Minimum Phase vs Linear Phase

[MrSoul4470]

If the EQing is coming from different plugins then I can hardly imagine that the resulting signal will not degrade.

 

[Julf]

(Novice reviving this thread, I hope this resurrection is OK.)
Could someone please confirm that it's correct/incorrect that if a hundred different EQs (all 64 bit plugins, minimum phase) are in a chain, with the last EQ undoing the cumulative changes of the previous 99, then the resulting signal will not be "worse" than the signal that entered the chain. (OR similar scenario, if you put 100 EQs in a chain, the result will not be "worse" than 1 EQ that represents the same filter effect as the sum of those 100 EQs). Thank you. I'm asking because of usage of multiple Goodhertz plugins plus Headphone EQ.
Does "stacking EQs" and "putting in a chain" have identical meaning?

The processing never has infinite precision, so there will be some degradation, but normally it will be way below audible limits and way below the noise and distortion in the source material and your speakers/listening environments.

 

[audixix]

Thank you. The reason I was asking is that I was wondering if there would be an advantage if there were one single plugin by the same company that is capable of "administering" all the desired effects and changes/filters/presets in as few steps as possible, in particular pertaining to the applied (cumulated) EQ curve, as opposed to the multiple plugins that I'm currently using in a chain. But it is my suspicion also that a small handful (e.g., 3 or 4) of plugins will not cause audible degradation for the simple reason that it is "unlikely" that one single plugin would, for technical reason, just happen to NOT produce audible degration, yet ENOUGH degradation that the cumulative effect degrading effect (assuming it will occur in the first place) of FOUR such plugins WILL be audible.

 

[MRC01]

To get more specific, for example, with minimum phase EQ in DSP do the phase shifts reverse each other? That is, pick an amplitude A, frequency F, and width Q. Apply +D at F and Q, and you'll get some phase shift. Now apply -D at freq F and Q, do you get the opposite phase shift and end up exactly where you started, minus numeric precision which should be in the inaudible dither/noise?

 

[MediumRare]

To get more specific, for example, with minimum phase EQ in DSP do the phase shifts reverse each other? That is, pick an amplitude A, frequency F, and width Q. Apply +D at F and Q, and you'll get some phase shift. Now apply -D at freq F and Q, do you get the opposite phase shift and end up exactly where you started, minus numeric precision which should be in the inaudible dither/noise?

You’re totally overthinking this. There is no answer, even if anyone knew what your chain, settings, and source material were. If you really want to know, run a dry sweep and a wet sweep and compare them both audibly and visually.

 

[NTK]

To get more specific, for example, with minimum phase EQ in DSP do the phase shifts reverse each other? That is, pick an amplitude A, frequency F, and width Q. Apply +D at F and Q, and you'll get some phase shift. Now apply -D at freq F and Q, do you get the opposite phase shift and end up exactly where you started, minus numeric precision which should be in the inaudible dither/noise?

Yes. Since for minimum phase systems frequency magnitude response is tied to phase response, if you inverse one (to undo it), you'll automatically inverse the other. See below:
(Source)

 

[Keith_W]

To get more specific, for example, with minimum phase EQ in DSP do the phase shifts reverse each other? That is, pick an amplitude A, frequency F, and width Q. Apply +D at F and Q, and you'll get some phase shift. Now apply -D at freq F and Q, do you get the opposite phase shift and end up exactly where you started, minus numeric precision which should be in the inaudible dither/noise?

Yes, provided that what you are measuring actually is minimum phase. There are conditions that can render what should be minimum phase to be non-minimum phase, for example reflections and room modes. So in a frequency response sweep, there will be regions that are minimum phase, and other regions that are non-minimum phase. It is not possible to predict where these regions are without running a sweep. See John Mulcahy's excellent explanation in the REW help guide.

 

[Chuckser]

This choice shows up in various places:
1. Many DAC's, including several reviewed here.
2. Most implementations of SOX
3. Rephase (default is minimum phase.)

