Go Back   Cockos Incorporated Forums > REAPER Forums > REAPER General Discussion Forum

Reply
 
Thread Tools Display Modes
Old 02-05-2019, 03:46 PM   #41
lolilol1975
Human being with feelings
 
Join Date: Dec 2015
Posts: 1,739
Default

ToneBoosters EQ has the same behaviour as ProQ.
lolilol1975 is offline   Reply With Quote
Old 02-05-2019, 10:45 PM   #42
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,690
Default

Quote:
Originally Posted by antti View Post
You'll see just badly distorted the shape ...
Any EQ curve is a "badly distorted shape" from some POV, as a perfect recording and playback should not modify the natural sound.
Hence the final overall frequency response of an EQ (setting) should be judged from the intended audio effect and considering any side-effects, but not from the seeming optical "beauty" of the frequency response curve.

-Michael
mschnell is online now   Reply With Quote
Old 02-06-2019, 01:02 AM   #43
antti
Human being with feelings
 
Join Date: Nov 2008
Posts: 29
Default

Quote:
Originally Posted by mschnell View Post
Any EQ curve is a "badly distorted shape" from some POV, as a perfect recording and playback should not modify the natural sound.
Hence the final overall frequency response of an EQ (setting) should be judged from the intended audio effect and considering any side-effects, but not from the seeming optical "beauty" of the frequency response curve.
Go ahead and try it. Pay particularly attention to what happens to the gain an octave below the center frequency.

Now I’m not saying that makes it useless but to claim it’s purely of theoretical or aesthetical importance is just silly.
antti is offline   Reply With Quote
Old 02-06-2019, 02:34 AM   #44
winbe
Human being with feelings
 
Join Date: May 2006
Posts: 1,361
Default

Quote:
Originally Posted by Judders View Post
It has a lot more options of filter type. A fairly big update was released recently, I found out last night, so if you downloaded it a while ago it's worth getting the new version (free update).

I much prefer it to ReaEQ, it does have decramped filters as well as a lot more features. Best way to decide is to experiment though.
Thanks, I used it yesterday, and the latest version is really nice, improved GUI, extremely intuitive / fast use. I immediately appreciated to be able to use steeper curves, and also to be able to monitor another track at the same time to see problematic overlapping frequences (I was dealing with a commmon kick / bass problem).
__________________
Cedric Simon aka FrozenjaZz
Fresh beats - www.frozenjazz.com
winbe is offline   Reply With Quote
Old 02-06-2019, 03:21 AM   #45
Judders
Human being with feelings
 
Join Date: Aug 2014
Posts: 11,044
Default

Quote:
Originally Posted by winbe View Post
Thanks, I used it yesterday, and the latest version is really nice, improved GUI, extremely intuitive / fast use. I immediately appreciated to be able to use steeper curves, and also to be able to monitor another track at the same time to see problematic overlapping frequences (I was dealing with a commmon kick / bass problem).
Yeah, it's my go-to for low overhead clean EQ.
Judders is offline   Reply With Quote
Old 02-06-2019, 03:39 AM   #46
Dannii
Human being with feelings
 
Dannii's Avatar
 
Join Date: Mar 2010
Location: Adelaide, South Australia (originally from Geelong)
Posts: 5,598
Default

After watching this video, Dan actually convinced me that there's no real need for me to rush out and buy ProQ right now. ReaEQ is my go-to EQ for non character duties and I already have SlickEQ (GE and M editions) which have similar characteristics to ProQ in natural phase mode.
I also have Waves F6 for dynamic EQ work and really like it. I have a few other EQs too but those I've mentioned are my primary ones.
__________________
Dannii is offline   Reply With Quote
Old 02-06-2019, 06:10 AM   #47
chip mcdonald
Human being with feelings
 
chip mcdonald's Avatar
 
Join Date: May 2006
Location: NA - North Augusta South Carolina
Posts: 4,294
Default

I would like to know if this is a side effect of math at 44.1 versus oversampling?

