Bitwig

Trancit · Post by **Trancit** » Sun Oct 27, 2019 9:11 am

I already often noticed the LP4 of the EQ´s in Bitwig make the sound more dull even if fully opened...

I investigated this a bit further and it turned out that the LP4 completely open (set to 20kHz) already starts rolling off the signal at 10kHz...

I emailed with Bitwig support and they told me this would be a wished behaviour and behaves like this by design...

I honestely dislike this behaviour in a clean studio EQ...
I know this from older vintage filter emulations, i.e. NI have this "feature" in their Pro-53 filter which is present in Kontakt too...
But in an EQ???
I would really prefer filters which do not cut the sound if opened completely...

What do you think?
Does this bother you?

SLiC · Post by **SLiC** » Sun Oct 27, 2019 10:04 am

I have never noticed this.

Some filters do colour even when open, more unusual for an EQ! I guess it’s just a case of not using them if you don’t want the sound they impart. Personally I have so many eq choices I only use the built in eqs for quick and dirty or as a modulated effect, I don’t use them for mixing, mastering or anything surgical where o guess this may be more of an issue.

Trancit · Post by **Trancit** » Sun Oct 27, 2019 10:44 am

SLiC wrote: ↑Sun Oct 27, 2019 10:04 am I have never noticed this.

Some filters do colour even when open, more unusual for an EQ! I guess it’s just a case of not using them if you don’t want the sound they impart. Personally I have so many eq choices I only use the built in eqs for quick and dirty or as a modulated effect, I don’t use them for mixing, mastering or anything surgical where o guess this may be more of an issue.

I work mainly with the built in ones... and on many sources I am used to cut off the very high end (above 18-19k) as it can give trouble with some encoding...
The LP2 isn´t steep enough and the only alternative would be for now to put multiple EQ - 2`s in series with a LP2
Therefore I would like o see some clean steep alternative

mclstr · Post by **mclstr** » Sun Oct 27, 2019 12:04 pm

Why would you enable LP and have a problem with rolling off everything above 10K? Your not going to enable it and cut anything above that. The 10k to 20k octave is barely audible by young people and not audible at all by older people.

Much anything above 10k adds an annoying fingernails on the chalkboard quality to the sound.
If you want that quality don't enable LP.

Trancit · Post by **Trancit** » Sun Oct 27, 2019 1:53 pm

mclstr wrote: ↑Sun Oct 27, 2019 12:04 pm Why would you enable LP and have a problem with rolling off everything above 10K? Your not going to enable it and cut anything above that.

If I understood you correctly (not sure though) why do you think so...
Every normal EQ is able to do so

The 10k to 20k octave is barely audible by young people and not audible at all by older people.

I am with you between 15k and 20k but the 10k up to 15k is clearly noticeable for me and I am 50 years old...

Much anything above 10k adds an annoying fingernails on the chalkboard quality to the sound.
...

Again above 15k yes and that´s the reason why I want to cut some away... this are only problem frequencies... but below 15k sounds easily muffeled...
Take a hihat loop and cut away above 15k with a 3rd party EQ LP4... you´ll hear not much of a difference...
Do the same with EQ-2/5 and it seems somebody put a blanket over the speakers... not the behaviour I like to have...

loungepanda · Post by **loungepanda** » Sun Oct 27, 2019 2:40 pm

Trancit wrote: ↑Sun Oct 27, 2019 1:53 pm Again above 15k yes and that´s the reason why I want to cut some away... this are only problem frequencies... but below 15k sounds easily muffeled...
Take a hihat loop and cut away above 15k with a 3rd party EQ LP4... you´ll hear not much of a difference...
Do the same with EQ-2/5 and it seems somebody put a blanket over the speakers... not the behaviour I like to have...

I just did this. A/B using EQ-5 and Neutron 2 EQ (seeing as it is also a surgical EQ) on a hi-hat loop. Result: The 4-pole filter in EQ5 did indeed sound slightly muffled compared to the 24dB filter in Neutron, however I did manage to get an identical result by simply increasing the Q in EQ-5 slightly.

