Modern Multi-Bit DAC vs Delta Sigma, specifically AKM's newest flagship, but also others

[mike7877]

To the mod who moved this OP to a massive thread about DAC sound signatures (and everyone else reading because I added more info to the thread in here too): this thread is not about sound signature, it's about the internal operation of recently manufactured multi-bit DAC chips (specifically the new AK4499 Velvet, but also others), which I've come to understand may be imposters compared to others. Imposters? Imposters! :not true multi-bit chips like the ones made in the past, but somehow a hybrid of delta sigma + multi-bit, which, being different, may affect performance. At the end of the thread I did have some words on the sound quality of my L70 Velvet, because to me, it sounded vastly superior to the [too many] other Topping DACs I have. Fun fact: the only reason I currently know the AK4499 Velvet in my E70 Velvet is a multi-bit design, is because I was looking for a possible explanation for my perceived [but maybe imagined..] difference in clarity between it and my DX1, E30 II, E30II lite, and E50 (I do have reasons for all of those except the E50, which I planned to return but missed the window for). ANYWAY! That bit at the end of the thread was very secondary to its primary focus, which, again (to be sure), is not discussing the implementation of the AK4499 Velvet in the Topping L70 Velvet, but the AK4499 Velvet itself: information on its design and basic operation, and maybe information on other, more recently manufactured multi-bit DACs if they're similar. I removed what I had at the end of the OP (regarding sound comparison) so there is no confusion.

Thread:
I've got AKM's 4499EX in my Topping E70 Velvet. I'm under the impression that the 4499EX is a two-chip solution - the 4499EX + comes with a 4191EQ

From AKM: "The AK4191EQ is a new concept Multi-bit stereo Premium Digital Data Converter employing VELVETSOUND™ technology. By using the AK4191EQ to process the digital signal for D/A conversion, we have minimized the effects of digital noise within the analog output, resulting in a perceived improvement of the ratio of signal to noise. The AK4191EQ has a built-in digital filter with multi-bit sigma delta modulator and 256 times oversampling processing. A wide variety of music can playback by inputting 1536kHz PCM data and DSD1024 data."

I understand this to mean the AK4191 is the chip that goes between the digital source (coax, spdif, USB, HDMI etc., etc.) and the AK4499.

What I don't get, though, is why there's a sigma delta modulator in it though... wouldn't that nullify the [alleged] benefit of the AK4499 being multi-bit?

Something else that's confusing is AKM says the 4499 and 4191 are separate to keep the analog filtering and digital processing in separate packages, but then they go ahead and say the 4919EQ processes the digital signal so that there's less noise on the 4499's eventual output... So if the 4191's input is already digital, what's this sigma delta going on for? T

And another thing: the low-pass filter on the analog audio output... you know, the thing that usually brings things down to -60dB to -100dB by 24kHz or so with a 44.1kHz... is that still in the 4499? Or is it in the 4191? So much of what AKM has said appears to be contradictory or just plain impossible!

Also, with the AK4499 Velvet (technically AK4499EX, and when packaged with the 4191EQ, called the AK4499EXEQ):
Top of Page 1 of datasheet for proof:

Where is its resistor bank? Is it in the chip???

I'm pretty new to exploring the intricacies of DACs past their advertised specifications and audible sonic qualities, so sorry for if any of the questions are pretty stupid... (Not an excuse but a reason: I had a desktop USB/toslink/coax DAC I was really happy with for almost 10 years, and it broke (accidentally, by my hand ), and now I'm interested in learning more about DACs. I do think I've been lucky to find my replacement as quickly as I have, but the journey isn't over yet -I'm still trying to learn more

And finally (& maybe most importantly...) : is the AK4499 a good example of a modern multi-bit DAC - basically made the same way but with slightly newer manufacturing technologies -, or is it an imposter - a delta-sigma DAC masquerading as multi-bit (like I said, I haven't been looking long, but I've briefly come across people saying some modern multi-bit DACs are weird hybrids or something... from what I've been able to extrapolate from what's been suggested here, there, and everywhere: a delta-sigma type DAC with unconventional output is run in a way (maybe at a higher frequency?) that it's able to output like a multi-bit DAC through multiple pins, but there are less pins and resistors and the clock is doubled up (well not doubled, but maybe 10x - it's like 5 bits instead of 20 or 24 bits). I don't know, it's messed up what I've seen and I'm tired of guessing so I'm hoping someone knows what's going on!

 

[ppataki]

I don't want to digress but do you know if the E70 Velvet supports DSD Direct?
Thank you

 

[Killingbeans]

What I don't get, though, is why there's a sigma delta modulator in it though... wouldn't that nullify the [alleged] benefit of the AK4499 being multi-bit?

