From the original SynthMaker forum (2004–2014), with its members’ names and dates.
Klentock1 ·
After weeks off experimenting and trying to get my head round dsp (I'm getting nowhere very fast but i can just about understand a little of the basics of it all) I thought I'd enquire on the forum to see if anyone could offer me tips.
I've took an example from one of my hardware synths as to what I'm after. Have a listen here.. http://www.box.net/shared/v7uvucf0gk
I'm not after a self oscillating filter or anything.
Just something that can handle quite a high resonance.
I've tried feeding the filter back into itself with it going into the bottom point of a stream subtract and adding a clipper in there just to smooth things out but it doesnt work.
I've also spent a few days looking at the stock SM filters and some filters floating around the forum but to no avail.
If someone could point me in the right direction as to how i can achieve this then it would be very much appreciated.
I'm not asking for anyone to give me a full blown example...just a little nudge in the right direction would be much abliged.
Thanks
Klentock
MichaelBenjamin ·
for filter experiments you could try to mix a lowpass with a bandpass for the resonance. maybe chain 2 bandpass to get a steeper slope. then you could eq that bandpass which delivers the resonance with a further eq. just some general suggestions.
Klentock1 ·
thanks very much for your reply.
I tried it with a highpass but not with a bandpass...i will check it out. thanks
Klentock1 ·
Ah excellent...i've got something very workable now.
Thank you very much!
Klentock1 ·
okay, more help needed! 😊
Certain frequencies are boosted when some notes are played..here is an example http://www.box.net/shared/cucr184ws0
Is there anyway to halt this from happening?
Tom7777 ·
Certain frequencies are boosted when some notes are played..
Yeah imo this is a bug and i have mentioned it to malc before but i don't think he thought it was so not much happened, i notice this with resonance alot on filters but some will do it and other won't. Seems like a bug anyways when some notes are fine and overs completely scream/boost.
MichaelBenjamin ·
you really think that is a bug in sm?
how do you come to that conclusion, could you explain?
MichaelBenjamin ·
hey klentock, i am glad it worked!
finally a suggestion by me that actually leads to something useful;) pff
Tom7777 ·
you really think that is a bug in sm?
how do you come to that conclusion, could you explain?
Yeah because it happens with a few different filters and its to do with resonance as it will only happen when it's high in most cases, i have come across it a few times with different projects.
Tom7777 ·
Hehe ok so just to show what im talking about, i made a quick example, this uses mycos optimized SVF and he is a good coder im sure, this will happen with other filters also though, so yeah im not making it up, and hopefully it doesn't just happen on my pc hehe, i put some simple instructions in the osm 🙂
MichaelBenjamin ·
hm i tested this, its as you describe.. but could that not be a normal reaction of a fixed freqency filter to different input freqencys.. so that the v key in your schematic has a freqency that resonates more than the b key?
Tom7777 ·
Im not sure, however the keys that sound ok all sound not too much of a difference but certain keys will jump up alot, the difference from harmonic reso to a screaming one just doesn't seem right.
Acrobat ·
Tom7777 wrote: Im not sure, however the keys that sound ok all sound not too much of a difference but certain keys will jump up alot, the difference from harmonic reso to a screaming one just doesn't seem right.
Hi Dan, for a chance are u listening with headphones or with monitors?
Cause typically rooms resonates at certain frequencies...
Tom7777 ·
headphones, do you not get the screaming sound with the example i posted then?
Acrobat ·
Tom7777 wrote: headphones 🙂
well if the filter is a combination of two chained codes, I think resonance will not work good ANYWAY
if it's just 1 code, there might be a problem. just my 2 cents.
Acrobat ·
Acrobat wrote:
Tom7777 wrote: headphones 🙂
well if the filter is a combination of two chained codes, I think resonance will not work good ANYWAY
if it's just 1 code, there might be a problem. just my 2 cents.
EDIT: - I mean self-resonance of course
Tom7777 ·
Well it's just Myco's optimized SVF, just 1 of them, take a look, i don't get why im being questioned about this lol, try for yourself 😉 Something seems to be up with it, if it's the normal thing i wonder if theres any way to re-adjust it so it doesn't overdrive.
Acrobat ·
Tom7777 wrote: Well it's just Myco's optimized SVF, just 1 of them, take a look, i don't get why im being questioned about this lol, try for yourself 😉 Something seems to be up with it, if it's the normal thing i wonder if theres any way to re-adjust it so it doesn't overdrive.
I' ll do that, anyway I tried the RBJ Lowpass ASM version but it didn't worked as the original Shifrin's.
It seemed to completely destroy the sound signal so I leave the coded one.
Tom7777 ·
-
kirkagur ·
I think that you need a full key track to make it work !!!!
MyCo ·
The self oscillating problem inside the SVF is because it is not a stable filter. It looks like they tried to fix it by tripple processing. The easiest way to get around it, is to limit the resonance input values below 1.0
MyCo ·
Acrobat wrote: I' ll do that, anyway I tried the RBJ Lowpass ASM version but it didn't worked as the original Shifrin's.
It seemed to completely destroy the sound signal so I leave the coded one.
The original works with a frequency input in Hz, while my version is constructed to work as replacements for other filters, so it uses a normalized frequency input.
Acrobat ·
MyCo wrote:
Acrobat wrote: I' ll do that, anyway I tried the RBJ Lowpass ASM version but it didn't worked as the original Shifrin's.
It seemed to completely destroy the sound signal so I leave the coded one.
The original works with a frequency input in Hz, while my version is constructed to work as replacements for other filters, so it uses a normalized frequency input.
It's not that problem, the fact is your RBJs seem to work bad with very fast changes in phase.
Could be related to the 128 hop, maybe avoiding the hop they work good but I can't manage ASM... 🙁
EDIT: - oh it's the "range" module I was missing, has to be copied along with the filter itself, forget about that, sorry!
Never understood why those SM multiplicators, why the "main" cutoff control weights more than the "modulated" one, anyway... 😕 x128 could mean 10 octaves plus eight semitones? and "x50" WTH it means? 😕
MyCo ·
they are just scale factors, the don't have to have a meaning at all. If you want an other modulation range, you could change the values. The scale factor of the cutoff input is 144 which results in a maximum frequency which is just a little bit above nyquisit even if you play the lowest possible note.
infuzion ·
Acrobat wrote: It's not that problem, the fact is your {MyCo} RBJs seem to work bad with very fast changes in phase.
Could be related to the 128 hop, maybe avoiding the hop they work good but I can't manage ASM...
Here is the part of the RBJ that controls the hop:
mov eax,ecx;
and eax,127;
cmp eax,0;
jnz hopend;
//hop(128)
The larger hop could cause that. Just reduce the 127 to the value of (hop-1)... I'm not sure if it has to be a power of 2, but try a hop value of "7" so the line reads and eax,7; & to see if that helps.
Tom7777 ·
Myco, you didn't reply to me about the filter resonance overdrive right and that was for Acrobat etc? If not the reso input is 1 on the SVF so not past it's limit, this seems like a bug but i may be wrong, i may possibly start a post in the bugs section, it's possible it just happens in SM and reacts differently in hosts but i have not tested yet. Imo it's a bug or at least seems that way and you can have some nice sounding filters and high reso fx but hitting the wrong key can spoil it imo 🙂
MyCo ·
yes, seems to be a bug, the resonance input should be limited to 0.99
But the filter isn't stable at high volume on input. So you still have got overdrive effects even on lower resonance eg. I've got it at 0.85
Wrote something here? To have a post removed or your name hidden, write to [email protected].
SynthMaker