If that is the case, I would then want to know if the same effect doesn't happen when you sum high frequency information that overlaps out to the Nyquist limit?
I hear the difference in the 10k example as a boost in the lower level content (the reverb tail). Since Worral's test is on dynamic material, I'd like to know


1) is this effect linear at all volume ranges, and also with 2 non-correlated signals (since rise/fall combinations at different levels may scale differently?)

2) Is the math that causes this phenomenon the result of summing that is also used in adding 2 tracks together with HF content?

In other words - are we possibly getting a small HF boost headed out to Nyquist as signal drops/modulates when summing 2 tracks together? This accumulating would explain a lot of voodoo... <g>


/ Yes, I could test this myself, but no, I'm probably not.
// apologies for a possible lack of coherence, I have only been awake for a few minutes.
__________________
]]] guitar lessons - www.chipmcdonald.com [[[
WEAR A FRAKKING MASK!!!!

Last edited by chip mcdonald; 02-06-2019 at 06:22 AM.
chip mcdonald is offline   Reply With Quote
Old 02-06-2019, 08:37 AM   #48
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,690
Default

Quote:
Originally Posted by antti View Post
it’s purely of theoretical or aesthetical importance is just silly.
I did not say this. I said that (for reasons I gave above) not modifying the volume close to Nyquist might be a potentially beneficial behavior that might be intended by the developers.

-Michael
mschnell is online now   Reply With Quote
Old 02-06-2019, 08:47 AM   #49
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,690
Default

Quote:
Originally Posted by chip mcdonald View Post
I would then want to know if the same effect doesn't happen when you sum high frequency information that overlaps out to the Nyquist limit?
Technically and by definition, there is no signal with frequency above Nyquist in a digital audio stream. As an EQ is not supposed to generate any signal that is not existing before, there will not be such a thing at all.

Nonlinear gear (such as compressor) can generate signals above Nyquist. But these need to be prevented from propagating (e.g. by limiting the speed of the amplifier control signal.) Oversampling is a means to improve this. Of course oversampling will switch to a different Nyquist, and the following downsampling will need to suppress any signal above the original Nyquist.

-Michael
mschnell is online now   Reply With Quote
Old 02-06-2019, 09:28 AM   #50
antti
Human being with feelings
 
Join Date: Nov 2008
Posts: 29
Default

Quote:
Originally Posted by mschnell View Post
I did not say this. I said that (for reasons I gave above) not modifying the volume close to Nyquist might be a potentially beneficial behavior that might be intended by the developers.
We know it isn’t and wasn’t, for two reasons:
1) For high shelving filter the cramping is equivalent to somewhat warped (f0 and bw change) continuous time filter response coupled with sharp additional high frequency boost (or cut if the shelf is a cut) near nyquist. The amplitude of this extra boost approaches half of the total boost amount (so extra boost of 5 dB for a 10 dB high shelf). Nobody in their right mind would claim this is desired.

2) The response of ReaEQ matches that of the Robert Bristow-Johnson ”cookbook” EQ filters he published back in 94 and which have been implemented a gazillion times. They were the only published simple closed form solutions until the last 10 years or so, so it’s natural Justin & co used them in ReaEQ. More or less every developer gets rid of the cramping once they’re capable of that (starting with Waves with Renaissance EQ around the turn of the millennium).
antti is offline   Reply With Quote
Old 02-08-2019, 07:50 AM   #51
chip mcdonald
Human being with feelings
 
chip mcdonald's Avatar
 
Join Date: May 2006
Location: NA - North Augusta South Carolina
Posts: 4,294
Default

Quote:
Originally Posted by mschnell View Post
Technically and by definition, there is no signal with frequency above Nyquist in a digital audio stream.
I know that. I wrote out *to* the Nyquist limit.