Best of both worlds.

I'm guessing that the final stage of the 4191EQ is a multi-bit quantizer, and the AK4499 converts those multiple bits to analogue.

It's not a "multi-bit DAC" in the sense that the original bitstream is converted directly via an R2R network, but it makes use of the same principle for the inevitable DA conversion that's needed after the quantizer output in a Sigma Delta DAC.

It's like paralleling serial signals to push performance to even more ridiculous hights (PCI Express for instance).

Makes zero sense to me that you would be able to hear any of this though.

 

[mike7877]

Best of both worlds.

I'm guessing that the final stage of the 4191EQ is a multi-bit quantizer, and the AK4499 converts those multiple bits to analogue.

It's not a "multi-bit DAC" in the sense that the original bitstream is converted directly via an R2R network, but it makes use of the same principle for the inevitable DA conversion that's needed after the quantizer output in a Sigma Delta DAC.

It's like paralleling serial signals to push performance to even more ridiculous hights (PCI Express for instance).

Makes zero sense to me that you would be able to hear any of this though.

I see what you're getting at - it's low level optimization. Digital compensation for the known behaviour of the analog stage. Like negative feedback for the output stage of a power amplifier, insofar as the DAC's input (DAC being amplifier in this case) is modified to optimize its output. But as the modification being done is in the digital domain, there ain't no analog negative feedback happening! lol

So getting better performance than you could with just the conversion circuit operating as it does with an input - allowing for better performance from the DAC for its actual analogue characteristics. Obviously this can only be taken so far... but done right, more could definitely be done with less (maybe a good comparison/parallel would be a speaker going from a passive crossover to an active one with DSP. Apparent bass extension and maximum continuous level can both be improved significantly with clever operating parameters

 

[MaxwellsEq]

To the mod who moved this OP to a massive thread about DAC sound signatures (and everyone else reading because I added more info to the thread in here too): this thread is not about sound signature, it's about the internal operation of recently manufactured multi-bit DAC chips (specifically the new AK4499 Velvet, but also others), which I've come to understand may be imposters compared to others. Imposters? Imposters! :not true multi-bit chips like the ones made in the past, but somehow a hybrid of delta sigma + multi-bit, which, being different, may affect performance. At the end of the thread I did have some words on the sound quality of my L70 Velvet, because to me, it sounded vastly superior to the [too many] other Topping DACs I have. Fun fact: the only reason I currently know the AK4499 Velvet in my E70 Velvet is a multi-bit design, is because I was looking for a possible explanation for my perceived [but maybe imagined..] difference in clarity between it and my DX1, E30 II, E30II lite, and E50 (I do have reasons for all of those except the E50, which I planned to return but missed the window for). ANYWAY! That bit at the end of the thread was very secondary to its primary focus, which, again (to be sure), is not discussing the implementation of the AK4499 Velvet in the Topping L70 Velvet, but the AK4499 Velvet itself: information on its design and basic operation, and maybe information on other, more recently manufactured multi-bit DACs if they're similar. I removed what I had at the end of the OP (regarding sound comparison) so there is no confusion.

Thread:
I've got AKM's 4499EX in my Topping E70 Velvet. I'm under the impression that the 4499EX is a two-chip solution - the 4499EX + comes with a 4191EQ

From AKM: "The AK4191EQ is a new concept Multi-bit stereo Premium Digital Data Converter employing VELVETSOUND™ technology. By using the AK4191EQ to process the digital signal for D/A conversion, we have minimized the effects of digital noise within the analog output, resulting in a perceived improvement of the ratio of signal to noise. The AK4191EQ has a built-in digital filter with multi-bit sigma delta modulator and 256 times oversampling processing. A wide variety of music can playback by inputting 1536kHz PCM data and DSD1024 data."

I understand this to mean the AK4191 is the chip that goes between the digital source (coax, spdif, USB, HDMI etc., etc.) and the AK4499.

What I don't get, though, is why there's a sigma delta modulator in it though... wouldn't that nullify the [alleged] benefit of the AK4499 being multi-bit?

Something else that's confusing is AKM says the 4499 and 4191 are separate to keep the analog filtering and digital processing in separate packages, but then they go ahead and say the 4919EQ processes the digital signal so that there's less noise on the 4499's eventual output... So if the 4191's input is already digital, what's this sigma delta going on for? T

And another thing: the low-pass filter on the analog audio output... you know, the thing that usually brings things down to -60dB to -100dB by 24kHz or so with a 44.1kHz... is that still in the 4499? Or is it in the 4191? So much of what AKM has said appears to be contradictory or just plain impossible!

