• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

A Thread Dedicated to Cirrus Logic CS43131

OP
Saidera

Saidera

Active Member
Joined
Jul 18, 2020
Messages
297
Likes
188
Location
Adelaide, South Australia, Australia
Thread Starter #21
https://www.cirrus.com/company/medi...udio-quality-to-mobile-and-pro-audio-devices/

https://www.planetanalog.com/digital-audio-evolves-back-to-the-single-bit-element-dac/

This is intriguing.

Cirrus Logic’s CS43130 was widely advertised to investors, so there were some inner details revealed, unlike CS43131, which came quietly.
Cirrus Logic designers improved the 1-bit DAC, they crafted 512 single-cell DACs into this architecture which reduces jitter degradation of good audio sound. The IC also filters out unwanted noise through each of the 512 DACs in sequence. In the earliest days of the compact disc, DACs did not employ oversampling, and thus most of the image energy removal was done with a high order analog circuit.



Multiple level one bit DACs is also a method used in S-Master amps.

But someone said: where's the detail on the CS43130? “512 single-cell DACs” sounds like an interesting approach but there's no clue about how it actually works. I tried finding some information on the web but what I found was that “512 single-cell DACs” is simply a bit of investor bait. Cirrus has a product page but it has no detail either.

Thanks for your interest in the CS43130 hi-fi DAC. It's always good to see discussion taking place! As you noted, the data sheet for this product is by request on our website, and we do that primarily because of competitive reasons. It's a support request form that gets channeled to a sales rep for follow up. Regarding the 512 single-cell DAC elements that form the DAC architecture of the CS43130, they're designed in a sequential method so that each element uses a clean clock source as it's reference and doesn't inherit jitter from the previous element, so that by the end of the 512 element chain, the audio signal is clean. In addition the elements sequentially filter out the unwanted frequencies to stop them affecting the wanted audio frequencies. The hard task is designing the element architecture, plus hooking them up together in such a way as to make this work and at the same time, make sure the actual sound produced from the DAC is as good as possible. If you'd like to follow up with me at I'd be glad to provide additional technical information and see what we can do to answer your questions.


Given this detail, one can begin to understand why the Cirrus way may be equated to S-Master more than ESS.
 
OP
Saidera

Saidera

Active Member
Joined
Jul 18, 2020
Messages
297
Likes
188
Location
Adelaide, South Australia, Australia
Thread Starter #22
I know there are simply too many adapters out there using CS43130/1, but Xiaomi HiFi DAC also stands out, as an almost identical copy of Meizu Hifi DAC. It's TPU and ABS, it's CS46L41+CS43131 as well! (see teardown seethrough below) It records ONLY 16/48. Truly a copy. But how on earth is noise to be isolated?

xiaomi.jpg
 

Katji

Major Contributor
Joined
Sep 26, 2017
Messages
1,490
Likes
945
Location
Mzansi
#23
I feel somewhat disadvantaged by having a Ti PCM5242 DAC now...but the overall package sounds better. - The overall package of DSP + amplifiers + DAC + speakers.
 

ElNino

Senior Member
Joined
Sep 26, 2019
Messages
474
Likes
589
#25
If you'd like to follow up with me at I'd be glad to provide additional technical information and see what we can do to answer your questions.
If you do manage to obtain more technical information, definitely please share it here. The reason I like the CS43131 is that it seems to combine some of the better modulator ideas from the old Wolfson team and a reconstruction filter with a gently assymetric impulse response, which the old Cirrus team occasionally used. I think it's the technical sleeper hit among the high-end DAC chips right now, and underpriced relative to the competition.
 
OP
Saidera

Saidera

Active Member
Joined
Jul 18, 2020
Messages
297
Likes
188
Location
Adelaide, South Australia, Australia
Thread Starter #26
It's obvious to everyone but CS46L41+CS43131 implementations record only in 16/48. We can deduce that CS46L41, the secreted USB Bridge of Cirrus, is aimed at the Android dongle market. Obviously CS43131 does not have any ADC.
 
