Does anyone know the formula SEIV uses to calculate score?

geoschmo · #1 January 24th, 2001, 02:51 AM

Thanks for all the feedback.

I spent a couple of hours Last night trying out different games settings and I have figured out the fomula exactly.

I also have done some work on what combination of ship size to component arangement gives you the most bang for the buck in score to maint ratio.

I want to win the twingalaxies tourney though so I won't be sharing my work here until after it's over. MUUUUUHAHAHA-HA-HA-HA!

Drake · #2 January 24th, 2001, 06:31 AM

Heh, good luck. I've done a few calcs of my own, and the biggest problem for me right now is how long it takes to process my turns because of all the ships in the game. I'm not sure I'll have enough time to finish a full 150.

Seriously though, it doesn't look like the game restricts you to the setting for max # of ships in startup. Even if you restrict yourself to 2000 ships & bases combined, you should be able to hit 50M+ by turn 150 using the biggest space station with the first Version of the master computer and lots and lots and lots of organic armor (again Version 1) to bring the total to 1510kt.

At a cost of 6500m, 4430f, and 1000r, a huge world should be able to pop one of these out a turn. Large and medium worlds every other turn at least. If you set things up right, hitting 2000 ships is a feasible target. Now worrying about the infrastructure to support all those ships... Well, I think I'll wait until next month before mentioning that, heh.

-Drake

Drake · #3 January 24th, 2001, 06:53 AM

On second thought, I've changed my mind. In the interest of fairness, I'd like to point out, in case you weren't aware, that you can reduce your maintenance costs to ZERO by upping maintenance to 120% and selecting merchants.

I also take organce, adv. storage, and the +25% planetary shipyard rate. Factor in +20% research, +10% growth, decent mineral and farming and +28% const rate, and you basically can take out the AI in about 20-25 turns usually without firing on their planets.

I'd pick either the Sergetti or the Amon'Krie symbols to play with, because they only surrender if you have 50x their score as opposed to 10x their score. By using their art, you ensure you won't be seeing them.

Other tips for the contest would be to make sure you have a good balance of races for the different atmosphere types (check race symbols at beginning of game), have over 35 systems in your quadrant, have 15 or 16 total players in your game (forcing them to surrender ALWAYS), and use a spiral galaxy type.

The spiral galaxies are typically better connected, so your missile escorts can take the shortest route to your enemy's HW. Keep close eye on your score, when it hits 10x theirs, demand surrender. Once you hit 200k+, you'll typically be able to demand their surrender as soon as you meet them, without bothering to blockade their HW.

Basically once you wipe out the AI, you're only concerned with the amount of construction you can get done per turn, and the resources needed to support that turn's construction. With zero maintenance, it simplifies things so that you don't find yourself overextended on turn 120 or something.

I'd basically suggest not bothering to research anything that didn't benefit either expansion, population growth on your existing planets(for higher construction), and resource production. Once those goals have been met, build the biggest size hull you can per turn, filled with organic armor.

I'll be interested in seeing what kind of scores you guys can generate with the tactic. I've been able to beat the AI in 16 turns with propulsion expertise, but not do that and keep the 0% maintenance, and the latter is more useful for purposes of the contest..

Keep me posted if you decide to give it a try & good luck!

-Drake

[This message has been edited by Drake (edited 24 January 2001).]

[This message has been edited by Drake (edited 26 January 2001).]

geoschmo · #4 January 24th, 2001, 06:48 PM

Good point about the spiral galaxies and using one of the Xenophobes as your own race Drake. That's two points I had not considered.

Zero Maint? I was using reduced maint but had not thought it all the way through to that extreme. That seems almost to be a bug if it is true.

Drake · #5 January 24th, 2001, 07:08 PM

I don't think it's really a bug, but probably more of an oversight that occured when they changed the effect of maintenance in race creation. I'm guessing they didn't intend for maintenance to fall below 5%, but forgot the merchant maintenance reduction.

I think the maintenance issue, coupled with the AI surrendering so easily, totally unbalances things with regard to the contest. I'd much rather see it be the player vs all other AIs, who have a huge bonus, and the player isn't allowed to communicate with the AIs at all. Either that, or base victory on something other than the score. Oh well, it's probably a bit late to change things that drastically.

Also, another tip - if you set your construction to 128% and pick hardy industrialists, on your standard HW with 2000M pop, you can build the shell of a colony ship each turn without having to go into emergency mode. Then just retrofit to include engines and a cargo module. It'll actually cost you less than building the full ship from scratch, due to the way retrofitting works. You'll only be delayed one turn, but after that you won't have a slowdown period. I wouldn't recommend doing that with your starting HW, because you'll probably need the quicker start to get your points up faster, but it works great with the homeworlds you'll acquire from getting the AI to surrender.

-Drake

[This message has been edited by Drake (edited 24 January 2001).]

Nyx · #6 January 24th, 2001, 09:30 PM

Warning, this one's really long.

Totally overriding the maintenance I'd call cheating.

Now this:

quote:
I also have done some work on what combination of ship size to component arangement gives you the most bang for the buck in score to maint ratio.

Is the sign of a good player.

Your trick about building colony ships faster is also a great idea. Exploiting the ability to retrofit a ship an infinite number of times in one turn, that's cheating.

AI and resource manipulation for your own benefit based on study, practice, and an understanging of the mechanics: Good gameplay.

Finding ways to avoid following the rules and circumventing the mechanics: cheating.

It is a moderately thin line, but you should all be able to recognize it. We expect anyone who considers themselves a champion at a game to be able to make that call.

quote:
I think the maintenance issue, coupled with the AI surrendering so easily, totally unbalances things with regard to the contest. I'd much rather see it be the player vs all other AIs, who have a huge bonus, and the player isn't allowed to communicate with the AIs at all.

