diff --git a/data/powerup.xml b/data/powerup.xml
index 479fb2832..4739fea76 100644
--- a/data/powerup.xml
+++ b/data/powerup.xml
@@ -84,7 +84,6 @@
is different. Typically, the higher the number of karts, the
stronger the difference in "powerup quality" between the first and
the last.
-
Each of those weight tags is stored into a WeightsData object.
At race time, a new WeightsData object is created from the list for
the current race type depending on number of karts in the race.
@@ -101,7 +100,6 @@
the first non-leader kart, and the last entry for the last
kart. So the 3rd till second-last entries are distributed
evenly instead of the 2nd till second-last).
-
For example, with 5 karts and 5 entries those points will
match positions 2, 3, 4. With 10 karts (and 5 entries), they
will correspond to 3,25 ; 5,5 ; 7,75. It is not an issue if
@@ -118,14 +116,12 @@
a triple item rather than a single one. The probability to get
an item is its weight divided by the sum of weights of all items
(single AND multi). It is recommended to keep that sum equal
- to 200 to easily keep track of probabilities.
-
+ to 1000 to easily keep track of probabilities.
'Global' items which affect all karts (switch, parachute) should
be quite rare, since otherwise the item might be used
too often (compared with many items which will only
affect a karts or two) - especially with increasing number of
karts in a race.
-
The distribution should give more similar items to different ranks
when there are few karts, but have more important differences when
there are more karts. -->
@@ -135,70 +131,69 @@
-
-
-
-
-
-
+
+
+
+
+
+
-
-
-
-
-
-
-
+
+
+
+
+
+
-
-
-
-
-
-
+
+
+
+
+
+
-
-
-
-
-
-
+
+
+
+
+
+
-
-
-
-
-
-
+
+
+
+
+
+