You'll need to use some kind of logarithmic scale at high levels. I tested this but it's a bit inappropriate at very low levels. So for levels 0 up to 15 I suggest a linear scale, then after 15 revert to the logarithmic.
For example the current to-hit in unarmed combat looks like this
Code:
UNARMED FIGHTING 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
To-Hit Modifier 0 +1 +2 +2 +3 +3 +3 +4 +6 +6 +9 +10 +10 +10 +12 +12
So I fit this to MIN(0.88*level,4.17*ln(level)) giving the values (after rounding):
Code:
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
0 +1 +2 +3 +4 +4 +5 +6 +7 +8 +9 +10 +11 +11 +12 +12 +13 +13 +13 +13 +13 +14 +14 +14 +14 +14 +15
The constants 0.88 and 4.17 I used to fit the curve as close as I could to the ADOM data. I wanted to maintain that level 15 gave +12 to-hit with both the linear and logarithmic. These constants can be modified to allow for to-hit, to-damage and DV, and race,class,starsign,weapon type, etc..
Going through the manual the current values seem to be almost linear - although some of them tend to "curve up" that is increase at an increasing rate, which could be a problem if you want to stick closely to these values. If you really want to stick to these closely, a cubic equation is probably necessary at levels 0..15 such that the to-hit at 0 is 0 with gradient 0, and the to-hit at 15 equals the given to-hit value, and "tapers off" slightly at the end. I.e. match the gradient of the cubic function to the gradient of the log function at that point.