Is Sabre ES9018 the best DAC chip right now?
Nov 2, 2017 at 7:01 PM Post #151 of 167
Just looked at the AKM differential output DAC's like the AK4490 and above are pretty intriguing.

The best THD+N figures I have seen so far.

@Jimster480 Can you comment on AKM DAC? (Assuming you have had an experience with AKM DAC based devices)
The only AKM DAC device that I have used (and know the AKM chip is inside) is the Fulla 2.
Its said that AKM chips are "brighter" but I would take that with a grain of salt.
I know that AKM is more versed for DSD music and its said that their AK4490 is more comparable to the ES9018/9028 chips.
I haven't done enough side by side comparisons of the two to say that one has more detail than the other because my Fulla 2 is not the same level of product as my SMSL M8 / M8A or Topping DX7 so the comparison would be unfair.
The THD+N does look good but remember that with many of these chips its a bit exaggerated because you won't reach those figures when playing 16/44.1 music, its just the maximum that it can reach provided that you are playing 24b music or high DSD.
 
Nov 7, 2017 at 2:48 AM Post #154 of 167
I was thinking about building a DIY DAC with possibly 2 of their DAC's per channel (probably will try 1 per channel first) and then analog summing it together or comparing the two sets of samples.

If you're seriously considering DIYing a DAC then have a look over at Hackaday.io, I'm in the process of writing up my latest (highly affordable) DAC design for anyone to have a shot at. It uses TDA1387 which is an excellent chip for DIYing because its a) only 8 pins so very simple and b) in an easy to solder package without very tight pin spacing and c) very, very cheap so no worries if you blow one or two up in your build process.

https://hackaday.io/project/27001-audiophile-sounding-dac-for-almost-no-money
 
Nov 7, 2017 at 3:06 AM Post #155 of 167
If you're seriously considering DIYing a DAC then have a look over at Hackaday.io, I'm in the process of writing up my latest (highly affordable) DAC design for anyone to have a shot at. It uses TDA1387 which is an excellent chip for DIYing because its a) only 8 pins so very simple and b) in an easy to solder package without very tight pin spacing and c) very, very cheap so no worries if you blow one or two up in your build process.

https://hackaday.io/project/27001-audiophile-sounding-dac-for-almost-no-money

Thanks man I'll check it out for sure.
I will diy a dac later, but I do want to probably use TI chips in a dual or quad mono format.
 
Nov 22, 2017 at 12:22 PM Post #156 of 167
I'll drop my 5 cents....
I used to have many dacs and will focus on portable implementations.
From all ES9018 machines (HM901, Ibasso, Fiio etc.) I can tell that sabre is a great chip, but really, really MA9 with Alex Mod and black elna caps Im using now is just incomparable. The rest of daps sound like middleware when compared... the richness of tones, bass depth and dynamics... PCM1704 blows, its just much better, sort of a "final solution", end-game, can play anything you throw on it.

Obviously a lot depends on a design and implementation, but if its done right, just go for the texas power...
 
Nov 26, 2017 at 12:45 AM Post #157 of 167
IMO Sabre chips are the worst of all DAC chips - incredibly digital sounding with sharp glare. Generally Delta-Sigma DACs are not very good in my perspective. The best DACs out there would still go to proprietary implementations or R2R/multibit. Once I went to multibit DACs, DS DACs were un-listenable. Of course there are exceptions to Sabre which can sound decent, such as the Auralic Vega, but they could never come close to R2R, even compared to $30k Sabre DACs out there. YMMV
 
Nov 27, 2017 at 8:51 PM Post #158 of 167
IMO Sabre chips are the worst of all DAC chips - incredibly digital sounding with sharp glare. Generally Delta-Sigma DACs are not very good in my perspective. The best DACs out there would still go to proprietary implementations or R2R/multibit. Once I went to multibit DACs, DS DACs were un-listenable. Of course there are exceptions to Sabre which can sound decent, such as the Auralic Vega, but they could never come close to R2R, even compared to $30k Sabre DACs out there. YMMV
I'm not sure what you are talking about.... honestly there is no difference with R2R and R2R actually has measurably worse performance vs DS.
The most complex R2R DAC's still measure worse than most of the cheat DS DAC's...
 