OP
Saidera

Saidera

Active Member
Joined
Jul 18, 2020
Messages
297
Likes
188
Location
Adelaide, South Australia, Australia
Thread Starter #27
If you do manage to obtain more technical information, definitely please share it here. The reason I like the CS43131 is that it seems to combine some of the better modulator ideas from the old Wolfson team and a reconstruction filter with a gently assymetric impulse response, which the old Cirrus team occasionally used. I think it's the technical sleeper hit among the high-end DAC chips right now, and underpriced relative to the competition.
Thanks! I'm sure Cirrus is happy to share general info and technical information, but I'm not counting on it. You know far more about these modulators and filters than me.
 
OP
Saidera

Saidera

Active Member
Joined
Jul 18, 2020
Messages
297
Likes
188
Location
Adelaide, South Australia, Australia
Thread Starter #28
Cyberdrive Clarity Aura (XMOS+CS4398); I never would have thought that such a product had been overlooked by me. Clearly I was misled by the sheer volume of high end expensive DSD DACs from the big audio brands, and never thought that dongles could achieve on-par performance. Someone did a simple teardown of Clarity Feather DAC

U9eb904d6bbd54ccab409e3e395e12c7cW.jpg
Uf1215d16fc7e4c22b269c48831ffae3fh.jpg


If it can do Direct DSD, its value would be extremely high. As it is however, to own the long-lived CS4398 is inherently meaningful. I do prefer USB drive DACs more than adapter dongles. However, one can never be absolutely sure that these chips are genuine...
 
OP
Saidera

Saidera

Active Member
Joined
Jul 18, 2020
Messages
297
Likes
188
Location
Adelaide, South Australia, Australia
Thread Starter #29
About USB Bridges

It’s common knowledge now, but IVX on diyaudio explained it really well.

He claimed at least that Savitech’s SA9302, which DC03/04 uses, doesn’t support DSD256, only DSD128. The SA9312, which SHDP uses, supports DSD256. Or not, according to IVX. This seemed strange to me anyhow, but anyway, the drivers are commonly seen as ‘BRAVO-HD’ ASIO, and often badly customised. On the other hand, you also have CS46L41, which IVX could not get a datasheet on. it is not made public on CS’ site, and a Chinese guy claimed that it only processes 24 bit internally, which means 32/352 is not possible hence no DSD256 support, hence no need for ASIO drivers either. CS46L41 is used standalone in one DAC, and as USB bridge in Meizu and Xiaomi DACs. One can use Realtek as USB bridge, saving the need for additional components? which Savitech needs in order to function (archaic tech). The USB bridge is possibly just as important as the DAC, for it controls what the DAC gets fed. If it is true that CS46L41 only does 24 bit processing, then the Meizu and Xiaomi DACs should be set to 24/352 or DSD64 to avoid going over the limits and getting cut back. Or is it still better to use 32 bit? CS46L41 is weird.

Back when I was working on implementing a Realtek DSD chip, https://www.diyaudio.com/forums/digital-line-level/351441-2020-choice-usb-audio-bridge.html put me off trying to deal with the blue crab.

On a related note, why does DSD256 ASIO on SHDP have issues on playback in HQPlayer and foobar? Despite ASIO, it seems to be using DoP or else it won't work...clearly SA9312 has issues too, like CS46L41 explained before.
 
Last edited:
OP
Saidera

Saidera

Active Member
Joined
Jul 18, 2020
Messages
297
Likes
188
Location
Adelaide, South Australia, Australia
Thread Starter #30
What really is direct DSD by Cirrus? If harmonic distortion is lower, that would appear to be a benefit over the PCM path. But its level is lower, so the measurements can never be as good as PCM. Is there no way to match the levels, without having to go through the DSD Processor or DSM? If they were to be matched without the DSD Processor or DSM, wouldn’t the DSD outperform PCM?

