Dispaying long number (with , in it)

[Uriziel01]

for example

i have

$pi=3.1415926535897932384626433832795

but $pi will have real value

$pi=3.14159265358979

i cannot use something like

$pi1=3.14159265358979
$pi2=3238462643383279

or in another example

$a=10
$b=$a/3

Anyone now how to make $b

$b=0,333333333333333333333333333333333333

not only

$b=0,3333333333333333

Any ideas ? Please:(

Edited March 23, 2008 by Uriziel01

[SmOke_N]

Maybe this post may give you some ideas:

http://www.autoitscript.com/forum/index.php?showtopic=56915

[Uriziel01]

Nice ! Txh alot but now i have another problem, im writing function that can quick calculate PI walue (http://en.wikipedia.org/wiki/Pi) my script looks like this

Dim $tStruct = DllStructCreate("double"), $i=0
for $ii=0 To 18
    DllStructSetData($tStruct, 1, DllStructGetData($tStruct, 1) + ((1/(16^$i))*((4/((8*$i)+1))-(2/((8*$i)+4))-(1/((8*$i)+5))-(1/((8*$i)+6)))))
    $i+=1
MsgBox(0,"PI value",DllStructGetData($tStruct, 1),1)
next

I think the problem is in (im loosing the precision in this pleace)

((1/(16^$i))*((4/((8*$i)+1))-(2/((8*$i)+4))-(1/((8*$i)+5))-(1/((8*$i)+6)))))

because it is buildet from normal variables not from DllStruct. Is somebody able to help me with this ? Im stuck

Edited March 23, 2008 by Uriziel01

[PsaltyDS]

Nice ! Txh alot but now i have another problem, im writing function that can quick calculate PI walue (http://en.wikipedia.org/wiki/Pi) my script looks like this

Dim $tStruct = DllStructCreate("double"), $i=0
for $ii=0 To 18
    DllStructSetData($tStruct, 1, DllStructGetData($tStruct, 1) + ((1/(16^$i))*((4/((8*$i)+1))-(2/((8*$i)+4))-(1/((8*$i)+5))-(1/((8*$i)+6)))))
    $i+=1
MsgBox(0,"PI value",DllStructGetData($tStruct, 1),1)
next

I think the problem is in (im loosing the precision in this pleace)

((1/(16^$i))*((4/((8*$i)+1))-(2/((8*$i)+4))-(1/((8*$i)+5))-(1/((8*$i)+6)))))

because it is buildet from normal variables not from DllStruct. Is somebody able to help me with this ? Im stuck

AutoIt uses signed 32-bit integers, which limits you to -2147483648 to +2147483647 because you get 31 bits for magnitude and lose a bit for sign. Switching to unsigned (which is what SmOke_N did with 'uint') only gets you one more bit to play with, still limiting you to between 0 and 4294967296. (You could use 'uint64' to double the bits, but there will still be a limit.)

The struct type you used, "double" is a 64-bit floating point. It can't represent greater precision than a 64-bit float supports. If you want to calculate Pi to more than about 16 places to the right of the decimal place, you'll have to use a different technique. You won't be able to represent your complete value as a number variable format that AutoIt recognizes. Instead you'll have to work the math for calculating the next digit, and then assemble it as a string.

There are already scripts to calculate Pi to a million digits or more, but they don't store that result as a number variable, but as a 1MB string.