solderdude
Grand Contributor
Why ?
Which components ?
You do believe in data sheets ?
What do these data sheets tell you ?
-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
DAC types and their sonic signature
- Thread starter Sparky
- Start date
Why ?
Which components ?
You do believe in data sheets ?
What do these data sheets tell you ?
1.
firstly every dac datasheet tell us how to design the low pass filter to minimise the ultrasonic noise.
I trust them if they think it's better to filter it.
Then we should check if the designer of the dac carefully designed the low pass filter.
Then we should check if it's adapted for all samplerates.
If at 44.1 khz the digital filter cut at 22khz and the analog filter cut off at 30 khz then at 192khz the digitall cut off freq will be around 96khz but the analog lpf will still be at 30khz and the filtering curve will be different.
If the analog lpf is designed to cut of at 80khz it will lead to a different curve that will maybe less adapted for cd.
That's why i asked to see the filtering curve at different samplerates and to see if high freq from high res generates.noise or not.
2.
On digital amps that modulates at high frequencies those ultrasonics in the output of the dac can interfere with the amp modulation and degrade its performance
3. It can produces emi
solderdude
Grand Contributor
1: You don't seem to understand the function of the extra low pass.
The extra analog low pass filter is to be set so the highest sample rate can pass the filter.
It does not change with the sample frequency and is not a steep filter either but is 6, 12 or 18dB/octave and as said there to remove MHz spurious.
It should be very clear to you, with the data that has been presented, that it is 'clean' between 20kHz and 80kHz.
It is VERY different from the switched capacitor post filter in a DAC chip and different from the digital reconstruction filter in the DS chip.
This depends on the filter selected and applied sample rate.
Your kite won't fly.
If you don't agree and have other evidence please show it.
Or measure it yourself, don't demand others to do it for you or to do tests that add nothing.
Most manufacturers follow the datasheets of the chip manufacturer and simply copy or even improve the design. Some manufacturers don't. These DAC's usually measure poorly on other aspects as well.
You are trying to blame components, others, those that measure, manufacturers and engineers in general for things you are convinced you can hear using incorrect testing methods.
Do you think the manufacturers of DAC chips which you seem to trust measure sighted and using their ears or do they use equipment and tests like Amir does ?
2: Yes, IF the there is HF content present AND the amplifier has a poor low pass filter then it is possible.
This is well known.
Will the same DAC sound poor on digital amps but not in analog amps ?
The extra analog low pass filter is to be set so the highest sample rate can pass the filter.
It does not change with the sample frequency and is not a steep filter either but is 6, 12 or 18dB/octave and as said there to remove MHz spurious.
It should be very clear to you, with the data that has been presented, that it is 'clean' between 20kHz and 80kHz.
It is VERY different from the switched capacitor post filter in a DAC chip and different from the digital reconstruction filter in the DS chip.
This depends on the filter selected and applied sample rate.
Your kite won't fly.
If you don't agree and have other evidence please show it.
Or measure it yourself, don't demand others to do it for you or to do tests that add nothing.
Most manufacturers follow the datasheets of the chip manufacturer and simply copy or even improve the design. Some manufacturers don't. These DAC's usually measure poorly on other aspects as well.
You are trying to blame components, others, those that measure, manufacturers and engineers in general for things you are convinced you can hear using incorrect testing methods.
Do you think the manufacturers of DAC chips which you seem to trust measure sighted and using their ears or do they use equipment and tests like Amir does ?
2: Yes, IF the there is HF content present AND the amplifier has a poor low pass filter then it is possible.
This is well known.
Will the same DAC sound poor on digital amps but not in analog amps ?
DKT88
Active Member
I don't know but if in some case it make some amps to sound bad we better know it.
Why don't you argue the same way with the measurements that are already there.
Measuring noise is not important to you ? It's better to check if a dac is 110 sinad or 115 sinad where most people will feed them with 16 bits audio that will result in 96db sinad?
If i follow your logical every test is needless.
We should check how the filter of dac performs.
Yes you can say they all perform good.
There can be some failures in design like the sinad show.
Ultrasonic test could show failure in the design too.
If you don't make the test you can miss failures.
Digital filter is a strong point where dac improve. Akm and ess has spent lot of research to improve it and you say we doesn't need to check if the designer of the dac use it good?