DMP-Z1 is also a consideration, since it uses AKM, does it implement direct DSD? If not, that is proof that it’s not a good pathway. It is already proof that S-Master in a larger DAC is not the best choice. Previously they also used ESS.

I like to search up products. The Huawei dac of 2018 Huawei CM21 used CS43130 and costed above 600 RMB, close to over 100 AUD, but one reviewer in Singapore did not find its sound impressive. Rather, its design and volume knob are so good and do not seem to have been matched by any subsequent product thus far. It won the iF design award. It took a year to develop. It couldn’t do DSD256 due to the chip, but then again, USB bridge has to support it.

No other dac appears to have used CS43130. When it comes to CS43198, there’s Topping D30Pro, which succeeds the D30 which used CS4398. Although these chips all have the mysterious direct DSD pathway, no advertising has mentioned it. The quad CS43198 in Topping D30Pro is interesting. Of course there are plenty of non-branded boards out there, but brand products are rare for CS43130 and CS43198. Recently A&K used dual CS43198 and some appear to be satisfied with its fairly customised sound yet its output impedance is an outlier in this market.

CS43131 is very widely used, but not so much, as a simple search will not reveal very many brand products. Established products from familiar names include Meizu Hifi, Tempotec SHDP, BHD or E series/Hidizs S8, iBasso DC0x, and maybe Xiaomi Hifi if you exclude that charging port.

If SHDP ends up stuttering or heating up and dying, or suddenly stops getting recognised by Win 10 (just uninstall device etc and maybe it’ll heal) and even A55 can end up not being recognised nor charging, that must be avoided. So these devices do tend to have a weak point at the USB stage. These aren’t the toughest devices. So each DAC I have is only used for a special purpose, to provide a different feeling due to the user experience, a different loudspeaker sound, different format support. The only DAC with a real warranty is the apple dongle. Which is why it’s the only one I connect to a PC 24/7, so that if it becomes defective, I can replace it. Even after 1 year! It's a shame that basically all the audio adapters have to be bought from overseas, with only the apple dongle having a true local warranty.
 
Last edited:
OP
Saidera

Saidera

Active Member
Joined
Jul 18, 2020
Messages
297
Likes
188
Location
Adelaide, South Australia, Australia
Thread Starter #31
Ranking the DAC chip makers

The Sony-centric view (my interpretation) goes like this from lowest to highest grade:

Realtek ALC/Sigmatel-IDT-Sony CXD (VAIO DSD PC)

Couple of dollars, simple to implement; 10 dollar DACs

Cirrus Logic CS (Apple, Meizu, CS43131 DSD)

Close to high range performance at prices like 9 USD, 15, 24, 40 (taobao), 60 AUD – so it’s a good price performance ratio and its sound is natural, similar to S-Master.

ESS Tech ES

Difficult to get under 40 dollars (Sonata HD II is an exception at $30)

Usually requires skilled implementation which drives up costs

Asahi Kasei Microdevices (AKM) AK

Last I saw was DC02, badly implemented and overheats, unsuited to dongles as it develops various issues; usually found in desktop DACs e.g. Topping or DMP-Z1

Sony S-Master CXD

Easily the best commercially produced DAC because the company itself designs it, however, has many shortcomings including DSD playback support and bad objective measurement results on ASR

Is there a need to try out AKM’s DSD Mode which many people like? I already have the DSD of low end devices. Regarding ESS, they do keep their info secretive, and their DSD path is quite complex and possibly inferior compared to AKM.
 
