, its kind of a fuzz floating around in front of the speakers with my setup. I reckon the ambience cues we depend on (to create the soundstage depth illusion) are below the tape hiss and these can get masked out by correlated noise.
 
Mar 9, 2015 at 12:28 AM Post #2,945 of 6,500
jogfi2002 said:
  Thanks for your advice, but I would rather to find a AC powered one instead of a USB powered one.
 
I am using Schiit Fulla now and not really satisfied with it...
Click to expand...


AC powered and reasonably priced? Gamma2.
 
I'm moving away from AC-powered DAC because I have to travel, but... if you have a bit of cash to blow, then yeah, admittedly, AC-powered DACs do sound good.
 
Incidentally, I'm also selling my main DAC as well. Though I'm not sure if you'll be interested. It's very specialized, and the difference between this one and every other DAC is very pronounced.
 
Mar 9, 2015 at 12:33 AM Post #2,947 of 6,500
purrin said:
 
Good luck soldering that Wolfson chip. That's when I hand over the board to the Vietnamese technician working at the board plant in Fremont.
Click to expand...

 
Oh ya... it's mostly surface mount. Hahaha.
 
Well, I've got a bit of experience working with surface mount now... essentially, flux, stable soldering tip, and all... so I'm a bit more comfortable, but yeah, it's kinda a nightmare to put together still. Especially for a beginner.
 
But... it sounds great. I don't think up to R2R vintage standards, but it still sounds good for something of that size and price. And it can be reasonably powered from USB, too.
 
I never got the chance to try the Gamma2 with LPS and Wyrd but Gamma2 + Wyrd was a very respectable setup IMO.
 
Mar 9, 2015 at 12:42 AM Post #2,948 of 6,500
purrin said:
  The D-S hash stuff, I wondering if it's something out of band that can be correlated to the experience.
Click to expand...

 
In the case of DSD, Parseval's theorem seems relevant. In practice - since the 1 bit DAC has only two places to go its output energy across the whole band (both in-band and out-of-band) is constant. It follows then at low-levels of in-band the OOB is at a maximum and conversely when something big happens in-band the OOB must diminish to compensate. So there's this inverse correlation between in-band and out-band energy.
 
Mar 9, 2015 at 1:09 AM Post #2,949 of 6,500
purrin said:
  I tire of this conversation. We'll be going round and round in circles. I'm not surprised the totalDAC is more resolving than the craptastic D/S that DACs you have heard. (BTW, you never answered zdfx's original question.)
 
Let's remember the overall theme and conclusion of this thread - more choice (better sound) and value:
 
1) Modern DACs have not really progressed in sound quality in 20 years. One step forward. One or two steps back.
2) To have more choice other than the D/S DACs that currently permeate the market.
3) To bring back multi-bit (R2R or otherwise) DACs that sound great.
4) While understanding that R2R will never be cheap - to have somewhat affordable multi-bit DACs that normal people (not wealthy people like you) can buy.
 
If you want to spend $6500-$10,000 for the privilege of a 13/14-bit (admitted by the guy who made it) multi-bit DAC, by all means, go for it. While I respect totalDAC dude's efforts, it's obvious his target customer base is people where 10,000euros is a drop in the bucket. 10,000euros is too rich for my blood. Seriously, the totalDAC shouldn't even be compared to the Yggdrasil because the price points are so different.
 
It my sincere wish that the Yggdrasil's technology will trickle down to their lower end products, and that in time, other manufacturers will take note. Perhaps Analog Devices or BB/TI might even notice and decide that there indeed is a market for good sounding high performance multi-bit chips that will be easier to work with for audio. 
Click to expand...

 
- There are many points I agree with actually and it's great Schiit is coming up with gear to stir the market a bit.
- One correction though: I really wish 10 grands were a drop in the bucket for my wallet, seriously :). Let just say I don't dissolve my hobby fund into too many directions and I do agree 10k in a DAC is silly...
 
arnaud
 
Mar 9, 2015 at 1:18 AM Post #2,950 of 6,500
bill-p said:
 
AC powered and reasonably priced? Gamma2.
 
I'm moving away from AC-powered DAC because I have to travel, but... if you have a bit of cash to blow, then yeah, admittedly, AC-powered DACs do sound good.
 
Incidentally, I'm also selling my main DAC as well. Though I'm not sure if you'll be interested. It's very specialized, and the difference between this one and every other DAC is very pronounced.
Click to expand...

Yeah I saw your post...It is absolutely out of my budget, haha..
 
Mar 9, 2015 at 2:27 AM Post #2,952 of 6,500
jogfi2002 said:
  Yeah I saw your post...It is absolutely out of my budget, haha..
Click to expand...

 
Yeah... haha. Well, like I said, Geek Out 450 is my choice for a budget option right now. I'm actually using it right now... not because I'm being cheap, but because there really isn't a lot of options for a decent USB DAC in the same size and package even up to $500, or beyond. The fact that it has a headphone amp built in is a nice bonus, though I do have an external amp to use with it either way.
 
AMB Gamma2 is my next choice if I could tolerate the size... but hey, the Geek Out saves me on bag space when I'm traveling, and that's where it shines.
 
I haven't heard the Fulla... admittedly, but I have heard quite a number of DACs under $200. The GO450 is still my best contender there. My friends who have heard the Fulla also agree that the GO450 is a decent one at that budget.
 
So... if you have access to Amazon, and the Geek is within your budget, I'd say... go for it. If you don't like it, they have an amazing return procedure. I'd think... there's no reason not to give it a try.
 
You must log in or register to reply here.

Users who are viewing this thread

Total: 5 (members: 0, guests: 5)
Back
Top