Circuit modeled filter, how to?

[aciddose]

of course you can not, but who gives a damn about that? should i state here now that "an apple is not an orange" ?

what you were saying is that "it is not simple to model this, your simplified method can not be applied in this circumstance" and that is not true. i'm not arguing that what you said about "apples are not oranges" is incorrect, who could possibly argue against that?

you were trying to argue that the simplified macro-models can not accurately model more complex electronic circuits, or that it is much more difficult to accurately model any more complex circuit. that also is not true - larger circuits to not become relatively any more difficult to model, difficulty increases linearly with respect to the complexity of the circuit.

other than modeling from the circuit, rather than modeling the fundamental functions (A->B->C vs. A->C) there is no reason to include any electronic component modeling in a dsp circuit model. your argument was that component modeling is necessary and that a basic macro-model is not a good starting point. your intent was to say that you can not go from A->B using a macro-model as a starting point. as if the macro-model path would lead to a dead-end.

if this was not your intent, ok, i'm sorry for misunderstanding you and having argued for such a long time. you have also been arguing however and have never once said "macro-models are accurate, macro-models work and macro-models are a good starting point".. your arguments have all been against that.

if it was not your intent, why would you at random bring up the fact that "A != B", or that "this incorrect function is incorrect". this makes no sense to me.

btw, "you cant get to outer-space only by walking."

[andy_FX]

of course you can not, but who gives a damn about that? should i state here now that "an apple is not an orange" ?

what you were saying is that "it is not simple to model this, your simplified method can not be applied in this circumstance" and that is not true. ...

Again, I never said that. Can you please stop paraphrasing what I am saying and just quote me if you want to quote me?

Andrew

[aciddose]

fine, then you have no disagreement with my original post, and everything you said opposing my suggested methods was not your intention.

you for some reason desired to say "this incorrect function is incorrect" without any purpose that i can understand. all i can see here is that you were implying that a beginner would be too 'dumb' to understand your example would be incorrect. you later implied that a beginner would be too 'dumb' to implement my function.

i do not agree that these people are too 'dumb' to do these things. yet you propose they start by learning all the functions listed on the wikipedia page i've linked?

something does not seem to make sense looking at the logic you've applied here...

[andy_FX]

fine, then you have no disagreement with my original post, and everything you said opposing my suggested methods was not your intention.

Are making things up again and paraphrasing and re-wording what you think I mean?

you for some reason desired to say "this incorrect function is incorrect" without any purpose that i can understand. all i can see here is that you were implying that a beginner would be too 'dumb' to understand your example would be incorrect.

I never called anyone dumb. I was making it obvious for people with little circuit skills that components in circuits that are not buffered from each other interact, so to model them you need to know some circuit math to model them properly.

you later implied that a beginner would be too 'dumb' to implement my function.

No, I never called anyone dumb. What I said was that to derive what you gave as an example to model the RCDD circuit you would need to know some circuit math.

i do not agree that these people are too 'dumb' to do these things. yet you propose they start by learning all the functions listed on the wikipedia page i've linked?

Here we go with the paraphrasing incorrectly again. I never called anyone dumb. Please provide me with the link to where I called people names. I have always been about helping people understand how to model circuits, and I have never assumed that people have any particular level of intelligence. I don't think any of the maths with any of this stuff is particularly hard.

something does not seem to make sense looking at the logic you've applied here...

My logic is just fine.

Andrew

[aciddose]

no, you are avoiding the fact that you were implying they're dumb, or that they need to learn something more than they know before they can start to be productive in this field. that is not true.

"... so to model them you need to know some circuit math to model them properly."

...and i have been arguing that NO YOU DO NOT.

[andy_FX]

no, you are avoiding the fact that you were implying they're dumb, or that they need to learn something more than they know before they can start to be productive in this field. that is not true.

"... so to model them you need to know some circuit math to model them properly."

...and i have been arguing that NO YOU DO NOT.

Did I scare you with the word math? Let me say instead "circuit knowledge" or how about simply: "you have to know about circuits to model them properly", which has the same meaning in what I intended in the above.

Andrew

[aciddose]

actually, you do not even really need to know that. i'd say at most you need to understand high-school level math, like graphing, basic linear algebra and trig. you then also simply need to understand modular systems and off you can go..

(maybe these things i'm talking about above are actually the same things you refer to when you say 'circuit theory'.. since many of them are also used there.)

nobody should be even considering writing something beyond the basic volume control vst if they are not at that point anyway.

i can agree that knowing some basic circuit theory will help a lot, but this is where "need" ends and "helps" begins. a beginner certainly in my opinion shouldn't start worrying about the very complicated functions you need to actually do circuit simulation until they've already learned and understood all the basic concepts.

i still think anybody could have just as easily written the two lines of code that i did, using the same methods. i didnt use any method other than what i described.. looking at the waveforms and guessing what function is taking place. all the knowledge about functions which influenced my guess are not related to electronics, they're basic functions that everyone should already know.

i know that most people dont, but perhaps we would want to suggest reading back through the high-school math textbook or taking a course at a local college for those who have forgotten or didn't pay attention in classes.

[Urs]

a beginner certainly in my opinion shouldn't start worrying about the very complicated functions you need to actually do circuit simulation until they've already learned and understood all the basic concepts.

afaiaa the original poster wanted to know about circuit modeling and not about those beginner methods that you always try to sell us here. Sorry but I find this annoying.

[Shy]

The original poster had "no idea on where to start", hence he's a beginner. The argument started with claims about what prior knowledge you have to have to be able to start. I don't understand what exactly you find annoying.

