• 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.

Next Level of Performance DACs.

Wes

Major Contributor
Joined
Dec 5, 2019
Messages
1,539
Likes
1,257
#41
If the quest for higher SINAD (or other benchmark) stops, what will be the next development? Its either features or performance that will incentivise someone to buy a new DAC. Multichannel may well be a next big thing but I suspect its a way off being a driver for mass sales .We have done the "balanced" thing. Suspect MQA is stagnating

I suspect the D90 V2 will be a spec bump, but no idea what the D100 will have that the D90 doesnt.
DSP is next I think
 

Jimbob54

Major Contributor
Forum Donor
Joined
Oct 25, 2019
Messages
3,602
Likes
3,067
#42
DSP is next I think
I can see that being the case. And nevermind the stuff RME does, maybe more the app driven more base user friendly type implementation that @IVX @e1da did on the powerdac v2.
 
Last edited:

KSTR

Addicted to Fun and Learning
Joined
Sep 6, 2018
Messages
530
Likes
1,226
Location
Berlin, Germany
#43
Hi Uli
Hi Klaus,
unfortunately the ADI-2 has only two channels :(
ADI-2 Pro has 4 channels, and you could daisy-chain the digital out to an ADI-2 DAC for 6 total, or use a second ADI-2 Pro for 8 total. You could even control the volume of two ADI-2 Pro FSR units with just one remote control, all 8 channels in sync.
Indeed there is some progress at RME. The ADI-2 now also seems to follow an external samplerate change correctly, the Fireface did not.
But why is it necessary to add an external USB isolator and another power supply? Couldn't this be done in the ADI-2 ?
It could, actually there is just enough space left in the enclosure to fit both the Intona PCB and the DC/DC. But that would significantly increase the price for only little advantage for most users. Neverless, true galvanic isolation and lowest coupling capacitance is king as my guess is many "digital" sound problems are actually ground contamination and RF demodulation in the DAC or later in the chain.
Indeed my hypothesis is: if a DAC is reacting on its analog output side on cables, clocks, isolators, power supplies and other tweaking on the digital sender side it is not properly designed.
My position as well. When going digital, invest in the best possible DAC(+Accessories), then the source can be just anything. But that is a "forbidden" position in the computer/digital audiophile community ;-)
 
Joined
Apr 4, 2019
Messages
21
Likes
14
#44
Filters, refined volume control and at some point ‘balance control’.
For those odd shaped listening rooms to obtain the most desirable euphony.
 

jae

Active Member
Joined
Dec 2, 2019
Messages
267
Likes
221
#45
Would also like to see some DACs with good ADC as well for mic or line input. Something to compete with RME ADI-2 PRO variant.
 

RayDunzl

Major Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
10,148
Likes
8,727
Location
Riverview FL
#46
Another challenge: feed two DACs by identical signals but one with inverted polarity. The sum is perfect zero on the digital side. Check the sum of the analog signals.
How to sync the two DACs?

Original
Inverted and delayed by one sample
mix

1592431184624.png
 

KeithPhantom

Senior Member
Forum Donor
Joined
May 8, 2020
Messages
303
Likes
209
#47
DSP is next I think
I think room correction DSP and headphone soundstage ones are going to be the future of DSP. But the plugins available and the problems we can solve with some processing is something that inspires me to invest on that front.
 

KSTR

Addicted to Fun and Learning
Joined
Sep 6, 2018
Messages
530
Likes
1,226
Location
Berlin, Germany
#49
Once you've correctly trimmed out the irrelevant linear deviations the "perfect sum" also happens on the analog side, any differences way below the noise floor. I'm discarding some legacy design DACs where one channel was one sample behind the other and similar broken designs.
Linear deviations come from a) the gains not being exactly the same (starting right in the DAC chip but more importantly in the I/V (if applicable), filter, buffer and gain sections on the way to the output sockets, b) from minute phase difference in the analog filters (as the phase starts to deviate way before the magnitude). There also might be contribution from different pickup of noise and spurious signal, mostly depending on PCB layout etc.
What's left is the combined nonlinearity of the conversion propers and the analog section, plus any"dirt mismatch". Usually way below -120dBFS (except mains hum, sometimes).
 

