Why not always cut the 20-30 Hz range?

[aciddose]

when you're applying a flat (butterworth or just a plain single stage, 6db/o highpass) filter below 30hz, the difference to the signal should be zero in absolutely all cases except where you have content below that point!

why? the filter doesn't do anything (aka, does nothing) to the signal above that point. for a non-butterworth filter, there will be a very small decrease in the passband near the cutoff, but this will be minimal and very, very smoothly fade to zero at +inf frequency. if that were ever to be an issue, well, use a filter with a maximally flat passband.

there are lots of eq filters which will not perform correctly, but that's a matter of the particular filter just not being cut out for the purpose.

if you apply a 20hz-30hz highpass to a sample nothing should happen. you should get an identical unmodified copy. if not, it means there was some really odd content in that sample.

[metamorphosis]

qa2pir: DOES IT SOUND GOOD? NO? THEN DON'T DO IT. YES? DO IT.
Only rule ever...

That's a good rule, but it's a bit of an oversimplification. Context matters -- "sounds good" where, exactly? There is no generic listening environment. What's the style of music? Some musics in some settings need more bottom, and some styles in some settings need less. Do you want a huge rumble and boom in the clubs or a theater? Boost the very lows. Do you want to keep clarity (i.e., do you want something that people will listen to), especially on normal equipment? Cut the very lows. Are you trying to get a brown note effect, or something oriented towards bass? Keep the very lows, and maybe cut some highs.

Good point dude - I do think that cutting the lows doesn't increase clarity unless the speakers are really bad. If the lows are well-balanced, there's no need to cut them entirely, IMHO-

[aciddose]

(first channel is unfiltered, second channel is highpass butterworth of 48db/o with +60db gain)

just to demonstrate. this sample contained a negative dc peak near the beginning at -58db. it sustains at about -70db. throughout the sample after the initial peak. this was probably created by the vca in the instrument the sample was taken from. -60db feed-through is "ok" but not really great.

anyway, the difference is so small that it only takes up three or four samples at 16bit. since the offset is sustained in the section at the end of sample which loops, it doesn't affect the loop at all. (although, harmonics above -60 are present anyway because the frequency isn't a perfect fraction of the sample rate)

this sample is from emu's 8mb gm set. obviously they didn't filter their samples at all before including them in a professional grade product

...they could have though and nobody would have been caused any trouble. although i doubt that an offset of -60db could cause any trouble being there, either.

if your samples don't do this, they must have an awful lot of dc content.

[narz3000]

People used to cut the lows for vinyl but these days I know several major label producers that don't low cut the master at all.

I don't. If you can harness the energy down there it can really create some energy that will tickle your balls in a club!

[aciddose]

as long as it's intentional.

do you like the effect of the thumping from a vocal track at 20hz though? that's the sort of thing that needs to be cut.

[narz3000]

as long as it's intentional.

do you like the effect of the thumping from a vocal track at 20hz though? that's the sort of thing that needs to be cut.

I don't use vocals in my music. It's too deep and existential.

[4damind]

I don't cut around 20Hz. Most sounds are always separated from the lower range and sounds with big subs (only the bass, sometimes the kick) have something like Renaissance Bass so it will be "moved" to a higher range.
Sometimes it sounds more pleasing for me having a high shelf on the highs so it will smooth a bit the often sharp sounding digital sounds in this area. But no cut, only a bit attenuation with a Pultec. With some of this tape or other vintage emulations there is often such a roll-off so it has the same effect.

If there is a problem with DC offset (bad converters etc) it makes sense to have a low cut if the DAW has not such a DC removal option.

[svenneke]

Hi,

I see all kind of information here, some right, some wrong.
Let me share some info with you.

First DC offset removing should not be done with an equaliser but with a DC removal plugin. If your DAW does not do that, which I doubt because most modern DAWs do, download audacity (windows, mac and linux compatible) and use the normalize function. This process allows you to remove the DC offset too. DC offset is a slightly level shift away from the zero crossing line towards the upper or lower side of a soundwave. An equaliser cannot change that shift because it is not a frequency, it's a shift in level away from zero. Only a specialised plugin can do that.

I almost always use a high pass filter on any track. The reason is that low frequency content takes away most of the headroom of a track. The amplitude of a 50 Hz wave need to be much higher than the amplitude of let's say a 3000 Hz wave to be heard equally loud. Therefore woofers are build because they can move more air. A tweeter can hardly be seen moving forwards and backwards and yet we can hear them very well.

If you look at the dynamic range a 16 bit file can hold, there is much less space for multiple soundwaves to use in the lower frequencies, speaking about volume, than in the mid and higher frequencies because the big low waves simply take to much space (amplitude value is high). If you reach the upper limit, you will get distortion because all other frequencies from your mix will add to that signal.

So cleaning up everything below what you need in each track, creates space for the rest to sum up. It's as simple as that. Look with a frequency analyzer at a recording of your voice speaking some words with P and B's in it. You will see the low content when those two letters are spoken. If you zoom in on the waveform in your DAW, you will find at those letters a few big long bumps in the signal where all the rest of the sound is modulated by. This is the impact of low frequencies.

