From Friday, April 19th (11:00 PM CDT) through Saturday, April 20th (2:00 PM CDT), 2024, ni.com will undergo system upgrades that may result in temporary service interruption.
We appreciate your patience as we improve our online experience.
From Friday, April 19th (11:00 PM CDT) through Saturday, April 20th (2:00 PM CDT), 2024, ni.com will undergo system upgrades that may result in temporary service interruption.
We appreciate your patience as we improve our online experience.
09-29-2006 10:59 AM
10-03-2006 11:00 AM
06-26-2009 05:24 AM
I am facing a similar speed issue with MathScript nodes. I have written a client application using two normal while loops - one producer loop reads the TCP port and writes to a queue at about 150Hz, the other consumer loop reads the queue and does some heavy going matrix calculations within one giant mathscript node, ideally within a ms or two.
The producer / consumer arrangment give me the buffer I need, because there is no handshaking with the server(!) So my mathscript nodes have to sit within loops and receive data a frame at a time. It might be a pain to do, but would normal wired code in a sub.vi be faster than mathscript?
06-26-2009 10:10 AM
grahamwebb wrote:... but would normal wired code in a sub.vi be faster than mathscript?
Yes!
What kind of operations are you doing? How big are the matrices?
06-26-2009 11:22 AM
Good! .. I am using 3x3 matrices for linked-segment modelling - addition, subtraction, transpose, multiplication, nothing over the top, lots of indexing and manipulating vector components. I've currently got 350 lines of Mathscript that I'd like to run at 120Hz. It will take a while to wire! so I wanted to check first what sort of speed increase to expect.
08-18-2009 08:44 AM
08-19-2009 01:41 PM
Hello,
The MathScript R&D team is continuing to work on improving the edit-time responsiveness of MathScript Nodes. According to the MathScript RT Module roadmap shown at NI Week 2009, we have currently scheduled some further improvements in this area for the 2009 SP1 release, targeted roughly 6-8 months from now. (Although, this is just the expected roadmap and things are subject to change.)
We would really like to make sure that we address your issues with these changes. If you are able to send us your application, we could make sure that our optimizations fix the issues that you're seeing. If you can't share your code for some reason, we would still appreciate any further details that you can give us. Are all the MathScript Nodes on the same VI or are they on different VIs? Do any of the MathScript Nodes have particularly complicated expressions or call deeply-nested user-defined functions? What kinds of actions are you performing when LabVIEW hangs? If you wait long enough, does LabVIEW become responsive again or do you have to restart? Any other details that you can give us would help us isolate the cause of the issues you're seeing.
With your help, I hope we can make the editing experience acceptable in the next release of LabVIEW.
Thanks!
jattas
LabVIEW MathScript R&D
08-20-2009 02:17 AM
jattas,
If you could provide an NI email address I'd be happy to send you a section of the code and discuss the problems further.
The nodes are all on one vi for development purposes, but will be 'subvi'd' later. They are performing simple matrix calculations. LabVIEW doesn't respond for a minute or so but then recovers. The fact the whole program doesn't respond is a pain because it isn't possible to work on other vi's while I wait.
Graham
08-21-2009 09:35 AM
Hello Graham,
I sent an email to the address that NI has on file for you. Please let me know if you do not receive it so I can contact you by other means.
Thanks,
jattas
08-27-2009 06:44 AM
Hi jattas, I haven't received an email. I have added a University email address to my profile that may be worth trying.