Really i don't understand your reasoning.
Quality of power supply and other designs can affect the performance of the digital filter
Last edited:
Indeed you don't.
- Joined
- Feb 23, 2016
- Messages
- 20,696
- Likes
- 37,432
I'm lost to what your point is? Is it an objection of some sort? Is it a critique about what is wrong or what is ignored?
solderdude
Grand Contributor
That's the real problem here.
As explained elsewhere you do not get the different filters in and outside of the chip and the consequences of their settings.
Reading datasheets as well as measurements and not fully understanding them plus the desire to draw incorrect conclusions based on what info you have results in these 'conversations'.
Now dragging in digital amps and more or less claiming these all behave similarly (which they don't) while it is the digital amp manufacturers responsibility to properly design the LPF on its input.
You seem to have the desire to 'link' the results of your incorrect way of sighted subjective testing with improper usage of user-selectable filters and design of the post filtering circuits.
There is no link here despite you demanding certain types of measurements that won't confirm your 'suspicions'.
The plots you desire have been shown here and there and don't seem to sink in.
restorer-john
Grand Contributor
I don't think so. He is probably able to "hear the difference" above 30Khz...
solderdude
Grand Contributor
He just demands to see plots up to at least 80kHz because he believes there may be 'something' there that should explain what he perceives using his terribly flawed subjective testing method.
What he is trying to tell Amir is that all of Amirs plots should at least span 100kHz so he can see if it is truly clean.
What he is trying to tell Amir is that all of Amirs plots should at least span 100kHz so he can see if it is truly clean.
OP
- Thread Starter
- #55
Forgive my ignorance but if all are meant to sound the same, why the different brands with their hardcore following who will swear blind on "sonic signature" differences.
This is exactly why I started this thread because I was convinced of a change in sound when using different DAC's.
If they're all meant to sound the same then what's the point if it all?
Again, I'm not trying to be clever or funny as I know absolutely nothing compared to you guys.
I'm here to learn and hopefully expand my knowledge of all things audio...
There's lots of non-audio examples of parity product where there is nonetheless fierce brand loyalty.
But parity in this case is purely the sound. There's differences in build, features, application flexibility, and (for lack of a better term) status appeal. Just because a Timex does a fine job of telling time doesn't mean people won't want Movado, Bulova, Rolex, or whatever. Same with DACs- I can get essentially perfect sound out of a $9 Apple dongle, but if I win the lottery, I'll probably get an RME ADI-2 because of the extra features and (TBH) the "coolness" factor.
The DACs that truly sound different are a pretty small subset.
But parity in this case is purely the sound. There's differences in build, features, application flexibility, and (for lack of a better term) status appeal. Just because a Timex does a fine job of telling time doesn't mean people won't want Movado, Bulova, Rolex, or whatever. Same with DACs- I can get essentially perfect sound out of a $9 Apple dongle, but if I win the lottery, I'll probably get an RME ADI-2 because of the extra features and (TBH) the "coolness" factor.
The DACs that truly sound different are a pretty small subset.
OP
Ahhh, I see your point. You start with a base to build on (the DAC chip) then engineer your hardware and software around that to mould it to your liking (from a manufacturer's point of view).
I hope that's what you meant anyway....
Close. Hardware, software, feature set, user interface, packaging. The last two are key differentiators.
OP
This is rather tame. Try to go onto other forums and use science....
Science means nothing now. People don't care. You can make your own facts today.
This forums thinking on audio is in the smallest minority. Vast majority of posts on reddit/headfi is subjective.
How is a person to understand that there is science when their whole life they've been bombarded with non scientific info on music.
Of course people think a dac has an effect, the player, the headphone cable, the alignment of the sun, etc. On sound
Also, given that in the scientific community, it's been found that many tests done were not able to be replicated due to power design of the test, are you surprised the average Joe buys a new dac/amp, "a/b" tests it in a completely poor manner and declares dacs/amps have a huge effect! It's really no surprise.
On headfi right now I'm lookin for cheap iem cable recommendations and xyz cable is good for bass, ABC cable is warm, jkl cable is cold, and lolrofl cable needs to be microwaved to perform best!
It's increasingly hard to use science to explain things to people even as our world becomes more driven by tech. Weird.
Similar threads
