What do you use to cut subharmonic frequencies.

[3*s]

Nothing but a Parametric / notch filtered EQ, should be used,
when your dealing with sub-Harmonics.

Interesting, please explain.

[Hunter]

I recall someone sugesting eqing a Hipass way down at 35 Hz or so followed by a Low shelf at 50-60hz.

I'm basically after the same sort of thing as you Chase as I've discovered that the sonitus EQ in sonar just wil not get rid of enough sub bass, no matter what I try. PITA.

Will try bugpass.

[HanafiH]

It depends what you mean by sub-sonics. As frequency gets lower and lower you can get a voltage offset from a wave whose length is greater than the length of your recording. This needs DC Offset removal and most sound editors provide some kind of utility to do this.

Intermediate subsonics are best removed using a highpass filter. The exact frequency depends upon the material itself, the type of filter design and its resonance. Generaly a non-resonant highpass set at 35Hz does a good general job. If you don’t have a subbass monitor, that’s probably the best you can do.

It’s worth looking at the spectroscope graph of some commercial releases similar to the area you’re working in to assess the current fashion. Generally tracks destined for club-scale massive amplification need very hard subsonic filtering to prevent dance floor patrons ears bursting and gushing blood everywhere. I made that last bit up. Ambient, which almost never gets massive amplification can get away with a wee bit more.

[Hunter]

Generally tracks destined for club-scale massive amplification need very hard subsonic filtering to prevent dance floor patrons ears bursting and gushing blood everywhere.

That's what I'm aiming at and i guess Chase is too both being drum n bass types. Actually I was at a Pendulum DJ set on saturday night and there was one tune where the bass was sooooo powerful it was almost unpleasant. made me want to pooh.

[Lunch Money]

Ah, the fabled brown note.

[Debutante]

Thanks for the simple but elegant idea of putting the filter in front of Inspector, Debutante. I can't believe it never ocurred to me before. I was suddenly curious to check out Tracktion's native hi-pass filter. Not as steep as I always imagined!!

Now I know that BugPass is the tool for me. Thanks again, Christian.

No problem! It's a great way to check out just how baddass your filters are. Try it with phasers an choruses and flangers too, although I don't really know how to interprate the display in these cases. If anyone does please give a shout. Sorry to go off topic.

[Christian Budde]

Everyone confused?

Do minimal phase filters lead to latency as well, just not noticeable?

They may introduce some latency, but in the range of only a few samples depending on the type and the order. Usually it's not noticable at all.

Does the phase (minimal or linear) of the filters work, in a way, like the phase shift (-> signal cancellation)?
Is linear like 180 degrees then?

180° degrees phase shift would mean inverting the signal
As above it also depends on the type and the order. For example a simple - minimal phase - lowpass (order 1) would introduce a phaseshift of 90° (correct me if i'm wrong) starting about one octave below until one octave above, but depending on the Bandwidth. So at DC the filter would have no phase shift = 0° at Nyquist it would have 90°. At the cutoff frequency it has exactly 45°.
The derivation (derivative? I'm not sure about the correct word) is called group delay (see earlier posting). So applying the derivation will give nearly 0 at DC, then (one octave below cutoff) increasing, maximum at cuttoff frequency and decreasing until 0 at nyquist.

I'm not sure if it is understandable. Anyway back to your question. This phase shift can usually not be recognised with your ears, because they are insensitive against the phase. For example lets have an allpass filter. This passes every frequency, but only process the phase. In this case the phase will spin around 360°. If you add the same source unprocessed twice, you will hear the canceling in the form, that at the cutoff frequency (where the allpass has 180°) cancelation occurs.

Totally confusing again, i'm sorry, next time i'll better post some charts.

Regards,

Christian

[HanafiH]

Here's an interesting wee trick you can try at home. Using almost any audio editor take a brief silent wave file and draw a single sample spike:



Now render this using a non-linear phase filter at 1K full boost:



The different frequencies in the original pulse have travelled through the filter at different speeds, causing the phase to shift.

Now try a linear phase filter:



This filter uses a mathematical trick to cause the phase shift to appear both sides of the original spike equally, make no overall shift in phase, but causing a fluffiness to appear in the sound.

[scuzzphut]

Nice illustration, Hanafih

Lunch - I just got the new version - I see it now I seem to recall trying out bugpass when it was first released and it didn't have Hz values?????

[The Chase]

Why don't you think it's good, Chase?

I said I loved it, but the latency really bothers me and makes it useless on things like drums in fast-paced electronic music.

[The Chase]

That's what I'm aiming at and i guess Chase is too both being drum n bass types. Actually I was at a Pendulum DJ set on saturday night

it didnt happen to be on the 17th in dallas, did it?

[kilroy]

I recall someone sugesting eqing a Hipass way down at 35 Hz or so followed by a Low shelf at 50-60hz.

I usually follow an 18th order Butterworth highpass filter at 28 Hz with a gentler 6th order filter in that 50-60 Hz range. This avoids a case where placing just one very steep filter further up the low bass region could exhibit the potential for audible perception of phase anomolies.

On too many mixes, the low bass is not very "phase clean" to start with. If things are sounding loose down there, and there does not seem to be the desired punch and focus in the deep bass, then listen carefully for any phase distortions. A phase clean low end is essential for those ultra deep, tight club sounds that behave well in the subs.

[Christian Budde]

I recall someone sugesting eqing a Hipass way down at 35 Hz or so followed by a Low shelf at 50-60hz.

I usually follow an 18th order Butterworth highpass filter at 28 Hz with a gentler 6th order filter in that 50-60 Hz range. This avoids a case where placing just one very steep filter further up the low bass region could exhibit the potential for audible perception of phase anomolies.

You mean 18dB and 6dB don't you? Because an 18th order Filter would give you about 54dB, which is very very steep.

Regards,

Christian

[Hunter]

That's what I'm aiming at and i guess Chase is too both being drum n bass types. Actually I was at a Pendulum DJ set on saturday night

it didnt happen to be on the 17th in dallas, did it?

No it was last saturday in Edinburgh. It's quite funny 'they' were simultaneously in Edinburgh, Toronto and San Francisco the same night


And it was nearly the brown note but not quite.


I heard a great story at Musikmesse this year from a fellow who is a professionial engineer and does these big sound system installations. He did one for an art exhibition in the states which was a big hall with 3 huuuge screens showing a few frames of this weird face on permanent loop. They had a row of monster subs all the way along each side of the room, out of phase, and i think every now and then the video would do a sudden change and the subs would go Fffffffwwwububububub and people would start feeling very strange. I think there may have been some brown note incidents.

[pitchdrifter]

Elephant has a DC filter with variable frequency in it.