DSD comparison to PCM.

[LynnOlson]

Hi PeterSt, this is a reply to your post #616.

The latest PFO article, although serious in tone, is partly a joke aimed at the mid-fi PCM DACs out there. From my perspective, these DACs do a pretty bad job with both PCM and DSD. The high frequencies of PCM are rendered as harsh and gritty, while DSD playback is syrupy and rolled-off sounding. However ... PCM converters followed by non-slewing electronics sound quite different; from my perspective, there's little to no "PCM" coloration as audiophiles think of it. DSD at the Playback Designs (and possibly, Invicta) level also no longer sounds syrupy and rolled-off sounding. So the real problem is with mid-fi converters, which do a very poor job of representing the recording.

Referring to the "transients" mentioned in post #616, I may not have been as clear in the PFO article as I could have been. I was not referring to musical transients at all, but the rising and falling edges that emerge from a R2R converter like the PCM-63, as directly measured with a 500 MHz Tek scope or a HP RF spectrum analyzer. The digital input to the converter was not fed musical transients, but a steady full-scale 20 kHz sine wave, and the current output of the PCM-63 was measured with a 50-ohm probe directly attached to the pin of the converter chip, with no I/V stage at all.

That's where I saw the nanosecond rise times; at the edge of each sample, before analog low-passing and waveform reconstruction. When fed into the HP analyzer, I saw a comb spectra extending out to 20 MHz, fading into the noise at the analyzer's noise floor of -100dB at 50 MHz. The PCM-63, with only a 50-ohm load and a short length of coax going into the scope, had very little overshoot, less than 5%, so it didn't appear to need deglitching.

The rise and fall times, though, as seen on the Tek scope, were extremely short, and were not affected by the sample rate of the PCM-63 ... pretty much the same at 44.1, 96, and 192 kHz. This wasn't surprising, since it looked like the rise and fall times were set by the speed of digital logic inside the PCM-63, instead of external clocks.

A little bit of math produced the rise and fall times needed by associated analog electronics, if slewing was to be avoided. It was a big number, in the range of 1000V/uSec, or maybe faster. Since it is very difficult to design analog electronics that have low distortion from 20 Hz through to 20 MHz, my feeling that passive pre-filtering is a very good idea.

Leaving a wideband signal alone is standard practice in traditional analog spectrum analyzers; there's no active preamplifier at all, the signal goes directly from the 50-ohm input, into a wideband attenuator, then right into the first mixer. The old-timers at Tektronix explained to me that a low-distortion wideband amplifier, with a bandwidth of 50 kHz to 1.8GHz, could not be built, so the signal had to go through an all-passive signal path before hitting the first mixer.

A similar problem confronts the I/V converter; the signal has too wide a bandwidth for opamp-style ultralow distortion amplification. There are zero-feedback circuits that can be borrowed from video and the vertical channel of scope preamps; cascodes and apparent long-tailed pairs where the first transistor is really an emitter-follower (collector goes straight to power supply) and the second stage is actually a grounded-base stage (base tied to ground, not feedback) and the output taken from the collector of the second transistor. By comparison, audio-optimized opamps are running out of gain by 1MHz, and barely functioning at 10 MHz. The distortion of these devices is extremely low at 1 kHz and 10 kHz, but that's not true above 1MHz, since the feedback is mostly gone.

I think it's good practice to design analog electronics that avoid gross distortion in normal operation, and if the input signal exceeds the slew rate by a factor of 50 or more, well, that's gross distortion, even though it's for a very brief duration.

The slewing has nothing to do with musical signal; it happens with all input signals, sine waves, triangle waves, square waves, etc. The unfiltered edges (of every sample) coming out of the converter are very fast before filtering is applied, regardless of signal input. This happens with any converter (PCM or DSD) that does not have a built-in opamp. If the converter has a built-in opamp, the slewing happens inside the opamp, where it cannot be seen.

The "joke" part of the article is that it's a crude workaround for DACs that have analog stages that are too slow. By making the too-slow analog electronics slew randomly, the subjective impression might be improved, although this is really a terrible solution. It's kind of like "improving" a car with bad brakes by throwing an anchor out the window, and hoping the anchor works.

 

[PeterSt]

Hi Lynn,

Although you positioned your article as a kind of joke, I don't like to see it like that, because sometimes making things "worse" work out for the better. Your article could be an example of that. And for example, supposed a speaker is not able to deal with more than 30KHz, we better take care that it is not given to it. So, when we take out 10KHz of that, will it be bad ? no, it can only be better. Still it's an unnatural thing to limit a 30KHz capable source to 20KHz.
But audio is full of these things, which often are not even trade-offs, like above example and assumed we don't perceive above 20KHz anyway.

