Page 3 of 8
Re: ODD's and EVEN's in ASM
Posted: Tue Oct 24, 2023 7:26 am
by Spogg
R&R wrote:
Clicks, clicks.. clicks
Seems hard to reduce when modulating some destinations... LP... or HP filter to subtract mod wave or something... Better than slew limiter...?
Any tips?
...
My current project (which may never get released) is using a basic LFO system to modulate a few targets. I have sine, triangle ramp and square. What I’ve done is bring in a LPF for the ramp and square. Since the maximum speed is 10Hz, I found by trial a fixed value which keeps the “punch” but doesn’t click. It uses a single pole low pass filter made by Martin (of course) and the frequency is set to 400Hz.
Nice clip by the way.
Re: ODD's and EVEN's in ASM
Posted: Tue Oct 24, 2023 8:31 am
by R&R
Good tip!
Might have to look into doing the same for some subtle smoothing, atleast for for my square:ish waveforms that lack smoothing or slope/edge controls... I use 80hz max for my LFO, "just" to make it difficult for myself... it feels like... looking at it in hindsight
Also found a
horrible sound error when modulating my MacGyvered implementation of sync saws in one of my Osc Modules. Either my solution can't handle the sharp/transient waveform from LFO or it's an out of range "frequency-thing" causing mayhem...
The simplest/shortest LP/HP filter I use is this asm snippet I made from a DSP snippet:
Code: Select all
streamin in;
streamin freq;
streamout LP;
streamout HP;
//LP=LP+((in-LP)*freq);
movaps xmm0,in;
subps xmm0,LP;
mulps xmm0,freq;
addps xmm0,LP;
movaps LP,xmm0;
//HP=in-LP;
movaps xmm1,in;
subps xmm1,xmm0;
movaps HP, xmm1;
Might be a terrible filter to use?
Spogg wrote:My current project (which may never get released) is using a basic LFO system to modulate a few targets.
Drone synth?
Spogg wrote:Nice clip by the way.
Thanks... I really should make real clips, level adjusted with instrument arrangement using a DAW... instead of just always running my plugin's .EXE and pressing "record" in Audacity...
I'm no musicmaker anyway so...
Re: ODD's and EVEN's in ASM
Posted: Tue Oct 24, 2023 12:32 pm
by Spogg
If your max frequency is 80Hz then a 400Hz -6dB cutoff would probably be ok.
I can’t comment on your filter because I don’t do filters. But Martin might have something to say…
R&R wrote:
Drone synth?
Might be, might be. Not saying.
Re: ODD's and EVEN's in ASM
Posted: Tue Oct 24, 2023 1:14 pm
by Tepeix
That's like what i call a zip filter.
I thing for the usage it might work.
I recommend to use this for the coefficient (so it stay the same at each sample rate) :
sin (Freq*pi/sample rate) then limit the result to max 1.
Comparing to a more classic one pole the cutoff will be a little lower.
The behavior at niquiest is a little different.
It's also different if we use higher order and resonance but without i suppose it's almost the same.
But why did you have a hp output ?
(Yep also, check if you don't have denormal with it.. Maybe not if the input is an lfo but possibly when the depht become zero)
Re: ODD's and EVEN's in ASM
Posted: Tue Oct 24, 2023 2:40 pm
by R&R
Spogg wrote:If your max frequency is 80Hz then a 400Hz -6dB cutoff would probably be ok.
I can’t comment on your filter because I don’t do filters. But Martin might have something to say…
Fortunately I don't modulate directly from LFO into my synths Osc Modules... rather the knobs inputs which mostly controls frequency and phases via remapping. Seems slew limiter does the job, but it destroys the wave more than a filter would... I'll likely end up modulating directly later on, in next synth so I'd better experiment with filters...
Beyond my main filters being mostly MVs asm's... I try to use his other ones since they are most likely to be solid... like the matched biquads.
But using the good:iest
filters might cost a bit of resources if used "here and there"
I "definately" don't do filters... being a few sandwiches short of a picnic when it comes to math
Spogg wrote:Might be, might be. Not saying.
I knew it!
A supercharged bagpipe sim...
Tepeix wrote:That's like what i call a zip filter.
I thing for the usage it might work.
I recommend to use this for the coefficient (so it stay the same at each sample rate) :
sin (Freq*pi/sample rate) then limit the result to max 1.
Comparing to a more classic one pole the cutoff will be a little lower.
The behavior at niquiest is a little different.
It's also different if we use higher order and resonance but without i suppose it's almost the same.
But why did you have a hp output ?
(Yep also, check if you don't have denormal with it.. Maybe not if the input is an lfo but possibly when the depht become zero)
I'll have to learn at bit more about filters.
Wish I was good at math so I could have joined in... in your filter posts. Looks interesting
Thanks, would have forgotten about sample rate
Yes I kind of feel that this simple filter could cause denormalization. I kept the HP part and just comment it away if not using it
Don't know if the HP part would work if used as DC block...
Re: ODD's and EVEN's in ASM
Posted: Tue Oct 24, 2023 3:41 pm
by martinvicanek
R&R wrote:I'll have to learn at bit more about filters.
I can offer this very basic FIlter crash course:
Years ago I posted it on the FS Guru site (RIP).
Re: ODD's and EVEN's in ASM
Posted: Tue Oct 24, 2023 4:02 pm
by R&R
Perfect!
Thanks Martin...
Re: ODD's and EVEN's in ASM
Posted: Thu Oct 26, 2023 11:44 am
by tulamide
Spogg wrote:Also, when you change anything, I believe it keeps the previous version in memory (maybe as a delta, I don’t know). That means a huge schematic will take progressively more memory and I’ve had “Out of memory” messages at times. My approach with huge schematics is to Save As frequently and close/open FS often, once I get into the “crash zone”.
Are you referring to the alpha? In that case, why didn't you report it? That's some serious info, that Maik should know!
Re: ODD's and EVEN's in ASM
Posted: Thu Oct 26, 2023 2:10 pm
by Spogg
tulamide wrote:Spogg wrote:Also, when you change anything, I believe it keeps the previous version in memory (maybe as a delta, I don’t know). That means a huge schematic will take progressively more memory and I’ve had “Out of memory” messages at times. My approach with huge schematics is to Save As frequently and close/open FS often, once I get into the “crash zone”.
Are you referring to the alpha? In that case, why didn't you report it? That's some serious info, that Maik should know!
No, only 3.06. I don’t develop in the alphas but I do help Johan and Ste sometimes and they use the alpha. So far I haven’t had the problem in the alpha, but I probably wouldn’t have done anyway. Also, it’s very unpredictable, so making a useful test case would be a thankless task.
I actually like the ability to undo after a Save or Save as, so for me it’s a small price to pay if I get the occasional issue as a result. I should add that when I have to restart FS it may well not error the next time.
Re: ODD's and EVEN's in ASM
Posted: Thu Oct 26, 2023 7:01 pm
by R&R
Well
endless addons and adjustments left to add/fix...
But here's an unfinished version 0.48 for anyone curious. Maybe someone has feedback on the UI etc...
More of a looker than a synth perhaps but it makes weird noises that's good enough for me
Saguaro.one v0.48 unfinished
edit: Only temporary... Link removed.
Under settings is the Random Detune and MIDI Unison (multi voice) option.
Also improved the LFO UI regarding uni/bipolar preview so it makes more sense whats happening, now especially with the recently added flippable LFO Depths knobs + the main LFO up/down option
Oh! and of course a couple of fresh new envelopes under the hood... Thanks for those Martin, much appreciated!