Feeding the hungry

Entelect Challenge 2022 Phase Balancing
1

Has anyone ran into the scenario where all the food on the map is depleted?

Do you guys think it will be part of the game and it will check which ai player adjusts best to this ?

or will the resources be bumped up to handle the exponential population grows rates ?

Because its a tricky challenge since if all the food runs out all players suffer catastrophic population loses and the regeneration rates cant keep up to rebuild the population to what it once was so the game basically just fizzles out.

Keen to hear your thoughts on this

2

Hi @THEKVN ,

I have been seeing the exact same issue. I was just about to make a post as well.
Almost halfway into the match, all food is depleted and every bot is running for the scraps as the food does not generate fast enough.
I played with some different strategies to not eat the food fast enough and end with a big population. but, when I play the slow population growth version of my bot alone, it will end with something like 20 000 population, BUT when I play it against the greedy version of my bot, it dies out due to the greedy bot taking all the food before the stable growth bot can get ‘any’ food.

So it’s either up to managing what’s left of your bot’s units to collect the resources that is left and keep sending small squads all over the map to get as much as possible,
Or the food nodes needs to increase, or the amount of food at a node needs to increase or the regeneration of the food needs to increase. I think there could be some nice ways for the EC engine team to accommodate the hunger problem with a nice regen strategy.

3

Hi @AtomicNomad

100% on the same page!

I’d say either increase regen rate by a lot , OR decrease the growth rate of the population so it doesnt get out of hand so quickly, OR maybe the EC team could have some game mechanic up their sleeve to introduce after the first tournament to balance it out.

I dont think it would so entertaining to just have the ‘greediest’ bot win because they gathered the most food in my opinion

4

I was wondering the same thing. Having a resource management game, but the way to win the game is to have the most of everything, rather than just managing it.
I would say that they need to increase the regen on the resources than capping the population, as there is a tier 5 specified, and currently I have not reached it yet with the amount of food that is not available to reach that level.

I guess they though of something like word events or something like that, such as food shortages or virusus or cold storms and such.

5

I think its part of the challenge,

If you get too greedy you could easily get burnt later on.
You should keep the map and its resources in mind while building.

If we ignore this part of the challenge it really makes strategy a bit flat.

that 10% population loss per tick can be pretty intense and it can burn you pretty fast.
But its also fun to run with…

6

But I have not run endgame much, Just working on small things.

And I was thinking perhaps event 2 will allow Armies. Where you can also assign population to attack other players,

But no idea yet.

7

My opinion is that it is a stretegy to mess with other players, because at the end it is the score that counts, not the person with the most population.

Obtaining tier 5 is very much possible but there wont be enough wood or food to sustain it, but like I said population is not everything, optaining a score above 200 000 in a 4v4 is very much possible with an ending population of less than 500

8

Yeah, that is indeed the case @Kortgat .

I think that in the end, the population is the highest priority and only that should count toward your score at the end, not the one who has collected the most resources and ends with the lowest population count. You can’t have a civilization with 100 000 wood/stone collected, but none to give it to.
The resources should only be there to sustain and build your population, and at the end, the biggest population must win.

9

Depends, a car company can have 50 employees but 100 000 cars that define the net worth of the company, but a company with 50 000 employees and 50 cars is bankrupt, so this is a resource management game,

I do see references to gold so i assume it will get more complicated after first tournament

10

What can be done is increasing the score multiplier for population a bit, since its only 10 to 25 or so, this will make a sla… person 25 times more valuable than food or wood

11

I feel having a population value of around 50 could be great as long as engine ticks do not change.

After building and finishing my bot and seeing how the population / resource score balance is stitched its really interesting that you have to factor in population VS Resource Hoarding.

It gives bots a bit of range to work with which will be even more interesting later when the engine expands.

I personally do not feel a “race to max population” is any fun.
We know that population isn’t sustainable, so how will you capitalize on this information.

I think in terms of engine balance everything can stay the same except:
A) Maybe value Population more (50 could be good).

We will also have gold later, most probably a useless resource worth lots of points that gives us something to think about.
I technically already accounted for Gold in my solution.

But I would not do something like shorten ticks AND increase Population value,
Else we loose so much depth.

Im excited for this event because people were talking about 20K populations and here I was still chilling on tweaking 50 units. But I pushed through the weekend.

Anyways,

I think doubling population could be great, I would not go as far to say it should go to 100, It could easily overvalue population if this update is rushed. As mentioned above, you need to have that population VS hoarding dilema or the depth of the challenge just dies.

It felt like 100 population stretched the “Population” side a bit too much leading to an undervaluation of resources.

I do however not agree with the following Statement:
“The resources should only be there to sustain and build your population, and at the end, the biggest population must win.”

Population means nothing if the bot cant sustain it over time.

I would say doubling population score is very possible and looks reasonable.
But any more will just take away from the challenge.

2 Likes
12

I found a interesting bug in the Engine code.

public static class IntExtensions
{
    public static int NeverLessThan(this int quantity, int lowerBound)
    {
        return quantity < lowerBound ? lowerBound : quantity;
    }
    
    public static int NeverMoreThan(this int quantity, int upperBound)
    {
        return quantity > upperBound ? upperBound : quantity;
    }
}

which is suppose to be (I think, from my limited understanding of the engine language):

public static class IntExtensions
{
    public static int NeverLessThan(this int quantity, int lowerBound)
    {
        return quantity < lowerBound ? quantity: lowerBound  ;
    }
    
    public static int NeverMoreThan(this int quantity, int upperBound)
    {
        return quantity > upperBound ? quantity : upperBound;
    }
}

What does this mean? Well, your growth rate and decline rate is not capped at 10%. If you have enough food and heat, you can potentially surpass a growth rate of 10%.

So it’s suppose to be a bit harder to reach a higher population limit. I think if they fix this bug, it should drastically change the scores and growth rates.

The reverse is also true, but to a lesser degree. Population decline is still relatively capped at 10% from what I could see.

Example:

Population = 50
Heat = 1160
Food = 880

You can expect a growth increase of 100 after the next day night cycle.

Population = 150

I might be off by a bit, but the above functions are definitely not working correctly. Going to see some interesting results in this tournament.

13

Thanks for pointing this out. I will feedback to the team and have them investigate.

14

I there, the calculations work as expected thank you. We are also going to be releasing a new release which does update quite a few calculations. Please keep a look out in the next day or two

15

Should you not wait till after the first tournament to apply any possible changes that can cause bots that have already been submitted based on current engine to perform worse than excepted.

Either postpone the first tournament by a week or wait till after the first tournament to interduce new changes, since new features will also be added after the first tournament.

2 Likes
16

We have postponed the tournament 1 by 1 week. Official comms will come out soon

17

I also experienced this, was surprized to see some of my growths,

For phase balancing I was actually going to suggest that they cap population growth so that things do not go as crazy, I found this by accident when i tweaked some values and the higher the value the faster I grew.

Anyways, would need to be looked at