DSD could be approached the same; It is cable of ~80KHz of frequency but in the mean time we run into all sorts of (filtering) problems when we'd really want to sustain that. And whatfore ? because some perceive higher frequencies through their bellies or something ? It only creates a high chance of perceiving distortion or high frequency "noise".

Of course I am fed somewhat by the knowlegde that already 16/44.1 is superb, and that combined with all the flawed hires which by now is on the market. I mean, with the knowlegde of 16/44.1 being able to work out superbly (and I am really serious) why bother to get more (beyond 22.05KHz) right than necessary. But since this is writing and not listening, how to judge this for real merits.
Anyway, a large part of my approach is based on this, and this is how I liked your article. It is no joke at all, for the targeted situations as you just described.


In the mean time I am not sure I understand this :

The rise and fall times, though, as seen on the Tek scope, were extremely short, and were not affected by the sample rate of the PCM-63 ... pretty much the same at 44.1, 96, and 192 kHz. This wasn't surprising, since it looked like the rise and fall times were set by the speed of digital logic inside the PCM-63, instead of external clocks.

So, rise and fall times of what would that be ? Even if I would envision an ever dropping back to 0mA before a next sample forms its output current, I still don't see how it can be INdependent of any clock (determining the speed of the sound, so to speak). Maybe if the currents are formed by some means of PWM ?
(but also see more below)

By comparison, audio-optimized opamps are running out of gain by 1MHz, and barely functioning at 10 MHz. The distortion of these devices is extremely low at 1 kHz and 10 kHz, but that's not true above 1MHz, since the feedback is mostly gone.

Yep, clear. Well, my little secret-thingy does 150MHz of bandwidth with -84dB THD at 1MHz and -30dB at 100MHz. Bandwidth can easily be higher of course but then you run into other sorts of problems.
Btw, by now I should be able to see some things myself through my new analyser (2,5GHz bandwidth), but so far I only spent a couple of hours on examining a 1KHz sine wave only (which is interesting enough already) and I will need quite some toying around before understanding all.

The slewing has nothing to do with musical signal; it happens with all input signals, sine waves, triangle waves, square waves, etc. The unfiltered edges (of every sample) coming out of the converter are very fast before filtering is applied, regardless of signal input. This happens with any converter (PCM or DSD) that does not have a built-in opamp. If the converter has a built-in opamp, the slewing happens inside the opamp, where it cannot be seen.

I don't know how we communicate, but yesterday I wanted to ask about exactly this, but could not find a nice hook to put the question to. And now you just gave me that hook;
Also related to my earlier text in this post, the transients (resulting in possible slewing) you may talk about occur from the sheer stepping (but it will be dependent on clock rate and frequency). So, these are just square litte signals and are as square as square can be, thus imply infinite frequency hence the problem you may be talking about. Now :

Let's say I got myself the most precise R2R chip which exist on the globe. One thing it exhibits is the most precise stepping I ever saw on a scope. My very first idea about it was : uh-oh, this will create proooblems. And exactly those subject to your article/last post. Just from theory, because the more sharp and not overshooting the stepping looks, the more high frequency it will imply (correct me where I go wrong please). This bugger has the opamp inside. And the specs of that ? 50uV/s. Quite similar to the advised opamp for the 1704. You said (last quote above) : "where it cannot be seen". So, what do you mean by that ? head in the sand ? Or won't it be a problem now because it is inside the chip ? (can't imagine the latter).
Again, I will be able to see for myself now (new analyser), but I always like to have my theories right first, and they don't look the best as of yet.

Thanks ...
Peter

 

[Ragnar]

Hi PeterSt, this is a reply to your post #616.

That's where I saw the nanosecond rise times; at the edge of each sample, before analog low-passing and waveform reconstruction. When fed into the HP analyzer, I saw a comb spectra extending out to 20 MHz, fading into the noise at the analyzer's noise floor of -100dB at 50 MHz. The PCM-63, with only a 50-ohm load and a short length of coax going into the scope, had very little overshoot, less than 5%, so it didn't appear to need deglitching.

That 5% overshoot represents a very significant error for 20 bit conversion; however, you didn't mention its time duration. I suspect the B-B dac is better than that by itself, and you were seeing the result of cable reflections or capacitive loading on the dac output.

