Welcome to the Lounge

5-Jun-23 11:10

Initially I was thinking why this hasn’t been done yet, I starting typing and then I realized why. I have in mind a general purpose problem solving algorithm that could be applied to RTS games too. To get it working you would need the following.
A list of problems that might get in while you try to achieve something
A list of solutions to each of those problems
It would operate like this:
Start the process PR1 which consists of the things that need to be done to achieve the main goal.
If something unexpected takes place, let’s say obstacle one in the way you have stop PR1, identify what type of problem obstacle one is from the list of problems and pick a solution from the list of solutions to that type of problem. Let’s say the solution is executing process PR2
While executing PR2 a problem may occur again. This means stoping process PR2 and finding a solution to the problem. The solution could be starting process PR3.
If we talk in terms of a RTS game PR1 might consist of units of type 1 UT1, trying to reach destination DT 1, obstacle one could consist of units of type 2 with a different color and with a fire range reaching at UT1, UT2 are located somewhere that UT1 can’t get to.
PR2 could mean airborne units that need to get to UT2. The problem that might get in the way while executing PR2 could consist of ground to air defensive buildings that are found somewhere down the road towards UT2 having the same color as UT2 I’m not going to cascade any further.

If PR2 is completed PR1 is resumed.
The problem is that in a RTS game things don’t stay as they are too much time. There is a chance UT2 might get moved by the owner from their initial location, that would leave PR2 in mid air.
What are your thoughts?

honey the codewitch5-Jun-23 11:50

honey the codewitch

5-Jun-23 11:50

This actually reminds me of complex questlines in Bethesda's RPG engines, like Underground Undercover in Fallout 4, as well as the intermingling of the questlines due to competing factions creating twists in the story depending on your actions.

Your method might be an easier way to accomplish it, but it actually might make the problems created by open ended / open world gaming with intermingling quests worse, since it would allow a freer sort of interaction, which leads to smurfing/exploiting opportunities and quest bugs.

Not that your system would have them as a matter of course, but the way it operated it would be difficult to devise a test matrix for it. Let's put it that way.

Edit: I used RPGs as opposed to RTSs as an example because my experience with RTS games is extremely limited.

There's smoke in my iris
But I painted a sunny day on the insides of my eyelids
So I'm ready now (What you ready for?)
I'm ready for life in this city
And my wings have grown almost enough to lift me

Calin Negru5-Jun-23 22:42

5-Jun-23 22:42

Traditionally RPG worlds are static. If you stop in the middle of a RPG level in a safe area and then leave your computer running for a few hours unattended, when you get back you will find things exactly the way you left them.
If you do the same in a RTS by the time you get back the AI player will have mined everything there is to mine on the map and had destroyed your base ten times over.

honey the codewitch6-Jun-23 0:10

honey the codewitch

6-Jun-23 0:10

If you leave your character sitting in the open world in the fallout series you will be eaten by a Yao Guai

There's smoke in my iris
But I painted a sunny day on the insides of my eyelids
So I'm ready now (What you ready for?)
I'm ready for life in this city
And my wings have grown almost enough to lift me

Slacker0075-Jun-23 12:27

Slacker007

5-Jun-23 12:27

Calin Negru wrote:
What are your thoughts?

Sounds, reads, and looks like programming question.

Going to grab a bowl of fresh popcorn and a seltzer and watch your programming question get answered in the Lounge.

Nelek5-Jun-23 14:11

Nelek

5-Jun-23 14:11

For me looks more like an description to start a debate / conversation?

I might be too tired though, so I'm going to bed. GN

M.D.V.

If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about?
Help me to understand what I'm saying, and I'll explain it better to you
Rating helpful answers is nice, but saying thanks can be even nicer.

Gerry Schmitz5-Jun-23 20:43

Gerry Schmitz

5-Jun-23 20:43

The computer doesn't do "missions"; the user may get a mission and the computer tries to stop the user; or it's user on user.

It comes down to just a few basics: moving; standing (holding, defending, delaying); firing; and hand-to-hand. Offensive, defensive; defensive-offensive.

"Before entering on an understanding, I have meditated for a long time, and have foreseen what might happen. It is not genius which reveals to me suddenly, secretly, what I have to say or to do in a circumstance unexpected by other people; it is reflection, it is meditation." - Napoleon I

Advanced TypeScript Programming Projects

Pete O'Hanlon5-Jun-23 23:20

Pete O'Hanlon

5-Jun-23 23:20

To be honest, what you have just described sounds like a fairly simple workflow system. Yes, the states you are describing may be quite complex, but you're effectively describing a series of workflows. This doesn't mean you need a workflow engine or anything like that, it's entirely possible to program this with events and states.

Calin Negru6-Jun-23 3:59

6-Jun-23 3:59

>sounds like a fairly simple workflow system

That’s something I have yet to learn

> Gerry Schmitz : the computer does’t do “missions”

Maybe I can come with an AI that does that
Of course my main goal is to write a common type RTS first.

