Talk:Review Algorithm/1.4.4/@comment-99.119.14.242-20130505004942

g = (m + q * m) * p * o * r * t S = [10 * g / h] * x / 10 It took me awhile to understand that you are not trying to get any particular absolute value for Design+Tech that results in g >= 10. Your score is relative to your BEST game's value. h is basically your highest g (production value) so far plus some percentage, which "raises the bar" for a 9-10 score each time you beat it. If I get a 9+ and repeat the same production values I will usually get score 6-8 simply because the bar is now 20-30% higher to get a 10.

You want to add only to g what you need to hit 9-10, if you add a massive new feature list (to boost D+T) it's a one-shot deal - now you have to train staff or add even more features for the next hit. Basically you need to keep training and adding people and features to hit the next "threshold" for a top game, but you don't want to do it all at once or you set the bar far higher than it needs to be (the "b" value inflates). You want to milk as many 9+ hits as possible as your production values increase.

There are a lot of smaller modifiers, but it looks like the main ones that you manipulate seem to be m and q. m is mainly increased with training (I prefer +speed so my T:D balance is mostly the same), and with features (benefit points = direct boost to D+T). Larger games and staff can do more D+T in absolute terms (higher m = higher g).

If your slider choices are proper then you get +0.2 q, if your overall employee stats are balanced for the game type you get another +0.1 q, this is as though you had 30% higher D+T. Now I realize you don't have to spend so much effort maintaining a strict tech to design ratio, the absolute value doesn't matter only the relative g value to the next hit.