A little bit of math produced the rise and fall times needed by associated analog electronics, if slewing was to be avoided. It was a big number, in the range of 1000V/uSec, or maybe faster. Since it is very difficult to design analog electronics that have low distortion from 20 Hz through to 20 MHz, my feeling that passive pre-filtering is a very good idea.

Lynn, you say you were feeding the dac a 20KHz full scale signal. Assuming an I/V converter had to output 2 volts RMS corresponding to 2.82x2 or approximately 5 volts p-p, for a full scale 20KHz signal to have <16 bit error, the opamp would have to slew the 5 volts in <500pSec or more like 10,000 V/uSec. Fortunately for us, and the music, we are not likely to encounter anything near full scale 20KHz signals, so with smaller current steps your 1000 V/uSec holds. This would be another reason to use higher sample rate signals, assuming the dac does not exhibit more errors, since the step sizes would be smaller and slewing less of a problem.

Leaving a wideband signal alone is standard practice in traditional analog spectrum analyzers; there's no active preamplifier at all, the signal goes directly from the 50-ohm input, into a wideband attenuator, then right into the first mixer. The old-timers at Tektronix explained to me that a low-distortion wideband amplifier, with a bandwidth of 50 kHz to 1.8GHz, could not be built, so the signal had to go through an all-passive signal path before hitting the first mixer.

In the late 60's I was working for an RF company making products for (primarily) the military. My boss developed and I built and tested, amplifiers with >1GHz bandwidth that had +30dBm HD3 intercepts and a 200pSec risetime for same signal. Of course today with SiGe technology much higher performance is possible.

The "joke" part of the article is that it's a crude workaround for DACs that have analog stages that are too slow. By making the too-slow analog electronics slew randomly, the subjective impression might be improved, although this is really a terrible solution. It's kind of like "improving" a car with bad brakes by throwing an anchor out the window, and hoping the anchor works.

My understanding of the ADI AD1955 dac is that a form of noise shaping roulette is applied to the very output current switches which would add a randomized noise component to the output. Although I am sure other manufacturers do this in one form or another, this may account for its superior sound in the world of sigma-delta converters.

Anchors aweigh!
Cheers

 

[Ragnar]

I'm LOVIN this here thread boys.

"Time is fun when you're having flies."

Kermit the Frog

 

[Ragnar]

Although you positioned your article as a kind of joke, I don't like to see it like that, because sometimes making things "worse" work out for the better. Your article could be an example of that. And for example, supposed a speaker is not able to deal with more than 30KHz, we better take care that it is not given to it. So, when we take out 10KHz of that, will it be bad ? no, it can only be better. Still it's an unnatural thing to limit a 30KHz capable source to 20KHz.
But audio is full of these things, which often are not even trade-offs, like above example and assumed we don't perceive above 20KHz anyway.

I am assured by a respected neuroscientist who's done the research that we (most of us - bats excepted) don't hear frequencies above 20KHz. Experiments show that through bone conduction we hear beat products from frequencies >20KHz.

Of course I am fed somewhat by the knowlegde that already 16/44.1 is superb, and that combined with all the flawed hires which by now is on the market. I mean, with the knowlegde of 16/44.1 being able to work out superbly (and I am really serious) why bother to get more (beyond 22.05KHz) right than necessary. But since this is writing and not listening, how to judge this for real merits.
Peter

I agree entirely. This is where the real efforts should be made, since >99% of digital material is already in this format. (I'm not including MP3!)

 

[LynnOlson]

Well, the source formats for most recordings made in the last twenty-five years is 44.1/20, 88.2/24, 96/24, 176.4/24, and 196/24 PCM. Is it unreasonable to ask to hear them in the original fidelity, as opposed to down-conversion to antique Red Book, or almost equally dated SACD/DSD-narrow? This would seem to be a reasonable business niche-market for FLAC downloads, or the Pono project, if it ever gets off the ground.

Think of the video equivalent: we're stuck with a VHS distribution medium, while TV and movies in the studio are all originated in high-def. Red Book was designed around 1980~82, while SACD/DSD-narrow dates from the late Eighties. Both are very old, and do not reflect modern design practices in converters or signal processing. We all think of the Burr-Brown family of 20/24-bit converters as antiques, but they're newer than DSD-narrow.

Think of the negative impact that Red Book has had over the last thirty years. For one thing, it is the longest time that a recording medium has stayed frozen in one format in the entire history of recording. Acoustic 78's were only in use for 10~15 years; electrical 78's were in use two decades; mono LP's were in use less than 10 years; stereo LP's were in use for 25 years, with vigorous competition from pre-recorded reel-to-reel and cassette tape during that time.