Now, imagine that this can happen in all live recordings with background rumble, resonances caused by guitar body's or drum sets to name a few and everything that produce low frequencies will mess up the bottom of your mix. It might look as if it's not there because you don't hear it or your speakers cannot reproduce it loud enough, but it is present in your data and it consumes energy and space. That energy is waisted and can cause mixes that do not sound as clean as they could be.
At the same time it prevents getting a decent loud mix and master. The cleaner each track is, the louder you can mix. (not that I am a loudness junk, au contraire mes amis)

Learn to look at your mixes as if you can fill a box; left is low frequency content and right is high. Bottom is silence and top is the loudest possible. If you learn to see that low frequencies will fill the left side of your box very rapidly and that you can fill plenty of stuff in the mid and certainly the right side of that box, your mixes will improve.

So yes, always clean everything below 30 - 40 Hz and even higher if your track allows you to do. Your results will only sound better and cleaner.

SVen

[4damind]

First DC offset removing should not be done with an equaliser but with a DC removal plugin.

Hmm... Afaik DC is only 0Hz so a good EQ with a high pass should work without problems.

[aciddose]

it makes sense to say you shouldn't "really" try to remove true (0hz) dc offsets using filters. however, what people refer to is ultra-low-frequency content like rumbling at 10hz.

a dc offset that may exist at 0hz in one part of a system quickly becomes a series of low frequency pulses as it's processed in another. for example, an oscillator with a slight offset will be gated by the amp envelope in a synthesizer. once you get to that point, it's no longer 0hz dc, but dc pulses.

also, there is no reasonable argument against using a highpass filter for dc as well. since a very low cutoff and a single 6db/o filter will work perfectly with a minimum of effort, where is the harm?

[svenneke]

Hi,

thx both for your feedback but there is a big difference between real DC offset and subsonic content (below the 20 Hz of our medium, not our hearing of course).

DC offset is happening in a signal when the voltage output at one point in the stage is not zero when it should be, but instead has a positive or negative (constant) value. DC offset, as I wrote, has no frequency. This is why it is normally not removed with a filter.

In the analogue way, you add a constant an equal opposite voltage to bring the zero point effectivly to zero.
In the digital world the shift is calculated and the difference is added or subtracted to bring the zero level really to zero.

If you don't do that you will loose headroom.

Now for everything below 20Hz in a recording or sample, you can use of course a (steep) filter.

To explain what DC offset is, imagine you take a very strong magnet and place it firmly at about 3 cm away from your woofer (don't do this in reality please). According to it's direction it will push your cone (magnet) away or pull it towards you. Your cone will stay there as long as your magnet is at the same place. That difference in distance from it's normal point of silence, is DC offset. Your cone has moved forwards or backwards but it will ADD the music to it's new zero position.

Let's suppose for example that your cone can move 0.5 cm forward and 0.5 cm backwards maximum before been blown out. If the magnet you had placed causes a backward shift of 0.2 cm from it's normal position at rest, then the backwards "headroom" to be able to move, would be only 0.3 cm anymore and the forwards movement could be 0.7 cm. Now, the 0.7 cm is no problem because it is more than it could do normally. However the backwards movement at maximum amplitude of your signal would be 0.2 + 0.5 = 0.7 cm. The result would be that you would blow out your cone instantly.

Read this article, it is very good:

http://www.harmonycentral.com/docs/DOC-1082

Fig.1 is an example of DC offset, fig.2 a mix of DC offset and a real frequency caused by subsonic content

Sven

[aciddose]

no there isn't a big difference between "dc offset" and "low frequency". as i said, a constant offset can become pulsed very quickly. what you fail to realize is that in order to be "dc" you must measure the offset for an infinite length of time. otherwise at the most, you have a pulse of low frequency content which is below 1/length_of_sample hz in frequency.

so in other words - there is no such thing as "dc".

also, why must you use a "steep" filter? there is no reason to set that requirement.

[svenneke]

no there isn't a big difference between "dc offset" and "low frequency". as i said, a constant offset can become pulsed very quickly. what you fail to realize is that in order to be "dc" you must measure the offset for an infinite length of time. otherwise at the most, you have a pulse of low frequency content which is below 1/length_of_sample hz in frequency.

so in other words - there is no such thing as "dc".

also, why must you use a "steep" filter? there is no reason to set that requirement.

Hi aciddose,

can you explain to me why a constant offset should become pulsed? If it becomes pulsed it is, contradicting your own words, no longer constant anymore, you get it? Then it becomes a frequency. If it is not pulsing or alternating, it is constant.
By definition a DC offset is a constant positive or negative voltage in an electrical circuit often caused by flaws in the design.
By definition an alternating current is caused by waves, frequencies, pulses, whatever you will call it, therefore the term AC.

Saying "there is no such thing as DC" is wrong, sorry mate.