UliBru

Member
Technical Expert
Joined
Jul 10, 2019
Messages
69
Likes
128
#50
I've never managed to trim out the deviations (linear and non-linear) to get a sum way below the noise floor.
If the gains are not exactly the same you may even correct it on the digital side.
Of course there are other possibilities, e.g. crosstalk and so on. IMO up to -60 dBFS (e.g. my RME) instead of your -120 dBFS.
So I simply wonder why to demand for better than -120 dB DAC resolution if the analog section behind the DAC chip worsens the result.
 
Joined
Sep 4, 2020
Messages
1
Likes
0
#51
My RME Fireface is e.g. reacting on different USB cables.
BTW with an Audioquest JitterBug plus 5m USB cable it even denies to play :)
But I mean DACs in general.
With a DAC immune to quality of input signal (it's digital, isn't it?) all tweaking actions before (from good old CD treatment to Windows in RAM) would be meaninglsess. Whereas the audiophile forums are full of tweaking discussions.
Are you being sarcastic when you say it reacts to that Audioquest jitter-inducer (as I like to call it) and your (passive?) 5m USB cable or do you really believe in USB audio myths as someone who owns a professional product like the RME Fireface?
 

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
1,175
Likes
1,311
#52
I realise that this might be an academic issue as far as audibility is concerned but there is a strange delight in the progress of technology and engineering prowess?
Given the choice of 3 DACs:

DAC A has a SINAD of 125, and costs $1,000
DAC B has a SINAD of 121, and costs $700
DAC C has a SINAD of 121, also costs $1,000, but has more inputs/outputs

Outside of those differences, they are the same.

Which seems like the best value? For me, it's between B and C based on what I need, while A is clearly the worst value. Engineering resources cost money, and we're already well past the limits of human hearing. As a consumer, a company using engineering resources to drive up SINAD even higher seems like a purely negative endeavor to me. I can't see any possible positive that can come from it. I'd much rather see them use those engineering resources to figure out how to make the same product cheaper, or use those engineering resources to provide additional features.

From an academic perspective, I agree that it might be fun to maybe see some Universities trying this, but I don't want companies that charge me money doing this :p
 

UliBru

Member
Technical Expert
Joined
Jul 10, 2019
Messages
69
Likes
128
#53
Are you being sarcastic when you say it reacts to that Audioquest jitter-inducer (as I like to call it) and your (passive?) 5m USB cable or do you really believe in USB audio myths as someone who owns a professional product like the RME Fireface?
I do not believe in audio myths.
But indeed I have noticed audio differences with different USB cables feeding my RME Fireface UC. A possible explanation: beside the transfer of digital information the cable can also transfer electrical noise, compensating current, act as an antenna and so on.

But independent of this I can clearly repeat the following test:
1. the Fireface is connected by a typical short USB cable. It runs as expected
2. with an Audioquest Jitterbug and the short USB cable the Fireface also works
3. using a 5m USB cable the Fireface does its job
4. using the Jitterbug plus the 5 m cable the Fireface stops working. So simply the signal quality must be reduced too much. IIRC someone has opened the Jitterbug,. It contains some electrical filters. Whatever reason, the USB comunication fails. So the signal quality seems to deteriorate too much.
5. using the 5 m USB cable plus another 5m extension cable (thus overriding the USB specification of 5m) the Fireface again works well
 

MediumRare

Major Contributor
Forum Donor
Joined
Sep 17, 2019
Messages
1,099
Likes
1,178
Location
Chicago
#54
Let’s see new transparent DAC products that incorporate usability features - like decent WiFi multi-room solutions at reasonable costs.
 
