low pass filter

the exactness of a square wave is due to the high frequencies...
as soon as you remove the higher frequencies from your square wave, it gets more and more ripply...
if you look at that link, but try your best to see it scrolling in reverse, you should get the idea...
but if you can do that, (in the forward direction), you'll notice that as the harmonics increase it's just like increasing your filter starting from the frequency of your square wave up...

mp3 wrote:Next time read the fine print.

what do you mean by this?...

mp3 wrote:pass-band ripple

i was not talking about pass-band ripple, nor did i mention it - stoopid!

Jimbrowski-One wrote:the exactness of a square wave is due to the high frequencies...
as soon as you remove the higher frequencies from your square wave, it gets more and more ripply...
if you look at that link, but try your best to see it scrolling in reverse, you should get the idea...
but if you can do that, (in the forward direction), you'll notice that as the harmonics increase it's just like increasing your filter starting from the frequency of your square wave up...

What the **** are you talking about? Were talking about filters dude... Take your square wave and be about your business.

Jimbrowski-One wrote:

mp3 wrote:Next time read the fine print.

what do you mean by this?...

The picture you linked to, if you scroll down the page, it says: "Additive synthesis of a square wave" Again, I fail to see how you are linking this picture to the clipping behavior of a lowpass filter.

Jimbrowski-One wrote:

mp3 wrote:pass-band ripple

i was not talking about pass-band ripple, nor did i mention it - stoopid!

Well you should be - stoopid! Its what is causing the filter to clip - stoopid!

Looks something like this:


Or this:


There's no headroom built into the filter. So when the passband ripples at certain frequencies (much lower than the squigglies in your pointless square wave), the filter clips. That's why they have an attenuation function, so you can add headroom if you need some. Stupid approach IMO.

Oh and just so you know, I've forgotten more math than you could remember.

Yeah, you guys both sound like a-holes...

Mazeone wrote:Yeah, you guys both sound like a-holes...

I know that's what I get for arguing with an idiot

Keep going! I find this very interesting. In my ears, the filter sounds something like this:

interesting...what does that mean in light of the following data?

^^^^^^^^^^^^^^^^^

nymenem LOLZ!!


yes its true mee too i have observed the clipping in a LOT of samples when i get down to filtering . i'm pretty sure that it has to do with the overshoot of the filter, regardless of the resonance setting... i notice this effect gets really bad when i wanna dial-in my hi-hats in JJOS1 where we still have the 2-filter option: i'll set up a hipass and a lopass together to get those hats sitting just right, but if the samples are already peaking too close to 0dBFS , therez crazy peaking and distortion... lately i just been tryin to leave at least 3db headroom in all my sample library, rather than normalising everything like i used to...and my mixes sound way cleaner as by-product of not boosting the noise floor in every sample, then turning it all down when mixing....

mp3 wrote:Theoretically it shouldn't happen, since the filter (without resonance applied)is supposed to subtract parts of the signal - thereby reducing the volume of the sample.

- Load up your favorite wave editor.
- Take a sample of a heavily distorted 808 kick.
- Normalize it to -3db.
- Filter it with a lowpass.
- Observe samples greater than -3db.
- Eat a...

If you forgot so much math, you need to go and learn some more. I don't know why you're even arguing with me. The math is obvious. My response was to the statement above that you made. An ideal filter will not necessarily reduce the "volume" of a sample.

Just think about it this way. If you have a wave of a certain frequency and when its peak occurs, the peak of another wave at another frequency occurs but in the opposite sign, the overall peak will be reduced. I thought that you would be able to see this from the picture that I attached. If you've done a lot of math then you should simply be able to conceptualize this... the picture is overkill.

Ok you win.

Better now?

mfc-42

good idea, except what about the other 29 filters you haven't replaced?

how about just fixing the damn filter???? filter isnt such a big deal imho even make it distort like the moogs, i'd say no problem..these shite filters dont even self-resonate with resonace maxed...maybe spending less time tryin to come up with all kinds of new features, and instead fixing the ones we already have so they actually work good.

i know its a crazy idea, but hey, its 2008, how about some common sense for a change?





















just kidding !!