OP
Saidera

Saidera

Active Member
Joined
Jul 18, 2020
Messages
297
Likes
188
Location
Adelaide, South Australia, Australia
Thread Starter #32
OP
Saidera

Saidera

Active Member
Joined
Jul 18, 2020
Messages
297
Likes
188
Location
Adelaide, South Australia, Australia
Thread Starter #33
OP
Saidera

Saidera

Active Member
Joined
Jul 18, 2020
Messages
297
Likes
188
Location
Adelaide, South Australia, Australia
Thread Starter #34
https://www.diyaudio.com/forums/digital-line-level/252266-cs4398-dsd-format.html

cs4398 Non-Decimating Volume Control is excellent but it needs software to control, it's better than variable resistor volume control?? most of the cs4398 design on the market uses only one oversampling ratio for all sampling rate just because of simplicity. It will hurt the audio quality seriously for the common sampling rate 44.1k/48k?? is it possible to use the XMOS for setting up the CS4398?? Does the cs4398 support DSD256??

https://www.diyaudio.com/forums/vendor-s-bazaar/274456-ak4490-usb-dac-dsd-support-5.html

PCM lacks in performance compared to the DSD??

PCM Data path goes through Digital attenuator, Interpolator, Delta sigma modulator and Switched capacitor filter.

DSD normal path goes through Digital attenuator, Delta sigma modulator and Switched capacitor filter.

DSD volume bypass path goes only through Switched capacitor filter.
 
OP
Saidera

Saidera

Active Member
Joined
Jul 18, 2020
Messages
297
Likes
188
Location
Adelaide, South Australia, Australia
Thread Starter #35
As Cirrus’ CS4398/43131 has likely been the most affordable and high-performing DSD chip (excluding Realtek) from 2015-present, it is of most interest to me. Many devices have used CS4398/43131. CS4398 was in Cyberdrive Aura/Feather with its overheating XMOS chip and very affordable price when competitors were much more expensive back in 2015. It was simply not so widely known, unfortunately. CS43131 with CS46L41 bridge achieved a finger-sized DAC with Meizu and subsequent copies. CS43131 with Savitech bridges enabled matchbox sized Sonata HD Pro and subsequent copies including the more pricey DC03/04.

The DSD issue was easily settled when I had tried out the PCM and DSD on CS43131. Volume matched by the chip, switching back and forth in HQPlayer, using similar filters and modulators, you’d be hard pressed to ‘hear’ differences. As it was not blind, I’d try higher volumes and harsher sequences and find no difference. It turns out of course that Cirrus subjects both to similar processing at the DSM and its DAC.

Only with CS4398 or CS43198 direct DSD mode would there be slight differences, and then volume matching would be impossible. One might wonder if direct DSD mode would measure more superior if it could be made louder by the chip. One may find that AKM’s DSD does sound different to its PCM, but that is not affordable. On the whole, there are very few statements suggesting that DSD is better, as it is clear that preferences are divided. For CS43131 purposes, either PCM or DSD would sound the same, except that PCM can achieve better measurements when driven loud. But at lower volumes, one wonders whether there might just be 10-20% less processing done on DSD relative to PCM. I might as well stick to DSD just for that inaudible difference.

There is the claim that a truly well designed recorder simply capturing and storing DSD, and then another playback device simply playing it back would achieve good sound. It seems so deceptively simple. And yet, DSD cannot be played back using a single resistor and capacitor. It appears that plenty of processing is necessary. It is very complicated. It is a way of thinking perhaps. If current DACs are 6 bit, one would like to get rid of as many stages as possible and just store that 6 bit audio for playback. Recorders would also have to record in 6 bits. Taking a step back from this, it would appear that PCM serves consumer needs sufficiently well. As long as it is done well, there is no need to eliminate stages. One should be pragmatic and focus on clear subtle differences one can actually hear.

Overall, due to my superficial chip-based view of audio, DSD and PCM are no different. Although Direct DSD mode sounds alluring, the benefits do not as yet outweigh the disadvantages. Truly good, real non-chip based DSD Recorders are far too expensive and difficult to design and basically all studios take no interest in them, save the few in Japan and especially the KORG, TEAC, SONY DSD trio affiliated firms. SBM Direct may carry over the DSD source qualities into PCM, but they are unlikely to be due to the inherent qualities of DSD as a format. More likely the qualities of the supreme DSD capable recorder with a rubidium clock, the ADA-7000R that weighs dozens of kilos. That was merely the opinion of a misguided Sony engineer and has nothing to do with common users. It may be correct of course. And upconversion to DSD to take advantage of the DSD pathway really depends on the DAC. For Cirrus chip users, either choice is fine.