I'm asking if the math that is doing what I presume is a quad/bi quad function to get more energy in the spectrum *towards* Nyquist also accumulate in a non-linear way when tracks are being summed, creating a similar cramping effect in the high end from multiple tracks with HF boosts either lower than the examples given in Worral's video or as mathematical artifacts from multiples accumulating.
__________________
]]] guitar lessons - www.chipmcdonald.com [[[
WEAR A FRAKKING MASK!!!!

Last edited by chip mcdonald; 02-08-2019 at 07:56 AM.
chip mcdonald is offline   Reply With Quote
Old 02-09-2019, 11:49 PM   #52
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,690
Default

Right you are.

While I don't consider the asymmetry of the resulting EQ curve with "band" as anything bad (when plotting it at a lin scale instead of log a "perfect" band curve asymmetrical, anyway ), adding yet another "peak" near Nyqust when a high center frequency is set is decently nasty.

They at least should prevent such center frequencies from being able to be set.

-Michael

Last edited by mschnell; 02-10-2019 at 04:57 AM.
mschnell is online now   Reply With Quote
Old 02-10-2019, 04:24 AM   #53
Bouroki
Human being with feelings
 
Join Date: Jun 2013
Posts: 79
Default

Something funny I've just noticed. Many (if not most) of the EQs that have decramped bells without oversampling (Orfanidis?) actually have what appears to be cramping high shelves! Most notably Toneboosters, Voxengo, Waves. Funnily, Kazrog ValvEQ also cramps even though it has 8x OS for the saturation (d'oh!). DMG Equick and OvertoneDSP AF2-10 were the only exceptions. In terms of sound, Equick high shelf sounded so much better than the others that I could 100% tell in blind tests.

Anyone (antti?) can shed some light on this?
Bouroki is offline   Reply With Quote
Old 02-10-2019, 04:59 AM   #54
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,690
Default

Quote:
Originally Posted by Bouroki View Post
... saturation ...
So it's not an EQ, but some kind of effect plugin.

-Michael
mschnell is online now   Reply With Quote
Old 02-10-2019, 06:04 AM   #55
Bouroki
Human being with feelings
 
Join Date: Jun 2013
Posts: 79
Default

Quote:
Originally Posted by mschnell View Post
So it's not an EQ, but some kind of effect plugin.

-Michael
I mean its eq section doesn't even make use of its own oversampling cycle.
Bouroki is offline   Reply With Quote
Old 02-10-2019, 07:35 AM   #56
cyrano
Human being with feelings
 
cyrano's Avatar
 
Join Date: Jun 2011
Location: Belgium
Posts: 5,246
Default

Quote:
Originally Posted by ashcat_lt View Post
I always lowpass below nyquist anyway.
Seems like a wise thing to do.

Me, I just don't worry that much. Sure, there's a difference in these fairly extreme examples. But which one is better?

First, you worry about a clean eq, then you throw on a dirty compressor? Or whatever else?
__________________
In a time of deceit telling the truth is a revolutionary act.
George Orwell
cyrano is offline   Reply With Quote
Old 02-10-2019, 08:33 AM   #57
antti
Human being with feelings
 
Join Date: Nov 2008
Posts: 29
Default

Quote:
Originally Posted by Bouroki View Post
Something funny I've just noticed. Many (if not most) of the EQs that have decramped bells without oversampling (Orfanidis?) actually have what appears to be cramping high shelves! Most notably Toneboosters, Voxengo, Waves. Funnily, Kazrog ValvEQ also cramps even though it has 8x OS for the saturation (d'oh!). DMG Equick and OvertoneDSP AF2-10 were the only exceptions. In terms of sound, Equick high shelf sounded so much better than the others that I could 100% tell in blind tests.

Anyone (antti?) can shed some light on this?
What eaxctly do you mean by cramping shelves? The same thing that ReaEQ does? Or the response just tapering flat right at nyquist?

