Extreme filtering software upscaling

[sir sanders zingmore]

13 minutes ago, legend said:

However on the Briggs-Myer

Not wanting to derail the thread but as you are  a scientist I’m sure this will interest you

https://mobile.abc.net.au/news/2019-01-07/history-of-the-myers-briggs-personality-test-and-why-we-trust-it/10593550?pfmredir=sm

[rand129678]

31 minutes ago, legend said:

I am a physicist by training so am EXTREMELY aware of the reductionist approach of changing one variable at a time!

 

However on the Briggs-Myer test I am also a INTJ who uses intuition and helicoptering rather than the typical ISTJ of engineer/scientist that likes to obsess about detail.

 

Hence I like to explore the edges of topics where often the more interesting connections occur, even if it does conflict with the first point.

 

And having been involved in audio design for 25 years or more I can also get the feel of whether something is better or not without DBT testing changing one variable at a time  - though of course the latter is necessary to be scientifically rigorous/certain!

 

I think the main point I was trying to make in the comparison of the Project DAC S2 DAC plus MQA and the Chord Qutest plus M-Scaler/extreme filtering was that for me they pointed in the same direction/type of improved sound quality (but of course varied in the extent due not least to cost differences which obviously affect technical implementation) and that this MAY have been due to their common emphasis on the importance of timing/impulse responses.

All noted. However on the topic of DAC design, everyone should check out the late Charles Hansen post (one of the most respected qualified experts in DAC, A-to-D converters and amp designs, in both pro and HiFi worlds). 

 

https://www.computeraudiophile.com/forums/topic/35106-how-does-a-perfect-dac-analog-signal-look-different-than-a-cheap-dac/?page=7&tab=comments#comment-713189

 

It’s worth a careful read a couple of times over and bookmarking for future reference.

 

His points that analogue section and power supply section design can swamp differences between digital filters when comparing DACs, often gets missed in this fun hobby of ours.

 

Edited February 10, 2019 by Music2496

[legend]

1 hour ago, Sir Sanders Zingmore said:

Not wanting to derail the thread but as you are  a scientist I’m sure this will interest you

https://mobile.abc.net.au/news/2019-01-07/history-of-the-myers-briggs-personality-test-and-why-we-trust-it/10593550?pfmredir=sm

Yes I have read this article a while ago.  Like most psychological/personality tests Briggs-Myer has its limitations - trying to reduce complex phenomena down to a single or small number of variables.

[Nada]

3 hours ago, legend said:

Yes I have read this article a while ago.  Like most psychological/personality tests Briggs-Myer has its limitations - trying to reduce complex phenomena down to a single or small number of variables.

You've completely missed the point there.

[Assisi]

1 hour ago, Nada said:

You've completely missed the point there.

@Nada

Can you explain more?

 

John

[davewantsmoore]

18 hours ago, Music2496 said:

ESS Sabre chips have always been a bit of a black box in some aspects.

I agree that people have approached them like a "black box", due to their lack of understanding of how they work.

 

 

Ask yourself.... Does that describe me?....    Because you appear to make some contradictory (ie. opposite) statements, given how the DAC works..... and especially given the context of this thread, I think it would be very very important to clarify them.   It would be a shame if we have people assuming that (in this case) the different input rates will result in less "work" and less "noise", if that is actually not the case.

 

I'll wait until I am invited this time.

Edited February 11, 2019 by davewantsmoore

[davewantsmoore]

9 hours ago, legend said:

INTJ

*waves*    

[Nada]

2 hours ago, Assisi said:

@Nada

Can you explain more?

 

John

Id rather not to be honest. Its off topic and could be misunderstood.

 

In the interests of science Ill ignore my better judgement and suggest anyone interested in stuff that works, scientific methods and psychology rating scales doesn't mention them in the same sentence as being similarly flawed as the Briggs-Myers personality drivel.  To be fair expertise in physics doesn't generalise at all well to  a working grasp of the psychological sciences. Somewhat ironically psychology explains that in such an area we should expect an over estimation of our abilities, in terms of confirmation bias and belief perseverance.  As fair disclosure im not a psychologist either so doubtless  similarly deluded No offence  meant to anyone other then myself and pseudoscientific bull faeces.

