KVR Audio

the V · Post by **the V** » Wed Jun 22, 2005 11:57 pm

Within the free ones, I found the already mentioned TobyBear's "DC KILLER!" to be the lightest and most effective, and I always stick with it when in need.

I recommend its "dynamic filter" algorythm: once I analyzed some original vs. processed difference wave, and I found this specific algorythm to be the one that, while removing DC, leaves most untouched the audible frequencies above.

greendoor · Post by **greendoor** » Thu Jun 23, 2005 4:57 am

Some of the Voxengo plugs, like Elephant, have High pass filters that kill DC. If it sounds good, why worry about it? Just fade in and fade out, so there is no click when the audio starts.

danielmm · Post by **danielmm** » Thu Jun 23, 2005 12:05 pm

....so, are we saying that if you use a high pass filter below 20hz that this means you do not require DC offset....I thought these were 2 different things???

dano

stefancrs · Post by **stefancrs** » Thu Jun 23, 2005 12:14 pm

danielmm wrote:....so, are we saying that if you use a high pass filter below 20hz that this means you do not require DC offset....I thought these were 2 different things???

dano

DC offset = 0hz signals. Highpass filters does not let 0hz signals pass through.

fabi · Post by **fabi** » Thu Jun 23, 2005 2:10 pm

C00kie wrote:Please file a bug report to the developer of the offending plugin!

But are you sure it's true DC offset (line above zero db even when silent signal passes) and not some sort of asymettrical behaviour (positive half of wave is amplified different from negative half)

it happened to me with pseudograins. even when silent, there was a line way above zero. might try to contact him. great plug, though. just don't have the time right now.

danielmm · Post by **danielmm** » Thu Jun 23, 2005 10:14 pm

stefancrs wrote:
danielmm wrote:....so, are we saying that if you use a high pass filter below 20hz that this means you do not require DC offset....I thought these were 2 different things???

dano
DC offset = 0hz signals. Highpass filters does not let 0hz signals pass through.

If the the average between the positive and negative wave peaks in your file is not zero, the result is inaudible low-frequency distortion. Although you won't be able to hear it in the .wav file, it will seriously degrade your compressed file, as well as introduce further distortion if other filters are applied to the uncorrected wave. The DC offset function corrects this problem, and should always be applied as the first step in optimizing an audio file..

Looked into it a little more and you are right....thought there was more to it than that though

dano

headquest · Post by **headquest** » Fri Jun 24, 2005 12:57 am

Hi,

If your host/recording software doesn't remove DC offeset for you (many do), surely this is best done offline/destructively in your sample editor? I know I wouldn't want this going on as a continuous drain on CPU...

Certainly I do this offline as a matter of course where its required (not very often here, but sometimes Subtractor patches come out wonky when bounced/rendered in my sequencer, for example).

Again, any sample editor worth its salt hsa this function...

IIRs · Post by **IIRs** » Fri Jun 24, 2005 1:43 am

headquest wrote:I know I wouldn't want this going on as a continuous drain on CPU...

Not a very significant one.. a 1 pole highpass will do.

popsych · Post by **popsych** » Fri Jun 24, 2005 5:51 am

As i said, DC is cumulative but in a decenent 32bit virtual mixer host (any VST host worth firing up is 32 bit internally) just one filter in your output bus should do.

Alternatively just dc offset correcting the filter on the two track in a wave editor will also do the same.

No point in removing it per track, unless only one track in question has it.

dc offset killer?