The second is inevitable for any digital IIR filter. The first I assume would be due to the fact that there have’t been any published methods to design a symmetric ”analog matched” shelving eq. David P. Berners and Jonathn S. Abel at Universal Audio published an iterative method to do that for a shelf having one resonant pole or zero pair. I have another method that works for the cases where the shelf doesn’t have any overshoot (such as with ReaEQ). Making a resonant shelf that’s symmetric around e halfway point and decramped is not an easy thing.
antti is offline   Reply With Quote
Old 02-10-2019, 09:17 AM   #58
Bouroki
Human being with feelings
 
Join Date: Jun 2013
Posts: 79
Default

Quote:
Originally Posted by antti View Post
What eaxctly do you mean by cramping shelves? The same thing that ReaEQ does? Or the response just tapering flat right at nyquist?

The second is inevitable for any digital IIR filter. The first I assume would be due to the fact that there have’t been any published methods to design a symmetric ”analog matched” shelving eq. David P. Berners and Jonathn S. Abel at Universal Audio published an iterative method to do that for a shelf having one resonant pole or zero pair. I have another method that works for the cases where the shelf doesn’t have any overshoot (such as with ReaEQ). Making a resonant shelf that’s symmetric around e halfway point and decramped is not an easy thing.
I mean the slope becomes gradually steeper and steeper as it approaches nyquist (so like ReaEQ I guess). However DMG EQuick does not! It remains the same steepness and it definitely sounds much better. EQuick is zero latency IIR minimum phase so I guess Dave has found his own method?
Bouroki is offline   Reply With Quote
Old 02-10-2019, 12:06 PM   #59
sostenuto
Human being with feelings
 
sostenuto's Avatar
 
Join Date: Apr 2011
Location: St George, UT _ USA
Posts: 2,881
Default

Enjoy this tutorial, but struggle mightily early-on.

So IIEQ Pro or EQuick to keep me close, and <$99. ??

Last edited by sostenuto; 02-10-2019 at 12:34 PM.
sostenuto is offline   Reply With Quote
Old 02-10-2019, 12:16 PM   #60
antti
Human being with feelings
 
Join Date: Nov 2008
Posts: 29
Default

Quote:
Originally Posted by sostenuto View Post
Enjoy this tutorial, but struggle mightily early-on.

So IIEQ Pro or EQuick to keep me close, and < $199. ??
When it comes to EQs, you can pry Pro-Q 3 from my cold dead fingers.
antti is offline   Reply With Quote
Old 02-10-2019, 12:33 PM   #61
sostenuto
Human being with feelings
 
sostenuto's Avatar
 
Join Date: Apr 2011
Location: St George, UT _ USA
Posts: 2,881
Default

FabFilter quality /capability not questioned.
Typo'd earlier post; now editing. Should have stated __ <$99.

Of course Pro-Q 3 purchase long ago would been much cheaper than bag of EQ wanna-be's now taking HDD space.
sostenuto is offline   Reply With Quote
Old 02-10-2019, 02:41 PM   #62
antti
Human being with feelings
 
Join Date: Nov 2008
Posts: 29
Default

Quote:
Originally Posted by sostenuto View Post
FabFilter quality /capability not questioned.
Typo'd earlier post; now editing. Should have stated __ <$99.

Of course Pro-Q 3 purchase long ago would been much cheaper than bag of EQ wanna-be's now taking HDD space.
PM me if 10% off (135e total) would make you pull the trigger.

My own opinion is of course that buying other EQs is a waste of money and you’re best off either sticking to ReaEQ / other free EQs or going straight to Pro- Q 3. The workflow advantages are just superb.
antti is offline   Reply With Quote
Old 02-11-2019, 05:09 PM   #63
nitsuj
Human being with feelings
 
nitsuj's Avatar
 
Join Date: Nov 2017
Posts: 292
Default

Greetings all! Great discussion. Just my cup of tea.

I'm still actively developing ReEQ, time permitting. At the moment I'm improving band listen/solo and I'm also adding extra filter types.

Dan's video was great, I particularly enjoyed the analysis of phase that he did.