The technological freeze of the non-upgradeable Red Book format has held back the progress of the high-end industry; a reasonable case could be made that much of the cable "tuning" that appeared in the Eighties and later are nothing more than attempts to conceal the faults of the prematurely introduced medium.

 

[LynnOlson]

Lynn, you say you were feeding the dac a 20KHz full scale signal. Assuming an I/V converter had to output 2 volts RMS corresponding to 2.82x2 or approximately 5 volts p-p, for a full scale 20KHz signal to have <16 bit error, the opamp would have to slew the 5 volts in <500pSec or more like 10,000 V/uSec. Fortunately for us, and the music, we are not likely to encounter anything near full scale 20KHz signals, so with smaller current steps your 1000 V/uSec holds. This would be another reason to use higher sample rate signals, assuming the dac does not exhibit more errors, since the step sizes would be smaller and slewing less of a problem.

Cheers

Good analysis, Renee, much appreciated!

My feeling is simple: why accept any slewing at all, even for a few nanoseconds? What is bad about passive filtering? It's standard practice in the RF world; well, this is a RF problem, why not solve it with RF techniques?

Once we accept slewing, then we get into the subjective thicket of what kind of slewing "sounds good". Depending on how it happens in the active circuits, it can sound slow, sludgy, and murky, or gritty, harsh, metallic, or just a subtle veiling of the signal, with no apparent tonal character. But why accept it, since it's not really the sound of PCM or DSD at all?

Why not filter the incoming signal passively (a simple 1st-order lowpass is adequate for the task), send it to the analog electronics of our choice, then have more filtering (if desired) following the I/V converter?

 

[Orb]

Good analysis, Renee, much appreciated!

My feeling is simple: why accept any slewing at all, even for a few nanoseconds? What is bad about passive filtering? It's standard practice in the RF world; well, this is a RF problem, why not solve it with RF techniques?

Once we accept slewing, then we get into the subjective thicket of what kind of slewing "sounds good". Depending on how it happens in the active circuits, it can sound slow, sludgy, and murky, or gritty, harsh, metallic, or just a subtle veiling of the signal, with no apparent tonal character. But why accept it, since it's not really the sound of PCM or DSD at all?

Why not filter the incoming signal passively (a simple 1st-order lowpass is adequate for the task), send it to the analog electronics of our choice, then have more filtering (if desired) following the I/V converter?

But then this omits Delta Sigma DACs, which ideally require roughly a 6th-order filter or better.
Cheers
Orb

 

[Ragnar]

But then this omits Delta Sigma DACs, which ideally require roughly a 6th-order filter or better.
Cheers
Orb

The higher order filters in a sigma-delta dac are used to decimate and noise shape the 16-24 bit incoming PCM to the 5-6 bits at a higher sample rate of the output current generators. In a sense, this eases the burden on the I/V converter by providing smaller step sizes, but adds a cloud of higher frequency noise (as Lynn has advocated.) So the I/V converter is still challenged.

What Lynn is suggesting is that the current be converted to a voltage using a simple resistor, paralleled with a capacitor to act as a low-pass filter so that the edges of the steps are exponential or rounded and demand less stringent slewing requirements in the following circuits. This could be a good idea providing the dac current generators are 'compliant' and can tolerate some voltage change on the output without creating an error in the intended output current. In most dacs the resistor must be very small which means a low voltage leading to a compromise in signal to noise performance.

 

[PeterSt]

Earlier in the thread I talked about my escapades with the passive I/V for the 1704 - how I used it for a year being happy - and how in the end that couldn't go for a production version because of too low output. I told about being able to squeeze out 600mVRMS. And lastly I said that once I reluctandly went the active route, it worked out for the better and found that the high frequencies ate the current with the passive solution.

What I did not tell is that I kind of forbid to use a preamp and/or any means of analog attenuation. It ruins the sound because if irregularities (not linear). But for this part in the thread it is about the preamp ...

The 1704 outputs 1.2mA and even when you stack a few it is clear that with the official 50R resistor you won't get much far.
No preamp = no gain either. So, get yourself a 700W Hypex instead of your nice 1W 300B ?

So there is more to it, once you start believing and perceiving that a preamp and such can only make SQ less, and never better. Uhm, assumed the impedance / current is OK, which of course is the DAC manufacturer's responsibility.