Trancit · Post by **Trancit** » Sun Oct 27, 2019 4:17 pm

loungepanda wrote: ↑Sun Oct 27, 2019 2:40 pm I just did this. A/B using EQ-5 and Neutron 2 EQ (seeing as it is also a surgical EQ) on a hi-hat loop. Result: The 4-pole filter in EQ5 did indeed sound slightly muffled compared to the 24dB filter in Neutron, however I did manage to get an identical result by simply increasing the Q in EQ-5 slightly.

Yes of course, that´s have the vintage filters behaved too...

Increasing the Q to about 1.5 brings about similar results...
This gets only problematic if you have to automate the filter down and the resonance gets into areas where it will be too noticeable... then you have to automate the resonance too...

These are at the very end workarounds you don´t need with a "normal" filter...

But I already see I am quite alone with that opinion...

No problem... I bow to the majority...

loungepanda · Post by **loungepanda** » Sun Oct 27, 2019 5:04 pm

Didn't save the project but I think the Q value was more like 0.95-1.0, so not really enough to create resonance issues, even when automating.

On a side note: I think that what EQ-5/2 could use is Q values for the shelves. Also phase modes.

Trancit · Post by **Trancit** » Sun Oct 27, 2019 8:33 pm

loungepanda wrote: ↑Sun Oct 27, 2019 5:04 pm On a side note: I think that what EQ-5/2 could use is Q values for the shelves. Also phase modes.

That´s true...

Tj Shredder · Post by **Tj Shredder** » Mon Oct 28, 2019 8:48 am

All digital filters will have problems in the highest octave below Nyquist. You could try to make a filter in the Grid. That is 4 times oversampled by default...

Trancit · Post by **Trancit** » Mon Oct 28, 2019 11:33 am

Tj Shredder wrote: ↑Mon Oct 28, 2019 8:48 am All digital filters will have problems in the highest octave below Nyquist. You could try to make a filter in the Grid. That is 4 times oversampled by default...

A problem every normal (not oversampled) track EQ has but nevertheless they all offer at least one clean digital alternative...

I just got another response from Bitwig that this behaviour would be normal because of being an analog model...
What me bother on this topic is that we got already analog models of the LP4 as well in Filter as in Ladder...
So in Bitwig we got 3 analog models and not a single digital "clean" LP4...
That´s a bit single sided and restrictive to me... in how many plugins do we need the (more or less) same filter without having other (native) alternatives??

Butwug · Post by **Butwug** » Wed Oct 30, 2019 1:09 pm

Yeah I’ve noticed that putting in an EQ and not even touching it, changes the sound. Now my default EQ5 setting is having all nodes off when I drop in a fresh EQ5. As far as 10-20k barely being audible, I disagree. 10-15k is pretty noticeable for most. 15-20k is another story.

mosska · Post by **mosska** » Fri Nov 08, 2019 12:16 pm

never used eq5..fabfilter's fan but usually use eq2 in 12db slope mode but a few days ago decided to lp4 and was surpised it rolls of high frequncies when open. I think it shouldnt be like that.

Hanz Meyzer · Post by **Hanz Meyzer** » Sat Nov 09, 2019 11:18 am

I really like the character of the Bitwig EQ, no complains here. I only wish it had a all standard features, like Q for low/high shelve (so you could set it so low that the band turns into a tilt filter), and a switchable analyzer slope from 0-4.8dB.

Also if they worked already on it, why not add peak/bell filters with 36-72dB slope, too? I think those sometimes are very useful, esp. for coloring a sound, e.g. with using keyfollow/shift.

Plus very steep LP/HP, like the butterworth ones in Renoise. They could decouple then the slope factor from type switch, so switching slope only with right mouse button or so.

Bitwig

KVR Audio

What do you think: EQ-2/5??