Hi!
Fooling around some more ,trying to program
a filter bass vsti kind of thing.
sounds like this now:
http://hem.bredband.net/b161498/filter3.mp3 (sweeping cut)
http://hem.bredband.net/b161498/filter31.mp3 (enveloped)
So how did this sound?
Any help to get some directions where to go
next would be fun!
//Daniel
Filter advice (again)
.
-
- KVRAF
- 8686 posts since 24 May, 2002 from Tutukaka, New Zealand
That sounded pretty darn good - I'd be happy with that kind of filtering on basslines that needed it.
It's a little difficult to give any in depth opinion on a filter without hearing it into self oscillation (or at least close to self oscillation) though.
What puts me off many VSTi filters tends to be the way the self oscillation is often a nasty thin whine - it's recognisable as almost a sine wave above the actual oscillator sound on some of the worst filters. A good filter should change its character, the harder you push it - I've had some VSTi's (that were quickly wiped off my HD!) where all that happens the more you overdrive the filter is that the nasty sine wave which is supposed to be self-oscillation simply gets louder, and the base oscillator sound gets quieter (And that's certainly not only the domain of digital filters - for example I never push my Juno hard with much resonance - Juno resonance is also nasty and thin and un-musical/characterless)
A good test of a filter is to get the standard WeeooW type sound - almost a vowel. As the filter is pushed harder, either the vowel changes somehow to a less and less recogniseable vowel sound, or it remains as a vowel type sound but changes the actual vowel whilst also getting squelchier - Does that make sense?
Think of the most characterful filters - say 303, OSCar, MS series, maybe even some Oberheim filters (although they were never my favourite ones). And I'm not saying they are the best filters - just that they have distinctive character (which is also why they are/were so popular too
They are all capable of getting very vocal/vowel sounds very easily - but the character changes as the filters are pushed - most of them become what you would probably call noisy even - definitely squelchier (maybe it has something to do with the bandwidth of their peak?) - but it's not that simple - they get noisier but it doesn't come across as a direct linear increase in noise - the noise somehow interacts with the oscillator sound rather than simply sitting on top of it.
And come to think of it - most of those better filters aren't the bog-standard 24dB/Oct - I don't know if part of it is the psychological thing that so many synths used 24dB/Oct that anything different stands out or not But Korg MSs had 12dB/Oct filters and were extremely noisy and characterful and POWERFUL (an MS10 in full flight beats the pants off the majority of 24dB/Oct comparable filters - which shouldn't be able to happen, but it does)
303s were 18dB/Oct (and like 303s or not, they certainly sounded different to the others of their time - 101s/202s are tame in comparison)
Yamah CS series had 12dB/Oct too - a completely different character from Korg MSs but capable of some wonderful sounds. Although the early Rolands had decent filters that were 24dB/Oct - Roland lost their way though - somehow their filters just got worse and worse as time went on - I quite like Jupiter 6s and 8s - they were nice and warm, but they weren't really anything to write home about in the filter department, and then the Junos just became bog standard.
So I'd say the first prerequisite is to be able to get a vowel sound, usually eeooow - much that I lke many VSTis, most of them can't do it convincingly. Then you work from there - must be able to introduce some noise into it without it just sounding like white noise. And squelch - I don't know how to describe that in words - but a good filter will squelch at low or high resonace.
Try out the ImpOSCar demo - that is certainly a good filter, and almost identical to the original. The MS20 - I don't think there's a demo is there? I'd post up some MP3s of my h/w 20, but I don't have a website etc. And Creakbox does a good emu of a 303. Fiddle around with the resonances of those 3 and you'll probably see what I'm waffling on about a bit better.
But, from the examples you posted, so far so good - nice sweeping etc, but we need to hear it being driven a bit harder for more proper feedback I'd say.
It's a little difficult to give any in depth opinion on a filter without hearing it into self oscillation (or at least close to self oscillation) though.
