Rolf Kalbermatter

QUOTE(anima @ Feb 1 2008, 09:49 AM) My crystal ball is out of order, so I can't see what you did and why you think it is slow. Just try to explain clearly and if possible show us some code. There is a lot one could do wrong but without having more specifics everyone here is simply dabbling in the dark. Rolf Kalbermatter

QUOTE(vronto31 @ Jan 31 2008, 12:43 PM) I haven't used it with SAP yet but another company wide database management system called BAAN. This used to be a rather popular system here in the Netherlands but is going to get replaced more and more by SAP it seems. However BAAN was built on top of the Oracle database engine and therefore the Oracle ODBC access proofed perfect for what we needed. Don't tell your IT guys that you are using LabVIEW, just ask them how to access their system from any client application through either ODBC or ADO. Once you know the specifics such as what ODBC/ADO driver to use and what parameters to use to connect and query the database you can just go about it much the same as you did with Acess. Rolf Kalbermatter

QUOTE(Aitor Solar @ Feb 1 2008, 03:53 AM) I can only second that. It's an esoteric function enough that I would guess that people using it in such ways do know what they are doing and otherwise are prepared to live with the consequences. I don't see a LabVIEW noob even knowing that it exists , let alone using it. As my workaround for the LuaVIEW Toolkit shows, treating the refnum as a 4 byte number and copying it into the 4 bytes of a strict typedef VI refnum still works correctly. So typecasting refnums at least in LabVIEW 8.5 is still simply a datatype change but does not change anything in the underlaying memory representation. I did assume that it might have been in preparation for more significant changes to the refnum system in LabVIEW in the future, that might break with simple typecasting. But your idea that it might be more something about guaranteeing that a strict typedefed VI refnum points to something to be sure in memory might be also a good guess. Rolf Kalbermatter

QUOTE(Aitor Solar @ Feb 1 2008, 02:34 AM) Well, I see. It still might be a problem in the future, or the refusal to typecast to a strict typedefed VI refnum is maybe a bug?? Na, I don't think so. Rolf Kalbermatter

QUOTE(R.Alves @ Jan 31 2008, 08:12 AM) No for processing images later you would definitely need some image analysis library such as IMAQ Toolkit from NI or the IVision Toolkit from Hytek Automation. Rolf Kalbermatter

QUOTE(souske @ Jan 31 2008, 10:21 AM) No sorry. It's some time that a played with the LabVIEW Embedded Module, and that was with LabVIEW 7.1 too. Also I didn't incorporate any external code at all. Just played with the provided example framework for the Linksys WRT-54L router. But no I didn't mean the Code Interface Node or CIN. As I have written many times here and in other places, CINs are legacy and NI is not going to go into any trouble to support CINs on newer platforms. They maintain them on the platforms where it is already working but for the rest CINs are dead. If and how the Call Library Node can work on your system is also a big question. If there is a clear and well defined standard for shared libraries on QNX it is quite possible that LabVIEW supports it. But I never went as far as using external shared libraries for LabVIEW Embedded so I'm not sure how they would support that. It looks likely to me that this would have to be part of the extra code one needs to write to interface LabVIEW Embedded with the external toolchain, but on the other hand I'm not sure how LabVIEW Embedded would provide hooks to do just that. What I meant is that I think to remember that LabVIEW Embedded should have a C Code structure. It would sit on the diagram, look like a Formula Node and allow to type in C code that is then interfaced to the variables at the border. This C code will then be passed to the external C code compiler and of course could also contain calls to library functions present on the target system including your own shared libraries or object modules. I'm not fully sure that exists, never used it, but believe that I read once something about it. It is in that way much like the VHDL Structure for LabVIEW FPGA where you can write in VHDL code to be passed to the FPGA compiler. Note: Ok I read on the NI site about it and it is called there the "Inline C Node". Rolf Kalbermatter

QUOTE(souske @ Jan 31 2008, 04:32 AM) How do you get your QNX system to work with LabVIEW? To my knowledge there is no direct support for QNX by NI other than possibly through LabVIEW Embedded. If you use LabVIEW Embedded there should be the C code node (a bit similar to the Formula Node) that allows you to link to external code by writing calls to external libraries. Those libraries of course must have been created outside of LabVIEW with the toolchain for your target. How you do write these libraries and device drivers is entirely outside of the scope of LabVIEW. It does not care nor could it! So for your main question you will have to take it up with the QNX support people for sure. If you are not using LabVIEW embedded, the only way I see that your LabVIEW software could communicate with your QNX system would be a server application on the QNX system that access your Hilscher card and provides an RPC interface through TCP/IP for instance to your LabVIEW application. Only LabVIEW Embedded could possibly produce code that could directly run on your QNX target. But unless NI provides a ready made example for integrating LabVIEW Embedded with your specific hardware plattform and OS however (they obviously can't and won't create this for every possible target system out there), providing the necessary glue code to integrate your platform and OS into LabVIEW Embedded is a tedious job that takes up some serious time even if you are completely familiar with both your target plattform and LabVIEW Embedded (and if you are not, expect even a lot more time to be needed). Rolf Kalbermatter