With regards to decramped filters which appears to have become a topic here, I took what might be classed as the easy way out using oversampling. I did see RBJ recommend oversampling to sort out cramping at Nyquist and I also know that was the route taken with Ableton's EQ8 which kind of inspired me in the first place (at least from a DSP implementation point of view). It adds latency and a x 2 hit on CPU processing for the filters but you get the curves and the phase close to analog.

I think the oversampling in ReEQ is close to Pro Q3's Natural Phase mode which also has latency but is close to analog. As Dan showed in his video, Zero Latency mode improves filter response but doesn't preserve phase. There's a debate to be had about whether this matters and if you can actually tell or not.

I did look at published work by Sophocles J. Orfanidis and also the BiQuadFits.pdf amongst others. I wouldn't be happy implementing a decramped solution without having a solution for all of the filter types including shelves and things like one pole filters.

If anyone (antti ?) would care to forward any code math for a decramped filter set I'd happily move implementation up the priority tree in order to offer it to the community in ReEQ. If not, I'll probably chip away at workflow, filter types and other usability improvements before tackling it.

FWIW, I'm also considering tackling linear phase and dynamic eq at some point although I think these are longer term. If anyone here has any idea about how to implement linear phase from, say where ReEQ is now, I'm all ears. I can't seem to find much material online.

---
ReEQ: https://forum.cockos.com/showthread.php?t=213501
nitsuj is offline   Reply With Quote
Old 02-11-2019, 11:06 PM   #64
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,690
Default

Quote:
Originally Posted by nitsuj View Post
FWIW, I'm also considering tackling linear phase and dynamic eq at some point although I think these are longer term. If anyone here has any idea about how to implement linear phase ...
Besides implementation details, it would be interesting to discuss the pros and cons for "linear phase" vs "natural" filters.

I understand that linear phase filters can be done as FIR filters using FFT, and "pure" FFT FIR filters obviously are limited to work only on frequency above the (inverse of the) latency they introduce.

-Michael
mschnell is online now   Reply With Quote
Old 02-12-2019, 03:04 AM   #65
antti
Human being with feelings
 
Join Date: Nov 2008
Posts: 29
Default

Quote:
Originally Posted by nitsuj View Post
I think the oversampling in ReEQ is close to Pro Q3's Natural Phase mode which also has latency but is close to analog.
As it happens, my old professor just directed me to a paper from the previous AES convention in May 2018: John Flynn & Joshua D. Reiss, "Improving the frequency response magnitude and phase of
analogue-matched digital filters".

It describes what is almost certainly what the Natural Phase mode does: Design a regular minimum phase matched IIR EQ and follow that up with a longish FIR that matches the complex frequency response, thus fixing phase (which is already almost exact at low frequencies) and any remaining frequency response error. Multiple EQ bands can be combined in the same FIR correction filter (the final frequency response is matched), so it seems to match very well with observed cpu usage behaviour of Pro-Q 3.


Quote:
Originally Posted by nitsuj View Post
As Dan showed in his video, Zero Latency mode improves filter response but doesn't preserve phase. There's a debate to be had about whether this matters and if you can actually tell or not.
I'm firmly in the camp that EQ phase response has always been a red herring and is only considered a problem due to wildly spread misunderstandings by people with no education in the topic. Plugin makers are naturally adjusting to customer demand but you can see Fabfilter for example are careful to avoid claiming matching it is actually needed.

Quote:
Originally Posted by nitsuj View Post
If anyone here has any idea about how to implement linear phase from, say where ReEQ is now, I'm all ears. I can't seem to find much material online.
Linear phase EQs seem to be more or less close to textbook frequency sampling FIR design method coupled with some smoothing to remove excessive ripple from the resulting response.

