Minimum Phase vs Linear Phase

[solderdude]

And now, minimum vs linear phase please.

This is what @mansr has to say about this.

 

[maty]

It is the first time I read the article.

As I said, the recordings with instruments whose sound / tone we have memorized, usually acoustic and electrified to a lesser extent, the minimum phase filter generates a more natural, credible or similar / near sound to live music.

The current products, with computer-generated instruments and voices software processed (Autotune), the filter choiced it will be more a matter of personal taste, which sounds more spectacular, I say.

That is why I asked if someone else has verified it my experience.

 

[maty]

https://www.google.com/search?q=Minimum+Phase+classical+acoustic+music

-> https://www.google.com/search?q=Bob+Katz+minimum+phase

https://www.digido.com/ufaqs/linear-phase-equalization/

However, when you boost (or cut) a linear phase equalizer, the original depth is retained, nothing moves forward or backward in the soundstage, but the frequency range itself is emphasized or reduced. Linear phase tends to sound “smoother and rounder and subtler”, perhaps also because of a reduction in transient response (which reduction does not occur to the same extent for all models of LP (linear phase) equalizers)...

Transients are very important with orchestral music.

 

[Ron Texas]

This is what @mansr has to say about this.

@mansr has a bunch of articles over there, and he has participated in this thread.

 

[Ron Texas]

I know of many people who imagine things that aren't there. Often they post on audio forums

I thought they ran for elected office or managed to weasel their way to the top in autocratic systems. LOL

 

[pwjazz]

there is a difference here

Most definitely.

 

[RichB]

If anyone has an Oppo UPD-205, try this experiment:

- Set in a sweet spot fairly close the speakers, in my case 6 feet from each in a 30' room.
- Play some content with clear instrumentals and vocals through the DAC with no processing engaged
- Use the Oppo Media Control app to switch between Min Phase Fast and Linear Phase Fast filters

The difference is not subtle. Perhaps, the ES9038PRO DAC is doing more or something wrong but I find better clarity with the Linear phase filter, especially with female vocals.

Honestly, anyone with a good system and decent hearing should be able to pass a blind with these filters but it would have to be using consistent content. Selecting favorite from track to track adds confusion.

- Rich

 

[LTig]

This is the hypothesis. In principle, the difference in compression in FLAC should not be noticed, but I noticed it many years ago. At that time with an AMD and XP?

Therefore something must be done badly, in the processors or in the OS or in the software players because it is illogical (unlike what happens with the MP3).

Now, with Win 10 highly optimized for multimedia the difference is more noticeable. Much of the improvement is due to the optimization of the access of the ALU to the internal records of the CPU, to the priority of the processes and to the management of the RAM.

I can only say it works. Sometimes I overdo it and there is too much detail in music and fatigue. It is a matter of trying to find the optimal point.

And on top of that, with the new JRMC v25 I had to tweak the rules because it sounded worse than v24. After that it sounds better. So its creator must be doing things similar to mine.

If anything of this survives a DBT then you should test all these different flac files against their source (WAV) file in a DBT (playing a WAV file clearly needs the least CPU resources, and the WAV file is the reference for perfect SQ).

If this DBT shows a difference change your DAC. Get an RME ADI-2 (PRO or DAC) and its ASIO driver. Play the RME bit test WAV files both as WAV and FLAC 0..9 and the DAC will tell you if they are correct. I'm very sure that this will always be the case.

 

[PaulD]

If anything of this survives a DBT then you should test all these different flac files against their source (WAV) file in a DBT (playing a WAV file clearly needs the least CPU resources, and the WAV file is the reference for perfect SQ).

If this DBT shows a difference change your DAC. Get an RME ADI-2 (PRO or DAC) and its ASIO driver. Play the RME bit test WAV files both as WAV and FLAC 0..9 and the DAC will tell you if they are correct. I'm very sure that this will always be the case.

I have done that test several years ago, with ALAC (just extended FLAC) vs. AIFF vs. WAV - no difference... I also checked the bit streams, also no difference.

 

[tmtomh]

As I said, the recordings with instruments whose sound / tone we have memorized, usually acoustic and electrified to a lesser extent, the minimum phase filter generates a more natural, credible or similar / near sound to live music.

The current products, with computer-generated instruments and voices software processed (Autotune), the filter choiced it will be more a matter of personal taste, which sounds more spectacular, I say.