 

Back to mega taps.....

[legend]

46 minutes ago, Nada said:

Id rather not to be honest. Its off topic and could be misunderstood.

 

In the interests of science Ill ignore my better judgement and suggest anyone interested in stuff that works, scientific methods and psychology rating scales doesn't mention them in the same sentence as being similarly flawed as the Briggs-Myers personality drivel.  To be fair expertise in physics doesn't generalise at all well to  a working grasp of the psychological sciences. Somewhat ironically psychology explains that in such an area we should expect an over estimation of our abilities, in terms of confirmation bias and belief perseverance.  As fair disclosure im not a psychologist either so doubtless  similarly deluded No offence  meant to anyone other then myself and pseudoscientific bull faeces.

 

Back to mega taps.....

I agree it is entirely off topic - and difficult to discuss briefly/properly.

 

My only general comment is that I was taught the B_M test at the Australian Graduate School of Management at UNSW by the prof of organisational behaviour who also had a Ph D in Chemistry so had a fairly good understanding of the methodology of both science and psychology - but perhaps he was also deluded.

[Ittaku]

Just now, legend said:

perhaps he was also deluded.

This part is on topic though. There's plenty of that going on in the audiophile world.

[davewantsmoore]

32 minutes ago, legend said:

but perhaps he was also deluded.

Like many things, it depends on exactly what/how you say about it, or refer to it, as to whether or not you are stupid or not   

Obviously things like it, are much less "absolute" than the physical sciences.

31 minutes ago, Ittaku said:

There's plenty of that

Indeed....  broad assumptions based on intuition are very common.

 

Resampling by a lower multiple (vs higher multiples) require "less work" and/or result in "less noise", or similar

Resampling using a non-integer multiple of the base rate, will result in worse performance

Etc.

Edited February 11, 2019 by davewantsmoore

[legend]

There was a great 'graph' in a book on the methodology of science called "Searching for Certainty" that had 'understandability/explaining ability' on one axis and 'prediction ability' on the other.  Quantum mechanics scored extremely low on the former but extremely high on the latter; most social sciences faired highly on the former but lower on the latter.

Edited February 11, 2019 by legend
clarification

[Nada]

19 minutes ago, legend said:

There was a great 'graph' in a book on the methodology of science (whose title & author I have forgotten) that had 'understandability' on one axis and 'prediction ability' on the other.  Quantum mechanics scored extremely low on the former but veexremely high on the latter; most social sciences faired highly on the former but lower on the latter.

That's an embarrassingly out dated concept. The wonderful thing about science is it keeps discarding once cherished hypothesis.

 

Paradoxically the maligned social sciences have high predictive value in showing why physicists so often do their best work in their early years and then go rapidly downhill.

Edited February 11, 2019 by Nada

[Assisi]

3 hours ago, Ittaku said:

This part is on topic though. There's plenty of that going on in the audiophile world.

@Ittaku ,

 

Who of us is deluded or is it all of us?  I read this thread so I might learn something but the complexity of filters and the explanations thereof etc is beyond me so far.  Thats fine to me though.

 

In a system comprising quality components that compliment each other is filtering and DSP necessary?  You should be able to answer this.  I have found that as the integral components in my system improve in quality the other add-ons and accessories often become superfluous to the final out come.  In fact there can be a dis benefit and less is better.

John 

[Ittaku]

22 minutes ago, Assisi said:

@Ittaku ,

 

Who of us is deluded or is it all of us?  I read this thread so I might learn something but the complexity of filters and the explanations thereof etc is beyond me so far.  Thats fine to me though.

 

In a system comprising quality components that compliment each other is filtering and DSP necessary?  You should be able to answer this.  I have found that as the integral components in my system improve in quality the other add-ons and accessories often become superfluous to the final out come.  In fact there can be a dis benefit and less is better.