If you have an output and you send no signal thru it, it should measure 0 Volt.
In some cases, when DC offset occurs, there is a constant output, be it a negative or a positive value, even if there is no signal. This is a direct current, DC.
If your so called pulsing offset should be measured, you will get an alternating current or AC, regardless how slow the pulses occur. Then you have a frequency and not a constant offset.

A DC offset can easily be seen on a hardware oscilloscope, I own one. In that case your wave will show up not centered around the horizontal axe, but instead higher above it or lower under that middle. To be able to compensate for this there is a trimpot on oscilloscopes present to allow you to adjust that shift and to center your wave to the zero axe.
So don't tell me that there is no such thing as DC.

Now, why a steep filter?

If you really feel the need to highpass @ 20 Hz and would use a gentle 6 dB one, a frequency of 10 Hz that is present in your data, will only be attenuated to half it's volume, clearly not enough to get rid of the low rumble it can cause and certainly not enough for everything between 10 and 20 Hz.
However if you take for example a 24 dB filter, then the volume of that same 10 Hz frequency is only 1/16th anymore. This way it will be way more effective to clean up the subsonic mess.
To be as much effective as a 24 dB filter set at 20 Hz to attenuate a 10 Hz frequency, a 6 dB filter should be set to a frequency of approximately 20*2*2*2=160 Hz. I don't think you will do that. You get my point?

Greetings
Sven

[EDIT typo]

[aciddose]

no there isn't a big difference between "dc offset" and "low frequency". as i said, a constant offset can become pulsed very quickly. what you fail to realize is that in order to be "dc" you must measure the offset for an infinite length of time. otherwise at the most, you have a pulse of low frequency content which is below 1/length_of_sample hz in frequency.

so in other words - there is no such thing as "dc".

also, why must you use a "steep" filter? there is no reason to set that requirement.

Hi aciddose,

can you explain to me why a constant offset should become pulsed? If it becomes pulsed it is, contradicting your own words, no longer constant anymore, you get it? Then it becomes a frequency. If it is not pulsing or alternating, it is constant.
By definition a DC offset is a constant positive or negative voltage in an electrical circuit often caused by flaws in the design.
By definition an alternating current is caused by waves, frequencies, pulses, whatever you will call it, therefore the term AC.

Saying "there is no such thing as DC" is wrong, sorry mate.

it appears you didn't bother to read what i said.

If you have an output and you send no signal thru it, it should measure 0 Volt.
In some cases, when DC offset occurs, there is a constant output, be it a negative or a positive value, even if there is no signal. This is a direct current, DC.
If your so called pulsing offset should be measured, you will get an alternating current or AC, regardless how slow the pulses occur. Then you have a frequency and not a constant offset.

A DC offset can easily be seen on a hardware oscilloscope, I own one. In that case your wave will show up not centered around the horizontal axe, but instead higher above it or lower under that middle. To be able to compensate for this there is a trimpot on oscilloscopes present to allow you to adjust that shift and to center your wave to the zero axe.
So don't tell me that there is no such thing as DC.

there is no such thing as dc. what happens when you turn the power switch on? wait a minute, are you saying that the level wasn't fixed for eternity? gasp.

Now, why a steep filter?

If you really feel the need to highpass @ 20 Hz and would use a gentle 6 dB one, a frequency of 10 Hz that is present in your data, will only be attenuated to half it's volume, clearly not enough to get rid of the low rumble it can cause and certainly not enough for everything between 10 and 20 Hz.
However if you take for example a 24 dB filter, then the volume of that same 10 Hz frequency is only 1/16th anymore. This way it will be way more effective to clean up the subsonic mess.
To be as much effective as a 24 dB filter set at 20 Hz to attenuate a 10 Hz frequency, a 6 dB filter should be set to a frequency of approximately 20*2*2*2=160 Hz. I don't think you will do that. You get my point?

Greetings
Sven

i don't think you have any idea who you're talking to. secondly, you forgot what you're talking about half-way through. with dc, the frequency is zero. therefore a highpass filter of 1/inf db/o at 1/inf hz will cut it to a level of 1/inf. you get my point?

[svenneke]

Aciddose

I did read everything that you wrote and yes I don't know who you are but that doesn't bother me at all and it really doesn't matter here. I come here to try to help people, not to waiste my time.

What is the point of asking what is happening when you turn the power switch on? Have you ever been able to measure a signal at the output of a device that is not switched on? I haven't.

Also saying this :"with dc, the frequency is zero. therefore a highpass filter of 1/inf db/o at 1/inf hz will cut it to a level of 1/inf", is incorrect.
In fact, what you are writing is this:
0 dB/0 @ 0 Hz will cut it to a level of ... 0.

0 dB/0? Are you serious?
Dividing by zero is not allowed in mathematics. You should know that.

I am not going into an endless debate here. All I am saying is that A DC offset can be measured as a constant positive or negative current at the output of a faulty device, if it is switched ON of course, even if you don't send any signal thru that device. End of story for me.

Bye