Invasion (positive or negative) alien

Daniferrito wrote:

As uteen said a few posts over yours, in-game years are not very reliable, as they are not always the same length. Some real-time measurements should be taken. something like game ticks or real-life minutes.

My example was real-life minutes, but real-life seconds may be doable on modern computers, which would be preferable. It will just be a few randomisations, so shouldn't take very long to compute.


@GmansWatching: That means planets orbiting closer to the star are significantly more likely to be invaded by aliens… Real-time is much better, and a game year is far too infrequent - it would result in ‘Alien invasion seasons’.

You know, every second is still way to slow for computers. I see game ticks way better than seconds. For all the randomness, we only need to generate a pseudo random number, and for every random thing to happen we only need an if check, which is quite fast (the fastest thing a computer can do). Unless we make hundreds of random events, we shouldnt need an extra loop to check for them.

My point: with a computer that can run thrive, every game tick is perfectly doable, as long as we don make hundred of random events.

@GmansWatching realistically speaking rates are too high.
we have to consider of the science we have: according to the Drake equation in our galaxy there are about 640 advanced technological civilization and this means that there are 1 every 4,000 light years.

If we take into account the equation, the game probably never meet any alien, and only in the best case, some planet with microbial life, or, rarely, multicellular life.

My percentages believe the right and from the point of view scientific, both from the point of view of the game.

What do you think?

Doggit wrote:

@GmansWatching realistically speaking rates are too high.
we have to consider of the science we have: according to the Drake equation in our galaxy there are about 640 advanced technological civilization and this means that there are 1 every 4,000 light years.

If we take into account the equation, the game probably never meet any alien, and only in the best case, some planet with microbial life, or, rarely, multicellular life.

My percentages believe the right and from the point of view scientific, both from the point of view of the game.

What do you think?

QFT, Seems good.

Also, what if we have a Galactic time? Where as it's based off of the creation of the galaxy.

I think that it should be calculated based on game, not real, time, and instead of arbitary percentages each stage it will be based on how much you leak into space.

Why set a percentage chance of aliens invading at all. Won't other planets be simulated and evolve on their own, and ultimately some will evolve sentient spacefaring species. If the AI controlling these species at any point decides to invade your specific planet, then that is that. You wouldn't need to introduce an independent, arbitrary percent chance of that happening.

You see, to further explain this, a reason games even add these percentages to games in the first place is to compensate for things the game does not directly simulate. I have a moderate degree of experience in this. For example, in a game like Europa Universalis III, or really any of the Grand Strategy Paradox games, battles are mostly reliant on percentage chances based on dice rolls and morale levels and other factors. At the end of the day, it is a competition of these percents. These percents compensate for the fact that the player is not actually individually commanding his units across the battlefield. However, a Total War battle fought manually is completely different, because it DOES simulate the individual soldiers on the battlefield. Whether the player positions his archers on the hill or in the valley makes a difference to the outcome of the battle, whereas in the Paradox Grand Strategy games the player does not have control over these things, so terrain is simply a number to be calculated as a percentage. Therefore, Total War battles fought manually don't use dice rolls and other generalized percentages, since the individual units are being simulated anyways, and there is no need to compensate.

Basically, to sum up, as long as other planets run their own courses at the same time that you play and even when you aren't their, then their is always a chance of an AI spacefaring species invading your planet. If that was not the case, and the parts of the galaxy that the player is not directly interacting with at the moment are essentially paused until the player return, or are just kept as static bits of memory, then the need for percentage chances arises, since the game does not simulate action in areas where you aren't.

tl;dr Alien Invasions are being simulated anyways, we don't need to introduce a random percentage to compensate for that.

So the percentages are useless?

Only if other planets are being simulated while you play. If they are just static memory, then percentages are needed. However, there are so many possible routes an AI can take, and setting percentages very poorly compensates for that since what can happen is limited to what percentages you set. For that reason I would prefer the former, as in other planets being simulated at the same time as you, unless it is too demanding for the CPU.

If both your planet and the planet being simulated are on the same time scale (both in civilization stage) then I think it could be done with some restrictions. If they are not, it is much nore problematic. For example:

I am evolving a species, and in 10 minutes of gameplay I advance 10.000.000 million years (or whatever), and in planet B, there was a civilization before the 10 minutes of gameplay. Then the computer needs to simulate all those years of advancement in the other planet. That can get costly for the processor.

Even more, we would want more than one alien planet being simulated. then the extra work for the processor gets higher.

If we stick to only simulating one planet, then we could go quite high on the simulation on that planet, but if we need to simulate many planets at one, then we need to lower the amount of things simulated.

Anyway, this could be further discussed once we haveall the planet simulation, and then we could decide wether we can run those extra simulations.

I'm sure we've had the discussion about the consequences of multiple saves…

Ah, it's one of my threads: Multiple Saves Dialemma
And an awesome thread, since I appear to have invented the God Machine in it…

…And discussed them for a full page-worth of posts…

…And then got back on topic…

…And then got off-topic again.

Well, that's wasn't very helpful.


But Nick has a good point - they hopefully will be simulated anyway, in the final game, so this isn't especially important. As long as we don't take the Spore path, that is.



#1400

This is not the same as multiple saves. It is somewhat related, but not he same.

Anyway, if we can simulate multiple planets, we will. If not, we won't. Once we have it working, we can decide.

Daniferrito wrote:

This is not the same as multiple saves.

Right…

How did I make that mixup…?

Weren't we thinking of something called (Don't take my word for this) speed evo which would essentially speed up evolution within seconds when you reached a planet so an old species could instantly become a new one? I know this is almost useless on this particular topic, but wouldn't that mean that the other planets are static memory and not in fact being simulated. But we could use speed evo here as well. We could set a percent value for alien invasions and the game speed evos a species from a planet from around you into invaders which then invade

Gryphogrox wrote:

Weren't we thinking of something called (Don't take my word for this) speed evo which would essentially speed up evolution within seconds when you reached a planet so an old species could instantly become a new one? I know this is almost useless on this particular topic, but wouldn't that mean that the other planets are static memory and not in fact being simulated. But we could use speed evo here as well. We could set a percent value for alien invasions and the game speed evos a species from a planet from around you into invaders which then invade

It's called super-auto-evo, but it works as you describe.
I suppose that could work, as long as it isn't too obvious when one civilisation attacks another, so we can add wars in when necessary and make it seem like it was always like that.