A question for the Amp Sim Programmers

[reg3n]

Of course the first one will be the most time consuming. After that it will depend on what you want to do. I mean, so far I've been trying to model real gear based on schematics. I have to admit that I got very frustrated lately by the fact that almost every schematics floating over the net are loaded with mistakes. I have been working on a Bogner XTC amp sim lately and my first version sounded like sh**t. After weeks of searching, I have been finaly able to gather enough info about wrong cap and/or resistor values in order to have, I think, a decent sounding sim. Still have some work to do on it but it's getting better. Anyway, just to say that I'm gonna finish the projects I'm actually working on and after that I think I'm gonna start building my original sims, not based on any real gear.

LePou

now THAT will be totally awesome, i love the amp emulations, but i think you're just about to take the next step, and we've already seen you're capable of doing amazing stuff. I wonder what would happen if you and Ryan (my tone god) got together and make a new amp soft from scratch... :zombie:

 

[DaWolf]

The whole programming thing is so way over my head, I have the highest respect for all programmers out there.
It`s amazing that a bunch of words and numbers can be made into something you can hear,see or play.
The only thing I ever did was a program with basic on the C64. It was much like a website with stuff you could select and it would show text or show an animation with sounds.

 

[-Loco-]

LePou why not get an electrical engineer to check over the schematic for mistaken capacitor and resistor values to confirm the accuracy of the design. Would that not save you more time? Yeah alot of the electronics is fairly simple, it's the coding that pisses me off but as EE we have to learn C and C++ so i'll probably get a better grasp of this stuff in the 3rd year of my degree. These are excellent resources guys

 

[LePou]

Loco, how can you be sure that the value of a component is wrong on a schematic apart from sneaking inside a real amp ? If the schematic says that their is a capacitor in parallel with a resistor somewhere and that capacitor is not actually there in a real amp, how can you know ?

 

[valdiorn]

You should know how to read schematics. You should know how to model individual componentens and their interaction (network theory). Or you need someone to do this for you. I think that the guy most fit for the job would be the electrical engineer. Because a software engineer is miles away from being able to do this sort of stuff, whilst most electronics engineers do know their languages (c,c++ and even vhdl/verilog). A mathematician does also not posses the needed knowlegde (but in real life would probably get it faster)

So ALL HAIL THE ELECTRONICS DUDE !!!!

Sarcasm? I really have to ask because what I said is true, so your joke is moot (if it was meant as a joke, because everything you said was actually correct )

I recommended an electrical engineer as best fit for the job, after all, the guy want to accurately emulate a guitar amplifier component by component:

doing everything that is possible to emulate the real saturation characteristics of real amps.

^ That ^ requires you to be an electronics dude... not a software engineer or a mathematician

If this wasn't a sarcastic comment on me, ignore this post

 

[TheWinterSnow]

Loco, how can you be sure that the value of a component is wrong on a schematic apart from sneaking inside a real amp ? If the schematic says that their is a capacitor in parallel with a resistor somewhere and that capacitor is not actually there in a real amp, how can you know ?

there are standard values for certain circuits, having series or parallel circuits and where they are plugged into the whole circuit will tell an engineer what that device is doing, like a gain stage, a filter, current limiter, all the functions and what they do will be known to an EE, then will they know which values are incorrect or if a certain area is wired wrong. Only someone without EE experience will not be able to find out what is wrong with the circuit.

 

[Wild Hades]

Only someone without EE experience will not be able to find out what is wrong with the circuit.

I think you're misunderstanding LePou on the word "wrong".

For example, you can have a plate resistor of 220K on a schematic while the real amp has a 100K after a revision, for example.
Now, both 100K and 220K resistor are electronically "correct", but changing the value causes differences in the final sound and if you follow the schematics, you're using a "wrong" component compared to the real amp.

It would still work, but in a different way...

 

[TheWinterSnow]

I am going to bump this for the purpose of rewording the question here. Right now I am going to be attempting to recreate my speaker saturation in VST plugin form as well as try to get an impulse loader for my loadboxes. I think the way I questioned the OP made it seem like I have no clue about EE design or programming. I am more or less more wanting to know the actual development stages of making a VST simulation:

the programs used
how to determine the circuits algorithm
how to implement that algorithm into source code
how to interface the circuit simulation to the host and hardware (the interface and DAW)

From what I gathered it looks like most amp sim programmers are using spice to create their models and let spice do the algorithm coding. That's cool, but I would like to see sources on how to do this, and again how to implement said algorithms into code, and eventually how to make the .dll file.

Right now for the past few years I have been looking up programming or rather VST and DSP and there are nothing on coding circuit simulation, just merely the theory of operation of DSP (that all KVR has on their "VST Resources" page) and while many many papers online exist going over the theory and summation equations, without explaining what half the variables are and how it is to be implemented into code, well, it doesn't do me much good.

 

[onqel]

U can use spice to calculate memoryless stuff afaik.. And then use a lookup table to use it in c++

 

[TheWinterSnow]

So that you responded Jon, riddle me this, I don't ask you to teach me to program anything of the simulation world, but I ask if you could at least simplify the process of what you do and give some sources to learning. As I have said before I am not in the dark, I have my degree in Electrical Engineering, design and build audio electronics (obviously lol) and have taken programming classes and even did programming for embedded systems, but fuck me if I can't connect basic programming, DSP theory and SDK and other libraries (JUCE, IPlug etc.) and connect the dots in making a final product. I have the parts but no instruction book on how to put it together.

My two projects are pretty simple too, like I said a voltage divider with one pot and a couple diodes for a VST plugin and a stripped down impulse loader for a DSP chip.

My other option is that I could find programmers to do the job for me, but I can't pay jack shit for that as it would only be a percentage of sales for a certain period of time or up until a certain amount.