The preferred outcome would have been to find proof that Cirrus’s DSD is the best audio playback that exists on this planet...
 
OP
Saidera

Saidera

Active Member
Joined
Jul 18, 2020
Messages
297
Likes
188
Location
Adelaide, South Australia, Australia
Thread Starter #36
https://www.aes.org/e-lib/browse.cfm?elib=12392
SESSION Z4: POSTERS: SIGNAL PROCESSING, PART 1 Saturday, October 11 2:00 PM— 3:30 PM
Z4-6: A Multibit Delta-Sigma DAC with Mismatch Shaping in the Feedback Loop—Bruce Duewer, John Melanson, Heling Yi, Steve Green, Cirrus Logic, Inc., Austin, TX, USA

1625222906026.png


Source: Principles of Digital Audio (K. Pohlmann) p. 721

Cirrus' Sigma-delta D/A converter = digital interpolation filter>sigma-delta modulator>switched-capacitor filter.

1. PCM or DSD input via serial port
2. DSD data is volume-adjusted and upsampled by a factor of 2
3. Data is applied to a sixth-order sigma-delta modulator with integrated second-order element mismatch noise shaping + second-order sigma-delta modulator as backup (replaces dynamic element matching (DEM)). DSD modulator also uses a fifth-order Butterworth lowpass filter with a corner frequency of 50 kHz. No decimation filter following the multi-bit conversion of DSD. It balances between quantization error and element mismatch error. It has low quantization noise, low sensitivity to clock jitter, and fewer idle tones compared to many one-bit converters as at 2003.
4. Analog out via 16-element switched-capacitor D/A converter operating at 6 MHz

This is most likely what basically occurs in CS4398. Now CS4313x is presumably a different creature.
 
OP
Saidera

Saidera

Active Member
Joined
Jul 18, 2020
Messages
297
Likes
188
Location
Adelaide, South Australia, Australia
Thread Starter #37
A cursory overview of Cirrus' hundreds of patents failed to yield any easy to comprehend methods of how their DACs work for DSD.

Old media releases: https://www.stereophile.com/news/10557/index.html

Extrapolating from the smart codecs lineup, CS46L41's 48kHz mic limitation is shared by most.
CS46L06 by Apple is hard to pin down, as it shares some from the smart and portable codecs lineup but removes some, and measures better too. CS42L73 shares the 48kHz limit and '1VRMS from a single 1.8V power supply'. I found that codecs are not hifi. CS46L06 by Apple is likely able to measure well by cutting off extraneous functions and keeping it simpler.

No reasons can be found for the 48kHz limit. CS46L06 takes the best parts of many chips it seems.
 
Last edited:
OP
Saidera

Saidera

Active Member
Joined
Jul 18, 2020
Messages
297
Likes
188
Location
Adelaide, South Australia, Australia
Thread Starter #38
OP
Saidera

Saidera

Active Member
Joined
Jul 18, 2020
Messages
297
Likes
188
Location
Adelaide, South Australia, Australia
Thread Starter #39
Finally I can try out Cyberdrive Clarity Aura DAC. $69.99 USD. CERTIFIED HIGH-RESOLUTION. Hi-Res audio is to MP3 Audio as Ultra 4K TVs are to Standard Definition. Just like HD revolutionized video, Hi-Res will revolutionize the world of audio. MINIMALIST FORM FACTOR. Feauturing Audiophile Grade Construction with an XMOS DAC, Cirrus Logic board and PCB production. Your music unleashed! Some of you may be wondering why you need a USB DAC when your computer/phone already has a built in headphone jack. As MKBHD explains in this video, most device manufacturers skimp on DACs in order to reduce manufacturing cost. As a result the default audio output of most computers/phones and tablets is not very good.Our goal is to push the entertainment industry into adopting Hi-Res Audio as the new standard! For too long have we been ok with listening to low-quality audio, that ends now!

