AHDSR into an AHD

[R&R]

Cool! Thanks Tepeix...

This gives a slighly better picture of the problem since i'm not familiar with asm, even less so any limitations of it...

as i suppose that you want to reset each voice independently..

No actually I just wan't to reset all playing streams at once with my trigger button (and they should also use same seed as input) and preferebly keep the seed input and functionality as MVs original asm. Frustrating, since the only thing I would really need is to retrigger the entire asm to "start over" using the same seed as input:ed from the poly input.

The seed will be a variable input using a knob etc. Atleast thats the idea I have

but we could reset it to zero.

All playing voices state:s to be reset to initial state using seed at once when sync is triggered is my idea

[R&R]

So maybe if I just do something like:

Code: Select all

stage0:
cvtps2dq xmm0,seed;
movaps state,xmm0;
cvtps2dq xmm0,seed;
movaps initstate,xmm0;


And then at stage2 do something like a comparison if an input trigger converted to "int" occured and then replace state with initial state that way?
Or are conditions (comparative integer operators) not available/viable in FS? Maybe that is what you are refering to Tepeix...?

[Tepeix]

Yes, but you could also use cvtps2dq xmm0,seed;
movaps state,xmm0; movaps state0,xmm0;

But yes, in 3.6 we do not have a comparator for int.
I suppose cmppd is only in the alpha.

But if you want to reset all the four channel at the same time the skip become more easy :

mov eax, reset[0]; cmp eax,0; jz reseting;
movaps xmm0,state0; movaps state,xmm0;
reseting:

Note that i do something a little bad in those line. Some would prefer this :
cvtps2dq xmm0,reset; movd eax,xmm0;
cmp eax,0; jz reseting;
movaps xmm0,state0; movaps state,xmm0;
reseting:


Yep, eax is supposed to be an int when using cmp.
That's in some way the poor comparator of int but working only with one channel at a time for those that are in 3.6. (and using skip..)

But as we compare against 0, the float will do the same as the int because 0 is the same bit structure.
And it always work for me.. But, it won't if we compare with 1 or any other number..

But if each channel at the same time, why to reset the random ?

[R&R]

The last fsm you posted looks very much like what I would expect it to look like, but it doesn't seem to work though. It doesn't reset/restart the noise it seems...

But if each channel at the same time, why to reset the random ?

Hardsync via triggerbutton(bool)/automation for use with my LFO The white noise can be considered a "by seed predefined number sequence". Not a pure random number generator.

The white noise is actually the source of random number sequences in MV's S&H and glide LFO modules... and maybe other things that it might be used for.

I also have an idea for a modulation-mix between his glide modules and pure waveforms. Don't know yet, if that particular idea will work...

[Tepeix]

Strange, it reset for me. Maybe it's because of the nature of the sync. (never used this before)
If it's a boolean it might be converted in some way..

[R&R]

Strange, it reset for me. Maybe it's because of the nature of the sync. (never used this before)
If it's a boolean it might be converted in some way..

Sorry! you are absolutely correct it does reset I forgot to modify my module fed by the noise

Thank you! Awesome...
This is excellent, now I can focus on the other modules...

[R&R]

I sent you and MV a PM, if you are curious of the modules you made and modified for me...

[martinvicanek]

I am always impressed when a creative mind puts together bits and pieces to build something novel, fresh and surprising. Perhaps it is an emergent property, a new quality that is not present in the components, but comes into being when these are assembled in a meaningful way.

[R&R]

I am always impressed when a creative mind puts together bits and pieces to build something novel, fresh and surprising. Perhaps it is an emergent property, a new quality that is not present in the components, but comes into being when these are assembled in a meaningful way.

What do they say? "Need is the mother of all innovation"

It's an interesting topic also In UX user experience design and within the field of testing there is often a need for users that have zero previous experience to test stuff.

I often realize how wide the knowledge gap is, and also the difference in angle and approach, when for example you or Tepeix responds with very advanced math solutions based of your interpretation of my (often "bad" and lacking) problem descriptions

In this case regarding FS... I think my ignorance and large lack of knowledge around alot of things can be both beneficial and a curse at the same time luckily I've had your help to solve stuff...

[R&R]

I've had to do alot of compromises in my synth...

I just finished with throwing in MVs Noise (Random) S&H now

Since stage0: doesn't care being fed value at "poly stream start" it seems... I split
the noise into two versions, unfortunately always running at the same time to avoid schematic reloads

Random_S&H.jpg (85.82 KiB) Viewed 8527 times

I also wanted a mode where seed is fed by note pitch 0-144 but that is probably just as much of a problem as
feeding any other poly to stage0: operations...?

Edit: Maybe the conversion to integer can be done at each sample instead... I'll have to look at it later.
Now on to more important issues to deal with