JackHamilton Posted January 31, 2004 Report Share Posted January 31, 2004 Having just gotten into a serious real-time application for the first time, there are certainly some caveats for the first time Quote Link to comment
Michael Aivaliotis Posted January 31, 2004 Report Share Posted January 31, 2004 Jack, welcome to the wacky world of real-time. I have programmed several real-time applications using PXI and Fieldpoint so I know exactly what you're going through. Your post is so timely for me since I am right in the middle of another RT system and have my own recent gripes. I didn't realize how much of a paradigm shift I had to make from standard LabVIEW coding practices. My first RT project was under quoted by about five-fold. Needless to say, I lost money on that first one. :2cents: Any VI written to be run in/on real-time must be coded with this intent ground-up. You Quote Link to comment
aledain Posted February 12, 2004 Report Share Posted February 12, 2004 I am interested in your TCPIP appraisal under RT. We've done several RT applications both on PXI and RT and have used TCPIP for all comms. We've written our own server that runs on the RT just fine. Maybe we've missed the point here or you guys approach it differently to us, but TCPIP under RT was fine. However I must admit that our most of our applications are embedded not RT'ed and if you RT them you need to make allowance for "extra" time just as you would under a windows PC. If you set the deterministic loop really tight, then there are no spare cycles for the comms loop (low priority). ie the OS preempts the comms and never serves it. Similarly this is the case for the DS connection too. However I have achieved a 125Hz (from memory), RT app communicating to a DS server and never missed any data points. Michael, I'd like to know how you're using VI server to get around the problems. cheers, Alex. Quote Link to comment
aledain Posted February 18, 2004 Report Share Posted February 18, 2004 I should have said that the 125Hz was on a compact Fieldpoint controller; obviously you could get much better out of a PXI version ;-) Quote Link to comment
JackHamilton Posted February 28, 2004 Author Report Share Posted February 28, 2004 I don't recall 'complaining' about TCP/IP, just stating that to communicate between a real-time VI and the host PC - you need to use TCP/IP. Quote Link to comment
aledain Posted March 22, 2004 Report Share Posted March 22, 2004 Sorry Jack, didn't mean to imply that you were complaining. I am interested in the RT stuff as a general platform for embedded programs and I know that some of you guys do things way better than I could. I am interested in finding out the best methods for comms (ie if there are better ways than setting up your own server, etc). BTW has anyone played much with the LabVIEW 7 RT Express communications stuff. Does this give more flexibility? cheers, Alex. Quote Link to comment
CraigGraham Posted March 23, 2004 Report Share Posted March 23, 2004 As someone who will probably end up doing some stuff with Compact Fieldpoint at some point, the mention a few posts back of VI Server got me interested. Can you just open a reference to a VI running on a Fieldpoint controller and set/read controls and indicators? If those terms have any meaning since there's no display so no point in there being any form of front panel once a VI is sent to the controller. Can you also use VI Server to push a new sub-VI down to the controller whilst the system's running, to do things like calling by reference? Quote Link to comment
JackHamilton Posted March 23, 2004 Author Report Share Posted March 23, 2004 Craig, That may work, But I take the position of 'Fear' and would program TCP/IP command capability to your routines. You can instigate command the data requests in a controlled manner. I'm not a Real-Time Power user by any means, but some of these nifty 'features' can have unintended consequences. Regards Jack Hamilton Quote Link to comment
Jim Kring Posted March 25, 2004 Report Share Posted March 25, 2004 Hello RT and TCP fans. I am currently working on an RT system and I had to create a custom TCP server. I decided to make it very general and allow registering call-back VIs to handle client connection. This design allows for multiple client connections and is super-clean :thumbup: . The call-back framework handles all the management (spawing and cleanup). Please take a look and feel free to comment and discuss in the forum (see hyperlink, below). :!: My only disclaimer is that I've only spent a couple days on it and it has not been tested for memory leaks, reliability, etc. Please report these back to me, if you would be so kind. Example - TCP Server, An OpenGOOP TCP Server Cheers, -Jim Quote Link to comment
Deon Posted January 26, 2006 Report Share Posted January 26, 2006 Hi Guys Just wanted to add my 2 cents worth! I am busy developing code for a FieldPoint Rt controller, and found DataSocket too slow, TCP too much hassle, and so settled on UDP to communicate between the Host PC and the RT module. Works very well. Also, as I'm programming in LV6.1, to pass data between the core & comms loops in th RT software, I used an ODM (Object Data Manager), which is simply a non-running While Loop with a shift register and a case statement to Set & Get the required data. Btw, I use complex clusters for all data communication and transfer. Thanks for a fantatsic site! Regards Deon. Quote Link to comment
Michael Aivaliotis Posted January 26, 2006 Report Share Posted January 26, 2006 I used an ODM (Object Data Manager), which is simply a non-running While Loop with a shift register and a case statement to Set & Get the required data. ODM! that's a new one for Jim's list. Quote Link to comment
wallyabcd Posted February 3, 2006 Report Share Posted February 3, 2006 Having just gotten into a serious real-time application for the first time, there are certainly some caveats for the first time Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.