Page 1 of 2
Peak EQ parameters inconsistent?
Posted: Wed Mar 11, 2020 2:13 am
by dotnet
First post on this forum, so – hi everybody!
I'm planning to put the room EQ settings from my Mac based setup into the Neutron player. When having a play with the Peak EQ settings I noticed that switching from the Q to the BW parameter type gives incorrect results. Going the other way seems to work fine. For example, if I enter BW=1.388 as parameter and then switch to type Q I'm getting 1.0 (correct). If I switch back to BW I'm now at 0.126 (incorrect). If I go back to Q from there I'm at 11.462 (which corresponds to BW=0.126).
Am I misunderstanding something here, or is this a bug?
Re: Peak EQ parameters inconsistent?
Posted: Wed Mar 11, 2020 6:10 am
by blaubär
Hi ! Good you're here !
Lacking a specification we cannot know for sure. I'd say if it's a feature it's a strange feature, though. If it's a bug it's a minor one. I suggest you ask the developer at
neutronmp@gmail.com .
Re: Peak EQ parameters inconsistent?
Posted: Wed Mar 11, 2020 9:12 am
by dotnet
Thanks Blaubär, I would have assumed that converting between two equivalent expressions of this parameter is a reversible operation.
I've entered all filters using Q instead of BW, and the result sounds the same as from Audirvana on the Mac. So, mission accomplished I guess.
Another weird thing I found (I'm really new to Neutron) is that when I select LPCM as output format for my DLNA DAC, all that's ever sent (according to the display in the player) is 16 bit 96kHz PCM, even for higher resolution tracks. If I select FLAC, what's being sent is 24 bit, 192kHz PCM, i.e. not FLAC but exactly what I want. This seems like a UI bug, the wrong radio button selection is passed on.
Re: Peak EQ parameters inconsistent?
Posted: Wed Mar 11, 2020 9:37 am
by blaubär
dotnet wrote: ↑Wed Mar 11, 2020 9:12 am
Thanks Blaubär, I would have assumed that converting between two equivalent expressions of this parameter is a reversible operation.
I've entered all filters using Q instead of BW, and the result sounds the same as from Audirvana on the Mac. So, mission accomplished I guess.
Yes, that's a reasonable assumption. I suggest you contact the developer and tell him about the bug.
Re: Peak EQ parameters inconsistent?
Posted: Wed Mar 11, 2020 9:42 am
by blaubär
dotnet wrote: ↑Wed Mar 11, 2020 9:12 am
Another weird thing I found (I'm really new to Neutron) is that when I select LPCM as output format for my DLNA DAC, all that's ever sent (according to the display in the player) is 16 bit 96kHz PCM, even for higher resolution tracks. If I select FLAC, what's being sent is 24 bit, 192kHz PCM, i.e. not FLAC but exactly what I want. This seems like a UI bug, the wrong radio button selection is passed on.
I don't stream so I cannot try this out. But I don't really understand this : FLAC is a file format, PCM is a stream format ... so what do you mean when you say you select FLAC for the output of a stream ? Perhaps a screenshot might be helpful.
Re: Peak EQ parameters inconsistent?
Posted: Wed Mar 11, 2020 10:22 am
by dotnet
- IMG_50B3A82BB782-1.jpeg (78.04 KiB) Viewed 3934 times
- IMG_5E8CDE550548-1.jpeg (41.09 KiB) Viewed 3934 times
Re: Peak EQ parameters inconsistent?
Posted: Wed Mar 11, 2020 10:34 am
by blaubär
OK, thanks ! I suggest you contact the developer with regard to this, too.
Re: Peak EQ parameters inconsistent?
Posted: Wed Mar 11, 2020 11:05 am
by blaubär
As to the definition of "BW" and "Q" : it would seen that "BW" means "bandwidth in octaves" and "Q" means "quality factor". Also see
Relation between Q factor and bandwidth BW
Bandwidth in Octaves Versus Q in Bandpass Filters
Bandwidth is expressed in several ways: in frequency, as being so many Hertz wide; or in octaves, as being so many octaves (or fractional octave) wide; or in decades, as being so many decades (or fractional decade) wide. Far and away, the most common audio usage is to express bandwidth in octaves.
The
conversion from BW to Q seem to be straightforward. From Q to BW one needs to solve a quadratic equation, perhaps the algorithm takes the wrong of the two possible solutions.
- bw.PNG (26.43 KiB) Viewed 3904 times
- N-Q.PNG (8.07 KiB) Viewed 3923 times
- Q-N.PNG (10.61 KiB) Viewed 3922 times
- octaves.PNG (5.1 KiB) Viewed 3921 times
Code: Select all
N = LOG(SQRT((((2+(1/Q^2))^2)/4)-1)+(1/(2*Q^2))+1)/LOG(2)
Q = SQRT(POWER(2;N))/(POWER(2;N)-1)
Re: Peak EQ parameters inconsistent?
Posted: Wed Mar 11, 2020 5:24 pm
by blaubär
The developer told me :
Neutron does not convert BW and Q but it is in to-do to implement it.
Re: Peak EQ parameters inconsistent?
Posted: Fri Mar 13, 2020 6:23 am
by dotnet
I've been using the sinh (BW -> Q) and asinh (Q -> BW) formulas myself for this, and it seems that Neutron does the conversion from Q to BW correctly. Anyway, I just entered all EQ parameters for 16 filters using Q (which is what I was getting from Room EQ Wizard anyway), and it sounds the same to my ears as what I'm getting from Audirvana using BW. Hence, it's no big deal, now that I know what to do.
Thanks for your investigation!