raw wrote:
That hand is sometimes unplayable, because - if we stick with the poker comparision - it is equivalent with being delt a 1 card hand and not 5.

Agreed. Also, a hand of poker can go a lot quicker than it takes to discover you're boned in a dry sector of galaxy. Some kind of method has to be construed to give players a fighting chance.



[Edit-Youdon'twanttospend30minsofyourlifejusttodiscoveryouneverhadthatmuchofachancetostartwith.AmoebaistheonlyraceIcanthinkofthatwouldbeabletoavoidthis]

But isn't overcoming obstacles and triumphing over adversity part of the game? Sometimes you get crap, other times you get Dust.

CybrSlydr wrote:
Whatever happened to playing the hand you're dealt?

That hand is sometimes unplayable, because - if we stick with the poker comparision - it is equivalent with being delt a 1 card hand and not 5.

Whatever happened to playing the hand you're dealt?

I completely agree with the OP. Most other 4x games have a post-generation balancing step. If some start position is poor, then the balancing step adds planets or resources in the home system or nearby. I have noticed in games against the AI (with pirates turned off) that some AI's just never get started. Investigating further, you can tell it is because they were in a corner, with no good planets around. For multiplayer, this is an absolute requirement to avoid ragequits.

Valazzar wrote:
Right now success rate is highly likely to be determined by what nearby planets you have when you start.

There should be a normalization of the galaxy, so that there is a reasonable balanced distribution of "good planets" for each races immediate vicinity.

If you start out with some tier 1 planets nearby, and another player has none (unless he's sower), then you almost automatically win, or at least dont matter in the game.

That cant be intended....

Or alternatively allow the player to mitigate inherent negatives by allowing them to spend money/other things on improvements (instead of turns/time). This would mean you've to rebalance the whole game.

What about a 10 turn Age of Discovery/Age of Decline period at the start of the game where rather than normalizing everyone's planet, everyone's FIDS is normalized to a certain amount. It would be based on FIDS output. After that 10 turns, FIDS returns to whatever the actual values are.



If you have a bad position/low FIDS, then your values are increased to X. You can plan how to maximize that time (research tech, spam colonies to make up for poor FIDS, whatever).



If you have a really great position/high FIDS, then your values are decreased to X (or Y). You still have the benefit of good position, but you have to decide how to prepare for it and minimize delays.



If you have an average position, no change.



Not really sure how large of an effect this would have in the long term though. But it at least allows for standard openings to be played consistently, which, let's face it, will happen (eg Improvement>Colony>Defenderx2 or whatever).

Photon_Ventdesdunes wrote:
The problem is after counting too much to have equal... there is no random anymore...



I was in a MP game this afternoon as Pilgrims. Crap everywhere... I used my colonyship after only 15 turns... I beat a Hissho 15 turns after and led the research area all game long.

The devs will sure try to fix to make huge problems disappear but I honestly think the random should stay random.

It's part of strategy to devellop arid or tundra science to be sure be able to colonize.

The thing is, random is great, for single player, and dont get me wrong, i dont want a mirror setup, thats just boring.

But thats because you can overcome the ai with your brain

Against a player, things are different, its no fun to play 3-4 hours and realize that you had already lost on turn 1, because....the galaxy was generated hugely in your opponents favor.

Valazzar wrote:
Could be better, have you tried doing any numbers on it?

not really. I was just going for pop size (4 for tiny, 10 for huge) times the tier number. it's not really exact, just to show what I mean. it's just that (disregarding terraforming) I like a medium desert better than a tiny jungle or huge lava.

The problem is after counting too much to have equal... there is no random anymore...



I was in a MP game this afternoon as Pilgrims. Crap everywhere... I used my colonyship after only 15 turns... I beat a Hissho 15 turns after and led the research area all game long.

The devs will sure try to fix to make huge problems disappear but I honestly think the random should stay random.

It's part of strategy to devellop arid or tundra science to be sure be able to colonize.