What puts me off many VSTi filters tends to be the way the self oscillation is often a nasty thin whine - it's recognisable as almost a sine wave above the actual oscillator sound on some of the worst filters. A good filter should change its character, the harder you push it - I've had some VSTi's (that were quickly wiped off my HD!) where all that happens the more you overdrive the filter is that the nasty sine wave which is supposed to be self-oscillation simply gets louder, and the base oscillator sound gets quieter
(And that's certainly not only the domain of digital filters - for example I never push my Juno hard with much resonance - Juno resonance is also nasty and thin and un-musical/characterless)A good test of a filter is to get the standard WeeooW type sound - almost a vowel. As the filter is pushed harder, either the vowel changes somehow to a less and less recogniseable vowel sound, or it remains as a vowel type sound but changes the actual vowel whilst also getting squelchier - Does that make sense?
Think of the most characterful filters - say 303, OSCar, MS series, maybe even some Oberheim filters (although they were never my favourite ones). And I'm not saying they are the best filters - just that they have distinctive character (which is also why they are/were so popular too
They are all capable of getting very vocal/vowel sounds very easily - but the character changes as the filters are pushed - most of them become what you would probably call noisy even - definitely squelchier (maybe it has something to do with the bandwidth of their peak?) - but it's not that simple - they get noisier but it doesn't come across as a direct linear increase in noise - the noise somehow interacts with the oscillator sound rather than simply sitting on top of it.And come to think of it - most of those better filters aren't the bog-standard 24dB/Oct - I don't know if part of it is the psychological thing that so many synths used 24dB/Oct that anything different stands out or not
But Korg MSs had 12dB/Oct filters and were extremely noisy and characterful and POWERFUL (an MS10 in full flight beats the pants off the majority of 24dB/Oct comparable filters - which shouldn't be able to happen, but it does)303s were 18dB/Oct (and like 303s or not, they certainly sounded different to the others of their time - 101s/202s are tame in comparison)
Yamah CS series had 12dB/Oct too - a completely different character from Korg MSs but capable of some wonderful sounds. Although the early Rolands had decent filters that were 24dB/Oct - Roland lost their way though - somehow their filters just got worse and worse as time went on - I quite like Jupiter 6s and 8s - they were nice and warm, but they weren't really anything to write home about in the filter department, and then the Junos just became bog standard.
So I'd say the first prerequisite is to be able to get a vowel sound, usually eeooow - much that I lke many VSTis, most of them can't do it convincingly. Then you work from there - must be able to introduce some noise into it without it just sounding like white noise. And squelch - I don't know how to describe that in words - but a good filter will squelch at low or high resonace.
Try out the ImpOSCar demo - that is certainly a good filter, and almost identical to the original. The MS20 - I don't think there's a demo is there? I'd post up some MP3s of my h/w 20, but I don't have a website etc. And Creakbox does a good emu of a 303. Fiddle around with the resonances of those 3 and you'll probably see what I'm waffling on about a bit better.
But, from the examples you posted, so far so good - nice sweeping etc, but we need to hear it being driven a bit harder for more proper feedback I'd say.
-
- KVRian
- Topic Starter
- 829 posts since 13 Jul, 2002
Many thanks for that superb info!
Yes I do get that nasty whine sound when you
push the resonance but I actually had a more of
noise burst at one time... got to do some back tracking here.....
I don´t know how many db this filter produce since
I only try to get the sound by listening but the
way it is built up it will be very easy to experiment with stuff like achieving noise ect..
I´ll be back in acouple of days! (hopfully with a
nice squelchier sound. )
//Daniel
Yes I do get that nasty whine sound when you
push the resonance but I actually had a more of
noise burst at one time... got to do some back tracking here.....
I don´t know how many db this filter produce since
I only try to get the sound by listening but the
way it is built up it will be very easy to experiment with stuff like achieving noise ect..
I´ll be back in acouple of days! (hopfully with a
nice squelchier sound.
)//Daniel