Also, with the AK4499 Velvet (technically AK4499EX, and when packaged with the 4191EQ, called the AK4499EXEQ):
Top of Page 1 of datasheet for proof:
View attachment 352942
Where is its resistor bank? Is it in the chip???

I'm pretty new to exploring the intricacies of DACs past their advertised specifications and audible sonic qualities, so sorry for if any of the questions are pretty stupid... (Not an excuse but a reason: I had a desktop USB/toslink/coax DAC I was really happy with for almost 10 years, and it broke (accidentally, by my hand ), and now I'm interested in learning more about DACs. I do think I've been lucky to find my replacement as quickly as I have, but the journey isn't over yet -I'm still trying to learn more

And finally (& maybe most importantly...) : is the AK4499 a good example of a modern multi-bit DAC - basically made the same way but with slightly newer manufacturing technologies -, or is it an imposter - a delta-sigma DAC masquerading as multi-bit (like I said, I haven't been looking long, but I've briefly come across people saying some modern multi-bit DACs are weird hybrids or something... from what I've been able to extrapolate from what's been suggested here, there, and everywhere: a delta-sigma type DAC with unconventional output is run in a way (maybe at a higher frequency?) that it's able to output like a multi-bit DAC through multiple pins, but there are less pins and resistors and the clock is doubled up (well not doubled, but maybe 10x - it's like 5 bits instead of 20 or 24 bits). I don't know, it's messed up what I've seen and I'm tired of guessing so I'm hoping someone knows what's going on!

If you haven't read it, this thread may be useful starting at this point:

AKM new flagship AK4191 + AK4499EX

Found it thanks! https://shenzhenaudio.com/collections/all/products/topping-e70v-ak4191-ak4499ex-dac-xu316-rca-xlr-output-decoder-with-remote-control-support-dsd512-pcm768khz-decoding-e70-velvet Makes the E70V-A70 stack pretty tempting for people in the market right now, I guess :D Indeed a...

audiosciencereview.com

 

[mike7877]

I don't want to digress but do you know if the E70 Velvet supports DSD Direct?
Thank you

I don't think it does... The datasheet for the 4499EX does not mention DSD once.
The datasheet for the previous 4499, mentions it like 70x

 

[mike7877]

If you haven't read it, this thread may be useful starting at this point:

AKM new flagship AK4191 + AK4499EX

Found it thanks! https://shenzhenaudio.com/collections/all/products/topping-e70v-ak4191-ak4499ex-dac-xu316-rca-xlr-output-decoder-with-remote-control-support-dsd512-pcm768khz-decoding-e70-velvet Makes the E70V-A70 stack pretty tempting for people in the market right now, I guess :D Indeed a...

audiosciencereview.com

I'm checking it out - thanks!

 

[Killingbeans]

I see what you're getting at - it's low level optimization. Digital compensation for the known behaviour of the analog stage. Like negative feedback for the output stage of a power amplifier, insofar as the DAC's input (DAC being amplifier in this case) is modified to optimize its output. But as the modification being done is in the digital domain, there ain't no analog negative feedback happening! lol

The 4191EQ is essentially outputting 7 weighted DSD signals, and the AK4499 converts that to analogue via a 7 bit resistor network.

Those signals are high speed, making the chip layout more critical, but that's dwarfed by the benefits of having to deal with the precision issues of "only" 7 bits.

 

[mike7877]

I don't want to digress but do you know if the E70 Velvet supports DSD Direct?
Thank you

After some reading now, I think it does:

 

[mike7877]

The 4191EQ is essentially outputting 7 weighted DSD signals, and the AK4499 converts that to analogue via a 7 bit resistor network.

Those signals are high speed, making the chip layout more critical, but that's dwarfed by the benefits of having to deal with the precision issues of "only" 7 bits.

Put even more simply: it's a 7 bit multi-bit DAC running at a higher frequency which essentially allows it to resolve more bits. Correct?

What are the precision issues that come up when you have, say, a 20 bit multi-bit DAC? Are they because the more resistors you have, the more exact their tolerance must be?

 

[solderdude]

7 bits at a higher sample rate (so DS) can have a much better resolution than 20bit multibit.
The reason for this is that the LSB and 2SB resistors can never be as accurate as DS LSB at high speed can be.
DS will be a bit noisier.