That is why I asked if someone else has verified it my experience.

You have said that, and the reason you've gotten few (no?) replies verifying your experience is that it's manifestly not true in any objective or even widely shared subjective way that acoustic and "less processed" electric instruments sound better with minimum phase filters. To the extent filter differences are audible - and I have found them to be, although very subtly so - I would argue that they do not correlate to musical styles or production types in the broad, simple way you claim.

If anyone has an Oppo UPD-205, try this experiment:

- Set in a sweet spot fairly close the speakers, in my case 6 feet from each in a 30' room.
- Play some content with clear instrumentals and vocals through the DAC with no processing engaged
- Use the Oppo Media Control app to switch between Min Phase Fast and Linear Phase Fast filters

The difference is not subtle. Perhaps, the ES9038PRO DAC is doing more or something wrong but I find better clarity with the Linear phase filter, especially with female vocals.

Honestly, anyone with a god system and decent hearing should be able to pass a blind with these filters but it would have to be using consistent content. Selecting favorite from track to track adds confusion.

- Rich

I have indeed heard some differences among the 7(!) filters the 205 offers, and I prefer the default Linear Phase Fast filter, in the sense that I've tried switching to others and have always gone back to that one.

However, my tests never have been blind, and so while I would like to think I could pass a blind test with these filters, or at least Linear Fast vs Minimum Fast, I wouldn't dare predict without actually doing the test - blind tests can be a lot harder than we think, and our confirmation bias and poor psychoacoustic memory can play some pretty convincing tricks on us!

 

[maty]

Of subtle nothing, the difference between minimum and linear is very noticeable. For years in my second system, with different processors, windows versions, ...

I have only done tests with MP3, OGG, FLAC and WAV. OGG -q9 sounds clearly better than MP3 320 kbps (LAME), so it is the format I use for the music I have on my mobile phone (usually convert from 24/96 FLAC of very good vinyl rips).

The music where you can clearly tell the difference in losses associated with MP3 is with the orchestral music, very easy test!

PD: always with high quality recordings and dynamic range, which is what I hear from the computer.

 

[Julf]

Of subtle nothing, the difference between minimum and linear is very noticeable. For years in my second system, with different processors, windows versions, ...

Can you please post your ABX logs?

 

[Julf]

If anything of this survives a DBT then you should test all these different flac files against their source (WAV) file in a DBT (playing a WAV file clearly needs the least CPU resources, and the WAV file is the reference for perfect SQ).

It is actually not at all clear playing a WAV file needs less CPU resources. With modern CPU architectures, the work of moving the extra, redundant data in and out can cause as much load as the actual processing of something like FLAC. Just curious - in what way is WAV " the reference for perfect SQ"?

 

[March Audio]

I thought they ran for elected office or managed to weasel their way to the top in autocratic systems. LOL

some are prolific on twitter...

 

[daftcombo]

the difference between minimum and linear is very noticeable.

What kind of difference?

 

[astr0b0y]

is that it's manifestly not true

I disagree. The reason is more likely due to being ignored.

 

[maty]

I should do the test again but the obvious difference is that I like the sound less, it differs a lot from how it should sound, less close to the live sound. So that we understand each other, that a violin does not sound like a viola.

It may also influence, I say, the phase response of the speakers. And of the amplifiers if they are class D. But I have not investigated about it.

I just noticed the difference so much that I do not even think about it.

 

[Julf]

I should do the test again

Yes, you should. Looking forward to you posting the ABX logs.

It may also influence, I say, the phase response of the speakers. And of the amplifiers if they are class D. But I have not investigated about it.

In what way is amplifier topology relevant in this context?

 

[tmtomh]

I disagree. The reason is more likely due to being ignored.

The two don't have to be mutually exclusive.

 

[pwjazz]

I should do the test again but the obvious difference is that I like the sound less, it differs a lot from how it should sound, less close to the live sound. So that we understand each other, that a violin does not sound like a viola.

It may also influence, I say, the phase response of the speakers. And of the amplifiers if they are class D. But I have not investigated about it.

I just noticed the difference so much that I do not even think about it.

Friendly suggestion - please use a single font size and avoid unnecessary bold or italic emphasis, it makes my head hurt and basically keeps me from wanting to read your posts.