Gerry Schmitz6-Jun-23 5:08

Gerry Schmitz

6-Jun-23 5:08

The closest to a (computer) "mission", is following waypoints to a particular destination (i.e. another point) ... which can be the location of an anticipated action.

Along the way, they may be: fired on; surprised; encounter an obstacle; charged by "mobile" units; etc.

What happens next depends on what "tactics" and probability calculations you've incorporated based on numerical superiority, type of ground, condition, past encounters: retreat; stand ground; counter attack; "panic"; etc.

At the end of an encounter, the chances that the party simply recommences their route march to somewhere, depends entirely on the "commander" (i.e. a "user").

(Actually, the "commander" from the computer's perspective is a roll of the dice when there are multiple options with varying probabilities: rolling a number less than, equal or greater than the associated probability. Win, lose, or maybe a draw. The "friction of war" ...).

PIEBALDconsult6-Jun-23 3:37

PIEBALDconsult

6-Jun-23 3:37

Calin Negru wrote:
What are your thoughts?

INIGO: You are using Bonetti’s defense against me, huh?

WESTLEY: I thought it fitting, considering the rocky terrain.

INIGO: Naturally, you must expect me to attack with Capo Ferro.

WESTLEY: Naturally. But I find that Thibault cancels out Capo Ferro, don’t you?

INIGO: Unless the enemy has studied his Agrippa. Which I have!

modified 6-Jun-23 12:21pm.

jschell6-Jun-23 5:09

jschell

6-Jun-23 5:09

Calin Negru wrote:
What are your thoughts?

First of course is a whole bunch of Deja Vu because seems like I remember seeing this same post a while ago.

Second, also of course, is that this is a fundamental problem for any game where there is a computer opponent. That is where humans start complaining about the AI (traditional game) being 'stupid' or citing a specific example where it seemed like it was human like. Also where game creators allow the AI to cheat with the idea of providing a better (harder) opponent.

Calin Negru6-Jun-23 21:44

Thoughts on current chatty AI

6-Jun-23 21:44

> First of course is a whole bunch of Deja Vu

You must have been traveling into the future

>Gerry Schmitz:

I think going around the problem enemy group is the first thing that should be tried when you have “something in the way “situations

Cpichols5-Jun-23 3:17

Cpichols

5-Jun-23 3:17

The reports of totally made up answers that are so confident and well-written have me thinking. These AI seem to have no concept of the difference between fact and fiction. Being asked to produce copy is the same thing to them as being asked for factual content. They regularly plagiarize, so taking bits from the questions and bits from other writings, they assemble responses as if they were just copy, even when asked to answer with simple facts, whole snips of historical documents/scientific studies, or calculations.

Perhaps what is needed is a sort-of 'scholar:' tag. So when asking for answers, it won't make things up.

Slacker0075-Jun-23 3:34

Slacker007

5-Jun-23 3:34

Common sense and baseline IQ will tell anyone that AI and AI "Chat" is still very new and is in constant development and progression.
Eventually, some day (soon?) it will be perfected.

To judge it now is premature at best.

WPerkins6-Jun-23 3:14

WPerkins

6-Jun-23 3:14

Sounds like something an AI would say!

Mark Starr6-Jun-23 4:20

Mark Starr

6-Jun-23 4:20

Yes, it is new. And it will change (significantly?) over time.
The problem, though, is an old one: people are lazy. Given a new tool to help them do a job, they'll quickly use it to do the job, without oversight or a critical eye.
So what difference does it make if you're getting misinformation from an automated device, or a politician, a newscaster, or your neighbor Paul. If you don't take time to verify, then what does it matter.

Time is the differentiation of eternity devised by man to measure the passage of human events.
- Manly P. Hall

Mark
Just another cog in the wheel

jschell6-Jun-23 5:16

jschell

6-Jun-23 5:16

They started AI research in the 50s.

Lisp is a programming language created about 1960 specifically for doing research in AI.

ChatGPT comes from OpenAI which was started in 2015 as a non-profit specifically to use existing (prior) knowledge of AI to research it.

ChatGPT is actually the third generation of something based on what they were working on.

So not really sure how any of this counts as "very new".

sasadler6-Jun-23 5:25

sasadler

6-Jun-23 5:25

I'll do like I did with digital cameras, I'll wait till things get 'good enough' before I invest my time with the bots.

Al Fargnoli6-Jun-23 8:48

Al Fargnoli

6-Jun-23 8:48

There's no intelligence in an LLM. It strings phrases together partially based on the frequency of those phrases appearing together in its training data. Here's an in-depth description: https://writings.stephenwolfram.com/2023/02/what-is-chatgpt-doing-and-why-does-it-work/. Yeah, they will add heuristics around it to make it seem less dumb, but how far is that going to take it? Have you seen the ChatGPT response where it's supposed to subtract 3 from 70, and it confidently answers "73"? And here is Donald Knuth's experience, Donald Knuth Asked ChatGPT 20 Questions. What Did We Learn? - The New Stack[^].