|
PureBytes Links
Trading Reference Links
|
The problem is not in Windows (specially not if you run NT).
The problem is if the programmers who wrote the TS Pro code (working for the
same company that for 7 years have not yet corrected the graphics trendline
bug) did not write the application with thought on high loads.
If the application is written taking high loads into consideration, then one
or more parts of the application is using "the remainder of the CPU" (idle
processing).
If the code is written with this in mind, no error will occur due to the
load that is not handled. Of course, some things will not be done, since the
load is too high, but what parts are not done will be selected by the
programming in TS Pro. Everything in Windows will still work, including all
messages.
Now, having said that, this will only happen if the code in TS Pro is
written in a good way by knowledgable programmers.
I more get the impression that there used to be knowledgable programmers
coding TS4, but all new features are done by new programmers who know more
about windows gui appearance than large multitasking projects, which
programs like Tradestation should be treated as. And of course, it has to do
with attitude, attitude towards testing, attitude towards chasing the last
bugs before releasing the program, attitude towards what to prioritise. The
company does not seem to be driving a very hard attitude towards "let us see
to it really works before releasing the stuff". If it does not, programmers
will take the simple solutions, not the best solutions.
> -----Original Message-----
> From: Lawrence Chan [mailto:stnahc@xxxxxxxx]
> Sent: den 20 mars 2001 16:19
> To: david b. stanley; omega-list@xxxxxxxxxx
> Subject: Re: TS Pro speed, let's compare notes again, READ.
>
>
>
> The issue is 100% cpu time usage and Windows.
>
> When Windows is forced to operate in 100% CPU usages,
> it will start to drop windows messages - the default
> way for most programs to operate within Windows,
> like getting mouse move, click, whether to scroll,
> etc.
>
> Worst yet, it is also the default way for many
> slow speed communication among programs. thus
> communication between two talking programs might
> drop too.
>
> And at 100% cpu usage, with heavy virtual memory
> swapping, Windows also kill ole communications,
> among applications and its own system services.
>
> For TSPro, its server is a cpu intensive task, thus
> during heavy market hours, it will take more CPU
> usage by design. Then to double the damage, TSPro
> need to talk to this server, and more busy the
> market, more talking is needed. Once cpu usage hit
> 100%, the talking between the 2 apps will no longer
> be guaranteed.
>
> As a summary, there is no way to point to a single
> source of problem for programs like TSPro why
> it fails when cpu usage hits 100%, because everything
> it uses failed to operate "normally" as the original
> programmers assumed - do you really think that
> the programmers will stress their programming machine
> in real-time to see whether their work will crash :)
>
> Try to reduce the load by having 2 nodes of TS2K
> in real-time instead, that way, you may save OR
> (oops, now whatever technologies) from running out
> of cash, plus, your tspro will work correctly.
>
> -Lawrence Chan
>
> --- "david b. stanley" <davestan@xxxxxxxxxx> wrote:
> > This is a question to those who understand
> > processors.
> >
> > Bilo is experiencing an obvious problem but what
> > about the hidden
> > problems
> > when the machine gets bogged down but doesn't shut
> > off? Is it possible for
> > stack overflow to occur thus pushing out surges of
> > processed data in the
> > wrong order?
> >
> > Bilo Selhi wrote:
> >
> > > i would like to compare notes on Pro speed
> > > and address the issue of Pro being slow on
> > > open, close of session and in fast markets.
> > >
> > > -today my Pro simply died on that fast market
> > > short covering rally. about one half into the
> > rally
> > > cpu got pegged at 100% and stay there till
> > > win2000 gave up the ghost. long beep, cpu
> > > and win2000 locked and i had to reboot mid
> > > short squeeze.
> > > - another observation. on just about every open
> > and
> > > every close Pro is unable to keep up with the
> > market.
> > > ( it keeps up only on very light open or close )
> > > all alert are delayed till up to about 5min ( 1
> > bar, intrabar orders )
> > > after market open
> > > or about 2-3 min after the market close,
> > > depending on how heavy the open is!!!
> > > i am greatly worried about this issue at this
> > time.
> > >
> > >
>
>
> =====
> Lawrence Chan http://www.tickquest.com
> Home of trading tools NeoBreadth and NeoTicker series
>
|