Just wanted to clear something up that's been confusing me. When people mention frequencies that that start high-passing or low-passing at, is that the frequency at which they place the frequency knob in the eq or is it the frequency which should end up getting filtered out entirely?

For the high pass filter in the attached image below, it looks like right about 4k is left unchanged, the frequency knob is set about 2k and then just below 1k is where everything will be filtered out, so which of these frequencies are folks referring to when they say I set a high pass filter at <blah> frequency? Just want to make sure I'm speaking the same language as everyone else.

Yea, you want to be careful with your cut off point.

When you play notes higher up or lower (depending on a high or low pass) those upper and lower notes, will be cut off.

I suggest you spend some time experimenting with how your playing responds to this.

If you have a lead or a bass line that goes all over the place, you typically want to the frequency zone of that particular bass line so it has it's own section, but any thing that is beyond or above the notes being played will get fitted nicely with every thing else.

Awesome, thanks all! So that means when folks are cleaning up say a lead synth, and say that they get rid of everything below 300 Hz, that actually means that they are just placing the -3db point somewhere around 300 Hz?

To be honest I've been thinking about that myself I was actually playing around with the EQs in FL studio and Studio One (what else would I do on a Friday evening?), and it seems that the cutoff doesn't really correspond to the -3db point in either of those (except for 6db/oct and 12db/oct in S1), and I'm not sure that the frequencies are really comparable between those two EQs. So I think you just have decide how much you allow 300Hz to be attenuated and then match the visualization with that.

This was hard to understand, can you please clarify?

http://www.extron.com/product/files/helpfiles/dsp_configurator/About_Filters_and_EQ.htm

I have this page book marked.

This really helps if you know a bit about math and physics

I internalized the eq's job so much so that i forgot how to explain it....


So basically, that number that you are assigning is the point at which the slope starts.

You define the Shape of your filter (slope/cut off)

You define the Q or bandwidth. <-----------------------Additive bumps if you will

You can then use both with gain or volume, with NEGATIVE or POSITIVE numbers for subtraction or additive properties.

You are basically defining the mid point for the Q or bandwidth, and the starting point of the slope for a filter.


The range affected for the BUMPS or DIPS, are Q and BANDWIDTH.

the FILTERS have a cut off range by the specific frequency you are cutting off.

Caps are used for emphasis and ease of viewing.

Just to clarify, we're talking about lowpass or highpass filters here.

Quote from: Marrow Machines on March 19, 2016, 06:39:54 pm

So basically, that number that you are assigning is the point at which the slope starts.

Yes, and the -3dB point is one way to define this. From your reference: "Cutoff frequencies begin at the half-power point, or '3dB down' point."
However, this is not true for all EQs; in FL studio Parametric EQ2 it seems that the frequency corresponds to a -6dB attenuation. And I think that for filters with steeper slopes, the frequency corresponds instead to the stopband edge, which in layman terms is a point after which everything is pretty much cut away.

the mid point was only in corresponding to the Q or BUMP, aspect.  It seems to me that the OPTIMIZED point on the curve of the bump, is the frequency you use for NEGATIVE or POSITIVE adjustments in volume/gain. As well as how wide the area is from that optimized point; in regards to bandwidth.

Never referenced any thing about the slope. Should of used curve in that statement.

This is beginning to be at the edge of my understand of math...and my ability to articulate.

Just wanted to clear something up that's been confusing me. When people mention frequencies that that start high-passing or low-passing at, is that the frequency at which they place the frequency knob in the eq or is it the frequency which should end up getting filtered out entirely?

Look at it this way: What is "entirely"? -20db? -60db? -120db?

There's setting on your spectral analyser that sets the minimal shown amplitude. Set it low enought and that frequency doesn't dissappear at all. While mixing, you're interested in what's happening in the first 10 - 20 db of dynamic range. (measured from the loudest sinewave)
...unless you do something like classical musc or jazz where the dynamic range is supposed to be much higher.

In this picture, the traditional -3db cutoff frequency is denoted by fc:


However, you can also specify the filter by the frequency where the stopband starts, i.e. after which the attenuation is as high as it gets. And my point was that, unlike I first assumed, the frequency control does not have the same meaning between different EQs, or even between different filters inside an EQ.

You're using information i do not have, i was just merely suggesting from instinct with out much data to back up my sayings.

That's an interesting graph though.

Hmm, seems like this post has gotten into a bit of a debate, and as a result has seemed to have gotten me even more confused :/

For a concrete example, let's say someone gives me the blanket advice of high-pass all kick drums at 100 Hz and low-pass all kick drums at 2 kHz. Do they mean?
a) The passband is from 100-2000 Hz.
or
b) The stopband should be from 0-200 Hz and 2 kHz-infinite, which would put the passband closer to something like 200-1500 Hz.

Not that it is a huge deal, I just want to be sure I'm speaking the same language as other folks.

You gotta know your tools if you want to answer that question for yourself.

If you want an easier route, use your ears and play it with the two conflicting elements and prioritize one in terms of the other. IE, if you want more sub bass than kick, cut the kick more. if you want more kick than sub, cut the sub more.

Quote from: Marrow Machines on March 23, 2016, 12:22:34 am

You gotta know your tools if you want to answer that question for yourself.

I don't see how the question is a matter of knowing you tools or not.

It's in fact the difference of how tools work that you need to understand. So UNDERSTANDING your tools is a big part of using your tools.

Other wise, if you knew exactly what your eq was doing, you wouldn't of asked the question in the first place, let alone the second question that you quoted me on.

I don't see how you can not think that it's not when it clearly is the problem.

Ark even brought up a point about differences in the graphs by different perspectives to ultimately give you a different result.


It may not differ by much, or it might, i don't know, but to throw out understanding is foolish. Let alone asking a question with out the purpose of understanding.

then don't waste your time or other people's.

How can you reasonably expect an accurate answer if you don't have some idea with what you're dealing with?


EDIT: I will say this, if you're not entirely confident in your understanding it's no use chasing a rabbit that you won't find unless you sit inside of a physics or math class room.

Just look up the schematics of your tool that's being modeled and see what you can muster from it.

other wise, we can't give you a clear answer if we don't understand the tool you're using.

Besides, your second question was mostly in terms of taste, excluding the knowledge of your tool.