Warrior General Games

I have a question about the outputs for the A.I. systems.

Say you have a total of 10 outputs for a select play. The %'s have to add up to 100 right. This is the way I've always done it.

However someone said you could have say 20 outputs and have all the %'s at 10% .... But if you add them up it's gonna be 200.

Does it matter??

Thanks in advance.

It doesn't matter, they will be adjusted to 100% for the game.

I have another question about this:

For a single input, does the AI read from the top down? For instance is each percent chance of an output rolled based on its percentage, in sequence, or is it an actual percentage chance?

In other words, let's say you have 10 outputs for an input, all set to 10%. Is the first output any more likely to be fired off than the 10th?

Originally posted by foofighter24
I have another question about this:

For a single input, does the AI read from the top down? For instance is each percent chance of an output rolled based on its percentage, in sequence, or is it an actual percentage chance?

In other words, let's say you have 10 outputs for an input, all set to 10%. Is the first output any more likely to be fired off than the 10th?

I wouldn't think so. Each time you press the update button, the list changes order, so top down readings don't seem to mean much for outputs.

Originally posted by foofighter24
I have another question about this:

For a single input, does the AI read from the top down? For instance is each percent chance of an output rolled based on its percentage, in sequence, or is it an actual percentage chance?

In other words, let's say you have 10 outputs for an input, all set to 10%. Is the first output any more likely to be fired off than the 10th?

I would assume that it works like most weighted RNG.


Like this:


Output #1 - 25%
Output #2 - 15%
Output #3 - 40%
Output #4 - 20%

The RNG does a roll, 1 to 100.

If it comes up 1-25, #1 is selected.
If it comes up 26-40, #2 is selected.
If it comes up 41 to 80, #3 is selected.
If it comes up 81 to 100, #4 is selected.


The order doesn't matter, unless Bort let his dog write that piece of code.

I have a question which is also sort of related...

if you put in a quarter-specific input, and set it's output(s) to less than 100% total, will it check that %, and if it doesn't roll that % then it will go on to the default playbook for that situation otherwise?

for example, say you set up a quarter specific input (e.g. 2nd and short in the 1st quarter), but then you only give it one output and set that output at 50%... would it run that 50% of the time, and the other 50% just go to the regular (non-qtr specific) output for that situation?

Or would it just adjust that 50% up to 100% like you're saying it does if you are over 100%, and override the default input for that situation every time?

Originally posted by tetura
I have a question which is also sort of related...

if you put in a quarter-specific input, and set it's output(s) to less than 100% total, will it check that %, and if it doesn't roll that % then it will go on to the default playbook for that situation otherwise?

for example, say you set up a quarter specific input (e.g. 2nd and short in the 1st quarter), but then you only give it one output and set that output at 50%... would it run that 50% of the time, and the other 50% just go to the regular (non-qtr specific) output for that situation?

Or would it just adjust that 50% up to 100% like you're saying it does if you are over 100%, and override the default input for that situation every time?

You would run that play 100%.

Are these answers specualtion, proven by experience, or from Bort's mouth? I am not trying to be an asshole, but it is kinda ridiculous how closed lipped Bort is on the mechanics of the game. These things should not require speculation.

my question can be tested. i just haven't done it yet. it is something we can figure out. but I'm also wondering whether you know from testing or from having read it someplace from bort. and yeah it would be nice if there was some central compiled guide for some stuff.

It doesn't matter if you're under or over 100%, it will get adjusted to 100%. If you only have one output, it = 100%, although I'm not sure what will happen if you set it to 0, I've never tried it.

Does it matter how I know? You asked, and were answered. If you choose to test it anyway, why ask in the first place? Just wasting both our time.

Originally posted by Baustin
It doesn't matter if you're under or over 100%, it will get adjusted to 100%. If you only have one output, it = 100%, although I'm not sure what will happen if you set it to 0, I've never tried it.

Does it matter how I know? You asked, and were answered. If you choose to test it anyway, why ask in the first place? Just wasting both our time.

Foo is right, lots of speculation is posted on here, it just helps to know what the source is...

And I asked because I haven't tested it myself, and was hoping to use it for the next game before I'd have a chance to try it.

So I appreciate your response, it was helpful and not a waste of time.

Originally posted by Baustin


Does it matter how I know? You asked, and were answered. If you choose to test it anyway, why ask in the first place? Just wasting both our time.

It matters because there is a ton of speculation about what does and does not effect things, and how one thing is effected by another on these forums. I find it insane that there is so little communication from the designer about how the Sim works. It leads to someone speculating, another person agreeing, and then you suddenly have people running aorund giving speculative advice like it is gospel.

btw, i did run a small test today. I set an input that had only one output set at 50%. That input occurred 3 times. Once it ran the output I had set, twice it ran other plays from a less specific input. Very limited sample, but Baustin if you are right, it would have run that output all 3 times. So I don't think that's what I does. It seems like it runs the output at whatever % it is set at, and the rest of the time it moves on to the next relevant input. I'll test it some more but that's what I found so far...

Originally posted by tetura
btw, i did run a small test today. I set an input that had only one output set at 50%. That input occurred 3 times. Once it ran the output I had set, twice it ran other plays from a less specific input. Very limited sample, but Baustin if you are right, it would have run that output all 3 times. So I don't think that's what I does. It seems like it runs the output at whatever % it is set at, and the rest of the time it moves on to the next relevant input. I'll test it some more but that's what I found so far...

Make up your mind, did that input trigger once, or 3 times? If you have more than one input that will trigger in a certain situation, it will use the usual criteria to decide which one triggers. Higher in the AI will trigger first, and quarter specific will override general. You can't have it both ways, either that input triggered, or it didn't.

I've been an OC since before there was an AI, and since it came out I've used one input with 3 outputs for a certain situation, the plays have changed but not the input parameters. I generally only use 1-2 of the plays, depending on the defense I'm going against. They're all set at 50%, I set the other one or 2 plays to 0 when I don't want them to run. When there's one play set at 50% and 2 at 0, that play runs 100% of the time that input triggers. It's always worked that way. I have never had a play that I didn't want to run, run in that situation. FWIW, that's in probably over 600 games.

Originally posted by tetura
btw, i did run a small test today. I set an input that had only one output set at 50%. That input occurred 3 times. Once it ran the output I had set, twice it ran other plays from a less specific input. Very limited sample, but Baustin if you are right, it would have run that output all 3 times. So I don't think that's what I does. It seems like it runs the output at whatever % it is set at, and the rest of the time it moves on to the next relevant input. I'll test it some more but that's what I found so far...

I would love to see this tested. If for no other reason than to get past the speculation and assumptions that people have.