My understanding is linear phase filters have pre-ringing. Minimum filters put all the ringing after the pulse. When lossy audio compression was under heavy development pre-ringing was often discussed and it could be identified in double blind testing. At the time developers were trying to reduce pre-ringing.

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.

So, what kind of filters should we be using, or is this from the marketing department? Does it make a difference if the application is a DAC, SOX or Rephase? Note the JRiver implementation of SOX is fixed, probably with linear phase and they are adamant about not making it configurable.

Might there be an issue with the RED BOOK de-emphasis filter?

 

[SIY]

Might there be an issue with the RED BOOK de-emphasis filter?

Pre and de-emphasis were relevant to a VERY limited number of CDs in the 1980s. I had a light on my CD player to indicate its use and not one CD in my collection triggered it. I seriously doubt it's been used in the past 30 years.

 

[AnalogSteph]

I wouldn't say it was that limited. The ratio of emphasis CDs in my collection is about 1%. And yes, the offenders are all from the 1980s, most commonly around 1985, the latest from 1987. Not used in the last 30 years is about right, looks like 1992 was the end of the line:

Pre-emphasis (release list) - cdHistory

www.studio-nibble.com

Pre-emphasis list?

I'm sorry, is there a way to scour the entire thread for a specific release? Like a CTRL+F type of thing. I didn't see it on any list, but I am not...

forums.stevehoffman.tv

I have a few of the more prominent examples, like TOTO IV or a few Vangelis albums. There's actually a few more not found in either list:
France Gall - Babacar (1987) - Apache / 2292-42096-2 (TOC: yes)
Tchaikovsky: Symphony No. 5, Lorin Maazel conducting the Cleveland Orchestra ((P)/(C) 1982, apparently a 1983 release), CBS / CD 36700 (TOC: yes)
Heinz Balli – Organ Concert At The Stadtkirche St. Nikolaus, Frauenfeld (1983), DENON 38C37-7068 (TOC: ?, subcode: yes; unfortunately I don't have that one here to check TOC but I do have a rip with cuesheet)
ZELDA - 空色帽子の日 (1985, reissue 1994) - Sony Records / SRCL-3053 (TOC: yes)
Tomoko Aran: MORE RELAX (1985) - Warner Japan / ? (is on digital platforms like Amazon sans de-emphasis, duh!)
(What, OMD's Architecture & Morality was an emphasis CD? Oh no. I'll have to check. Hmm, mine - CDID 12, barcode 0777 7 86484 2 0 - does not have the emphasis flag set in the TOC according to EAC. Sound wise, it certainly could be. I'll have to fire up CUERipper and check the subcode. Hmm, nope, no FLAGS PRE anywhere in sight. By contrast, the above Tchaikovsky has that on every track in the cuesheet. Could this be one of the rare unflagged emphasis discs? How insidious. I tried it, and with DE applied, it certainly sounds a lot more like the 2003 remaster on Amazon, and still has plenty of treble left... I guess the flags accidentally got lost when the disc was reissued, given that it seems to have been released in correct form originally.)

Anyhow, since the pre-emphasis was applied using IIR (well, analog minimum phase) filters I would very much hope that any standard implementation of de-emphasis would be IIR as well.

 

[Ron Texas]

Might there be an issue with the RED BOOK de-emphasis filter?

I can't answer that. These days I'm using a MiniDSP 2x4. There is no harshness.

 

[AnalogSteph]

PS for posterity:

Heinz Balli – Organ Concert At The Stadtkirche St. Nikolaus, Frauenfeld (1983), DENON 38C37-7068 (TOC: ?, subcode: yes; unfortunately I don't have that one here to check TOC but I do have a rip with cuesheet)

Nope, nothing in TOC (which is how it's supposedly correct but annoying).

 

[thulle]

Pre and de-emphasis were relevant to a VERY limited number of CDs in the 1980s. I had a light on my CD player to indicate its use and not one CD in my collection triggered it. I seriously doubt it's been used in the past 30 years.