The AK4191EQ is a new concept Multi-bit stereo Premium Digital Data Converter employing VELVETSOUND™ technology. By using the AK4191EQ to process the digital signal for D/A conversion, we have minimized the effects of digital noise within the analog output, resulting in a perceived improvement of the ratio of signal to noise. The AK4191EQ has a built-in digital filter with multi-bit sigma delta modulator and 256 times oversampling processing. A wide variety of music can playback by inputting 1536kHz PCM data and DSD1024 data."

is all marketing nonsense to justify the higher price one pays for 2 chips instead of just 1 that does the same.

 

[mike7877]

My understanding of a basic 16 bit multi-bit DAC is that its very first analog output is 16 pins connected to 16 resistors. The pins power the resistors depending on the data - the data is sampled 44,100 times per second and is comprised of 16 bits. The 16 bits represent the height of the waveform, and it is plotted 44,100 times per second (in the horizontal). Each resistor is such a value that when they are powered according to the data stream, their output current, when totaled, is representative of the original, analog waveform. This current source is then converted to voltage source, because audio drivers' variable resistance over their operating range requires voltage amplifiers to keep frequency response flat. So the current source from the DAC is converted to voltage source, and then to increase power (dramatically from like 3mW) to the correct level for listening (generally 0.1-5W RMS), 2-4 more stages to increase voltage and/or current are added in various components between the initial current output of the DAC, and the terminals on the back of the speaker.

So how does a 7 bit multi-bit DAC turn into something with >125dB dynamic range in the digital domain?

 

[solderdude]

The same way any 4, 5 or 6 bit DS DAC does.
By using DS modulation and below a certain level not using the more significant bits.

Like Philips found out you can reach 16 bit resolution with a 14bit DAC by oversampling 4x (so increasing the switching range).
In this DAC it is (max) 256x oversampling (off 44.1 bitrate I presume)

When you move the quantization noise above 20kHz and you measure using a filter that cuts off above that range you can get impressive numbers.

Delta-sigma modulation - Wikipedia

en.wikipedia.org

 

[voodooless]

The datasheet for the 4499EX does not mention DSD once.

Because you’re looking at the wrong datasheet. The AK4191 does the modulation bits. It can do direct DSD. Which is rather silly if you have a 7-level DAC on hand… why not actually take advantage?

 

[mike7877]

The reason for this is that the LSB and 2SB resistors can never be as accurate as DS LSB at high speed can be.

Is this because the resistors are in the chip?
Externally, resistors can be made to be extremely accurate.
And can't they use lasers now to trim resistors to within like 50 microns? The resistor can be made very wide and tall and thin on a circuit board, and small amounts from the edges can be vaporized away to slowly reduce resistance. I'm sure 304,187.412 ohms is possible on a piece of PCB no bigger than 5mm x 15mm. The PCB can be a daughter board in the DAC. (spitballing ideas, but don't they do things like this already in different form factors?)

 

[mike7877]

Because you’re looking at the wrong datasheet. The AK4191 does the modulation bits. It can do direct DSD. Which is rather silly if you have a 7-level DAC on hand… why not actually take advantage?

I've got all the datasheets now for the new 4499 and 4191. A shot of them connected is in post 9

 

[solderdude]

On-chip resistors can be laser trimmed so extremely accurate.
External ones need to be measured and selected.
There are also different methods of creating 'resistor ladder' DACs and usually the discrete resistor DACs use methods that moves the problem away from the 0-crossing.
Classic R2R DACs have issues with tolerances of the MSB (which is the culprit) as that one has to be the same as all the other bits combined.

 

[mike7877]

The same way any 4, 5 or 6 bit DS DAC does.
By using DS modulation and below a certain level not using the more significant bits.

Like Philips found out you can reach 16 bit resolution with a 14bit DAC by oversampling 4x (so increasing the switching range).
In this DAC it is (max) 256x oversampling (off 44.1 bitrate I presume)

When you move the quantization noise above 20kHz and you measure using a filter that cuts off above that range you can get impressive numbers.

Delta-sigma modulation - Wikipedia

en.wikipedia.org

So this is a 7 bit multi-bit DAC being used in DS mode?

How many bits are AKMs DS DACs then, like the 4493S?

 

[mike7877]

On-chip resistors can be laser trimmed. External ones need to be selected. There are also different methods of creating 'resistor ladder' DACs.

The L70 (headphone amp twin of the E70/E70 Velvet, which I also have) has one of these internal resistor ladder networks I think. I believe I saw a picture of about 25 resistors for all 200 positions between 0.0dB and -99.5dB

 

[solderdude]

Yep.

A lot of DS DACs usually are 5 to 6 bit. It really is not such a big deal.
The biggest challenge is to get the maximum speed higher so DS can be at a higher switching speed. One runs into practical limits.

There is nothing really 'special' about 7 bits DS.