Discussion on DAC(Digital to Analog Converters)

dqwong

High Supremacy Member
Joined
Jan 21, 2002
Messages
34,868
Reaction score
7,930
I have been researching alot into DAC(AKM/Crystal Semiconductor/ESS/Ti/Wolfson) and here's some interesting quotes I found from Charles Hansen(Founder and Designer) of Ayre Acoustics

As time went on, the main demand has been for smaller, cheaper DAC chips with lower power consumption. This is due to the iPod craze. There is still a market for high-performance audio DAC chips, but there are only a handful left. All but one (the Burr-Brown PCM1704) use some form of a delta-sigma design that typically has only one to six bits, and relies on oversampling and noise shaping to attain reasonable performance.

In general, the more bits it has, the better the performance will be. However with a ladder DAC, all of the bits beyond 18 or so are called "marketing bits" as there is no audio-grade ladder that can exceed 18 bits of resolution. For example when Burr-Brown replaced the "20-bit" PCM1702 with the "24-bit" PCM1704, not one single specification changed. The only difference was that you could feed it digital words that were 24 bits long.

source:
Q&A with Charles Hansen of Ayre Acoustics | AudioStream



The Analog Devices AD1955 is essentially similar to the Burr-Brown devices. The Cirrus Logic CS4398 has a switched-capacitor analog stage, which in conjunction with an internal op-amp provides a voltage output. I have never been a fan of op-amps. The AKM AK4396 is similar to the Cirrus Logic part except that it has a patented method (6,693,574) of canceling some of the out-of-band noise created by the sigma-delta modulator. Some people, particularly Alex Peychev of the now defunct APL Labs, loved this part for that reason.

None of these parts provides any better low-level resolution than the Burr-Brown parts.

The ESS9018 is all the rage these days among certain people. The DAC section itself performs comparably to the PCM1792. However, the chip has a smorgasbord of other functions built in -- asynchronous sample rate converter, programmable digital filter, 8 channels of DACs (for easy design of a 7.1 channel receiver), and a multi-input S/PDIF receiver. Consequently, this part is expensive -- nearly 4x the price of the PCM1792. Since I don't need those other functions I don't feel that it offers a good value for my needs.

The AK4399 is similar to the AK4396, but is a "32-bit" part. This was probably the first of the "32-bit" parts. The specs are not improved over the 24-bit parts, it is simply a marketing gimmick that produces little, if any, audible improvement.

source:
Oversampling: Who Does It Best? - Page 5
 
Last edited:

kenz

Master Member
Joined
Dec 8, 2003
Messages
2,555
Reaction score
0
Really boils down to individual experience, sound preference, music genre taste, mood and a whole lot of other marketing stuff that are "internalized".

In my own personal experience, it boils down to the tuning and architecture implementation, to deliver the final output. Every thing in-between can a topic of debate or discussion, will amount to nothing if the output doesn't fit the user's taste.
 

dqwong

High Supremacy Member
Joined
Jan 21, 2002
Messages
34,868
Reaction score
7,930
Really boils down to individual experience, sound preference, music genre taste, mood and a whole lot of other marketing stuff that are "internalized".

In my own personal experience, it boils down to the tuning and architecture implementation, to deliver the final output. Every thing in-between can a topic of debate or discussion, will amount to nothing if the output doesn't fit the user's taste.

Essentially it boils down to: One man's meat is another man's poison

I am just wondering how much $$$ must one invest before one is nearing the summit of high fieldity. It's just that at the highest end of audio gear, most of the time you are just paying for the design of the casing(aesthetics) more than anything else.

e.g. The sonic performance between DX100 vs HM901 vs AK240 vs Fiio X5

And between HD800 vs LCD-X vs PS-1000e
 
Last edited:

weng

High Supremacy Member
Joined
Jan 1, 2000
Messages
25,014
Reaction score
1
imo too component level liao. may or may not be meaningful to judge a product quality. just like what kenz said, final output still matters.

eg what centrance mentioned in this page:
CEntrance » Blog Archive » Inside DACport, by Popular Demand


"DAC design is not about the parts, but about the overall circuit design, down to the power supply, the neighboring components, and the individual traces on the PCB."
 

dqwong

High Supremacy Member
Joined
Jan 21, 2002
Messages
34,868
Reaction score
7,930
imo too component level liao. may or may not be meaningful to judge a product quality. just like what kenz said, final output still matters.

eg what centrance mentioned in this page:
CEntrance » Blog Archive » Inside DACport, by Popular Demand


"DAC design is not about the parts, but about the overall circuit design, down to the power supply, the neighboring components, and the individual traces on the PCB."