OP
T
Joined
Jul 5, 2019
Messages
37
Likes
23
Location
United Kingdom
Thread Starter #55
Given the choice of 3 DACs:

DAC A has a SINAD of 125, and costs $1,000
DAC B has a SINAD of 121, and costs $700
DAC C has a SINAD of 121, also costs $1,000, but has more inputs/outputs

Outside of those differences, they are the same.

Which seems like the best value? For me, it's between B and C based on what I need, while A is clearly the worst value. Engineering resources cost money, and we're already well past the limits of human hearing. As a consumer, a company using engineering resources to drive up SINAD even higher seems like a purely negative endeavor to me. I can't see any possible positive that can come from it. I'd much rather see them use those engineering resources to figure out how to make the same product cheaper, or use those engineering resources to provide additional features.

From an academic perspective, I agree that it might be fun to maybe see some Universities trying this, but I don't want companies that charge me money doing this :p
My mind agrees with you but my heart yearns for the best - I have an aversion to mediocrity!
 

IVX

Senior Member
Forum Donor
Joined
Mar 3, 2019
Messages
437
Likes
1,250
Location
South of China, SHZ area
#56
to get THD+N -123-124db need to use DAC's output level 5VRMS or 10VRMS because APx555 has the best performance there. BTW, at 10.2VRMS APx555 even worse than the previous generation of AP analyzers. In my opinion, if somebody said he got the industry lowest result in THD+N ever, he has to show that on the trusted equipment or method to prove.
2020-09-09_11-24-14.png
 

JohnYang1997

Major Contributor
Technical Expert
Manufacturer
Joined
Dec 28, 2018
Messages
4,642
Likes
7,090
Location
China
#57
to get THD+N -123-124db need to use DAC's output level 5VRMS or 10VRMS because APx555 has the best performance there. BTW, at 10.2VRMS APx555 even worse than the previous generation of AP analyzers. In my opinion, if somebody said he got the industry lowest result in THD+N ever, he has to show that on the trusted equipment or method to prove.
View attachment 82080
Actually you can measure up to 5.3V before it changes to 10V range, and likely 10.5V before it changes to 20V rangs. But yeah.
 

IVX

Senior Member
Forum Donor
Joined
Mar 3, 2019
Messages
437
Likes
1,250
Location
South of China, SHZ area
#58
I do that with my grandma SYS2522, even 3VRMS on the "2.5VRMS" range to see THD+Noise about -120db(however, it makes worse THD). Seems the bottleneck SYS2 and APx5 is too high noise, the THD performance itself a lot better than THD+Noise. I believe a lot of DACs on ES9038 or AK4499 have THD+N already better than -124db level, and the only weakness of APx555 doesn't let us see that.
 

JohnYang1997

Major Contributor
Technical Expert
Manufacturer
Joined
Dec 28, 2018
Messages
4,642
Likes
7,090
Location
China
#59
I do that with my grandma SYS2522, even 3VRMS on the "2.5VRMS" range to see THD+Noise about -120db(however, it makes worse THD). Seems the bottleneck SYS2 and APx5 is too high noise, the THD performance itself a lot better than THD+Noise. I believe a lot of DACs on ES9038 or AK4499 have THD+N already better than -124db level, and the only weakness of APx555 doesn't let us see that.
As shown in the DAC8 Stereo measurements. The measurement result is actually better than loopback. So far I'm only able to replicate the exact performance with 4x 9038q2m one channel. I'm curious to see what difference will be with better DAC(9038pro one channel for example). At this moment there may be DACs that are better than -124dB. It's difficult to measure the noise when DAC is on full output. The noise is different from idle/no output/low output. This phenomenon is very exaggerated on cs43131/43198. It's possible to use DC to test but it needs verification (still doesn't work on cs43198).
 

HooStat

Active Member
Joined
May 11, 2020
Messages
255
Likes
151
Location
Calabasas, CA
#60
Are there any advantages to putting the DAC "inside" the amp and optimizing the amp's output? I am just thinking that the amp output is what the end-game is, and there is probably more progress that can be made on the amp output.
 
Top Bottom