A DSD DAC with no chip

Mike Lavigne

Member Sponsor & WBF Founding Member
Apr 25, 2010
12,471
11,367
4,410
Really, I didn't know that. Does it say that somewhere on their website?

from the Playback Designs website;

Playback Designs’ philosophy is very simple, but again, very difficult to achieve. In order to achieve the best possible sonic performance it is vital to understand all the details of the theories behind digital and analog signal processing. Relying on component manufacturers to solve signal processing challenges generally ends up in compromised performance. Playback Designs embarked right from the beginning on a path where it applied its vast technical expertise in analog and digital signal processing to design solutions that give “discrete” a new meaning: where standard chips for digital signal processing were not good enough, Playback Designs uses general gate arrays that can be programmed with their own discrete and proprietary algorithms, where standard analog integrated amplifiers (OpAmp) were not good enough Playback Designs uses discrete analog components to better the overall performance. Playback Designs did not leave any step untouched in the theory of Digital to Analog Conversion nor did it leave it to 3rd party vendors for any solutions.

this is just one paragraph.

for more complete information go to www.playbackdesigns.com , select 'start here' from the top bar, then 'Technology and Design' from the drop-down, then scroll down.
 

Ken Newton

Well-Known Member
Dec 11, 2012
243
2
95
Whatmore:

Think of it this way; A 1-bit encoding system, as is DSD, only requires the A/D and the D/A conversion to represent (quantify) two states. Call those states high and low, or 1 and 0, or positive going and negative going, or rising and falling, but only two states need be recorded, and later reproduced. Therefore, a simple logic gate or two-state (binary) driver could form the D/A conversion circuit. Often, such 1-bit drivers are implemented within an custom ASIC device, or an FPGA device, where many such two-state D/A circuits can be inexpensively implemented in a massively parallel array.

Having only two states available obviously produces a lot of analog amplitude error, known as quantization noise. For a signal resolution equivalent to that of CD, a 1-bit converter can only be correct at 2 levels out of approximately 65,000 possible levels, assuming identical sample rates - sort of reminds me of that old saying about a stopped clock being correct twice a day. The trick with 1-bit DSD, as with multibit (typically, 5 - 6 bits) sigma-delta conversion, is to somehow separate the quantization noise from the desired signal buried underneath. A digital processing technique called (quantization) noise shaping performs this task by greatly oversampling the signal and utilizing the resulting ultrasonic spectrum to relocate the undesired quantization noise there. Now, all we have to do, in essence, is filter away the ultrasonic spectrum, and along with it goes the quantization noise (the analog amplitude error).

It sounds like Lampizator is simply planning to buffer the two state DSD data stream, low pass filter it, and then amplify the resulting music signal, all using tube based circuitry. So, there still is a D/A conversion taking place even in such an minimalist implementation, it's just that it can be as simple as a single silicon logic gate or binary driver. I suppose, Lampizator can claim to have no silicon D/A circuit if they use a tube to buffer the silicon logic gate or binary driver carrying the two-state data stream. However, exactly what Lampizator is planning is obviously just conjecture on my part.
 
Last edited:

Ken Newton

Well-Known Member
Dec 11, 2012
243
2
95
Hi, Richard. Thanks, for inviting me here a while back. I had been lurking until now.

By the way, I don't know if you're aware, but there's been new activity on your "Digital, but not by the numbers" thread at DIYA which you may want to address.
 

opus111

Banned
Feb 10, 2012
1,286
3
0
Hangzhou, China
I saw that Trevor posted what (to me and at least one other) looked like an attempt at a troll, so I'm not going to respond over there. Not that I don't enjoy ridiculing trolls, rather that the moderation favours them so any clear rebuttal would probably get deleted. At least one mod over there takes quite a dim view of trolls getting outed for trolling, being cut from the very same cloth himself :D
 

Whatmore

Well-Known Member
Jun 2, 2011
1,011
2
438
Melbourne, Australia
I saw that Trevor posted what (to me and at least one other) looked like an attempt at a troll, so I'm not going to respond over there. Not that I don't enjoy ridiculing trolls, rather that the moderation favours them so any clear rebuttal would probably get deleted. At least one mod over there takes quite a dim view of trolls getting outed for trolling, being cut from the very same cloth himself :D


I'm very confused. My name is Trevor. Are you saying my post was trolling??

I may be wrong, you may not mean me as I don't know where my name apoears on my profile. Hence my confusion.
 

zenelectro

Well-Known Member
Dec 16, 2012
4
0
306

As per others replies, simple fact is you can shove a DSD stream through an analog low pass filter and voila - music.

I believe the late Allen Wright might have been the first to do this with his SACD player upgrades.

To say the DAC is devoid of SS parts is incorrect. There needs to be -something- to decode
the DSD data stream from USB IP. This USB->DSD converter, in effect, has now become the DAC.

The question then becomes, how well has this been implemented?

Z
 

DonH50

Member Sponsor & WBF Technical Expert
Jun 22, 2010
3,947
306
1,670
Monument, CO
A single-bit DAC output still has plenty of potential for error, from duty-cycle distortion and ISI to jitter, to power supply noise, to signal-dependent rise/fall times (also part of DCD/ISI), etc. etc. etc. It is not a panacea.
 

opus111

Banned
Feb 10, 2012
1,286
3
0
Hangzhou, China
The susceptibility to jitter is particularly acute with a single bit DAC of this form. Consider that with a symbol rate of 2.8MHz, digital silence is encoded with a 50% duty cycle waveform with ~700nS on time and ~700nS off time. Any jitter will impact the duty cycle - 700pS peak jitter for example will induce an error at a level of -60dB. If we want 120dB dynamic range from such a DAC then the peak jitter would need to be kept below 700fs.
 

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. This is THE place where audiophiles and audio companies discuss vintage, contemporary and new audio products, music servers, music streamers, computer audio, digital-to-analog converters, turntables, phono stages, cartridges, reel-to-reel tape machines, speakers, headphones and tube and solid-state amplification. Founded in 2010 What’s Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals, we enjoy learning about new things and meeting new people, and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing