Metaverse Scores

Also: I've noticed metaverse totals are not just a sum of your scores, nor are they an average. Anyone figure out exactly how they're calculated?

just curious, how long does it take to play through a gigantic game? i've only been playing on medium galaxies.

the thought of having to deal with hundreds of constructors travelling to dozens of starbases gives me the heeby-jeebies.

cheers

h

Took me two days, about 12-14 hours of game time. The last twenty or so turns mostly involved me clicking through the screens as fast as I could so I could hit turn. I was sending out stacks of 16 constructors (I only bothered moving them out of star systems when they hit 16) to build partypalace-terror stars, so it was only a matter of time before I culturally dominated everyone, and I wanted to get it over with.

#1 by Ambassador JavaScout - 3/29/2003 1:51:25 AM
Total Points for games
----------------------
SQRT(Number of games)



~SDC~

I am having the same displeasure with the scores. Here are the 3 games, with approximate scores, map size, difficulty level, and victory type:

1 - Gigantic, All AI Bright, Military (2 turns from cultural, about 5 turns from tech), 2200 points

2 - Gigantic, All AI Intelligent, Military (4 turns from cultural, about 10-20 from tech), 1900 points

3 - Gigantic, All AI Genius, Tech (6 turns from cultural, 10-15 from military), 1100 points

As you can see, my scores went DOWN as the AI levels went up. I think the scoring needs to be re-evaluated. My game against the Genius AIs was much more difficult, and took a lot more strategy to win.

I really think the scoring calculation needs to be re-evaluated.

-Dave

~SDC~

Scores are HEAVILY affected by population! IT is a bit annoying, but if you win a gigantic galaxy on Genius and win the exact same galaxy on Bright (but take longer and thus have a higher population) your bright victory will have a higher score. The scoring does badly need modifying to reflect the difficulty level not just the finishing conditions.

I also do not like the current metaverse scoring method. i love playing tiny and small galaxies on incredible AI for a quick 1-2 hour game. Never submit them anymore though as my metaverse score would drop.

Paul.

~SDC~

'I also do not like the current metaverse scoring method. i love playing tiny and small galaxies on incredible AI for a quick 1-2 hour game. Never submit them anymore though as my metaverse score would drop.'

The problem is that if there isn't some sort of weighting for the number of games, then you can easily end up with problems similar to the ones caused by 'cheat servers' on Stellar Frontier. People submitting so many low score games so quickly that they end up on the top by default.

~SDC~

Different ways of overcoming this potential problem need to be considered. Stardock has been very vocal about it's desire not to punish real players in an effort to stop piracy (hence no CD required in CD player to play game). To then punish real players on the metaverse in an attempt to stop potential cheating / score bashing would be against that policy.

Perhaps the easiest suggestion is only to let winning scores count on the metaverse.

Other suggestions include

- Balance the game scoring better so that quick easy games generate very low scores.

- Split the scores into categories for intelligence or galaxy size.

- Only count best 30 games from past 3 months in your metaverse score.

Paul.

~SDC~

Sure the scoring is kinda weird ... a cultural victory on a medium map brought me 3429 points .... it was my first game with all races being 'fools' !!

I had previously assumed AI difficulty played a part in the score calculations. If that's not the case, I can already tell that I'm probably going to end up with a very large score if I win the current game I'm playing. I took the maximum planet quality bonus along with a reproduction bonus and.. something else I think. So I suspect if I manage to win I'm going to have a stupidly large population. On top of, currently, the most powerful military. And the most influence. Think I'm "only" running second for tech, though.

This definitely needs changing if the metaverse scores are actually going to 'mean' anything. And I'd even suggest periodic resets of the "high score" (not the running totals) just to take into account changes to scoring methods, patch changes, emerging strategies etc...

~SDC~

'I had previously assumed AI difficulty played a part in the score calculations.'

It is, FWICT, but I'm not sure if it's large enough weight to account for the higher level AI's improved playing.

~SDC~

Face it, scoring is an artificial art even more difficult than programming an AI- no one since the beginning (before even Civ1) has ever done it very well... they are full of artificial values that don't reflect the "experience."

~SDC~

"Face it, scoring is an artificial art even more difficult than programming an AI- no one since the beginning (before even Civ1) has ever done it very well... they are full of artificial values that don't reflect the experience."

...but we can still strive for a MORE accurate reflection of the games that are played. The inability to make it perfect should not deter us from trying to make it better.

