The Worlds of Kaomaris

Guys

Currently a players score is calculated by adding Corn, Wood, Gold & Diamond value together.

To this is added a calculation of an approximate defence value per province held.

This defence value is the PDEF of the province (taking into account terrain, upgraded troops and creatures) multiplied by the total number of troops and then modified by effectivness.

The total should give an approximation of how difficult someone is to eliminate based on their current troop deployments.

Scores will vary dramatically from turn to turn. If you move catapaults from a mountain area to a plain then your score will drop and vice versa.

Is everyone happy with this?

Can it be done better?

Suggestions please.

Gaz

ps I will probably be making adjustments to the code throughout the game but will not access player data at any time. Any changes required to players data will be done by Korexus.

pps I am fairly confident that the last few coding bugs have been removed over the last couple of days but you never know for sure.

pps Post number 400 !!!! Yippee

I preferred the old system, where the score was independet of the location of the upgraded troops and their grouping (i.e. fixed score bonus for each upgrade).

If however people prefer the new system, where you can use the scores to gain information about the location of the enemies upgrades, then I would also be happy to stick to with the new one.

Xarfei

I don't like the score to necissarily tell too much in terms of how everyone is doing, so I like the newer scoring system over what Xarfei mentioned as the old scoring system.

Ulty

It seems that the problem with the new scoring system is the dependence on the armies location. Apart from that the new scoring system seems better and gives a more accurate view of who is the strongest player in the game.
The problem with the old scoring system was its dependence on the spell id. This allowed people to guess which spells players started with by looking at their starting scores. To fix this problem AL added a random number to the score, but that way he made the score random...
So I suggest we use the new scoring system but remove the effect of terrain from the calculation. That will make the score less accurate, but it will be more consistent between turns and it will be less revealing.

I agree with Mullog, the terrain is the key problem with the PDEF. Having a high score because you have many upgrades is perfectly reasonable, being able to hide it by keeping them in a different terrain seems unfair.

Also remember that while the armies defense power may be lower in a different terrain, they could still be about to rampage through 7 provinces of the terrain for which they are suited, thus they're attacking power is being ignored. Mullog's solution covers this too.

As for spells. Adding the spell ID as points was completely random to begin with, adding a random number on the end just confirmed the score calculation as a Gassner classic. Far more sensible to me would be to say, for example, all creature spells are worth 100 points, all red spells 50, all white spells 75 and all black spells 150. (Numbers open to amendments.) Or even just say 100 points per spell...


korexus.

I'd like to see the current method work in practice for a couple of turns.

If the score changes due to troop locations are not too chaotic, then I wouldn't mind the drawbacks that the other players mentioned.

trewqh wrote:I'd like to see the current method work in practice for a couple of turns.

Then take a look at the Advanced Omega test, it was being used there

If the score changes due to troop locations are not too chaotic, then I wouldn't mind the drawbacks that the other players mentioned.

It was pretty chaotic as I recall. Unfortunately I don't remember which turn had the massive jump, but it was somewhere around the middle, I think.

To reuse the example I gave at the time: using the current formula, 100 level 1 catapults are worth 1,400 points in a swamp and almost 10,000 points in a mountain. When 20,000 is considered a very healthy score this is quite a difference. (Obviously catapults have the biggest jump because they are the most powerful upgrade, but it is catapults and knights which are the most common.)



korexus.