Well I am thinking interms of Audio GD DACs:

All of the Audio GD DACs have the almost same type of beefy power supplies/circuitry/dsp/usb/production quality and all. Thus it now all boils down to the "DAC chip" level for the sound signature.

http://www.audio-gd.com/Products-EN.htm

They offer 3 different types of DACs inside their current DACs:

1) ESS Sabre 9018
2) Wolfson WM8741
3) Ti/Burr Brown PCM1704

Now pondering whether to bite the bullet for the more expensive R2R dac or stick with cheaper sigma-delta dac.


http://www.head-fi.org/products/audio-gd-master-7/reviews/8087
http://www.head-fi.org/t/693798/thoughts-on-a-bunch-of-dacs-and-why-i-hate-chocolate-ice-cream
 
Last edited:

Seth Lee

Senior Member
Joined
Mar 29, 2013
Messages
881
Reaction score
0
Well I am thinking interms of Audio GD DACs:

All of the Audio GD DACs have the almost same type of beefy power supplies/circuitry/dsp/usb/production quality and all. Thus it now all boils down to the "DAC chip" level for the sound signature.

ºÍ§Ó­µ响

They offer 3 different types of DACs inside their current DACs:

1) ESS Sabre 9018
2) Wolfson WM8741
3) Ti/Burr Brown PCM1704

Now pondering whether to bite the bullet for the more expensive R2R dac or stick with cheaper sigma-delta dac.


“Nothing more, sonically, I could ask for.â€* - Currawong’s Review of Audio-gd Master 7
Thoughts on a bunch of DACs (and why I hate chocolate ice cream)

think the Chord Hugo has elevated the game beyond the implementation of "off the shelves" DAC chips like Wolfson, ESS, Burr Brown, Cirrus, Ti and what have you with their AIO chip.
 

Questor

Senior Member
Joined
Mar 21, 2004
Messages
515
Reaction score
0
I agree the most with kenz.. while the chip has some determinant, it does not dictate what the eventual sound is like. The implementation is more important.

Personally I wouldn't get Audio GD, but that's because of hearing experience rather than specs.

There is no 'summit' of high end. It is a journey and you'll never reach the end.. you just have to choose where you want to drop off. There are DACs costing $10 to $100k.. as you go higher end, there are less and less off-the-shelf chips being used.. dCs, emmlabs, chord, etc all use their own proprietary methods.
 

rickysio

Master Member
Joined
Mar 26, 2009
Messages
3,205
Reaction score
0
Problem with Audio-GD is that it's one man churning out millions of designs in very short times... I don't really think that's a very good style of business to buy into.

At the end of the day, happy can liao. Focus too much on the chip and you miss the jungle for a tree.
 

dqwong

High Supremacy Member
Joined
Jan 21, 2002
Messages
34,868
Reaction score
7,930
Problem with Audio-GD is that it's one man churning out millions of designs in very short times... I don't really think that's a very good style of business to buy into.

At the end of the day, happy can liao. Focus too much on the chip and you miss the jungle for a tree.

Audio GD is like buying from a hand soldered company: :s22:
audio_gd_ref7-1_14.jpg


source:
The Flagship Audio Gd: PCM1704 based Ref 7.1

However on the other side of coin is that Audio GD is a company that is constantly improving their product line, and pretty much value for money(production $ into internal components vs exterior aesthetic).

As much as I am a "newer tech is always better" person, I think audio quality is all about good electrical design and along with good component materials.
 
Last edited:

dqwong

High Supremacy Member
Joined
Jan 21, 2002
Messages
34,868
Reaction score
7,930
think the Chord Hugo has elevated the game beyond the implementation of "off the shelves" DAC chips like Wolfson, ESS, Burr Brown, Cirrus, Ti and what have you with their AIO chip.

