Digital that sounds like analog

DonH50

Member Sponsor & WBF Technical Expert
Jun 22, 2010
3,947
306
1,670
Monument, CO
Quick points:

  1. Butterworth filters have lower roll-off than elliptical for the same order, and group delay is not constant (phase issues).
  2. Most op-amps do not have the bandwidth (small or large signal) to deal with the fast DAC output glitches (which are almost impulses) and do not behave well when slew-limited. Some behave quite badly, in fact.
  3. I am confident 0.1 dB ripple is inaudible and buried by room and speaker variations. Still, somebody will probably say they can hear it...
 

JackD201

WBF Founding Member
Apr 20, 2010
12,308
1,425
1,820
Manila, Philippines
Could someone please explain the "glitches" discussed in layman's terms :)

Being a non-engineer it's starting to read like the chips are somehow making the analog output stages clip in some way.
 

JackD201

WBF Founding Member
Apr 20, 2010
12,308
1,425
1,820
Manila, Philippines
Unfortunately it's still over my head Don. I guess what I'm asking is what does this "mis-timing" end up sounding like. In some cases, does it sound similar to, say, overloading an input stage of an analog device? That kind of distortion?

I ask because this is the kind of "digititis" I'm sensitive to when it comes to most consumer CD/DVD players that's typically found in big box stores and electronic sections aka what the general public uses to listen to CDs. It's a harshness that sounds like the opamps are strained.
 

Kindhornman

New Member
Dec 3, 2012
21
0
0
Los Angeles, California, USA
So is this glitching as postulated in post #21 analogous to Jitter in the timing clock? Are these just two ways of defining the same phenomena? I had never heard the term used this way. A glitch as commonly used is something that is just a relatively small error used in many other instances and I am just trying to understand the term myself.

Richard,
The ferrite beads that you are using, those are actually magnetic beads is that correct? So the ferrite's magnetic field follows the same function as the magnetic field that is create in a wound inductor? How do you tune or select such a device, do they have specific values as would be the case with a specific inductor value where you are creating a resonant circuit or trap? And why a series of beads rather than one, are they spread over a frequency range or just a series creating a higher order attenuation?
 

LL21

Well-Known Member
Dec 26, 2010
14,411
2,509
1,448
Glitch, very hand-waving explanation: Think of DACs as switching small units of voltage, or charge, that is passed to the output. Because switches in the DAC do not turn on and off instantaneously and in perfect synch, there is a small period of time when both are on, or (depending on the design) both are off. Parasitic elements like stray capacitance around the switches and in the switching devices themselves cause some energy to "bleed off" to the output during these times, and the result is a fast "glitch" until the switches are fully off/on and stable. Depending upon the design and signal, these glitches can be very large (1/2-fullscale) and although they are very fast (because the switches change state quickly) it is putting wideband noise into the output at the clock rate of the DAC. Furthermore, because the amount of charge depends upon how many switches are changing state, which is in turn a function of the signal. That means the glitch energy may depend upon the signal, which due to various factors in the DAC adds signal distortion as well as noise to the output.

HTH - Don

Hi Don, following on JackD201's questions about what this sounds like...effectively what i think you are saying is that glitching is stray capacitance 'bleeds into' the output. Because switches are pretty fast, it may not be that much...but it happens. And it can happen more or less often depending on how many switches are opening/closing due to the signal...ie, the glitching volume can be signal dependent, among other things.

If excess capacitance (energy effectively?) is straying into the signal...we are getting varying levels of excess energy in the signal going thru the output. Since the output is an analog wave, is this comparable to have a capacitor problem in an amplifier? (you can see where i am going with this...which is that some of us may be more familiar with what an amp sounds like with capacitor problems of greater magnitude, than a DAC with mini-capacitor problems.

I am NO techie...so i am literally trying to tie together breadcrumbs to make a slice of knowledge...
 

Kindhornman

New Member
Dec 3, 2012
21
0
0
Los Angeles, California, USA
By the way hello to all on this thread and on this site. Richard and I have been talking on the DIY site and he told me about this site also. I am a noob as far as electronic circuit design and am here and there to learn. My area is in both waveguide design and also loudspeaker design. I work much more in the mechanical side of things and so I am trying to broaden what I know and what is possible.

Steven:)
 

DonH50

Member Sponsor & WBF Technical Expert
Jun 22, 2010
3,947
306
1,670
Monument, CO
The glitch is like a noise spike added at every sample (every time the sample changes). It can be (usually is) signal-dependent. Because it is nonlinear and correlated to the signal it can sound like many things, alone or in combination, including a harsher buzzing sound, and can increase distortion and raise the noise floor.

