Frequency EQ noise artifact?

Can anyone please tell me what am I seeing here?
Looks like Frequency (and all other stock EQ’s) adds some kind of noise to the signal. Here is the test
Sine wave at -6db goes into three EQs each at the time
APQualizir 2, Fabfilter Pro-Q2 and Frequency.
Every eq has +24db boost at 520hz
Please, have a look at the spectrum analyzer. what kind of noise is that? Only Frequency and all stock Cubase EQ’s has this artifact, no matter what settings I use it always has that noise when boosting.
Also here is interesting video demonstration where you can see how this noise behave when moving EQ band around:
and here are screenshots:

My guess? Steinberg’s subliminal message to keep us buying mediocre software upgrades.

Bump - I’d be interested to know what this is.

And here I thought it was on par with the rest. Is this also for the channel strip EQ?


Does anyone think this is important? -120 dB … how many simultaneous instances would it take before it started becoming audible?

a lot. but yet, none of the third party plugins I tried has this issue, so I guess those developers think it’s important. otherwise, it could be all over. it’s not important for Steinberg as long as we purchase every new update.

If you’ve got a large track count, using stock EQ’s on them with compression it might be an issue. But tbh, more than if it’s audible or not I’d like to know why it’s there.

How often do you boost things by 24dB?

There’s just no reason why it should be worse than other modern EQ plugins, it’s not like it’s more CPU efficient. Steinberg pls

It’s not about a single channel signal boost. If you’ve got multiple channels with that noise you don’t want stacking to give you more noise than you need to have.

Ive also noticed this when boosting to higher gain- bot Frequency and channel eq produce the same artifacts on a spectrum meter at about -120 to -110 db. Its not too concerning since boosting to such gain is very rare, but interesting to see that other options dont produce such artifacts.

DJW: While i found frequency quite low on CPU when its window is closed, CPU usage really jumps when you open its window with spectrum analyser active (approx 10-15% of cpu) so if you compare it to for instance Fabf q2 (there is no significant CPU rise when its spectr. analyser is runng) yes its quite a CPU hog.

there are enough noise generator plugins out there (all sorts of vintage emulations). EQ like this must be clean no matter what.

I don’t believe this is the same kind of noise as emulation noise, the latter tending to be harmonically related to the source signal. This would just be digital noise if it were amplified enough.

I’d be interested in hearing …does anyone think this low level of noise, -110 dB, would ever be audible in a “normal” session, even if used on multiple tracks with compression?

Thanks -

You would need like 100 or 200 tracks with unrealistically high EQ boost playing at the exact same time just for the noise to become slightly audible during quiet sections.

Steinberg should still look into this, but it’s essentially a non-issue.

it’s inaudible at this volume, of course, it’s masking by other tracks sounds and so on. But the fact is, The quality of this EQ is not as good as third party plugins.
I guess we just need to deal with it. Stock plugins are not as good as they could be, and look like Steinberg don’t care about it.
It’s inaudible so let’s leave it as it is. Use same old code for ages in all upcoming EQ’s, but make sure to have different skins.

9.0.20 this “feature” still there

Looks like this thing has been fixed with 9.5 update.
Thank you.