I wouldn't say it was that limited. Not used in the last 30 years is about right, looks like 1992 was the end of the line

I just had my first contact with a CD with pre-emphasis and did a bit of a deep-dive on how to handle ripping it. I came across this 2005 post:

Page Redirection

In my collection of ~ 5000 classical CDs I have discovered some 150 CDs with pre-emphasis so far (1 among ~33 - is this rare???).

Attention: It's often said that only recordings mastered before 1986 use pre-emphasis --> definitely wrong !!!
Some labels do use it just up to date (but not regularily) e.g. Erato and the Swedish label BIS (on almost every second CD) !!!

[...]

Pre-Emph is (still !) used frequently on Supraphon, BIS, Erato, New World Records and other smaller labels - especially from Northern & Eastern Europe.
( I believe you'll find it often, too on smaller jazz labels...)

 

[MC_RME]

Whoever uses Pre-Emphasis in 2025 does not do good service to his customers. With the very popular (and good, I might add) ESS9039Q2M DAC chip we have the first one (AFAIK) that no longer supports De-Emphasis. Others might follow...

 

[MRC01]

Whoever uses Pre-Emphasis in 2025 does not do good service to his customers. With the very popular (and good, I might add) ESS9039Q2M DAC chip we have the first one (AFAIK) that no longer supports De-Emphasis. Others might follow...

So if one plays an old CD having pre-emphasis with this DAC, it will sound excessively shrill/bright unless he has a way to apply the de-emphasis curve?
I suppose one would do this upstream from the DAC in DSP. First extend the 16-bit to 24-bit so you don't lose resolution when applying de-emphasis, then feed the 24-bit de-emphasized data to the DAC.

 

[MC_RME]

Well, emphasis is a limited treble boost, and 'shrill' seems a bit exaggerated. In fact with older recordings they do sound more like modern ones, except for the still missing bass. EQ can be used to replace the missing de-emphasis. Still, the conecpt of emhasis in audio makes no sense anymore in 2025.

 

[MRC01]

Well, emphasis is a limited treble boost, and 'shrill' seems a bit exaggerated. In fact with older recordings they do sound more like modern ones, except for the still missing bass. EQ can be used to replace the missing de-emphasis. Still, the conecpt of emhasis in audio makes no sense anymore in 2025.

Emphasis starts around 1 kHz and becomes +/- 10 dB at 20 kHz. Here's the curve. I linked that article as it has details enabling one to build his own de-emphasis with DSP.

I've listened to CDs that have it, but the player didn't apply it (whether the fault of the CD or the player). I also have a test CD with tracks having emphasis, one flagged the other not, which makes it easy to hear what it sounds like. IMO, "shrill" and "way too bright" are not exaggerated. The article linked above uses the terms "strident" and "harsh", which I would agree with. But it does depend on how sensitive one's hearing is to high frequencies.

 

[tmtomh]

A CD with pre-emphasis played back without de-emphasis is incredibly easy to detect by listening as @MRC01 notes - the effect is not subtle. If you feel that it is subtle or difficult to detect, then rip the CD and make two copies - one as-is, and one using software to apply the de-emphasis curve. When you play those two versions on your computer (or whatever), and switch back and forth, the difference will be quite stark, even if you have reduced high-frequency hearing because of age.

As to whether modern players should include pre/de-emphasis support, personally I like to see that feature, but I have to agree with @MC_RME - I don't think it's particularly realistic to expect manufacturers to support it. None of us blinks an eye when modern players don't support HDCD for example, and HDCD is a newer (or I should say less-old) feature than pre-emphasis. I have at least a couple hundred early CDs produced from 1983 through the early 1990s, and exactly one of them has pre-emphasis.

 

[olds1959special]

I use linear phase. My speakers are designed with correct phase alignment in mind, so why would I want to use anything other than linear phase?

 

[MRC01]

I use linear phase. My speakers are designed with correct phase alignment in mind, so why would I want to use anything other than linear phase?

Much (though not all) of the interaction of your speakers with your room is minimum phase, so if you're correcting room modes with parametric EQ, you may want to use minimum phase.