Linear Phase filter settings. Topping: How we (ASR) can help?

[sajunky]

Here is a problem raised in the other thread: Minimum Phase vs Linear Phase

I haven't done any blind testing, but there appears to be a certain harshness in loud and high notes of some female vocalists which goes away when the DAC is set for minimum phase. The Topping D30 has this problem, in my opinion. When I use my Topping DX3 Pro with a linear phase filter, it sounds just like the D30. Switch to a minimum phase filter and the harshness goes away. This was noted on both LSR305 Mk II's and LS50's powered by a Crown XLS 1502. The difference is subtler wiith my Grace M9XX which might mean something else is going on.

The answer is given by a member @Veri that chosing filter shouldn't matter if "all 5 filters were 'proper' without severe roll-off or aliasing" Post #41

Now lets present my point. According to the above, Linear Phase filter sounds he same in both D30 and DX3 Pro, but is inferior to the Minimum Phase filter in DX3 Pro. I only have Topping D30 and I complained already that while a simple type of sound is presented clean, in the passages where there is a sound with complex harmonics like choir or a loud orchestra there is lot of distorsion which I interpreted as a jitter problem. The same opinion in different words is presented above. In addition to the above, I reported a problem I didn't see in any other DAC. A perfectly tuned piano on the D30 sounds like individual strings were in complete dissonance (detuned). It is like DSP processing on D30 is shifting frequency of individual strings to present it more different of each other. In result is an cacophony effect.

I understand that Topping do it intentionally to achieve the best measurement tests results with complete disregard how it sounds. They know that Amir do not test secondary filter settings, so they tune it for the best sound. Linear Phase filter is a primary one (and the only one on D30), so they push these settings above any sanity check. In result we see a perfectly flat frequency response, it is were "Excellence of engineering" is, but sonically it sucks.

A question is what we can do to promote on one side "Excelence of engineering" and on the other side the best audiophile sound quality without of any sign of listening fatigue. If they are conflicting, there must be a balance between.

 

[RayDunzl]

The filter in my DAC "sounds good", produces good measurements, and I don't experience DAC Filter fatigue, although I may experience source material fatigue occasionally.

It produces the symmetrical pre/post ringing when tickled with an "illegal" pulse, so I suppose it is a linear phase filter.

I don't worry about it.


Impulse calculated from sine sweep at preamp output:

Impulse calculated from sine sweep output by speakers (MatinLogan reQuest) at listening position via UMIK-1:

Pre and Post DRC, this time with the JBL LSR 308, same DAC and preamp.

Impulse calculated from sine sweep output by speakers at listening position via UMIK-1:

If you're worried about the output wobbles around the impulse I might suggest looking at something other than the DAC filter.

 

[sajunky]

If you're worried about the output wobbles around the impulse I might suggest looking at something other than the DAC filter.

Do you still not understand? It is a measurement of a pulse response. Topping will produce it well, as engineers know how you measure it.

On the other side, well tuned piano produces two or three semi-independent ~sine waves that naturally tend to synchronise together. From the Topping it comes out in dissonance, like from detuned piano. Whe all know about that Topping DSP engine tries to enhance major tones to make it perfectly clean with a perfect frequency response, such settings look well in our measurements, but result is that frequency of individual tones is altered with a bad sonic effect. Please show us how you would measure it. Before you come out with a clue, maybe it is worth to ask Topping why Linear Phase filter in D30 and DX3 Pro produce harshness and a Minimum Phase filter from the same device do not. Thank you.

 

[solderdude]

well tuned piano produces two or three semi-independent ~sine waves that naturally tend to synchronise together. From the Topping it comes out in dissonance, like from detuned piano.

Whe all know about that Topping DSP engine tries to enhance major tones to make it perfectly clean with a perfect frequency response, such settings look well in our measurements, but result is that frequency of individual tones is altered with a bad sonic effect.

[more citation needed]

 

[sajunky]

[more citation needed]

Blind test? .LOL.

 

[SIY]

Do you still not understand? It is a measurement of a pulse response. Topping will produce it well, as engineers know how you measure it.

On the other side, well tuned piano produces two or three semi-independent ~sine waves that naturally tend to synchronise together. From the Topping it comes out in dissonance, like from detuned piano. Whe all know about that Topping DSP engine tries to enhance major tones to make it perfectly clean with a perfect frequency response, such settings look well in our measurements, but result is that frequency of individual tones is altered with a bad sonic effect. Please show us how you would measure it. Before you come out with a clue, maybe it is worth to ask Topping why Linear Phase filter in D30 and DX3 Pro produce harshness and a Minimum Phase filter from the same device do not. Thank you.

Pulled that one out of your ass, did you?

 

[Veri]

We all know about that Topping DSP engine

Topping DSP engine?? And we all know about this DSP engine too? To think I was left out on all the fun! Here I was thinking all they do was customize the DAC chip's values... which can transparently be verified from the I2C bus to see if they pulled any weird "DSP" shenanigans in there...

No secret sauce. Just like @mansr objectively took apart the dragonfly cobalt. Careful, the article there begins with "Objective"

 

[solderdude]

Blind test? .LOL.

well tuned piano produces two or three semi-independent ~sine waves that naturally tend to synchronise together. From the Topping it comes out in dissonance, like from detuned piano.

