|
PureBytes Links
Trading Reference Links
|
Jay,
I think that you may be confusing indicator code with
function code. Indicator code typically contains the
commentary information you're talking about (at least
in TS4 - in 2000i the commentary is in separate functions).
So this code is not something you have to
worry about in terms of system executable size.
What I'm really referring to in terms of size optimization
is the code itself. There are more memory intensive
and less memory intensive ways to code things...
that's really what I'm referring to - making the code
efficient (perhaps at the expense of clarity) by eliminating
unnecessary variables and/or inputs, etc.
Also, none of this has anything to do with series
variables vs.simple variables. In fact, you don't
really control this. All you can control is series vs.
simple function inputs...
Functions called multiple times are built in to the
copiled code once.
X = Value1[1] is, in general, valid syntax. Again, series
vs. simple is not going to effect the size of compiled
code.
OM
---- you wrote:
> The Omega Man had written:
>
> >One of the keys to reducing the size of your EasyLanguage executable (and
> avoiding the need to write a dll) is reducing the number of inputs and
> variables you use.
> >Also, try to reduce or eliminate the use of functions which have a large
> number of inputs.
> >Also, remember that many of Omega's built-in functions have not been
> optimized for size so you may be able to create your own, smaller versions
> of them.
>
>
> You have begun to answer my questions in your interesting reply that I have
> summarized
> above, but let me ask you to elaborate. Many of the Omega-supplied
> functions include a
> lot of extraneous "stuff" - the commentary that newbies can all up to
> explain what it means
> to have one moving average cross another - or whatever. But, for those who
> simply want the
> function to return the value of the m.a., this commentary code is not used.
> I assume that
> the code required to support that feature counts toward the 64K limit. And,
> that this is
> what you mean by ".....you may be able to create your own, smaller versions
> of them."
>
> It seems that I recall hearing that if a system/indicator calls the same
> function from
> multiple places, then that function's size is counted multiple times. I
> suppose that
> has to do with the "simple vs. series" thing, where series functions need to
> retain the
> prior values of variables. Does that imply that multiple calls to a simple
> function
> would not result in its being counted multiple times? Is the E.L. compiler
> that
> clever?
>
> It's interesting that named variables count toward the 64K, but "Value1" and
> "Condition1"
> -style variables do not. Does that suggest that Value1 is a simple, and not
> a series
> variable? Ie, is a statement like: X = Value1[1]; valid?
>
> Thanks
>
> Jay Mackro
>
>
>
|