Not related to capacitor or timing problems in the manner suggested in the previous posts, if I followed them in my two-second skim. I wsa talking about stray capacitance around the switches and should have just left that out (too deep). Let's try again: You have two "special" light switches. Turn them on and off together and the light goes on and off just like there was one switch. Both have to be in the same state for the light to be fully on or off. Switch one, then the other, with a little delay, and the light turns on, then turns off briefly, then back on after a brief delay when the switches are "together" again. Or vice versa (off, then on for a brief period when the two switches are out of synch, then off again). Those brief little periods of light or no light when the switches are out of synch represent the glitch out of a DAC. If you throw both switches at exactly the same time, no glitch. Mistime just a little, and a little glitch in the light appears.

Now make the switches dimmers as well, and the light level may vary a bit as you change switch states. The glitch depends on the state of the dimmers and how fast you throw the switches (how close together).

Very handwaving, hope no techie types read this... :)
 

Kindhornman

New Member
Dec 3, 2012
21
0
0
Los Angeles, California, USA
Thank you Don for the dumb'ed down explanation for the term. So how does this glitch material show up in the frequency response in the audio band? Are these little spikes at random in the response or is there something else to look for?
 

NorthStar

Member
Feb 8, 2011
24,305
1,323
435
Vancouver Island, B.C. Canada
By the way hello to all on this thread and on this site. Richard and I have been talking on the DIY site and he told me about this site also. I am a noob as far as electronic circuit design and am here and there to learn. My area is in both waveguide design and also loudspeaker design. I work much more in the mechanical side of things and so I am trying to broaden what I know and what is possible.

Steven:)

---- Hi Steven, and a warm welcome here at WBF. :b

Bob

_________________

* Sorry Richard, I made an exception here.
 

LL21

Well-Known Member
Dec 26, 2010
14,411
2,509
1,448
The glitch is like a noise spike added at every sample (every time the sample changes). It can be (usually is) signal-dependent. Because it is nonlinear and correlated to the signal it can sound like many things, alone or in combination, including a harsher buzzing sound, and can increase distortion and raise the noise floor.

Not related to capacitor or timing problems in the manner suggested in the previous posts, if I followed them in my two-second skim. I wsa talking about stray capacitance around the switches and should have just left that out (too deep). Let's try again: You have two "special" light switches. Turn them on and off together and the light goes on and off just like there was one switch. Both have to be in the same state for the light to be fully on or off. Switch one, then the other, with a little delay, and the light turns on, then turns off briefly, then back on after a brief delay when the switches are "together" again. Or vice versa (off, then on for a brief period when the two switches are out of synch, then off again). Those brief little periods of light or no light when the switches are out of synch represent the glitch out of a DAC. If you throw both switches at exactly the same time, no glitch. Mistime just a little, and a little glitch in the light appears.

Now make the switches dimmers as well, and the light level may vary a bit as you change switch states. The glitch depends on the state of the dimmers and how fast you throw the switches (how close together).

Very handwaving, hope no techie types read this... :)

Thank you. That helps...i guess the key is most of have probably not only heard this distortion/effect (perhaps not realizing what it was)...but we may actually have found it is this element of digital reproduction that we do not like (again without realizing what it is). the only way to know for sure would be to hear a system with it...and the same system without it.

But in any event, the total non-techies are starting to get a glimpse of what is happening...and it does start to form a picture of digital playback.

the primary follow up question is...what % of digital reproduction 'problems' do think these glitches represent? I know, i know...roughly to give us a sense of the scope of this problem. Is this 1 of 10 issues...or is this the single largest issue, say "50%" of the problem?

thanks for any 'rough' guidance...
 

DonH50

Member Sponsor & WBF Technical Expert
Jun 22, 2010
3,947
306
1,670
Monument, CO
Thank you Don for the dumb'ed down explanation for the term. So how does this glitch material show up in the frequency response in the audio band? Are these little spikes at random in the response or is there something else to look for?

They will raise the noise floor, of course. Any distortion terms will show up like any other distortion, adding spurs at multiples (harmonics) of the signal frequencies.

I can show pictures of the impact but probably not for a bit as I am swamped with stuff this week (work every day, gigs every night but tonight, plus gigs and other things non-stop through the weekend).
 

DonH50

Member Sponsor & WBF Technical Expert
Jun 22, 2010
3,947
306
1,670
Monument, CO
the primary follow up question is...what % of digital reproduction 'problems' do think these glitches represent? I know, i know...roughly to give us a sense of the scope of this problem. Is this 1 of 10 issues...or is this the single largest issue, say "50%" of the problem?

thanks for any 'rough' guidance...