Well that really resonates with me, since I absolutely suffered immensely at the hands of 24/48 limited Realtek crab sound. Yet at the time I also had access to an Xperia phone and VGN-FE48 capable of DSD. The 'default audio output' is fine in some products. We're ok with listening to low-quality audio, so long as it measures alright. I've no confidence in how Cyberdrive might measure.

Well I got it for $16 AUD and many in Japan have as well, far earlier than me. They provided a USB Micro > USB Type-A and C adapter. Before installing the Cyberdrive Audio Driver v3.12.0 it showed up as capable of 24/352 as well, but after ASIO was installed it could only do 24/192. DSD DoP was not possible. But DSD256 over ASIO is possible.

I got it for its low price and its CS4398: Cirrus' Sigma-delta D/A converter (digital interpolation filter>sigma-delta modulator>switched-capacitor filter) A chip released when I was still learning to speak.

The crucial volume control testing revealed that Clarity Aura can adjust DSD volume via its two buttons. Unlike TT's SHDP, the increments are very small and it takes a long time to go from low to high volume. Playing both PCM and DSD it took over 30 presses of the + button to hear a large difference. That effectively rules out CS4398 Direct DSD mode being implemented here. Of course one could program via the XMOS perhaps but I'm not all that keen on Direct DSD anyway. I'm quite certain this Clarity Aura XMOS idea surfaced on diyaudio, so it's possible that this product is semi-DIY. All of that heightens its appeal.

1. DSD input via serial port
2. DSD data is volume-adjusted and upsampled by a factor of 2
3. DSD modulator uses a fifth-order Butterworth lowpass filter with a corner frequency of 50 kHz. No decimation filter following the multi-bit conversion of DSD
4. Analog out via 16-element switched-capacitor D/A converter operating at 6 MHz

So one can get a CS4398 implementation costing hundreds or this $16 adapter introduced in 2015. Pre-2017 and the CS4313x, this adapter was the best buy. Even in 2021 it is immensely good value and the ASIO driver works across every software flawlessly. Too many dongles only rely on WASAPI which has loads of issues. Clarity Aura drives this RP-TCM125 fairly well. It's only a bit longer than SHDP. It has the same width as SHDP. Unfortunately it doesn't work with many old USB-microB capable smartphones. Its plastic body is smooth and shiny like those cosmetics containers. SHDP has an aluminium body. There's a blue LED you can see through the plastic. below the + button. The XMOS chip doesn't get as hot as I expected.

Overall, although CS43131 enables finger sized DACs, CS4398 thumbdrive sized DACs are cheaper and serve most users' needs sufficiently.
 
Last edited:
OP
Saidera

Saidera

Active Member
Joined
Jul 18, 2020
Messages
297
Likes
188
Location
Adelaide, South Australia, Australia
Thread Starter #40
https://www.audiosciencereview.com/...ds/hardware-teardown-of-topping-d30-dac.2230/
https://www.audiosciencereview.com/.../topping-d30-dac-measurement-and-review.2016/
As one of the most praised CS4398 implementations, D30 even had a teardown discussion! Apparently CS4398 has a patented 'tri-plate capacitor and quieter rails'.

D30 is a dual CS4398 DAC. Clarity Aura doesn't appear to have a clock other than the JD9 24 MHz one for XMOS 6U6C5. The larger board size and a 'moat' helps separate the digital and analog sides. Clarity Aura is all squashed together. Not sure what amplifier is used in Clarity Aura, but definitely no high-end capacitors. Like most cases, here Clarity Aura's performance is most likely hurt by its headphone amplifier.

D30Pro is a huge difference from D30 though.

So far it seems XMOS and the Thesycon ASIO driver are not reprogrammable things. XMOS is widely known, arguably well documented, and easy to start. However its pricing, power consumption (110mA), with many surrounding parts make it unsuitable for portable DACs. Yet we have Clarity Aura.
 
Last edited:
Top Bottom