I was doing some testing earlier today and the galaxy generation just isn't very good:



http://cloud.steampowered.com/ugc/595846901580296401/068029E2D23C9C6E6129ACFD7A4F0851FA7C40FD/

They could add a slider for the point system in game set up. The designated starting areas (be them constellations or systems in a certain vicinity of the home system) would have the same number of systems, and the points value would be only for these systems. They could be as large as a whole arm, or corner of a disc, or smaller with more randomized systems in between each starting area. Each starting area would have the same points value, and it would randomize between each of the systems. Number of Planets, Planet types, Planet sizes, Moons, and positive anomalies would have positive points, while negative anomalies would have negative points. Note that I do not mean the same exact planets would be randomized over the starting areas, but perhaps in one starting area, there would be more positive anomalies, but smaller, lower tiered planets and less of them, while another system would have some nice Large Planets, but more negative anomalies. The points for each variable would be based on the effect each item has in game. This means everyone would have an even start. Note that neither here, or before when I mentioned a points system did I include strategic or luxury resources being normalized in each players starting zone. These should continue to be random throughout.

I would rather have reality over specially designed systems for players who can't deal with it. After all, our own solar system has gas giants, frozen dwarf planets, molten rocks, and asteroids. That's life, deal with it.

Why not have very similar starting systems, then differing constellations for people to expand to?

vaendryl wrote:
I like the point system, but I feel planet size versus planet tier should be multiplicative in value.

for example, a huge jungle would be 10x5=50 whereas a tiny barren would be 4x1=4

that feels closer to how valuable they are.

Could be better, have you tried doing any numbers on it?

sagittary wrote:
What about a 10 turn Age of Discovery/Age of Decline period at the start of the game where rather than normalizing everyone's planet, everyone's FIDS is normalized to a certain amount. It would be based on FIDS output. After that 10 turns, FIDS returns to whatever the actual values are.

That seems awfully hard for a player to predict, or even understand.

I like the point system, but I feel planet size versus planet tier should be multiplicative in value.

for example, a huge jungle would be 10x5=50 whereas a tiny barren would be 4x1=4

that feels closer to how valuable they are.

Right now success rate is highly likely to be determined by what nearby planets you have when you start.

There should be a normalization of the galaxy, so that there is a reasonably balanced distribution of "good planets" in each race's immediate vicinity.

If you start out with some tier 1 planets nearby, and another player has none (unless he's sower), then you almost automatically win, or at least dont matter in the game.

That cant be intended....

Valazzar wrote:
Regarding how to solve it, then what i imagine is some kind of normalization by internally assigning point values to various sectors of the galaxy where players start, doesent have to be the entire arm in a spiral galaxy for example (might be a good idea though), but "near the homesystems".

Then either remake the galaxy until you get one that is "within parameters", or redistribute planets and systems until it does fall within parameters.

Parameters might be a minimum number of tier 1 planets (1 or 2), or a point value derived from total number of planets, their size, tier and various modifiers.

Redistribution is exactly what other 4x games do. I know for a fact this is what Civ IV does because I have written these point total functions for TC mods. For ES, I do not think the galaxy should be regenerated. But, using point totals, evaluate the arm or the two-jump area around each player. If a player's point total is below some threshold, then improve a world in this area (replace gas giant with terran, etc) until the point total reaches this threshold. The exact point values are debateable, but Valazzar's value seems reasonable.



It would be a large project, but it is *possible* with the modding tools to read the galaxy map and perform this calculation. As a first step, this result could be displayed for a map so we can see the huge variations. Then as a second step, the galaxy map could be improved. If many multiplayer players come to complain about this, we can present a fan-based solution.

Im not terribly worried about the "long term game", but i am worried about the game up until turn 50 or so. because after that you can do something about your situation.

Regarding how to solve it, then what i imagine is some kind of normalization by internally assigning point values to various sectors of the galaxy where players start, doesent have to be the entire arm in a spiral galaxy for example (might be a good idea though), but "near the homesystems".

Then either remake the galaxy until you get one that is "within parameters", or redistribute planets and systems until it does fall within parameters.