Maybe I only wanted to say : No element in audio can be designed as that element only, and everything must be taken into account. Trade-offs are always there and without exception it is always about choosing the best of the worse. Also, once you have the big picture of all what should work, you are 1000s of hours further before you can see/hear the result, and often there is no going back. This needs creativity I think unseen in other professions as well as and infinite time, patience and money too.
If you are ready to accept this all, what remains are challenges. Well, do I love challenges.

Peter

 

[Orb]

The higher order filters in a sigma-delta dac are used to decimate and noise shape the 16-24 bit incoming PCM to the 5-6 bits at a higher sample rate of the output current generators. In a sense, this eases the burden on the I/V converter by providing smaller step sizes, but adds a cloud of higher frequency noise (as Lynn has advocated.) So the I/V converter is still challenged.

What Lynn is suggesting is that the current be converted to a voltage using a simple resistor, paralleled with a capacitor to act as a low-pass filter so that the edges of the steps are exponential or rounded and demand less stringent slewing requirements in the following circuits. This could be a good idea providing the dac current generators are 'compliant' and can tolerate some voltage change on the output without creating an error in the intended output current. In most dacs the resistor must be very small which means a low voltage leading to a compromise in signal to noise performance.

Ah I see, was late when I posted and thought Lynn was talking about replacing existing function-architecture doh

Slightly off-topic as you mention the interpolation-noise filter (noise shaping loop with that higher frequency noise as you say) has essential/core benefits with regards to SD DACs that also include higher nth filter to overcome DC input deadbands and in-band tones and greater resolution (in context of SD DAC and oversampling).
But maybe I am missing something with regards to slewing when considering modern SD DACs are hybrid "multi-bit" (rather than the original 1-bit), op-amp slew rate/non-linearity applicable to ADC compared to SD DAC so slightly different criteria, and possibly idea of SC filter - anyway I am thinking of this from a multi-bit SD DAC rather than 1-bit.

Ragnar, any idea how I/V conversion is done by some of the big name DACs?
I am wondering if some of the problems relate to I/V being done on the chip and cannot be bypassed (for some anyway) and possibly some audio manufacturers with a scope of cheap implementation, meaning some implement this much better than others.

Thanks
Orb

 

[LL21]

Ragnar, any idea how I/V conversion is done by some of the big name DACs?
I am wondering if some of the problems relate to I/V being done on the chip and cannot be bypassed (for some anyway) and possibly some audio manufacturers with a scope of cheap implementation, meaning some implement this much better than others.

Thanks
Orb

+1...would be interesting in reading more about this..

 

[Ragnar]

+1...would be interesting in reading more about this..

Of the four major manufacturers of high-end integrated dacs, Analog Devices, ESS, Cirrus and TI, their top performers have differential current output, the only way to achieve s/n approaching or exceeding 120dB. The dac numbers are: ADI - AD1853, AD1955; ESS - ES9012, ES9018; Cirrus - CS4392, CS4398; TI (B-B) - PCM1792A, PCM1794A. I am not that familiar with Wolfson, but they also make a product in this category.
All of the above parts have prices >$10 in modest (audiophile product) quantities. Keep in mind that this will add $50-100/stereo output to the final product cost.

Lower cost parts as used in most high volume, competitive products incorporate internal I/V conversion done with minimalist CMOS topologies, the same semiconductor process used for the digital portion of the dac. Some incorporate switched capacitor filters, so external filtering can be minimal or eliminated entirely. These parts all cost <<$2 in high volume.

I will tell you that a high performance opamp topology is not achievable in a CMOS process; the environment is very noisy and CMOS itself is noisy, but the target market wants the features at low cost.

 

[Ragnar]

Of the four major manufacturers of high-end integrated dacs, Analog Devices, ESS, Cirrus and TI, their top performers have differential current output, the only way to achieve s/n approaching or exceeding 120dB. The dac numbers are: ADI - AD1853, AD1955; ESS - ES9012, ES9018; Cirrus - CS4392, CS4398; TI (B-B) - PCM1792A, PCM1794A. I am not that familiar with Wolfson, but they also make a product in this category.
All of the above parts have prices >$10 in modest (audiophile product) quantities. Keep in mind that this will add $50-100/stereo output to the final product cost.

Lower cost parts as used in most high volume, competitive products incorporate internal I/V conversion done with minimalist CMOS topologies, the same semiconductor process used for the digital portion of the dac. Some incorporate switched capacitor filters, so external filtering can be minimal or eliminated entirely. These parts all cost <<$2 in high volume.

