making a flowstone game

[wlangfor@uoguelph.ca]

Hi all, making a game with flowstone and it will involve learning a lot of math but I will try



here is the beginning which made Me feel it's possible: (gdrive)
https://drive.google.com/file/d/1Gc6I7YvzWJQGf3v_lCn6e_-WzeY6watk/view?usp=sharing

and here is the youtube video:
https://youtu.be/iAV0RpG9JoQ

so I'll be looking for useful ruby gems. Please let Me know if you have found any I can use.

[Lex248]

I also made a game on the FlowStone.

Memory Avatars (blocks)
http://nalexsoft.blogspot.com/2019/05/m ... atars.html

[wlangfor@uoguelph.ca]

I also made a game on the FlowStone.

Memory Avatars (blocks)
http://nalexsoft.blogspot.com/2019/05/m ... atars.html

that's really cool

I'll check it out.

BTW, to anyone reading I have been having an issue regarding feedback. And that is in the way of having eighty unopened mono connections.

They're not problematic when in the case of only twenty at a time, but if I'm to have more than that, there is a spike i memory usage. I am fearful if I were to use that much memory in the case of a game or a very serious sampler that there would be issues on the end user's computer and I wondered if anyone knew an optimization trick.

TIA if anyone knows how:



and here is the schematic to study the problem:

one pad paradise experiment version 2.rar

forty samples 1 mb - 1.2 secs 1 percent cpu but spike cpu

(3.91 MiB) Downloaded 907 times

BTW, I'll be looking up some simplified trigonometry in order to maybe optimize the math for a basic strategy 3d like 2d game. I'll definitely post regarding those results.

[adamszabo]

What is this even trying to be?

[wlangfor@uoguelph.ca]

What is this even trying to be?

It's like sins of a solar empire - it's like ages of empire in space with spaceships. Pretty ingenious but it seems like the programming team was replaced an the last version of it stopped working right.

It's too bad and that had been My feeling that I could make a new sub version sort of.

I highly recommend either the first or the sequel but like I said, diplomacy, the last version does not work. Seems like an issue with ships in other galaxies and the db involving a firefight overloading everything.

[wlangfor@uoguelph.ca]

So, I wanted to add about the most recent audio schematic added. It's really cool that you can have so many samples live and running with next to no memory required. But there in lies the issue with the feedback.

Keep in mind too, that in this example (derived from loopeytunes btw) the sample plays all the way through to the end.

And that's a very different thing, most samplers just play on note down. So there's added functionality which is very useful for the individual who wishes to make something above and beyond; like for instance a piano.

I of course am allied with some other elite programmers and promised them I'd learn about samplers so that's an ulterior motive; but I'll share the end product.

KK, so here is the other stuff about the game:
it will give you an idea about what I have planned.



It's not as simple as you would think but I was thinking against trigonometry due to the fact that basing math upon the telemetry might become extremely convoluted. But perhaps that when I create the move the cube which turns and zooms and turns with mouse interaction I can decide and perhaps you can decide what will ultimately make this product work better.

[adamszabo]

But why are you using all that midi to poly and lot of mem stuff? What are the triggers supposed to do? I think there is a thousand times simpler way to achieve what you want because right now it looks like a huge mess unfortunately

[wlangfor@uoguelph.ca]

But why are you using all that midi to poly and lot of mem stuff? What are the triggers supposed to do? I think there is a thousand times simpler way to achieve what you want because right now it looks like a huge mess unfortunately

tbh, much less than most I've tried. There's less asm,dspcode and though I haven't used stock interpolated there's next to no memory draw.

Frankly, it's an experiment to see how little cpu might be used to compile a large database of samples together. I'd been noticing for instance when making something like a piano, that may have for instance velocity zones, especially in the case of a SFZ sampler. It becomes ridiculous. So, humble as it may be it's using a bit less CPU, but I've found when combining the use of many samples together (when stereo is added to the audio out - many times over) that it is getting feedback.

This will ultimately be used for Ruby to call triggers on mouse interaction, like say for a button or for a computer player's speech. But, keeping in mind that I will likely replace wave with wave array and many times over like the included example. If you can make something neater, i'd like to see it. But will they individual volume controls without transmuting velocity is what I wonder. And won't you have to separate the midi mapping in some manner to achieve it? It'd be educational if you could provide a simple example.

[wlangfor@uoguelph.ca]

I put together some stuff to illustrate the strengths of this method. I've been able to make the same method now only use 1% of cpu (or less) in both 32 bit and 64 with even 80 samples. That could even be quadrupled as well. I think it comes down to what your needs are. I mean, if you are making an instrument that uses samples and you want to recreate a grand piano then the amount of octaves and needs are horrendous.

Sufficed, I think that this method is the beginning of a practical solution towards those goals. i was able to take both screenshots and a proof of concept and of experiment in mp4 form.

here's the images:


and here is what the schematic looks like:


Thanks to Adam for pointing out that it looked somewhat sloppy and I cleaned it up.

so, here is that file.
https://drive.google.com/open?id=1yFT1J2w7EvSgTCVYjpYsWTEvNVAXzPXT

and to better illustrate the strength of only ustilizing 1% of memory in either fl studio 10 or 20 and in either 32 or 64 bit here is the link to that MP4:
https://youtu.be/QmleRprYg28

[wlangfor@uoguelph.ca]

I was thinking about this new experiment in depth. I realize it seems perfectly suited to a game engine; but then, we do everything here.

So, eventually I'd like to use the same idea to work on a sampler that would be a bit more efficient using the same layout. And I guess what it comes down to is translating the time it would take to process delays and effects such as adsr and translate them to milliseconds. Then I suppose that it comes down to possibly using ruby for the timer so as to ensure that the pulse included to activate that portion is activated for the valid time during FTRT sampling.

I'll have something on the studio EQ, and something on the reverb a bit sooner than I get back to this thread but it shouldn't be more than three days I'd think. GL with your sampling.