Quote:
Originally Posted by nitsuj View Post
If anyone (antti ?) would care to forward any code math for a decramped filter set
I may end up writing a paper on it, so you'll have to wait until that.
antti is offline   Reply With Quote
Old 02-12-2019, 08:46 AM   #66
nitsuj
Human being with feelings
 
nitsuj's Avatar
 
Join Date: Nov 2017
Posts: 292
Default

Quote:
Originally Posted by mschnell View Post
Besides implementation details, it would be interesting to discuss the pros and cons for "linear phase" vs "natural" filters.

I understand that linear phase filters can be done as FIR filters using FFT, and "pure" FFT FIR filters obviously are limited to work only on frequency above the (inverse of the) latency they introduce.

-Michael
Yes, although my motivation for implementing linear phase would be that people want to use the eq in that way. Often, so I hear, for some cases when mastering. There's plenty of material online talking about issues like pre-ringing etc.

I guess what I'm looking for is references to converting IIR filters to FIR filters.
nitsuj is offline   Reply With Quote
Old 02-12-2019, 09:28 AM   #67
antti
Human being with feelings
 
Join Date: Nov 2008
Posts: 29
Default

Quote:
Originally Posted by nitsuj View Post
I guess what I'm looking for is references to converting IIR filters to FIR filters.
I don’t think people actually do this but instead natively design a FIR filter that matches the spectrum of a set of analog filters. Afterall, the latency, and hence FIR length, doesn’t change with the number or types of EQ bands.
antti is offline   Reply With Quote
Old 02-12-2019, 11:14 AM   #68
X-Raym
Human being with feelings
 
X-Raym's Avatar
 
Join Date: Apr 2013
Location: France
Posts: 9,875
Default

Offtopic: Dan has just release a video about ReaComp: https://forum.cockos.com/showthread....04#post2094004
X-Raym is offline   Reply With Quote
Old 02-12-2019, 02:50 PM   #69
nitsuj
Human being with feelings
 
nitsuj's Avatar
 
Join Date: Nov 2017
Posts: 292
Default

Quote:
Originally Posted by antti View Post
I don’t think people actually do this but instead natively design a FIR filter that matches the spectrum of a set of analog filters. Afterall, the latency, and hence FIR length, doesn’t change with the number or types of EQ bands.
Ah, I don't think I worded that correctly. What I meant was, I'll have to work out how to generate the filter values for all the filter types I'm supporting for the FIR.
nitsuj is offline   Reply With Quote
Old 02-12-2019, 03:04 PM   #70
antti
Human being with feelings
 
Join Date: Nov 2008
Posts: 29
Default

Quote:
Originally Posted by nitsuj View Post
Ah, I don't think I worded that correctly. What I meant was, I'll have to work out how to generate the filter values for all the filter types I'm supporting for the FIR.
The question you want to ask is "How do I calculate the spectrum of the continuous time filter for each type?". And the answer turns out to be remarkably simple since you're only dealing with the analog prototypes, for which the H(s) is much simpler.

The rest is "merely" designing a FIR filter to match the combined response of all enabled bands. Most good DSP textbooks should have enough information on frequency sampling FIR filter design to get you started there.

We actually do something related at the company I work for, but in our case we measure the system frequency response and design a compensation filter for that. In practise we calculate the wanted inverse filter response magnitude spectrum, take IFFT and transform it to minimum phase (we know the system is almost perfectly minimum phase and also require low latency).
antti is offline   Reply With Quote
Old 02-18-2019, 11:06 PM   #71
yep
Human being with feelings
 
Join Date: Aug 2006
Posts: 2,019
Default

Quote:
Originally Posted by nitsuj View Post
...FWIW, I'm also considering tackling linear phase and dynamic eq at some point although I think these are longer term...
Personal opinion, FWIW:

Linear-phase and dynamic eq are specialty tools, and there are 3rd-party solutions for those that need them.

Regular, plain-jane minimum-phase EQ is basic functionality for a DAW, mixing console, or even a typical cassette 4-track, back when those were a thing.

