The following are possible process control and visualization software that might be adapted to work with the GCU. Advantages/disadvantages of each package or new software options should be added below.
Process Control/HMI
Proview - http://www.proview.se/
Scicos - http://www.scicos.org/
Modeling
OpenModelica [modeling language] - http://www.ida.liu.se/labs/pelab/modelica/OpenModelica.html
Ascend - http://ascend.cheme.cmu.edu/
Graphing
LiveGraph - http://www.live-graph.org/
kst - http://kst.kde.org (Linux/Windows, please report back if you get this working, looks powerful)
RRDTool - http://oss.oetiker.ch/rrdtool/index.en.html (needs script to upate database, writes graph to file, for e.g. web serving)
Veusz - http://home.gna.org/veusz/
GNUplot - http://www.gnuplot.info/
- forum post on using GnuPlot for real time plotting: http://www.gekgasifier.com/forums/showthread.php?t=395
R - http://www.r-project.org (statistics package, graphing and data analysis)
Helpful R language reference: http://cran.r-project.org/doc/contrib/Short-refcard.pdf
Comments (3)
michaelshiloh said
at 4:48 pm on Sep 22, 2009
bear - i have a student doing independent study with me who might be perfect for doing the visualization software. i'm interested in using processing for my own selfish reasons. call or email to discuss.
bk said
at 5:01 pm on Sep 22, 2009
Thanks Michael, yes, lets talk.
bk said
at 12:23 pm on Sep 26, 2009
Veusz looks like a promising means to display data. It is cross-platform, can auto-update from a data source, but is not explicitly set up to display realtime data.
Jeremy Sanders, author of Veusz responded to an email asking about Veusz:
"I just found Veusz and it looks great. Cross-platform and powerful. I see there's a capture option with auto-updating, which looks great. I'm looking to just read the last data written to a file (say the last 300 lines) and graph just that (no need to store previous data either). Is there a reasonable way to do that?"
Sorry for the delay in replying - I'm a bit busy at work at the moment. There's no explicit option for only storing the last X values - it would be a good thing to add to the program and probably wouldn't be too hard to do.
What you can do to simulate this is to create a new dataset based on a previous one but only having 300 values. Go to Data->Create. Enter a name for the dataset, e.g. "newdata", choose data from expression, then enter an expression in the value text edit such as
olddata[-300::]
And put this in the error text edits if your dataset has error bars.
You can then plot newdata rather than olddata. You can enter any python numpy expression when creating new datasets based on old.
Unfortunately this will give an error if olddata does not already exist, so you would need to do some capturing first or import or create a minimal dataset.
You don't have permission to comment on this page.