Well, let's say we hold a player vs player contest. We can't rightly say that the winner is the top player can we? S/he's only the best who showed up to that tournament. High score is the best tool, but it's not a perfect tool--even in arcade and pinball (though it's close to perfect in some of those cases).

Simply defeating the AI? Please, even all AIs vs the human with all AIs getting high bonuses, isn't much of a challenge. Even if it was, you know we'd have more than one person win. What criteria do we use to determine the victor then? It's like calling someone who finished Mario Brothers a world champion just for completing the game. Overcoming the game is nothing, you need to ovecome the other players. We could combine it with fewest turns, then highest score Last, but even then it's iffy. So someone's fast and aggressive, are they the best just for that? Is crushing several AIs faster than they can act truly the sign of a better player than one who can build up and control the greatest portion of the galaxy and get the maximum value out of his holdings? I don't necessarily think it is.

I actually like the way this contest is turning out. The top players are finding fascinating techniques and we're into the empire building side of the game as much as the enemy crushing side. The winner will be the person who crushes the AIs fastest, and then develops the most in what time remains. S/He's got to excell at both aspects of the game, not just one or the other.

quote:
Oh well, it's probably a bit late to change things that drastically.

Technically, not. Our rules and terms & conditions clearly state we can alter anything at any time in the interests of fairness and avoiding cheating. However, this has to be weighed against the severity of the disturbance to the players.

This is where the PC platform is a problem compared to our origins in the arcade world. Arcade ROMs rarely changed and basically had only a couple of settings. We declare the original factory settings to be the official and only accept scores on those settings played on actual arcade machines. Simple. Even easier than consoles. PCs... Boy they're a mess.

Not to mention how with new games the bugs and cheats are being found continually. There's really not much left for people to discover about Pac Man. Three people can score "perfect games" meaning get every dot on every level, eat every bonus fruit, and eat all four ghosts off each power pellet all on their first man. It takes about 4 hours (nearly 6 the first time, but they've perfected the technique since then). These guys know EVERYTHING about the game. I've seen a tape where Billy Mitchell just walks away from the machine for over 15 minutes to take a break and the ghosts don't get him. He knows the patterns well enough to know where and when such things are possible. No rules changes will be made in that game, I assure you.

Unlike the arcade games, we can't hold a PC world championship a year or two after release since no one is still playing the game. We have to aim at the first couple of months. But that means the game is in flux with regard to bugs and cheating. It's an ugly situation for the rules. especially here in SE4, where we usually play Last-man standing but this contest is high score. Bugs we never saw before because we were never looking are turning up.

I have no problem with the AI surrendering, because in a high-score contest, you're not competing against the AI, you're competing against the other humans in the contest. If you can get them to surrender faster than me, then you're a better economist player than I am. If I can enslave them faster than you can force surrenders, then I've got the advantage. This is all acceptable and even desireable. You should see what some of the people can do in the top tier of gaming. Using the computer as a resource is a key strategy to achieving the best scores in any game.

quote:
Either that, or base victory on something other than the score.

We thought about first person to a set score (we were thinking 5 million or so), and we may yet introduce that contest at a later date. I've almost got our founder to agree to a SE4 team ladder and/or contest. We could really do just about anything you guys ask for. It all depends on how much of a success this one turns out to be.

------------------
Compete in the Space Empires IV World Championship at www.twingalaxies.com.

[This message has been edited by Nyx (edited 24 January 2001).]

Drake · #7 January 25th, 2001, 01:26 AM

I understand where you're coming from, and mostly agree with your points, Nyx. However...(another long post, sorry)

I think the issue with maintenance is still a little murky. I'd agree that 0% maintenance was probably not intended and is somewhat exploitive. What's a fair amount then, the 5% you can get not counting culture modifications? If so, can you then take 115% on maintenance to get 10% and reduce that by another 5% by picking merchants to net yourself 1000 points in race creation?

quote:
I have no problem with the AI surrendering, because in a high-score contest, you're not competing against the AI, you're competing against the other humans in the contest.

I guess the problem I have with the AI surrendering is that is does so based on the difference in score. If there aren't any rules against it, I'll find myself building more ships than I need in a certain turn, simply because I know it'll inflate my score, and the AI will capitulate. This can give me an unfair advantage when comparing results against players who aren't aware of the quirk in the AI. I mean, I found myself not wanting to colonize a planet I would normally colonize right away because if I did, I'd lose the 3k pts from the ship and that would drop my score below the 10x threshold! I'm not really concerned about making it fair for the AI, it's going to lose anyway.

In the buildup stages, there's no point for me to build anything that isn't as effective as possible in getting a higher score. I feel like it's a little exploitive being able to just dump tons of organic armor and organic weapons on hulls and not having them cost any more in minerals than one with just a master computer. I don't actually need these ships/bases I'm building, they're purely for score.

quote:
It is a moderately thin line, but you should all be able to recognize it. We expect anyone who considers themselves a champion at a game to be able to make that call.

I guess what I'm getting at is I find this to be a little... well, arrogant. The tone implies that there's a line that should be clearly drawn, and if you don't see it, you're a 'cheater' not a 'champion'. No hard feelings here, but while most would agree on the infinite retrofit being a cheat, I think you'd see a lot of different opinions on things like maintenance, taking advantage of AI, and the like. That's mainly why I brought up the issues in the first place, to see what people thought and hopefully generate some discussion on the matter. Thanks for taking the time to weigh in.

-Drake