Nov 28, 2017 at 3:55 AM Post #159 of 167
I'm not sure what you are talking about.... honestly there is no difference with R2R and R2R actually has measurably worse performance vs DS.
The most complex R2R DAC's still measure worse than most of the cheat DS DAC's...
Correct. I completely agree on that R2R measures comparably worse than DS. Not everything that is objectively measured can be directly correlated to how much you enjoy a piece of gear, if that were the case, we would have much less variety in products. Measurements are important, don't get me wrong, Tyll from Innerfidelity has a great article elaborating on that.

I just feel that R2R captures natural-ness of presentation much better than DS. They give a 'live' presentation that leans more towards analogue than digital, and I am not the only one that feels so. To each his own, some may like the more digital presentation, but for me, I will never go back to DS.:)
Before this becomes a subjectivist vs objectivist name calling session, let us all just keep an open mind and try out stuff before putting it down.
My best audio experiences came from spontaneous meets.
 
Nov 28, 2017 at 4:11 AM Post #160 of 167
100% agree with boxxi!!!!!!!
 
Apr 17, 2018 at 3:18 PM Post #163 of 167
The make and model of the dac chips used in a dac is largely irrelevant. It’s how a particular chip is implemented that is responsible for sound quality - that is, the design of the digital and analog parts of the circuits and how they work together. Different chips enable different design options but a dac chip alone has no sound character as such.

That's like saying a cartridge and stylus have nothing to do with the sound quality of vinyl playback - it's all in the phono preamp.

The DAC chip is what takes a series of numbers and generates an audio waveform from it.
I promise you the quality the sound is a function of the quality of the component actually rendering the waveform.

It is true that the implementation, including filtering and analog output stage also have a significant effect on the sound.
 
Apr 17, 2018 at 3:26 PM Post #164 of 167
Except jitter isn't measured either in terms of DAC specs.
That is THD Because Jitter = distortion.

No, Jitter creates distortions that are not harmonic in nature.

I suggest a reading on what harmonic distortion is, starting with what harmonics are in relation to fundamentals, even-order and odd-order harmonics, etc.

THD is, in fact, arguably the least offensive type of distortion, as it tends to sound quite musical. In fact, there is a broad trend in professional digital studio recording today to add harmonic distortion to recordings via DAW plug-ins as a sonic enhancement to make in-the-box recordings sound better.

Much worse are IMD and phase distortion, the latter of which is caused by jitter.

THD is only the predominant spec buzzword because it was part of the Federal Trade Commissions standards for rating stereo amplifier power circa the early 70's.
 
Jun 6, 2018 at 4:44 PM Post #165 of 167
Except jitter isn't measured either in terms of DAC specs.
That is THD Because Jitter = distortion.
SNR = noise floor

And its not distortion that the D30 has or even the M8 or M8A or Modi Multibit have... the details are simply different on all of them.
The M8/M8A are basically identical, but the Modi multibit has the least details with a sound that is more like an oldschool stereo/cd player vs modern technology. None of this is shown though in their specs, and this is what I am talking about.
I went on a quest of DACs to see what the ruckus was about and if it was just marketing or measurable in any way.
I tested DAC's with Wolfson (now Crystal), TI, ESS, AKM and Schiits ADI Multibit implementation. Despite comparable on paper specs they all sound different and reproduce music differently with varying levels of detail.
Also part of this is due to the analog stage, but I don't really think detail retrieval has to do much with the analog stage.

Hey, Can you comment on which ones you found to have most details
 

Users who are viewing this thread

Back
Top