ANOTHER Example of Why I HATE DSD and Why Customers Who Bought Sony's Boloney Are So Annoying

[Ryan Berry]

On 3/7/2020 at 1:05 PM, Slaphead said:

I mean is DSD really still a thing?

 

Unfortunately, yes.  In a way, it's worse than before as people continue to redefine DSD to give it new life with standards never defined by Sony.

[Ryan Berry]

9 hours ago, Miska said:

 

Because SDM (DSD is a marketing name for one flavor) is so much more modern than PCM and can be native to many actual D/A conversion sections unlike PCM. Making a good native PCM DAC - R2R ladder is very hard and expensive. And to make such perform well it still needs some similar technologies as already utilized for SDM.

 

16x oversampling filters don't get yet anywhere close to rates needed by modern delta-sigma modulators (> 10 MHz), so rest of the oversampling and delta-sigma modulation is up to tiny DSP on the DAC chip. I'm not so big fan of such all-in-1 solutions...

 

Instead I prefer DAC to be just that - digital to analog converter. Converting input bits to analog bit-perfect, as good as possible, without touching the input data. DSP is better performed elsewhere, outside of the box where sensitive analog signals are.

 

P.S. Quick calculation is that for example my EC modulators perform about 6.5 billion calculations per second per channel for "DSD256" output. This is modulator alone, without 256x oversampling filters... So quite a bit of calculations.

 

 

And impedes manufacturers from doing what we know makes a large positive change in the units.  I fail to see the advantages of attempting to drive the DSD standard on listeners.  All it's really doing in the end is bypassing filters, which means people are often comparing apples to oranges and not realizing it. Add in times when things are getting "dressed up" in DSP on the software end, that even tricks people into thinking the DSD is somehow better.

[Ryan Berry]

7 hours ago, Miska said:

It doesn't, instead it leaves you more money to focus on the actual conversion and analog sections and doing things you know. And better allows you to omit cheap COTS DAC chips and go for discrete designs of your own. I much more see that use of COTS DAC chips impedes you from doing many things.

 

In some ways we agree.  Many of the processes you're talking about is not done in the DAC chip in our units.  Allowing the DAC to handle everything is easy to do, and I'm sure it's really tempting for some to slap a chip on a board, feed it power, clock, and signal and call it a day, but we've found that it doesn't sound particularly good.  In other ways we're probably not going to see eye to eye (or perhaps ear to ear) here.  I think one of the tricky things is that it can be difficult to understand what is actually happening inside of a unit without knowing the designer's implementation.  We're fortunate enough to be able to turn off various functions of the DACs we use, disable or enable filtering, listen to various external programs handling the upsampling, etc. before we make decisions.  Ultimately, we picked what sounded the best to us that also gives listeners the most consistent results after very exhaustive testing.  Happily, we can do all that we're doing AND focus on the items you mentioned, so it's a win-win.

I'll agree on one part with R-2R DACs...they're damned expensive to do well.  Charley roughed together an idea that cost well over $1000 in parts alone just to put together.  It's not whether or not it was doable at that point, but a question what would people actually gain in the process vs. what would be gained elsewhere in the design.  Like I said, we don't completely disagree on several points, just feel there are better gains in different approaches.