[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Push-Pop vs. GV Performance effects



PureBytes Links

Trading Reference Links

Greetings:

I am trying to get a multiplier of tick chart indicators on charts. Push-Pop
is an excellent program, and after some effort, I got multiple tic
indicators working. But before I continue, would the C++ speed of Global
Variables be a better match for what I am trying to do? Push-Pop is
excellent due to its time/date stamps, but the Pascal programming is not as
efficient as C++. I have heard some speed concerns before but do not know to
what extent - ?

Both Global Variable and Push-Pop cannot be used for historical testing, and
Pushpop did 'jag' the longer tic indicators due to the smaller tic frames
increments. So I assume GV will do the same, since TS does the 'jag' as well
on multi minute charts. Pushpop's Jags can be worked through by 'double' JMA
(averaged) smoothing, and other fine tuning tricks I'm sure. And I did
notice Realtime, PushPop indicators were smoother than reading from
historical data after rebooting - ?

The best or smoothest multi time frame indicators are by Sacred Chaos, but I
want Jurik indicators and C++ speed, so... I don't have Juriks code to
transform the MTF function. My best 2 alternatives for tic on tic indicators
seem to be Pushpop (which works) and Global Variable (which I assume would
work, but have never seen tic on tic examples and GV has no time/date stamp
capability).

Ultimately PushPop's Pascal speed is an issue since it would be used
excessively. But I'm not sure if GV even does tic on tic chart indicators,
and getting wrappers really sux since all TSworld indicators are in eld's (I
have TS2K). So working with Global Variables will be much slower than
PushPop which was tailored for TS2K.

At this point, could someone give some directions for what I'm trying to
get. My current setup would require 4 screens, 2 subcharts/screen, 3
oscillators per subchart (which means double this to send and receive), &
WinXP Pro.


Thanks,
Phil