|T O P I C R E V I E W
||Posted - 20 Apr 2010 : 02:59:23
In a program that is used and developped for many years now we use the Realtime charting tools for Windows in c++. We still use revision 2.52.
Now we are building a new version that is supposed to run on Windows XP. Somehow, it seems that at unpredictable moments, the data that is drawn on the graphs is incorrect. This results in a realy 'wide' line in the graph, where it should be a nice sine funtion.
As you can see, in the first part of the graph, everything looks normal. In the second part of the graph, the error occurs.
In this program, the sampling of the data, and the drawing of the chart are different threads. The strange thing is that as soon as the drawing of the chart is restarted, everything goes back to normal, and the problem disapears.
This leads us to believe that the problem is in the Quinn-Curtis functions, or in our calls to those functions.
Can you give me some pointers to where to start searching?
Thanks in advance!
P.S. The older version of this program runs on Windows 2000 and doesn't have this problem.
|1 L A T E S T R E P L I E S (Newest First)
||Posted - 20 Apr 2010 : 08:48:46
There is no support we can supply specific to Rev. 2.5. Support for that revision was dropped more than 10 years ago. The last revision of the software was 3.0, in 1998. You should have upgraded long before now. You can still upgrade to Rev. 3.0 from our web site. There is a special price if your use our download ordering system at: http://www.quinn-curtis.com/Ordering.htm
Your description sounds as if multi-threading is your problem. You cannot call any of our Charting and Real-Time graphics routines from a thread different than the thread which created the original chart. This always applied to the software, and still is a requirement in Revision 3.0. It may be that something in XP exacerbated the problem, regardless, it has always been a requirement. So you are going to have to figure out how to keep all calls to our software in the same thread.