Yeah I'm familiar with that one. This happened around the time of AWTresspasser if you remember those times. They were actually a bit before my time but I did catch traces of this upheaval in various searches that I've made.
That is a good point though that society is probably just population. I didn’t catch that implication previously.
There are actually a number of implications that can be gleaned from this. The first one was that they were essentially taking the discrete integral of the area under the associated curve with of course division by the turn number. This was assumed to cause the score to asymptotically approach a limit. The problem is that even though you are dividing by the turn number you are summing up the total number of slices which of course go up with turn number. The net effect is a sum that is unbounded by the number of turns which was proven by AWTresspasser’s 209 year 6 million point game.
The other point which we have discussed before is more of a supposition on my part and that is that the original scoring method of the sum of the raw values divided by turn number is what forms each component score which is then multiplied by the difficulty to the 1.1 power and divided by another turn number to provide it’s contribution to the total score. This is why good players that build up relatively high values early in the game can have a total score twice as great as the sum of the four component scores whereas average players may eventually reach the same levels of raw score but since they tend to reach these levels later in the game their total score may only be half that of the sum of the four components. This kind of 4 to 1 swing between games that otherwise have very similar “raw score” totals is something I’ve seen demonstrated time and time again.
Like I said we’ve both had this discussion more than once, this is more for the benefit others that may read this, however this is by no means the first such time we’ve tried to convey this concept.
One other thing to mention is that there actually does have to be some kind of scaling involved between population and the other three components simply because populations are in the billions and the other components are in the millions when they score similar amounts. It could be as simple as dividing pop by 1000 (probably is). Also this highlights why research is so useless a place to optimize score since if income and research are similarly scaled then income will always be more significant than research since research is limited to be a relatively small percentage of total income.
As far as my original question a new Diplomat asked this question at the Core and I definitely remember hearing someone spout off the ratios. But this was probably close to 2 years ago. I was hoping that perhaps Kryo remembered more precisely.
Ahh. Found the reference I was looking for. It was at
GCII Victory Conditions - Points.
According to Magnumaniac
Currently the ratios are:
Tech 1.0
Alliance 1.5
Culture 1.75
Military 2.5
Now can anyone validate these ratios? Kryo?