No idea, sorry, but I suspect it is very small. Modern DAC architectures greatly reduce such glitches, and the focus on linearity and low distortion has also helped decrese their impact to (relative) insignificance. For high-speed (GHz+) converters, the type I previously worked on, it is still a major issue.
 

opus111

Banned
Feb 10, 2012
1,286
3
0
Hangzhou, China
I guess what I'm asking is what does this "mis-timing" end up sounding like. In some cases, does it sound similar to, say, overloading an input stage of an analog device? That kind of distortion?

Your earlier comment about being like the sound of clipping was pretty accurate. Clipping is when a signal exceeds the voltage ability of the device, these glitches are exceeding the time ability of the device. Too quick, as opposed to too large for the amplifier to respond to. An analogy with a phone preamp might help - the glitches are like ticks and pops and clicks from an LP - a good phono preamp doesn't overload from these, but a poor one does. The difference is that these glitches are orders of magnitude faster than anything that could be reproduced by a cartridge but OTOH their amplitude is very small compared to the wanted signal.

I ask because this is the kind of "digititis" I'm sensitive to when it comes to most consumer CD/DVD players that's typically found in big box stores and electronic sections aka what the general public uses to listen to CDs. It's a harshness that sounds like the opamps are strained.

Yes I think this is right - the opamps being strained are the ones on the DAC chips firstly, then subsequently those in the following active filters.
 

LL21

Well-Known Member
Dec 26, 2010
14,411
2,509
1,448
No idea, sorry, but I suspect it is very small. Modern DAC architectures greatly reduce such glitches, and the focus on linearity and low distortion has also helped decrese their impact to (relative) insignificance. For high-speed (GHz+) converters, the type I previously worked on, it is still a major issue.

thanks...if you do think digital reproduction has a long way to go...where do u think in the design we should focus to make the great leaps and strides in improvement?
 

opus111

Banned
Feb 10, 2012
1,286
3
0
Hangzhou, China
i guess the key is most of have probably not only heard this distortion/effect (perhaps not realizing what it was)...but we may actually have found it is this element of digital reproduction that we do not like (again without realizing what it is). the only way to know for sure would be to hear a system with it...and the same system without it.

I agree, its something I listened to in digital music for years - the noise modulation, the greying out of all the tonal colours. I never realised that's what was happening. But now, wherever I go - like coffee shops or restaurants or DVD shops and they're piping out music, I hear it. Once I recognized it, I realized its totally pervasive, ubiquitous.

But in any event, the total non-techies are starting to get a glimpse of what is happening...and it does start to form a picture of digital playback.

I think this lack of glitchiness is what makes your Zanden sound very attractive to you - it uses one of the least glitchy DACs and follows that with a passive filter to further soak up any remaining ones. That's the way to build DACs which sound great I've realized :p

the primary follow up question is...what % of digital reproduction 'problems' do think these glitches represent? I know, i know...roughly to give us a sense of the scope of this problem. Is this 1 of 10 issues...or is this the single largest issue, say "50%" of the problem?

In terms of all systems, most systems nowadays are D-S and D-S DACs have their own problems, glitching as such isn't one of them. With multibit I'd say its 90% of the SQ issues, just a finger in the air though. Its why TDA1541 beats PCM1704 and why the sound has so much to do with the analog stage right after the DAC.
 

opus111

Banned
Feb 10, 2012
1,286
3
0
Hangzhou, China
No idea, sorry, but I suspect it is very small. Modern DAC architectures greatly reduce such glitches, and the focus on linearity and low distortion has also helped decrese their impact to (relative) insignificance.

I agree that modern DACs have made great strides in improving glitch performance - just from reading ADI's TxDACs datasheets. But modern multibit DACs don't exist in audio - they're all oldish (like at least a decade, maybe over two decades, old). That's one reason I'm really keen to listen to the sound of a modern multibit DAC like those ADI sells into comms applications. The only problem is they often top out at 14bits so a little DSP is required to get the audio band noise down to 16bit levels.
 

Kindhornman

New Member
Dec 3, 2012
21
0
0
Los Angeles, California, USA
Richard,
As an aside do you happen to know what type of DAC was used in the old Theta branded CDP machines? I want to remember that this was a PWM type. Have we come very far since then or as you stated above are well still using the same basic DAC chip sets that were around 20 years ago? I believe that the basis of the Theta was oversampling the output waveform before most understood anything about Jitter in the clock.
 

Kindhornman

New Member
Dec 3, 2012
21
0
0
Los Angeles, California, USA
Richard thank you for the link. I will read it after dinner. They were one of the first to try tackle some of the early sound problems in CD players and talk about jitter and clocking errors if my memory is still clear.
 

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