I recently moved my studio and changed around a bunch of stuff, including upgrading the monitors and switching consoles and a bunch of things. Since moving and switching to Barefoot monitors, I have repeatedly noticed this barely-perceptible thing in the extreme highs, a kind of digital-sounding something, almost like modem noise just past the threshold of audibility. It's barely audible to my ears, and doesn't show up on every project, and nobody else has noticed or complained, but it's ugly, when I do hear it.

I was suspecting my console, or even the building electrical (a lot has changed).

But now, after watching this video, I am starting to think that what I might be hearing is a combination of better monitoring, and a lot of extreme boosting near Nyquist using ReaEQ, and then pushing that into heavy compression, which is something I do kind of a lot.

A pretty common scenario for me is to record stuff either with ribbon mics or with condensers with the highs rolled off via shelf or lo-pass (console or digital), and then boost something up in the 12-15k "air" frequencies to kind of "un-muffle" the sound, while still suppressing the peaky/harsh stuff in the 5-10k range. I do this with drum overheads, guitar amps, acoustic guitars, anything prone to presence-range harshness, and I tend to be pushing all that stuff into some kind of bus compression, along with track insert compression.

This video has got me thinking that maybe the weird high-end glitchiness I am hearing is due to pushing these cramping artifacts into multiple layers of compression. That would explain why I'm not noticing them when first applying the boost, and why I'm more hearing it as a generalized ugliness at mixdown, since it's probably happening across a whole bunch of tracks in what might be a 100-track project. All of my layers of bus compression are just gradually pushing a very subtle set of artifacts up into an audible range, or something like that...

In any case, back to the topic, I would personally vote for making ReaEQ behave like an analog EQ before adding linear-phase or dynamic eq. I think I'm going to stop using ReaEQ as my default channel EQ, which is a shame, because I love the UI and functionality.
yep is offline   Reply With Quote
Old 02-19-2019, 07:00 AM   #72
nitsuj
Human being with feelings
 
nitsuj's Avatar
 
Join Date: Nov 2017
Posts: 292
Default

Quote:
Originally Posted by yep View Post

This video has got me thinking that maybe the weird high-end glitchiness I am hearing is due to pushing these cramping artifacts into multiple layers of compression. That would explain why I'm not noticing them when first applying the boost, and why I'm more hearing it as a generalized ugliness at mixdown, since it's probably happening across a whole bunch of tracks in what might be a 100-track project. All of my layers of bus compression are just gradually pushing a very subtle set of artifacts up into an audible range, or something like that...

In any case, back to the topic, I would personally vote for making ReaEQ behave like an analog EQ before adding linear-phase or dynamic eq. I think I'm going to stop using ReaEQ as my default channel EQ, which is a shame, because I love the UI and functionality.
I agree. In the meantime you might want to give this a spin if you haven't (disclaimer: I'm the author)

https://forum.cockos.com/showthread.php?t=213501

In default mode it preserves analog filter shape and phase at the cost of a little latency. It's a JSFX, so performance is dependent on the machine you're running it on.
nitsuj is offline   Reply With Quote
Old 02-19-2019, 12:00 PM   #73
Jeffsounds
Human being with feelings
 
Jeffsounds's Avatar
 
Join Date: Feb 2013
Location: Northeast Michigan
Posts: 3,460
Default

Quote:
Originally Posted by nitsuj View Post
I agree. In the meantime you might want to give this a spin if you haven't (disclaimer: I'm the author)

https://forum.cockos.com/showthread.php?t=213501

In default mode it preserves analog filter shape and phase at the cost of a little latency. It's a JSFX, so performance is dependent on the machine you're running it on.
Nice! But I'm getting 14.0% CPU usage average with ReEQ on Win7 machine.
__________________
"TV has become nothing more than a Petri dish where this country grows its idiots." -Dr. John Becker
My First CD On Spotify - Side O' The Highway
Jeffsounds is offline   Reply With Quote
Old 02-19-2019, 01:17 PM   #74
clepsydrae
Human being with feelings
 
clepsydrae's Avatar
 
Join Date: Nov 2011
Posts: 3,409
Default

Quote:
Originally Posted by nitsuj View Post
It's a JSFX, so performance is dependent on the machine you're running it on.
Is there a machine dependency related to JSFX's as distinct from other plugin types?
clepsydrae is offline   Reply With Quote
Old 02-19-2019, 07:19 PM   #75
yep
Human being with feelings
 