I didn't know that Topping creates dissonance... that a piano is reproduced detuned ?
How can a digital filter change the frequency of 2 or 3 tuned strings but not one of them ?
Do you have any proof in the form of captures (at high sample rates) of a single piano hit where one can see that only one of the strings is accurate (when linear phase filter is used) and all 2 or 3 strings are perfectly in tune when the minimum phase filter is used.
I would like to see evidence of this.
(Blind test? .LOL) doesn't say much and isn't any kind of evidence.
What you claim is very measurable.. I want to see the results with 2 different filters.
Oh... and don't bother with ... you prove it, you measure it... its your (unsubstantiated?) claim.

Whe all know about that Topping DSP engine tries to enhance major tones to make it perfectly clean with a perfect frequency response, such settings look well in our measurements, but result is that frequency of individual tones is altered with a bad sonic effect.

I don't know this ... How do they 'enhance' major tones and where is proof found that frequencies of individual tones are altered ?
Do you have any evidence of this ?

 

[pierre]

Do you still not understand? It is a measurement of a pulse response. Topping will produce it well, as engineers know how you measure it.

On the other side, well tuned piano produces two or three semi-independent ~sine waves that naturally tend to synchronise together.

maybe, do you have examples?

From the Topping it comes out in dissonance, like from detuned piano.

Really, if it is obvious please measure it with REW

We all know about that Topping DSP engine tries to enhance major tones to make it perfectly clean with a perfect frequency response, such settings look well in our measurements, but result is that frequency of individual tones is altered with a bad sonic effect.

Who is we? Does the topping have a DSP? I guess it has some filters that you can choose. I would be surprise it also has an eq for flattening the frequency.

 

[sajunky]

I didn't know that Topping creates dissonance... that a piano is reproduced detuned ?
How can a digital filter change the frequency of 2 or 3 tuned strings but not one of them ?

Two or three, but not one, .LOL. How it is a matter to you?

 

[sajunky]

Pulled that one out of your ass, did you?

That's getting fun....

 

[solderdude]

Two or three, but not one, .LOL. How it is a matter to you?

Why would you bring such total nonsense up as an argument against a certain DAC ?

 

[sajunky]

Why would you bring such total nonsense up as an argument against a certain DAC ?

Why you replied?

 

[solderdude]

Why you replied?

Why not ?
I was interested to see how you would wiggle yourself out of the nonsense you wrote.. but it is clear you can't.

 

[sajunky]

Why not ?
I was interested to see how you would wiggle yourself out of the nonsense you wrote.. but it is clear you can't.

You didn't answer my question "how it is a matter to you". You wrote something, it needs to be clarified first before going any further. Otherwise there is no common ground, but I see you are giving up.

 

[Thomas savage]

You didn't answer my question "how it is a matter to you". You wrote something, it needs to be clarified first before going any further. Otherwise there is no common ground, but I see you are giving up.

You made some striking claims that you presented very much as matters of fact.

You need to present the qualifying evidence to substantiate theses things . ..

Or it's just ' what you reckon ' or what you made up in your imagination and presented as fact .

If you can't do this you have nothing more to say on the matter.

 

[solderdude]

I was just wondering how a DAC would change frequencies and how one filter type would make tones 'dissonant' and other filter types won't.
Where you got that info from as 'we all know this'.
I am one of the 'we' on this planet and so I was compelled to ask the question... so ask for a citation of your general knowledge.
That was how it became a matter to me ... the we all know this bit and I (and countless others) didn't.

Where is the data that makes you conclude what you did. You must have some 'evidence' ?

 

[BDWoody]

Why not ?
I was interested to see how you would wiggle yourself out of the nonsense you wrote.. but it is clear you can't.

Amazing how detailed his technical answers can be when there are actual answers to be given. As soon as there is a challenge to support the claims from the crazy train, the answers start to sound like trying to reason with a stubborn teenager who thinks he's being clever or profound...

I expect the next response to be something along the lines of 'your mama.'

 

[sajunky]

I was just wondering how a DAC would change frequencies and how one filter type would make tones 'dissonant' and other filter types won't.

I was also wondering, it could be a side effect of things like this:

[specific DAC model] is the plant or actuator, the goal of regulation functions is minimizing of 2nd and 3rd harmonics difference vs reference. If the input reference signal is the clear sine(1kHz), the system will adjust the DAC to reduce 2nd and 3rd harmonics, however, if your input reference is the combination of 1kHz 99.75% + .2% of 2kHz + .1% of 3kHz (kinda SE harmonics emulation), then DAC's THD will be the same as your reference. In other words, you can modify the THD profile of your DAC randomly.

I don't mention the name, as I don't want to discourage the enthusiastic character nor disturb business startup, but it makes me worry what other manipulations were in place already under NDA's. The author is also proposing embedding of analyser - real time measuring DAC distorsions and inserting it (like we use a convolution function) to tell DSP engine how to compensate these internal distortions.

 

[solderdude]

But that won't affect fundamental frequencies to change. It merely has to do with the amplitude of its harmonics.
It also isn't a 'step' thing either which one would expect when one makes the claim that 3 strings in a pinao which are individually tuned to have a certain relation to create a 'note' would change the frequency.

This can easily be tested by sweeping or slowly wobbling a second frequency slowly around a fixed frequency.