DC offset on bass

[Dr. Fluffenstein]

Is it bad to have DC offset throughout an entire bass note (around a 16th long)? If so, what can I do to get rid of it? The specific bass I'm referring to was made in Zebra CM and has one oscillator in the 40Hz range, while the second one is an octave higher. Both oscillators are filtered slightly and I adjusted their symmetries a bit. When I look at the waveform the part below the 0 line is about twice bigger than the part above. Any ideas or suggestions?

[somaaudio]

i think you can eliminate DC offset with a really steep high pass filter @10Hz maybe with a linear Eq.
Hope it helps,

Cesar

[Dr. Fluffenstein]

Thanks, that worked out pretty well . I used EQ8 though and was still left with a bit of offset (on the other side of the line this time). Are there any free EQ's I can try out that can cut that low?

[leedsquietman]

I just use Soundforge's remove DC offset function, 2 button clicks and you're done.

Voxengo's Elephant v3 limiter also has a DC offset removal (and dithering) option.

EQ8's filters aren't very steep so it takes more effort to use it as a DC offset removal tool.

[somaaudio]

Thanks, that worked out pretty well . I used EQ8 though and was still left with a bit of offset (on the other side of the line this time). Are there any free EQ's I can try out that can cut that low?

You can use ozone , neon , fab filter Pro Q or any linear phase eq ( download a demo ) if you dont have any of these. Maybe use two or three bands of EQ8 set at the same frequency to make the curve steeper. Linear phase eqs work best for this as they dont color the sound or cause phase problems.
Regards,
Cesar

[Dr. Fluffenstein]

Thanks for the tips guys, that's what I ended up doing with EQ8. I stacked 8 of those "eq balls" at 30Hz, and there was still a tiny bit of offset. 3 didn't really do the trick and the more I added the less offset there was. I'm concerned about cutting so steeply though, I always thought gentle roll-offs were the way to go for cutting off the extreme low end. I can't imagine a 96dB cut being a good thing, but someone please correct me if I'm wrong.

[hoffman2k]

Have you tried the DC switch on the utility plugin?

[simpli.cissimus]

Have you tried the DC switch on the utility plugin?

Obviously not ... :D:D

[Dr. Fluffenstein]

Thanks, forgot that DC switch was there . Unfortunately it didn't work that well with this particular bass. The only fix I found was changing the 'waveform symmetry' parameter in the synth itself, but that changed the timbre quite a bit. It did work on other basses though. Here's a screenshot of the waveform:

http://img689.imageshack.us/img689/770/bassoffset.jpg

[Synthbuilder]

When I look at the waveform the part below the 0 line is about twice bigger than the part above.

It's not how big the waveform is below the 0V line but how big it is and how long it stays there. Short big excursions can be less negative overall than more sustained medium height excursions. For average DC content you are looking at area enclosed within the waveform. Remembering that the area below the central 0V line counts as negative and will subtract from the bits above the line.

From that image it's hard to tell for sure but I don't think it has too much DC content to worry about. You can see it spends quite some time above the line.

But rolling off below 20Hz with a shelving EQ could reduce any DC that is there. If you have significant DC then you'll see that adding the high pass filter will produce a waveform that appears to move upward (or downward) as the note progresses. If it is that bad it'll probably click like anything on the attack of the note anyway.

[docprosper]

I would assume that clicking would be only one aspect of the problem, the other being that a strong DC offset would drive the speakers (we tend to forget the final part of the chain...) in a less than ideal fashion. The signal ends up driving a physical element in an AC fashion about a centerline; adding a DC element changes this centerline. This would (I think) reduce speaker efficiency but might also cause distortion at high DC offset levels. Food for thought.

Also, when I work with samples I tend to pull them through a sample editor (I use Audacity, which is freeware) and normalize and remove DC offset before using in Live. In Audacity, these are embedded in a single function. I also take this opportunity to fade in/out the beginning/end of each clip to further reduce the risk of clicks.
-Hamish