John 

Filtering during upscaling or oversampling within a DAC is an essential but small part of your overall sound. The changes I'm ascribing from extreme filtering are real but remain subtle and are really the last pooftenth percent after everything else that is far more important. I refer to other areas of audiophilia that people are delusional about, not the filtering. However, I will not speak openly about them here for fear of being lambasted by the community with what would be very unpopular opinion. At best I'd get a lot of dislikes, at worst I'd get a lot of people flaming me and risk being banned. There's a good chance you've seen all those things already mentioned and debated to death on other threads anyway.

Edited February 11, 2019 by Ittaku

[davewantsmoore]

10 minutes ago, Assisi said:

explanations thereof etc is beyond me so far

It's quite a complex topic.    I think a good example of "the delusion" is trying to generalise/simplify it down too far.     Lest we end up with general advice which in some situations would be right, but other times totally wrong.

23 minutes ago, Assisi said:

In a system comprising quality components that compliment each other is filtering and DSP necessary?

Yes, the filters/DSP which are being discussed are a fundamental part of how digital to analogue converters work (and so are usually done inside the DAC).    What is being talked about is doing them differently (in order to improve performance), and doing them outside the DAC.

[Assisi]

1 hour ago, Ittaku said:

I refer to other areas of audiophilia that people are delusional about,

@Ittaku

That was what I thought you were alluding to .  I am well aware of delusional perspectives.  At least once it has been said that I was delusional.  Just look at the thread on Audio Myths and Misconceptions and the number of times the word delusional is used including by the OP.  I was being naughty asking you the question.  My apologies.

1 hour ago, davewantsmoore said:

Yes, the filters/DSP which are being discussed are a fundamental part of how digital to analogue converters work (and so are usually done inside the DAC).    What is being talked about is doing them differently (in order to improve performance), and doing them outside the DAC.

@davewantsmoore

 

So, filtering and or DSP are internally fundamental to the operation of a DAC.  The one I have now and the one before it and the one before that and the list goes on all do it internally.  In only one of them could I choose a filter.  I never did.  None of them had user control of the DSP that I am aware of.

 

I use Roon and I did on a couple of times experiment with DSP.  There was a difference but I am not sure that there was a benefit, so I turned it off.  Dave you have taught me about the distinction between a difference and a benefit.  I now know that many things are different but not necessarily beneficial.

 

I have very good components so what I hear is definitely wonderful without adding extra processing.  As I said I am becoming focused towards less is more

 

The question is filtering and or DSP external to a good quality DAC with serious internal implementation, of a benefit and if so to what extent, what are the processing overheads and can there be an impact on the final result?

John

Edited February 11, 2019 by Assisi

[Ittaku]

1 minute ago, Assisi said:

The question is filtering and or DSP external to a good quality DAC with serious internal implementation, of a benefit and if so to what extent, what are the processing overheads and can there be an impact on the final result?

DSP when you're talking about filtering is just one small example of DSP. DSP (digital signal processing) can really mean doing anything to the music while it's still in digital form. In that regard, I think that DSP has a lot to offer and use it extensively in my system before my DAC while it's still in digital form. I've mentioned it a few times, but the DSP I use is both a crossover and room correction device. It allows the main speakers much more scope to operate in their best range by offloading the lowest most extreme frequencies only to the subwoofer. But the room correction is really where it shines, where massive dips and troughs in the frequency response are smoothed out to something far more respectable. With an open baffle speaker and relatively challenging room conditions, these peaks and troughs are inevitable, and the sound is pretty bad without the room correction. Does it worsen the sound? This is very debatable, and some people believe that any time you modify the signal you are going to make it sound worse. With my understanding of musical information whilst it's still in the digital domain, I think this is not true when done at a high enough bitrate and frequency (most work in 32 bits internally.)

