EQs and difference in weight and transient
-
midi_transmission midi_transmission https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=298730
- KVRian
- 988 posts since 13 Feb, 2013
No have not tested it Ozone 9, as far as I know it's a not avoidable problem.pixel85 wrote: ↑Fri Mar 12, 2021 5:05 pmHave you tried Ozone EQ in Linear mode? Recently I did a test in RX (to see also spectral view) with a steep HPF on synthetic Kick and I couldn't hear any pre-ringing. I was really surprised. Maybe it's an isolated case.midi_transmission wrote: ↑Fri Mar 12, 2021 4:35 pmPre-ringing, I would never use an linear EQ for transients in the lows for example
google for 'pre ringing impulse response' with picture search and have a look at the impulse response
So far Ozone 9 is the winner but I didn't stop looking for maybe even better options.
* The test was made to find EQ that I can use to clean foley / field recordings with minimum possible impact on the sound as those files will be used in sound design and then compressed in the game engine (not much control over the compression process there).
** Even more surprise I had when I discovered that Analog mode added signal at 30Hz (Cut off freq of HPF) fading out for about 3 seconds
But if it's noticeable depends on on a lot of things of course and also on the monitoring and material. When you have a lot of noise in the signal it can cover it. Sometimes I hear it only with loud volume, sometime it's very easy to spot. I guess the conclusion like always is, try what works best in each situation, if it sounds right...
I would never say don't use a linear EQ for bass or that it's always worse, but it's usually not the tool what I like on bass, because small phase shifts are usually not a problem on a signal signal compared to pre-ringing.
But linear EQs are a must in other case, like multiple mic recordings of the same signal.
Just be aware of the possible advantages and disadvantages (pre-ringing vs. phase shift) and don't think it's always better than a minimal phase eq. They are similar but different, like a compressor vs. limter.
- KVRAF
- 2254 posts since 10 Jul, 2008 from Orbit SW US
As stated preringing with linear phase eq is at present unavoidable. i'd be happy if someone proves me wrong.
gadgets an gizmos..make noise https://soundcloud.com/crystalawareness Restocked: 3/24
old stuff http://ww.dancingbearaudioresearch.com/
if this post is edited -it was for punctuation, grammar, or to make it coherent (or make me seem coherent).
old stuff http://ww.dancingbearaudioresearch.com/
if this post is edited -it was for punctuation, grammar, or to make it coherent (or make me seem coherent).
-
- KVRist
- 99 posts since 5 Jan, 2006 from Kingston, Sweden
Weiss EQ1
Vertigo VSE-2
Airwindows Baxandall
Trident CB9066
Schaack AnalogQ (analog phase)
Magpha
Crave (transparent mode)
Vertigo VSE-2
Airwindows Baxandall
Trident CB9066
Schaack AnalogQ (analog phase)
Magpha
Crave (transparent mode)
Dub it up blacker than dread
-
- KVRAF
- 3318 posts since 16 Jan, 2005 from Ottawa, Ontario
-
- KVRian
- 986 posts since 8 Mar, 2009
How are we defining weight and transients here? as far as i understand with completely minimal EQ's is both are completely dependent on the curves/Q and magnitude of the band, which determine the phase, group delay, and impulse response for the input sound..if some of these EQ's are completely MP (i don't know if they are or not) then you should be able to match the impulse with any other entirely MP EQ as the implementation is the same on the completely passive side of things.
I know in theory a lot of analog EQ's are MP but i have never gotten a plugin EQ to pull out transients like an API 5500 can for instance regardless of the curves. I've gotten a bit closer with freephase in EQuilibrium(only EQ i use now)but there's always a tradeoff between ringing either side of the cutoff point, and the accuracy of that EQ node to accomplish something similar. Both can be improved by increasing the kernel length, but that isn't practical for sound design or mixing in realtime
I'll check out the Weiss though. Curious to see if its just MP or not
I know in theory a lot of analog EQ's are MP but i have never gotten a plugin EQ to pull out transients like an API 5500 can for instance regardless of the curves. I've gotten a bit closer with freephase in EQuilibrium(only EQ i use now)but there's always a tradeoff between ringing either side of the cutoff point, and the accuracy of that EQ node to accomplish something similar. Both can be improved by increasing the kernel length, but that isn't practical for sound design or mixing in realtime
I'll check out the Weiss though. Curious to see if its just MP or not
I
-
- KVRAF
- 1863 posts since 11 Apr, 2008
I'm just reading a topic on gears**** (forbidden word! ) and it seems that pre-ringing can be avoided in certain cases. It all depends on what filter, what frequency and what signal is processed.CrystalWizard wrote: ↑Fri Mar 12, 2021 11:27 pm As stated preringing with linear phase eq is at present unavoidable. i'd be happy if someone proves me wrong.
This could explain why in my test I couldn't hear it.
Fabien from TDR is of course the best knowledge compendium in that matter
-
- KVRAF
- 6428 posts since 17 Dec, 2009
Avoided =! below significant audible threshold. If you use gentle filtering and restrained gain then of course, it will ring less - but in that case, there would also be much less phase distortion with a minimal phase EQ.pixel85 wrote: ↑Sat Mar 13, 2021 8:44 amI'm just reading a topic on gears**** (forbidden word! ) and it seems that pre-ringing can be avoided in certain cases. It all depends on what filter, what frequency and what signal is processed.CrystalWizard wrote: ↑Fri Mar 12, 2021 11:27 pm As stated preringing with linear phase eq is at present unavoidable. i'd be happy if someone proves me wrong.
