The expressions doc gives bitwise evaluators, but doesn't show how to set bits. Is this supported in Mugen yet?

I know two ways to accomplish such a thing:
First:

[State WhatEver, I don't Care]
Type = VarSet
trigger1 = Some Trigger Here!
var(0) = var(0) | floor(exp(var(1)*ln(2)) + 0.5)

Where var(0) is the var you want to set the bit, var(1) is the bit you want to set (may not be bigger than 31 nor smaller tha 0)

The other way is more straight:

[State WhatEver, I don't Care]
Type = VarSet
trigger1 = Some Trigger Here!
var(0) = var(0) | (The result of Power(2, Bit Value (0..31), such as 32, 256 and so on)

The first one is more generic. The second method allows you to set a fixed bit.
You may want to add a & mask to reset var(0) or make trick combination of the opperators to get nice results such a XOR.

Have fun, and always keep in mind this: exp(power * ln(base)) > base ** power.

This post has been edited by SMI on Nov 13 2003, 10:30 AM

Didn't realize it was so involved, but that's highly nifty to know. That's gonna
go in my toolbox.

Is there some benefit to using a generic bit? Is there anything you can do with
it that you can't do with an int?

Btw, bitwise XOR is already an operator in the expressions doc.

SMI, and 2**x won´t work?

where x is the (x+1)th bit you want to set.

I haven´t tested it yet, but it makes sense and it´s equal to the formula you gave.

Renzo: I don't trust the ** operator. The solution I provided gives more accurate results.

Enkephalin:
The results are the same, but the generic solution makes it easier to debug and maintain the code.

This post has been edited by SMI on Nov 13 2003, 05:19 PM

I do not get the question... the advantage of using bitwise operators is that they allow for further tweaks and better code performance by using optimization of simple repetitive conditions. Another advantage is data compression in case you find 60 int vars too few of a number for your needs.

I meant was there advantage to setting a generic bit as opposed to a fixed bit. I was thinking I could cram a whole bunch of attributes into one bitmap.

the advantage is the usual coding dilema, performance against clarity of code; but since they are constants and ebyte might some day implement code optimization in their parser i suggest you to use the generic bit approach.