Need help with variables

[Sheepmancer]

• • Unit Group - Pick every unit within 200.00 of Temp_Point matching (((Matching unit is Alive) equal to true) and ((matching unit is an enemy of (owner of (triggering unit)) equal to true) and ((matching unit is a building) Not equal to true)) and Do Actions

I have this unit group action, wich works, however instead of a constant 200 range i want to change the range acording to the level of the ability being cast.
I tried making a variable, Frostbite_Aoe, and set Frostbite_Aoe[0]=0,Frostbite_Aoe[1]= The aoe of the ability on level 1,Frostbite_Aoe[2]= The aoe of the ability on level 2, Frostbite_Aoe[3]= The aoe of the ability on level 3, ETC.

Now heres the problem, i can't put a variable in the place of the 200 value.
The Variable type is intreger and i set it to array.
Should i choose a difrent type of variable than integer, or is what trying just impossible?
If anyone knows tell me!

 

[Flux]

Use a real-type variable.

 

[Sheepmancer]

Use a real-type variable.

Ty, it works now +rep

 

[Wietlol]

Integers are round numbers, while reals (or floats) are decimal numbers.
You can convert one to the other using the functions "Conversion - Convert Real To Integer" and "Conversion - Convert Integer To Real"

 

[Kaijyuu]

The reason we don't always just use reals is mostly because they're inaccurate. You can't accurately represent some numbers on a computer (like 1/3), only approximate them to a "reasonable" accuracy, and this occasionally has noticeable errors.

Another reason to have integers is for some low level computer stuff (like array indexing) which if you know how computers work, wouldn't make sense with reals, but that's a bit more complicated.

 

[Wietlol]

"and this occasionally has noticeable errors."
You are talking about an inaccuracy error of -0.00005% to +0.00005%
It shouldnt be noticeable.

 

[Dr Super Good]

The reason we don't always just use reals is mostly because they're inaccurate. You can't accurately represent some numbers on a computer (like 1/3), only approximate them to a "reasonable" accuracy, and this occasionally has noticeable errors.

Its mostly because they are slower to compute than integers as they are more complex.

Another reason to have integers is for some low level computer stuff (like array indexing) which if you know how computers work, wouldn't make sense with reals, but that's a bit more complicated.

One could create an instruction set that implicitly converts floats to integer values when using them as index or other whole number needing things. However it would have many issues and would almost certainly be slower.

It shouldnt be noticeable.

For most purposes it is not noticeable. However for others it is.

 

[Kaijyuu]

For wc3 in particular, it gets noticeable with very high health/mana values. A while back I was using extremely high max mana units (mostly to just hide the mana numbers under the portrait), and was getting significant errors when tracking how much mana they spent on spells (a 100 mana spell would end up costing something like 64 mana). Fixed by reducing their max to more sane values.