Which code is faster?

[snorkle]

is it faster to have a byte array and use leftb$,rightb$,midb$ etc., or to have a string and use left$,right$ etc... ?


why is it faster to use the $ /w those commands?

 

[JohnYingling]

If you want string processing use MID$, LEFT$ which return strings (not MID ,LEFT which return Variants). Use byte processing only for binary data encoded in strings e.g. BLOBs, images from a database. Each VB string character is two bytes of Unicode data so if you process it as byte, you better know what you are doing.

http://WWW.VBCompare.Com

 

[snorkle]

john - thank you for your response. i never noticed they returned variants.

what i am doing is loading a file into a byte array. then i wanted to assign different sections of that array to other arrays. i'm doing that like this:

dim b1() as byte
dim b2() as byte
dim f as long
Dim f As Long
f = FreeFile
Open "c:\something" For Binary As #f
ReDim b1(LOF(f))
Get #f, , b1
Close #f
b2 = MidB(b1, 10, 5)

is it faster to use the arrays like this, or to use strings in a simular way?

 

[JohnYingling]

I do not know what kind of binary data you have. if it is truly ASCII characters then read it into a string. If not, you are better off staying in the byte array. Look at StrConv function. It does a lot of stuff I'd rather not mess with as far as Unicode. If you have character data go with strings. Look also at AscW and ChrW.

http://WWW.VBCompare.Com