Parameters might be a minimum number of tier 1 planets (1 or 2), or a point value derived from total number of planets, their size, tier and various modifiers.

An example of a point system (which is just an example, i dont expect the point values to hold up).



Point values:

Tier 1 planet = 10 points

Tier 2 planet = 7 points

Tier 3 (+Asteroid) planet = 4 points

Tier 4 (+Gas Giant) planet = 1 points



Huge Size = 10 points

Large Size = 7 points

Medium Size = 5 points

Small size = 3 points

Tiny = 1 point



Number of systems = -2 per system (the more systems, the worse it is)



This would make a huge lava planet, equal to a tiny terran planet (not sure thats how the balance should be, but just going with it for now).



So an example:

Player 1:

Home system is 4 planets, home planet is not counted, so 3 planets.

Home system consists of:

Home planet = not counted

Medium Desert = 9 points

Huge Lava = 11 points

Tiny Arid = 4 points

Total value of home system = 24 points

The other 4 systems in the vicinity have :

1 huge ocean = 20 points

1 medium tundra = 12 points

3 large barren = 24 points in total (3x8)

2 small arid = 20 points (2x10)

1 medium lava = 6 points

1 Large arctic = 10 points

Total of vicinity systems = 92 points

There are 5 systems in the vicinity, including home system = -10 points

So the total points score for player 1 is 106 points





Player 2:

Home system is 3 planets, home planet is not counted, so 2 planets.

Home system consists of:

Home planet = not counted

Large lava = 8 points

Medium Asteroid Field = 8 points

Total value of home system = 16 points

In addition he has 2 other systems in the vicinity

1 large gas giant = 8 points

1 medium jungle = 15 points

2 tiny arid = 16 points

1 medium asteroid field = 8 points

1 large lava = 8 points

1 medium barren = 6 points

There are 3 systems in the vicinity, including home system = -6 points

So the total points score for player 2 is 71 points



Now i imagine the balancing system could balance in several ways:

A) Shuffle planets around until it balanced, in this case move about 15 points of planets worth, from one player section to the other.

B) Generate new planets until Player 2 was on par with player 1

B) As B, except remove planets from Player 1 at the same time.

C) Have a predetermined "optimal" which could be deviated by say 10 points (so 80 to 100).

- If that optimal was 90 points, then one would remove planets from player 1 until he reached 100 points (90 + 10)

- In this case Player 1 would have the medium lava planet worth 6 points, removed. if the numbers had been different, he could have been reduced to a bit more than 100)

- Player 2 would have added roughly 9 points worth of planets, this could be done by randomly rolling a new planet and adding it until his point value got past 80, that could be a medium tundra for example, thats 12 points.



So for arguments sake, lets hold on to this example and see how it looks then:





Player 1:

Home system is 4 planets, home planet is not counted, so 3 planets.

Home system consists of:

Home planet = not counted

Medium Desert = 9 points

Huge Lava = 11 points

Tiny Arid = 4 points

Total value of home system = 24 points

The other 4 systems in the vicinity have :

1 huge ocean = 20 points

1 medium tundra = 12 points

3 large barren = 24 points in total (3x8)

2 small arid = 20 points (2x10)

1 Large arctic = 10 points

Total of vicinity systems = 92 points

There are 5 systems in the vicinity, including home system = -10 points

So the total points score for player 1 is 100 points





Player 2

Home planet = not counted

Large lava = 8 points

Medium Asteroid Field = 8 points

Total value of home system = 16 points

In addition he has 2 other systems in the vicinity

1 medium jungle = 15 points

1 medium tundra = 12 points

1 large gas giant = 8 points

2 tiny arid = 16 points

1 medium asteroid field = 8 points

1 large lava = 8 points

1 medium barren = 6 points

There are 3 systems in the vicinity, including home system = -6 points

So the total points score for player 2 is 83 points





Im not sure this is going to hold up, but its an example...a proof of concept if you like. would need to make some tests to see if it repeatedly gives results that are useful, or if it needs tweaking.