-Dave

~SDC~

With the grand score of about 214 or something and one game submitted, I can say that I look forward to and expect the scoring to take into account:
1. All games played, win or lose
2. Difficulty level
3. galaxy size
4. number of turns played
5. Total pop, R&D levels, military, etc.
6. perhaps a small bonus for being the winner

I think there should be a large bonus for being the winner. After all, you play games to win, no? And I think that the number of turns polayed should not factor into the equation at all. If it must be included to give some added weight to players who play on small or tiny maps, fine, but it should be minimal. I think the scores should be weighted by (in order of importance):

0. Victory
1. Difficulty Level
2. Total Pop, Tech level, Economy
3. Percent of total resources controlled by you
4. Military Prowess (w/o ship killed differenetial)
5. End of game Alliances/Diplomatic ability (for those who like to pursue peaceful victories)
6. Map Size
7. Possibly a way to reward points for drastic comebacks (ie from 7th to 1st in various categories after a certain date)

That is kind of rough, and I will put more thought into it later and suggest more, but I think this covers the basics of what really shows how good of a player you are, and how good of a game you had.

-Dave

~SDC~

I also think all of the existing scores whould be re-calculated once this issue is resolved.

-Dave

~SDC~

I just finished my Gigantic game, got through to a Cultural Victory in the end, with a score of 1441.
No real idea how I could've increased the score, though, since it's not really itemised anywhere...

~SDC~

I feel that all the scores should be scaled to mapsize. Playing a Gigantic map should not yield greater scores imho.

0-Victory should not need to be counted separately in a good scoring settings since winning only works of you have a high score. You won't win a cultural victory if you aren't the biggest anyway
1-Difficulty level should be related to mapsize. The larger maps are easier to start on and in some sense easier to play. It's also easier to isolate yourself on such maps.
2-Totals divided by number of stars
3-% of controlled resouces x time controlled. So resources controlled since start count stronger
4-Nice one. Take into account any advantages (attack vs defense). Shooting a lot of freighters with an Excalibur is a bit cheesy
5-Any civs still alive at the end should be added to your score perhaps?
6-Map size should be used as scaling
7-wautch! Tricky to keep track of that but would indeed be fun. I like my games that way. Slowly working my way back from the lower rankings.

~SDC~

It may be easier to "isolate" yourself on a big map, but that just leaves a lot more room for the AI to play in, build a great big fleet and come pound on you.

I think maybe Huge and Tiny could be similar score modifiers, but Gigantic should be worth more. It takes a heck of a lot longer, you have to really go after the victory if you don't want one of the AI to come stop you, and so on. Smaller maps are more luck than skill, often. Bad starting points can equal a loss no matter what you do. :notsure:
Plus, giving more points for the small maps will just encourage the cheese players.

~SDC~

Maybe metaverse scoring should be broken down into various groupings, depending on difficulty and map size. Perhaps a few more sorting filters would do the trick?

~SDC~

'0-Victory should not need to be counted separately in a good scoring settings since winning only works of you have a high score. You won't win a cultural victory if you aren't the biggest anyway'

A diplomatic victory can be totally irrelevant to your stats (I've actually accomplished this while only being 5th before), and cultural victory can come from someone that's not otherwise one of the top empires.

~SDC~

Acturus, in that case a small bonus would be nice I just haven't experience that myself yet

I like the idea of separate scoring for different size maps. That way also small map players can have a say in things

~SDC~

I don't think everyone will ever be happy with a single scoring method. I'm not sure about total pop, tech level and econ in scoring. Consider game A.

After 14 hours playing on a gigantic galaxy I'm in a winning position and the last enemy planet falls. Score = 2000 points

Now instead of finishing I give a planet back to the last race and continue 'playing' for another 2 hours increasing my tech, population and economy. Every time the final race gets culturally conquered I give him another planet.

After 16 hours I have huge population, tech and economy. Score = 5000 points.

WHY????? What did I do for the final 2 hours to justify that score increase? A small kid could have achieved the final 2 hours.

Now you see why I believe the populatiuon and tech level should not necessarily count.

I would love to see a max score set for a given set of starting conditions. Dragging a game on to increase score should not be possible.


Paul.

~SDC~

I agree with you Paul.
I think that's why they have a strong weight on the timing of your win in there.
The faster you win the more points. Am I wrong there?

~SDC~

I think it's clear from the above that Solitair is abusing the Metaverse scoring system. I thought something was going to be done to stop people abusing the system?

I'm not saying Solitair should kicked up the bum or anything (it's not like he's alone, or at the top of the scores anyway), but it does make a bit of a mockery of the system. The Metaverse can clearly tell when a score is from the same game (it only counts the highest score) so why can't it just reject scores coming in from the same game? You can't submit a score before the game has been finished and going back to an earlier save changes the natural course of the game.

~SDC~