No matter how good sounding or how well designed filters on the hugo might be:

The overall build quality of the Hugo leaves me totally unimpressed from a durability stand point:
Chord_Hugo_Inputs.jpg


They make Ibasso look like Swiss Made.(yes thats how poorly made the hugo is, especially when you consider the price it is sold at)

image-1398068981-4.jpg
 
Last edited:

dqwong

High Supremacy Member
Joined
Jan 21, 2002
Messages
34,868
Reaction score
7,930
Ben Zwickel of Mojo Audio

We’ve found that the power supply makes the biggest difference in any component. Even the most modest of chips and circuitry can have exceptional performance with the right power supply. Performance is influenced to a greater extent by the power supply than by any other factor. When it comes to DAC chips, however, this is a bit of a complex topic. To oversimplify, DAC chips can be divided into two major categories, modern single bit closest approximation DACs and vintage multibit R-2R ladder DACs. In my experience, each does different things better, and I have not yet heard a DAC chip that does all things correctly. The modern closest approximation DACs have better extension, better dynamics, and lower background noise, whereas the vintage R-2R ladder DACs have better time, tone, timbre, and emotional content.

A significant part of our recent prototyping has been about searching for a modern DAC chip that would give us the best of both worlds. So far we have not found one that does.

Of course modern DACs play modern formats like DSD and native high-res files. However, I would estimate that over 90 percent of all recorded music available today is only available in 16-bit digital, so personally I prefer a DAC that is optimized for 16-bit as opposed to one that sounds amazing with a handful recordings done in modern high-res formats.

An Interview with Ben Zwickel of Mojo Audio - Page 6 of 6 - Dagogo | A Unique Audiophile Experience
 

t258jgn

High Supremacy Member
Deluxe Member
Joined
Oct 11, 2008
Messages
44,834
Reaction score
516
The overall build quality of the Hugo leaves me totally unimpressed from a durability stand point:

Labour cost very high in UK. Everything use high end components + 50% profits = SG price will end up $3+k.:s22:

Nobody will buy le.
 

dqwong

High Supremacy Member
Joined
Jan 21, 2002
Messages
34,868
Reaction score
7,930
Labour cost very high in UK. Everything use high end components + 50% profits = SG price will end up $3+k.:s22:

Nobody will buy le.

I can understand if they skim cost on the exterior casing of the hugo. However the input jacks look horribly cheap and fragile, like your S$20 usb soundcard you find at challenger. Even my PC motherboard onboard soundcard's jack looks more well made/durable. This is totally unacceptable in a S$2k priced product, even for S$200 product. All they need to do is increase the price by S$100-$200 to use abit better jacks, I also won't mind.

Consider a <S$200 soundcard from Asus has better jacks:
xonardx4-7.jpg
 
Last edited:

weng

High Supremacy Member
Joined
Jan 1, 2000
Messages
25,014
Reaction score
1
Haha not to join in the "fun" but really would like to comment about the apparent exterior built quality of the hugo. To me it looks and feels like those soundtech av splitter standard. Sound is another thing.

A0U0_12989562077149414094PqFgJxgG.jpg
 

t258jgn

High Supremacy Member
Deluxe Member
Joined
Oct 11, 2008
Messages
44,834
Reaction score
516
I can understand if they skim cost on the exterior casing of the hugo. However the input jacks look horribly cheap and fragile, like your S$20 usb soundcard you find at challenger. Even my PC motherboard onboard soundcard's jack looks more well made/durable. This is totally unacceptable in a S$2k priced product, even for S$200 product. All they need to do is increase the price by S$100-$200 to use abit better jacks, I also won't mind.

Consider a <S$200 soundcard from Asus has better jacks:

It's like $490 spent on tokgong components & left $10 spent on IO jacks :s13:
As long as sound is tokgong can liao
 

kaixax555

Honorary Member
Joined
Nov 6, 2007
Messages
126,059
Reaction score
223
Hmmm ultimately how the DAC sounds like and performs boil down to the design of the circuitry and other components

Take a look at iBasso D10 and Headamp Pico...both use the same (or at least same family) of chips yet they don't share similar signatures nor sound quality
 

dqwong

High Supremacy Member
Joined
Jan 21, 2002
Messages
34,868
Reaction score
7,930
Hmmm ultimately how the DAC sounds like and performs boil down to the design of the circuitry and other components

Take a look at iBasso D10 and Headamp Pico...both use the same (or at least same family) of chips yet they don't share similar signatures nor sound quality