Join Date: Aug 2006
Posts: 2,019
Default

Quote:
Originally Posted by nitsuj View Post
I agree. In the meantime you might want to give this a spin if you haven't (disclaimer: I'm the author)

https://forum.cockos.com/showthread.php?t=213501

In default mode it preserves analog filter shape and phase at the cost of a little latency. It's a JSFX, so performance is dependent on the machine you're running it on.
That looks very cool! Thanks!
yep is offline   Reply With Quote
Old 02-20-2019, 11:45 AM   #76
mawi
Human being with feelings
 
Join Date: Apr 2011
Location: Germany
Posts: 1,185
Default

Doesn't the PluginDoctor show the phase upside down? Doesn't it have to be inverted?

Sorry, I'm confused because Christian Budde's vst plugin analyser shows the phase upside down.

Which phase display is the right one?

Last edited by mawi; 02-20-2019 at 12:58 PM.
mawi is offline   Reply With Quote
Old 02-23-2019, 02:41 AM   #77
pipelineaudio
Mortal
 
pipelineaudio's Avatar
 
Join Date: Jan 2006
Location: Wickenburg, Arizona
Posts: 14,047
Default

Whoa the toneboosters one is also a dynamic EQ! Is this EQ really ok and decramped?
pipelineaudio is offline   Reply With Quote
Old 02-23-2019, 10:26 AM   #78
Magicbuss
Human being with feelings
 
Join Date: Jul 2007
Posts: 1,957
Default

Quote:
Originally Posted by pipelineaudio View Post
Whoa the toneboosters one is also a dynamic EQ! Is this EQ really ok and decramped?
I'm sure it is. Pretty much all well regarded commercial EQ's are decramped or have oversampling. Toneboosters is a DSP guru. I think he literally writes books on the subject. All of his stuff is really good.
https://au.linkedin.com/in/jeroenbreebaart

Here are the sub $100 EQ's that I think compete with any of the more expensive ones.

IIeq https://ddmf.eu/iieqpro-equalizer-plugin/
TB EQ4 https://www.toneboosters.com/tb_equalizer_v4.html
Crave EQ https://cravedsp.com/crave-eq
Slice EQ https://kilohearts.com/products/slice_eq
Acon Equalize https://acondigital.com/products/equalize/

Honorable mention to the TDR EQ's. Slick EQ sounds great but is only 3 bands. Nova is a good dynamic EQ but its parallel only. IIeq has a great feature to switch from parallel to series.

I would use any of these instead of ReaEQ and of course I do because I own IIeq, Acon and TDR EQ's. That said the new ReEQ looks promising as a free ReaEQ replacement. https://forum.cockos.com/showthread.php?t=213501
Magicbuss is offline   Reply With Quote
Old 02-23-2019, 11:26 AM   #79
pipelineaudio
Mortal
 
pipelineaudio's Avatar
 
Join Date: Jan 2006
Location: Wickenburg, Arizona
Posts: 14,047
Default

Thank you! I like TB4 because it is kind of a zero latency version of Nova

It does munch a lot more CPU than ReaEQ but nowhere near most
pipelineaudio is offline   Reply With Quote
Old 02-23-2019, 12:13 PM   #80
pipelineaudio
Mortal
 
pipelineaudio's Avatar
 
Join Date: Jan 2006
Location: Wickenburg, Arizona
Posts: 14,047
Default

What is series and parallel in this context? IIEQ doesn't have dynamic functions like nova and eq4 does it?
pipelineaudio is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -7. The time now is 02:46 AM.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.