I will tell you that a high performance opamp topology is not achievable in a CMOS process; the environment is very noisy and CMOS itself is noisy, but the target market wants the features at low cost.

Left off a few high performers: ESS - ES9008, ES9016; Wolfson - WM8740, WM8741, WM8742; AKM - AK4399.

 

[Orb]

Of the four major manufacturers of high-end integrated dacs, Analog Devices, ESS, Cirrus and TI, their top performers have differential current output, the only way to achieve s/n approaching or exceeding 120dB. The dac numbers are: ADI - AD1853, AD1955; ESS - ES9012, ES9018; Cirrus - CS4392, CS4398; TI (B-B) - PCM1792A, PCM1794A. I am not that familiar with Wolfson, but they also make a product in this category.
All of the above parts have prices >$10 in modest (audiophile product) quantities. Keep in mind that this will add $50-100/stereo output to the final product cost.

Lower cost parts as used in most high volume, competitive products incorporate internal I/V conversion done with minimalist CMOS topologies, the same semiconductor process used for the digital portion of the dac. Some incorporate switched capacitor filters, so external filtering can be minimal or eliminated entirely. These parts all cost <<$2 in high volume.

I will tell you that a high performance opamp topology is not achievable in a CMOS process; the environment is very noisy and CMOS itself is noisy, but the target market wants the features at low cost.

Thanks Ragnar.
Regarding switch capacitors, yeah why I sort of indirectly mentioned SC filter previously in my post regarding why concern of slew rate as this helps to mitigate much but appreciate not all use this..

Regarding my question.
Sorry what I meant by audio manufacturers were also those using said DAC chipset/architecture, such as Naim who implement their own I/V conversion (differential input stage-voltage gain stage-buffer/output stage, using 16x oversampling combined with 6th order analogue filter with zero feedback.
Their solution is pretty neat, and I would hope the more respectable high end digital audio manufacturers would also implement a good I/V solution.

Cheers
Orb

 

[opus111]

Of the four major manufacturers of high-end integrated dacs, Analog Devices, ESS, Cirrus and TI, their top performers have differential current output, the only way to achieve s/n approaching or exceeding 120dB.

Only TI, ESS and ADI have parts with current out, the others (Cirrus, Wolfson and AKM) are using switched capacitor output filtering across their whole range, necessarily voltage domain.

Many moons ago, Cirrus had a part without the on-chip filtering, CS4303. It had output RTZ coding of a 1bit DAC.

 

[Ragnar]

Only TI, ESS and ADI have parts with current out, the others (Cirrus, Wolfson and AKM) are using switched capacitor output filtering across their whole range, necessarily voltage domain.

Many moons ago, Cirrus had a part without the on-chip filtering, CS4303. It had output RTZ coding of a 1bit DAC.

Thanks Opus, for the correction. I hadn't looked at their parts recently; somehow I'd remembered them making current output parts.

So that thins the pack.

 

[Orb]

Only TI, ESS and ADI have parts with current out, the others (Cirrus, Wolfson and AKM) are using switched capacitor output filtering across their whole range, necessarily voltage domain.

Many moons ago, Cirrus had a part without the on-chip filtering, CS4303. It had output RTZ coding of a 1bit DAC.

Opus,
you think SC filter mitigates some of the issues regarding slew rates; such as linearity issues based upon finite slew rate, CT active filter opamps requiring very high slew-rate for avoiding signal related slewing?
Appreciate your not really a SD DAC multi-bit fan though
More seriously I can appreciate why as it does seem they all have a similar perceivable behaviour that is hard to describe (for me anyway).

Cheers
Orb

 

[Orb]

Opus,
wonder if you also know why Burr-Brown moved away from the sign-magnitude design (PCM1704), which they felt was better than traditional delta sigma DACs.
I thought if they were that confident in the architecture compared to SD they would continue developing it, unless comes down to Texas Instruments.

MSB Tech still using sign magnitude technique?
Cheers
Orb

 

[opus111]

you think SC filter mitigates some of the issues regarding slew rates; such as linearity issues based upon finite slew rate, CT active filter opamps requiring very high slew-rate for avoiding signal related slewing?

Nope, SC is used to reduce jitter sensitivity, but I'm with Ragnar on this - never heard a CMOS opamp I've liked the sound of.

I think BB/TI moved away from multibit because they really believe S-D is superior for audio purposes (after all it boasts better numbers across the board), and its also a lot cheaper to make ( = presumably better margins).