The difference between Ibasso and headamp pico is more due to the output capacitors:

picoamp
open_2.jpg


ibasso dzero
D-ZERO-C.jpg


which the situation is even more complex for DACs:

There's so many factors that affect the final sound:

1)Power Supply: Noise Filter/Rejection, Stability, Transient Response Time
2)USB/Input receiver implementation, Jitter Rejection
3)Master clock
4)DAC Chip
5)Brickwall/Analog/Digital Filtering
6)Output Circuitry topology/Opamps/capacitors.

video on same topic:


Audio Jitter is machiam like the hand reaction of the conductor of a symphony. If the violin comes a few picosecond later, it won't sound right to the discerning classical music lover. :D
 
Last edited:

Questor

Senior Member
Joined
Mar 21, 2004
Messages
515
Reaction score
0
Audio Jitter is machiam like the hand reaction of the conductor of a symphony. If the violin comes a few picosecond later, it won't sound right to the discerning classical music lover. :D

I don't think that's what jitter sounds like (you can play with a Audiophilleo usb/spdif converter which mimics jitter), though improving jitter does help with timing. Extending or shorter decay is usually caused by something else, IMO.

If jitter were the biggest issue, chord's qbd76hd would have been closest to perfect with its buffer.

Since you like quotes, this is one from Charles Hansen - "The end result will depend on how well your external DAC rejects jitter. Unfortunately there is no such thing as an S/PDIF DAC that completely rejects jitter, except for the Chord. It has a big buffer and it takes 4 seconds after you hit play before you hear music. "
Computer Audio Asylum

The reality is that it isn't the best still, and even with the buffer, you can improve the sound by improving the transport. I kept it for several years before moving on after upgrading it twice.



Not really sure where this discussion is going. I'll just say it's good to know and understand the theory, but listening is still the best determinant because there are too many unknowns in audio that are discernible at the moment, unfortunately. More expensive or better spec doesn't mean better sounding. And in a complete hifi system, there are probably equally or more important components to look at, such as transport, usb power and pre-amp. IMO.

Remember back in the day when async usb was supposed to solve our computer audio issues, and yet now actually we are finding computer audio is far more sensitive than one could have imagined? :)

We have to remember music is something emotional and personal. It isn't a game to see what is faster/more accurate. Just like how the best ingredients doesn't translate to the most satisfying food.. sometimes a really nice chicken rice beats everything.

I have to question if when you listen to live music, can you really hear all those micro details of the violin? Personally even with unamplified performance for a crowd of 10, seated just steps away, I can't hear as much as from my hifi system... is it important to be able to hear that much to enjoy? Not saying that it's not good to pursue that as a hobby, but that other elements of the music can be more important that we should never forget about.

Often for people in this hobby, I always see comments that they have higher performance after upgrading, but music is not enjoyable for some other reason (which may spur on more upgrading). To me, that's someone who went off the path.. because music should ALWAYS be enjoyable as a core criteria.
 
Last edited:

dqwong

High Supremacy Member
Joined
Jan 21, 2002
Messages
34,868
Reaction score
7,930
There's still alot of mystery behind jitter like why or how it happens.

Athough now we have come to a point where there's an atomic clock for audio purposes: :s22:

10M Rubidium atomic clock | Antelope Audio Blog

Things will keep changing for the better or worse.

We went from Vinyl to Tapes to CDs to SACD to Mp3 to AAC to DSD128/256/512

I agree that audio is like food. Some may like the sound from their television speakers more than the sound that's coming out from a 20k home theater.

Just like any other hobbies like photography or cars, there's no such thing as a perfect camera or a perfect car.

We as humans is always thinking of ways to build/buy a better mousetrap. :)

As I am upgrading my equipment to a higher tier, I am hearing more details in my music that wasn't noticeable before, sometimes it's for the better, sometimes it's for the worse. Too much detail/resolution/accuracy isn't necessary a good thing.

Ignorance is bliss like they say, especially in this hobby where there's no end to "upgrading". :D

keep-calm-and-enjoy-the-music-7.png
 
Important Forum Advisory Note
This forum is moderated by volunteer moderators who will react only to members' feedback on posts. Moderators are not employees or representatives of HWZ. Forum members and moderators are responsible for their own posts.

Please refer to our Community Guidelines and Standards, Terms of Service and Member T&Cs for more information.
Top