 

The advantage in terms of what I get from room correction and crossing over is worth 100x more sound improvement than if I'd just plugged the signal straight into my DAC. That said, there are finite limits to how much correction you can apply, and troughs that are nulls in the room are actually impossible to fill in - they just consume more power, create driver excursion and non-linearity, and are still nulls. So it has to be applied thoughtfully. I also find the frequency responses that have less dB modification do sound better so I think it's best to be prudently applied rather than very aggressively. The dspeaker device I use does most of this automatically, though I always fine tune it myself afterwards. The major(?) disadvantage is that DSP devices tend to be limited in the sample rate they run at. My dspeaker runs at 44/48/88 or 96 internally and while it takes higher rates, it downsamples them. Many other devices actually run only one frequency and resample everything to that, like 48! Fortunately I can't tell 96 audibly apart from 192 or higher anyway normally so it hasn't worried me. These upscaling experiments though are showing something different though when more taps were added and the sample rate was pushed higher - but it's impossible for me to separate whether that was because of the more taps, the higher sample rate, or the combination of more taps and higher sample rate, as I don't have a way of testing those, but others have tested higher and swear it sounds better.

[legend]

10 hours ago, Ittaku said:

Filtering during upscaling or oversampling within a DAC is an essential but small part of your overall sound. The changes I'm ascribing from extreme filtering are real but remain subtle and are really the last pooftenth percent after everything else that is far more important.

I think part of the problem with 'improvements' (to me reduction in distortion = difference between output and input) is that their  perception is not linear ie the closer the audio reproduction system gets to reproducing the initial recording and its emotional impact then the more obvious even 'small' (in absolute terms  - but not in relative terms) improvements may be.  Perhaps this is related to the logrithmatic nature of our perception of sound (and other sensory inputs such as light) and so how our brain's work?  Whatever, to me the  extreme upscaling (and alternatives such as the M-Scaler) give a very noticeable improvement in my system.  

Edited February 11, 2019 by legend
typo

[legend]

7 hours ago, Ittaku said:

 

The advantage in terms of what I get from room correction and crossing over is worth 100x more sound improvement than if I'd just plugged the signal straight into my DAC. That said, there are finite limits to how much correction you can apply, and troughs that are nulls in the room are actually impossible to fill in - they just consume more power, create driver excursion and non-linearity, and are still nulls. So it has to be applied thoughtfully. I also find the frequency responses that have less dB modification do sound better so I think it's best to be prudently applied rather than very aggressively.

Agree entirely.  Another disadvantage/danger of room correction is that is only works at one position (or a small variation that is a fraction of the wavelength being corrected) because if you (partly) fill in a trough at that position by adding gain at that frequency it will also add that gain at other positions where there is already a peak.  This is not a problem if you are single listener but is for family listening or at GTGs and shows,

[davewantsmoore]

8 hours ago, Assisi said:

The question is filtering and or DSP external to a good quality DAC with serious internal implementation, of a benefit

That's a complex question ... and depends a lot on the specifics of the DAC and the specifics of the filter.....  It could be, Yes.

[davewantsmoore]

3 minutes ago, legend said:

This is not a problem if you are single listener

Eh?

Peaks/dips in the 'off-axis' response of a speaker are a not a problem?    

 

You are undoubtedly aware that many people consider them to be one of the biggest problems (!?)

[legend]

8 hours ago, Assisi said:

@Ittaku

 

I have very good components so what I hear is definitely wonderful without adding extra processing.  As I said I am becoming focused towards less is more

 

Someone on this site has/had a saying along the lines "things need to be as complex or as simple as they need to be, not more or less".  I think this is true and in some cases complexity is required to solve a problem/make an improvement as in the case of extreme upsampling.

[legend]

1 minute ago, davewantsmoore said:

Eh?

Peaks/dips in the 'off-axis' response of a speaker are a not a problem?    

 

You are undoubtedly aware that many people consider them to be one of the biggest problems (!?)

I was speaking purely in the context of room correction,  I assumed a single listener is on axis.

[davewantsmoore]

30 minutes ago, legend said:

I was speaking purely in the context of room correction,  I assumed a single listener is on axis.

Yes, I know you were.

 

If you correct the single listener directly arriving sound ....  Then as you said, you create peaks/dips in other positions (ie. at other listening angles).

 

This is a significant problem for that single listener ..... as they do not just hear the directly arriving sound.   They also hear the reflected/delayed sound (which contains the peaks/dips).

 

 

This is why speaker directivity (goes by other names, like "polar response") is so important ....  The more similar the speaker radiates with angle, the less likely correcting the directly arring sound, will result in peaks and dips at other listening angles.