|
PureBytes Links
Trading Reference Links
|
-----Message d'origine-----
De : Mark Johnson <yenmstr@xxxxxxxxxxxxx>
À : 'pierre.orphelin' <pierre.orphelin@xxxxxxxxxx>; omega-list@xxxxxxxxxx
<omega-list@xxxxxxxxxx>
Date : jeudi 29 avril 1999 00:18
Objet : RE: Best alternative to TradeStation 2000i?
The answer will be sharp and fast.
1)Most of Mark Johnson points rely on the assumption that :
data1 is a privilegied one in TS when considering function.
This is false.
Only data 1 has a special status regarding to systems.
ESPL language do not solve the problem (that is not a problem, as we will
see).
2)
There is also problem for some of you to understand series and simple
functions that seems to be black magic.
May I suggest that you attend to Mark Jurik session, devoted to the topic in
Las Vegas ( he performs two sessions).
Also some good reading on TS Express old issues.
This is not a valid reason to throw EL away.
If anything of above works for you, try black magic and sacrify a white
chicken on the TS4 security block and say three time in a row " yk2 patch".
3) TS2000 new features allow to write functions that you may apply to
arrays.
This means that you can ignore the classical maxbarsback concept shared by
all data N and write your own one bar back user functions.
The calculation being applied to an array that you may load with a one bar
maxbarsback setting ( consider a circular array), the problem disappear in
TS2000i.
4) If you could spend some more time to study the language ( that you
obviously do not master ) and its subtilities that the time you spend to try
to destroy it on this list, things could be better for all of us, yourself
included.
5) No programming language will help you if you consier things an prblem
from an apriori and biased point of view.
This applies to anyone, any software, any language.
Rgds
-Pierre Orphelin
Représentant exclusif de Omega Research en France.
web: http://www.sirtrade.com
====================original snipped message============
>Easy language is simpler because it increments the
>bar number for me?? What's hard about bar = bar + 1?
>
>Because easy language is dependant on data1 bars --
>I suffer. I can only make trades off data1 because EL
>only understands bars from data1. If I have other data
>series better hope that all the bars match with Data1 or
>else.
>
>Oh yea, EL keeps all there vars in arrays. Makes life so
>much easier for me with all this series and simple functions stuff.
>
>Like I have to have a Average function that is both series and simple.
>Now how confusing is that??
>
>Much rather just create a Average function with a For Loop at least there
>is no black magic involved and my code is clean.
>
snip
> Yes it is a very obvious mistake. How can I work with bars from
Data2,Data3,Data4??
> The answer is you can't -- unless EL has a reference to a bar in Data1.
>
snip
>> > rAverage := rPrevXA; <========= previous values are stored in EL
>
snip
> Yep the wonderful maxbarsback. Makes life easier for me. If I want
my different data
> series to have different lookbacks to bad in EL. Have to use the
global maxbarsback
> across all data series. So damm limiting.
>
snip
> Nope no code at all to write the above function -- just the know how on
how to understand the black magic in
> series and simple functions.
>
>Mark J.
>
|