Brent My original email below explains that the delay could be longer than 2 microseconds if the VI is interrupted. As I indicated, in my case I needed a delay of at least 2 microseconds so that's OK for me. If you need an exact delay, this is almost impossible to guarantee. As Greg McKaskle pointed out though, for short delays, it is unlikely that the VI would be interrupted although undoubtedly this would happen once in a while and that may or may not be good enough for your application. For me, I use delays in communicating with hardware via a parallel port. I'm talking to a device that times out if I don't send data in 18 msec so if my many waits of 2 microseconds or 20 microseconds get extended because Windows interrupted my application, there's a good chance my application will run again before the 18 msec expires. I added logic to begin transmission again if 18 msec has expired but tests show that this never happens despite many million executions. It's a lot more reliable than I had hoped. This part of the software runs in above normal priority in the data acquisition thread under LV 6.0.2.
R. Glenn Givens P.Eng. Innovention Industries Inc. Burlington, ON, Canada www.innovin.com -----Original Message----- From: Brent DeWitt [mailto:[EMAIL PROTECTED] Sent: January 9, 2004 23:46 To: R. Glenn Givens; [EMAIL PROTECTED] Cc: LabView Discussion Group Subject: RE: Wait 0.5 milliseconds? Could you explain in more detail? I don't understand how software timing to the resolution of 2 microseconds can be implemented in a multi-tasking operating system (take your pick: Windows, Linux, Apple). But then I'm not an OS or LabView guru. Brent DeWitt -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of R. Glenn Givens Sent: Friday, January 09, 2004 12:51 PM To: [EMAIL PROTECTED] Cc: LabView Discussion Group Subject: Wait 0.5 milliseconds? We do waits of 2 microseconds up to 1 millisecond. Of course, if the VI is interrupted, the delay could be longer but that's OK in my case. We use a For loop with N being a value we set during software installation by testing the speed of the computer over a fraction of a second. It works very well for our purposes. R. Glenn Givens P.Eng. Innovention Industries Inc. Burlington, ON, Canada www.innovin.com ............................................................. Subject: Wait 0.5 milliseconds? From: "E. Blasberg" <[EMAIL PROTECTED]> Date: Tue, 6 Jan 2004 15:22:31 +0200 Hi All, I'm going to assume that the answer is "no" (at least if one is not using LabVIEW Real-time), but does anyone know if it's possible to set a delay time of less than 1 millisecond? Here's the problem: we set a switch (using a DIO card), wait n MICROseconds (called the Switch Settling Time) and then trigger a Receiver to measure a value. Currently this is done in C++. Naturally I'd like to be able to say that LabVIEW can "do you anything you can do better". Sadly, I think this time I can't. Any comments? Thanks (in advance), E. Blasberg iDAQ Solutions Ltd.