[Urs]

The original poster had "no idea on where to start", hence he's a beginner. The argument started with claims about how much prior knowledge you have to have to be able to start. I don't understand what exactly you find annoying though.

He didn't know where to start on circuit modeling. That doesn't imply that he's a beginner in the sense that he has no prior knowledge. Unlike some other people posting here the OP has already released several successful products. Which indicates that he knows/uses all the stuff that AD tries to teach him. Andy on the other hand has posted some really usable hints at the topic. Hints that are good enough to get an experienced dsp developer started on the topic of circuit modeling. I have no idea why AD needs to spread his doubt on that over and over again and again and even start becoming offensive. That is something that annoys me. I would have prefered to read more about Andy's insights rather than naive waveshaper theories that pop up all over this thread. Noone who's listened to Andy's stuff will want to miss out on that because it's awesome.

[hifiboom]

regarding the original thread topic, it would be somehow cool, to implement a software system, where you have various lowest circuit atoms (like capacitors and that stuff) and where you could recreate the circuit visually in digital domain.

I think about something like ms visio where you place elements into an project arrangement context and then connect them in via cables.

Finally the software would replace all visually circuit elements with VST code to simulate the circuit in digital domain. So soemthing like a synthedit for circuit level.

(btw you should be able to switch on oversampling on all single elements individually ...or grouped on the whole circuit)

It would be possible to build, at least try out different circuit models quite quickly, even if you don't have an idea what the circuit is doing exactly at start.

If and how you render the final code down for optimizations is another story.

Even stuff, like code optimizing could be done by such a system automatically, if someone defines some detection rules for replacing code parts.

Later you could use the final outcome in a bigger context...

[aciddose]

i think the quote "There is no 'real function'." from this thread is pretty cool. best example of a discussion between an abstract and non-abstract thinker, ever.

andy, do you have any demos of strobe that do not suck? this is supposed to be a sh-101 emulation i thought, yet none of those sound anything even remotely like a sh-101. sounds like some random parameter modulations on shitty pro-5 quality sounds.. not intending to offend here, i'm being honest.

i actually want to do some A/B with this sh-101, however those 'demos' make it impossible. would really appreciate if you could post/link something good.

i was thinking something like:
http://xhip.cjb.net/temp/public/sh-101.wav.mp3

warning, this one not a sh-101: http://xhip.cjb.net/temp/public/boo1.mp3
(just the type of sequence/sounds i'm talking about)

wanted to see what all urs' fuss was about.

[Shy]

Also the sample I posted earlier in the thread is an SH-101. Again, no emulation of any kind sounds remotely like it.

[boimb]

The original poster had "no idea on where to start", hence he's a beginner. The argument started with claims about how much prior knowledge you have to have to be able to start. I don't understand what exactly you find annoying though.

He didn't know where to start on circuit modeling. That doesn't imply that he's a beginner in the sense that he has no prior knowledge. Unlike some other people posting here the OP has already released several successful products. Which indicates that he knows/uses all the stuff that AD tries to teach him. Andy on the other hand has posted some really usable hints at the topic. Hints that are good enough to get an experienced dsp developer started on the topic of circuit modeling. I have no idea why AD needs to spread his doubt on that over and over again and again and even start becoming offensive. That is something that annoys me. I would have prefered to read more about Andy's insights rather than naive waveshaper theories that pop up all over this thread. Noone who's listened to Andy's stuff will want to miss out on that because it's awesome.

yeah yeah yeah... i think andy is as least as bitchy as aciddose, period.
even more i think this aciddose dude, is not naive at all. as far as i can judge i think aciddose is pulling some very sensitive strings regarding this andy dude . some people underestimate the knowledge of this guy. thinking he's some teenager , maybe he is, idunno, but he frikking bites, and in the end his demo's ruleeeee,which is evidence enough

feed the troll

[andy_FX]

andy, do you have any demos of strobe that do not suck?

I'll hook you up with a copy and you can play to your hearts content. Final graphics are still in progress but there won't be any changes to the core dsp.

this is supposed to be a sh-101 emulation i thought

No it's not. It is inspired by synths such as the Roland SH-101, the SCI Pro-One, the Moog Prodigy etc, but not a model of any of them. It's not just the kind of sound of the filter etc I'm after, it's also the types of modulation and the limitations present in the originals that I like to have as well. I want to keep strobe simple to start using, but get deeper as you go.

yet none of those sound anything even remotely like a sh-101. sounds like some random parameter modulations on shitty pro-5 quality sounds.. not intending to offend here, i'm being honest.

Here is an example of me manually sweeping the filter of an sh-101 and a simple model of the sh-101 filter I did. Since I'm using midi the resolution on the digital sweeps isn't as good as analog, but you get the idea. The digital version is x2 oversampled of 44100:
http://vellocet.com/dsp/analog/roland-s ... analog.mp3

Here is a page with some self oscillation harmonic plots, which I find useful only as a double check once you have modeled the circuit of the filter, but thought might be useful to others:
http://vellocet.com/dsp/analog/SelfOscHarmonics.html

I think the key sound to strobe is the sub oscs. There is a saw and sqr with pwm, then a tri, sin, saw and sqr with pwm set of subs each of which can be switched to -1, -2, or -3 octaves. There is self sync, which is cheating a bit, but very useful, oscillator stack. The filter has around 30 filter responses created by making sums of the different taps. I especially like the notch filters where I have a notch placed exactly an octave down or up from the resonant peak. There is also a cool one with two notches, one an octave down, one an octave up, and a resonant peak in the middle, so this is kind of 6 poles of action out of a 4 pole filter Here are some graphs of the responses:
http://vellocet.com/dsp/CascadedFilterResponses/

Andrew