KVR Audio

jens · Post by **jens** » Sat Jul 06, 2019 11:02 pm

djeroen wrote: Sat Jul 06, 2019 10:03 pm
jens wrote: Sat Jul 06, 2019 9:01 pm Yeah, the high-q mode increases the CPU usage up to a factor of more than 4 (again depending on the settings) - I think it really must be broken somehow
Not broken at all! It switches to 4x oversampling so yes you will see an increase in CPU load

Sure - I am just confused that it is such a huge increase. I don't think I have ever seen such a big increase going from no to 4x oversampling in a plugin and your plugins are usually highly efficent...
(I mean: how can 4x oversampling cause the CPU usage to increase by even a little more than a factor of 4?)

And what about the drastic change in sound (volume, drive)?

djeroen · Post by **djeroen** » Sat Jul 06, 2019 11:12 pm

jens wrote: Sat Jul 06, 2019 11:02 pm
djeroen wrote: Sat Jul 06, 2019 10:03 pm
jens wrote: Sat Jul 06, 2019 9:01 pm Yeah, the high-q mode increases the CPU usage up to a factor of more than 4 (again depending on the settings) - I think it really must be broken somehow
Not broken at all! It switches to 4x oversampling so yes you will see an increase in CPU load
Sure - I am just confused that it is such a huge increase. I don't think I have ever seen such a big increase going from no to 4x oversampling in a plugin and your plugins are usually highly efficent...
(I mean: how can 4x oversampling cause the CPU usage to increase by even a little more than a factor of 4?)

And what about the drastic change in sound (volume, drive)?

At 4x oversampling, every sample is effectively processed 4 times (internally it runs at a 4 times higher sampling rate). In addition, one needs anti-aliasing filters when doing oversampling properly. So that's where the extra bit comes from.

Therefore one would expect a 4+ times increase in CPU load for 4x oversampling.

If other plugins show an increase of less than 4x the CPU load, it typically means that there is a lot of other stuff going on that is not oversampled (and takes up a significant CPU load that is not dependent of oversampling).

And as with any non-linear processing: oversampling removes aliasing. That's why you'll get a slightly different sound when running the non-linear circuit modelling without oversampling. It's therefore recommended to keep the HQ mode enabled if you enable circuitry modelling.

jens · Post by **jens** » Sat Jul 06, 2019 11:26 pm

Understood - thanks for the explanation!

NeonNightflyer · Post by **NeonNightflyer** » Sat Jul 06, 2019 11:27 pm

NeonNightflyer wrote: Sat Jul 06, 2019 3:12 pm I love what the new update promises, but I'm having an issue where any existing EQ4 automation I have in any given project is overridden when I install the updated version (e.g. frequency automation for a LPF is changed to a completely different parameter). This happened with the previous two updates as well, but not the updates before those. I'm on Mojave using Logic 10.4.5 (but this happened with the previous version of Logic as well).

I'm holding off on updating EQ4 for the time being. Any ideas on how to address this?

Hi Jeroen,

Have you heard of this issue before?

Thanks!

djeroen · Post by **djeroen** » Sat Jul 06, 2019 11:29 pm

NeonNightflyer wrote: Sat Jul 06, 2019 11:27 pm
NeonNightflyer wrote: Sat Jul 06, 2019 3:12 pm I love what the new update promises, but I'm having an issue where any existing EQ4 automation I have in any given project is overridden when I install the updated version (e.g. frequency automation for a LPF is changed to a completely different parameter). This happened with the previous two updates as well, but not the updates before those. I'm on Mojave using Logic 10.4.5 (but this happened with the previous version of Logic as well).

I'm holding off on updating EQ4 for the time being. Any ideas on how to address this?
Hi Jeroen,

Have you heard of this issue before?

Thanks!

No, not really. That said with one of the earlier updates some of the parameter enumeration IDs might have changed accidentally. With the last two updates, however, we've made sure that this isn't happening. This should however only affect hosts that still use parameter ID enumeration.

NeonNightflyer · Post by **NeonNightflyer** » Sat Jul 06, 2019 11:41 pm

Okay, thanks!

neverbeeninariot · Post by **neverbeeninariot** » Sun Jul 07, 2019 9:15 am

Plug-in installer 1.3.0 changelog

Equalizer v4.1.9

Improved consistency between high-quality and eco mode for circuit modelling

Butwug · Post by **Butwug** » Sun Jul 07, 2019 9:16 am

Oh nice update. I like the circuit drive option. Rounds off the highs nicely and the lows as well. Makes it sound warmer/fatter. Plugin is getting better and better.

Edit* Thanks for fix on Reelbus! I should have waited 30min to download the update

djeroen · Post by **djeroen** » Sun Jul 07, 2019 9:29 am

Just released: update for ReelBus to fix the label visibility.

neverbeeninariot · Post by **neverbeeninariot** » Sun Jul 07, 2019 9:41 am

Quick semi-unscientific CPU test on the modeling & HQ settings (i7 4790, Reaper x64).

3 bands dynamic EQ, 2 downwards (1 shelf, 1 bell), 1 upwards (bell)

All off: 0.2%
HQ off, modeling on: 0.2 - 0.3%
HQ on, modeling on: 0.5 - 0.6%

Vortifex · Post by **Vortifex** » Sun Jul 07, 2019 1:29 pm

Halonmusic · Post by **Halonmusic** » Sun Jul 07, 2019 3:40 pm

This eq is getting better and better.

spigmu · Post by **spigmu** » Mon Jul 08, 2019 12:14 am

djeroen wrote: Sat Jul 06, 2019 11:29 pm
No, not really. That said with one of the earlier updates some of the parameter enumeration IDs might have changed accidentally. With the last two updates, however, we've made sure that this isn't happening. This should however only affect hosts that still use parameter ID enumeration.

Jeroen, this is something new to me, so can you say which DAWs would be ones to still be using this? I'm confident it won't affect me (ProTools and Reaper) as this is something that has never bit me, but I want to ask just for better knowledge of this thing.

Thanks!

Tappistry · Post by **Tappistry** » Mon Jul 08, 2019 3:09 am

Still no config option for changing mousewheel behavior of adjusting Q value of filters? For some reason, this EQ is still reversed in that behavior from almost every other EQ plugin I've tried, from stock ReaEQ to others.

BerryLaCroix · Post by **BerryLaCroix** » Mon Jul 08, 2019 4:01 am

Tappistry wrote: Mon Jul 08, 2019 3:09 am Still no config option for changing mousewheel behavior of adjusting Q value of filters? For some reason, this EQ is still reversed in that behavior from almost every other EQ plugin I've tried, from stock ReaEQ to others.

Hmm not seeing that here?

In Logic Pro X, moving the mouse wheel up increases Q.
In Pro Tools, moving the mouse wheel up increases Q.
In TB EQ4, moving the mouse wheel up increases Q.

I'd say that's pretty consistent?

ToneBoosters E.Q.4