This could explain why in my test I couldn't hear it.
Fabien from TDR is of course the best knowledge compendium in that matter
It's pretty corelated.
Neither phase distortion nor pre-ringing is extra-problematic when using 6dB/8ve filters and +/-3dB gain.
When you start doing 24dB and 12dB +/- you get either peaks and smearing, or audible preringing.
-
- KVRAF
- 1863 posts since 11 Apr, 2008
Well, it's more than just that.
I don't want to quote FabienTDR here without his permission but there's a fresh topic on another forum where he's explaining more precisely how LP filters work and what they do. Topic name is " Linear phase pre-ringing: Does it effect all frequencies?" for anyone interested.
I don't want to quote FabienTDR here without his permission but there's a fresh topic on another forum where he's explaining more precisely how LP filters work and what they do. Topic name is " Linear phase pre-ringing: Does it effect all frequencies?" for anyone interested.
-
midi_transmission midi_transmission https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=298730
- KVRian
- 988 posts since 13 Feb, 2013
I've also experienced (the 2 times I had the opportunity to use a good one...) that a practical difference to digital eqs is, that analog eqs often have a perceived speed how they grab transients for some technical reason I don't know. I think this is one reason why they are still in popular for some cases as a special tool, but there is a reason why clean digital eqs got popular before the daw age too. like MP and LP, everything has advantages and disadvantages...TIMT wrote: ↑Sat Mar 13, 2021 6:29 am How are we defining weight and transients here? as far as i understand with completely minimal EQ's is both are completely dependent on the curves/Q and magnitude of the band, which determine the phase, group delay, and impulse response for the input sound..if some of these EQ's are completely MP (i don't know if they are or not) then you should be able to match the impulse with any other entirely MP EQ as the implementation is the same on the completely passive side of things.
I know in theory a lot of analog EQ's are MP but i have never gotten a plugin EQ to pull out transients like an API 5500 can for instance regardless of the curves. I've gotten a bit closer with freephase in EQuilibrium(only EQ i use now)but there's always a tradeoff between ringing either side of the cutoff point, and the accuracy of that EQ node to accomplish something similar. Both can be improved by increasing the kernel length, but that isn't practical for sound design or mixing in realtime
I'll check out the Weiss though. Curious to see if its just MP or not
Last edited by midi_transmission on Sat Mar 13, 2021 6:03 pm, edited 1 time in total.
- KVRAF
- 6961 posts since 28 Dec, 2015 from Atlantis Island
To make a long story short: just use Tokyo Dawn Labs plugins.
https://sonograyn.bandcamp.com/music Experimental Ambient
https://martinjuenke.bandcamp.com/music Alternative Instrumental
https://martinjuenke.bandcamp.com/music Alternative Instrumental
-
- KVRAF
- 6428 posts since 17 Dec, 2009
Which section is it under?pixel85 wrote: ↑Sat Mar 13, 2021 5:53 pm Well, it's more than just that.
I don't want to quote FabienTDR here without his permission but there's a fresh topic on another forum where he's explaining more precisely how LP filters work and what they do. Topic name is " Linear phase pre-ringing: Does it effect all frequencies?" for anyone interested.
-
- KVRian
- 986 posts since 8 Mar, 2009
I don't think there is, or at least not anything that would be useful in the practice of utilising them in a mixdown. "pre-ringing"as it generally is referred to is just the group delay occuring before the audio input. I think giving it a different name confuses matters, or implies the artefact is different, but it's exact same one produced by minimum phase IIR filters, but it happens before your audio input. It really should be called "pre-group delay"
The only thing you can do to limit the group delay "ringing"before the audio is by windowing it or reducing the length of the tap, but then you decrease its accuracy in terms of attenuation or boost and depending on the windowing function, can introduce ripple into the passband if the impulse is truncated and the windowing is lax
You can intuit all this just from fiddling around with the FIR section of EQuilibrium. It's a great educational tool, as well as a do it all plugin EQ
I
-
- KVRAF
- 3089 posts since 4 May, 2012
Well explained.TIMT wrote: ↑Sat Mar 13, 2021 6:36 pmI don't think there is, or at least not anything that would be useful in the practice of utilising them in a mixdown. "pre-ringing"as it generally is referred to is just the group delay occuring before the audio input. I think giving it a different name confuses matters, or implies the artefact is different, but it's exact same one produced by minimum phase IIR filters, but it happens before your audio input. It really should be called "pre-group delay"
The only thing you can do to limit the group delay "ringing"before the audio is by windowing it or reducing the length of the tap, but then you decrease its accuracy in terms of attenuation or boost and depending on the windowing function, can introduce ripple into the passband if the impulse is truncated and the windowing is lax
You can intuit all this just from fiddling around with the FIR section of EQuilibrium. It's a great educational tool, as well as a do it all plugin EQ
FIR filter topology is very easy to follow and there are excellent design programs like the free ones produced by Iowa Hills: http://www.iowahills.com/ - which really speeds up implementation for those who want to dabble. A lot of useful information on that site for understanding and making digital filters in general.