QUOTE(Harvey @ Jan 31 2008, 04:20 AM) You sort of could write one, by converting a carefully selected timestamp into a string and then searching the specific positions and separation characters. The problem however is that this will never be able to solve all problems. This is because the DB engine and the according ODBC or ADO connector can or can not use the local date/time format on the machine they are running. I do know that the MS ODBC/ADO connector always observe the local format of the client machine and the SQL server can be configured to do that or not. Other SQL database engines will most likely have their own ideas about if and where to adhere to local format settings. There is supposed to be an ODBC standard format for date strings, but I had at least with SQL Server trouble to get that working between different computers. The solution in that case was to use the SQL Server CONVERT() function in the SQL statements to tell it explicitedly the string format the date was in. QUOTE Other problem is with the numbers point or comma '100,00' Converted into '10000' Same issue really. All MS ODBC/ADO client connectors will recognize the local decimal point. So for them you would use the standard LabVIEW formatting functions making sure "use local decimal format" is enabled in the Options, or using the Format Into String nodes add a %; to the beginning of the format string to tell it to use the local decimal fomat. Other database client libraries however may or may not observe the local decimal format. Rolf Kalbemratter

QUOTE(R.Alves @ Jan 29 2008, 02:28 PM) Duplicate post! Look http://forums.lavag.org/Package-for-Ip-cameras-t10006.html' target="_blank">here. Rolf Kalbermatter

QUOTE(Neville D @ Jan 30 2008, 03:49 PM) Which RT system has a GigE interface? Without it won't work anyhow. Rolf Kalbermatter

It's considered polite to put up a link for cross posts. Here is the same text again. It seems not possible to find any detailed specs for this camera on the internet. That means I can not even guess how to access them at all. But since it has a built in web server you can always access the images through HTTP. This could be either done through a IE ActiveX control or similar embedded in your VI front panel or by using the Internet Toolkit HTTP VIs to retrieve the image programmatically. OpenG has some rudimentary HTTP routines too, which should probably just work fine too. Rolf Kalbermatter

QUOTE(TobyD @ Jan 29 2008, 02:55 PM) And if the first is true I wonder if the name is just to try to get more likely help as a HH :laugh: Rolf Kalbermatter

QUOTE(R.Alves @ Jan 29 2008, 02:37 PM) Well, nice idea but there is a problem with that. There is nothing like a single IP camera communication protocol that I'm aware of. If a camera has the possibility to be controlled over IP directly it is sometimes RSTP or a proprietary protocol, that you can access through an Active X control only, and whose protocol is not normally documented or it contains an embedded web server that can be accessed through a web browser. For the proprietary protocol through ActiveX, you would have to integrate it through use of the Active X functionality in LabVIEW and this will be of course camera specific, and the second you could access trough the HTTP VIs that are part of the Internet Toolkit and of course are camera specific too, since the web server structure is also different on each camera type and sometimes varies between versions of the same camera too. RSTP is not very common and only really allows control of the camera itself, but is not the streaming protocol itself. If the camera supports it you can request with the DESCRIBE command an url from which you can retrieve the current image, which will be usually an FTP or HTTP url, and that can be handled with the Internet Toolkit. But considering the NON-standardization among IP cameras such a Toolkit will never be able to support more than a few specific cameras, so it will be hard to justify creating something like that unless you happen to have a specific application, and that means often that one develops them for a specific project and therefore can't just put it up for free download because of right of ownership. There is of course also GigE, which is a standard and I think NI is supporting that already. Rolf Kalbermatter

QUOTE(Tim_S @ Jan 28 2008, 07:53 PM) Why do sara and jennifer post quite similar requests here. I get this feeling that sara learned from the reactions and now got an identity change and tries to at least show that she did a little work before asking questions. Still!! Rolf Kalbermatter

QUOTE(Yen @ Jan 29 2008, 01:23 PM) It won't! The IMAQ Control is not really a fully built in Control but added to the LabVIEW system as an extra through external files. Those external controls do not support the object methods necessary to edit the control. And you can customize sub controls of normal built-in controls! You can't open a customize window for a second control when one is already open, but when there are editable sub controls (such as a digital control in a slider for instance) you can open that in a second editor window. I know from the original creator of the Control Editor that he considers that as a work that was done back at that time which should have been redisiged since a long time. One limitation was that that editor was not really reentrant for instance which made it necessary to disallow multiple editor windows being opened at the same time, except for the case of editing sub controls. Another limitation is that it is fragile at its best. Unexpected edit operations or controls that behave strange in some circumstances can make it go complete belly up together with LabVIEW and all your code. Rolf Kalbermatter

QUOTE(Yuri33 @ Jan 28 2008, 05:28 PM) It works in LabVIEW up to 8.2.1 but fails in LabVIEW 8.5. The typecast function will refuse a refnum connected to it with a broken wire. We found this in our LuaVIEW Toolkit package and had to develop a workaround for it. I sort of understand the idea behind, since a refnum is in fact something like a pointer to an object and messing with pointers can be a VERY bad thing. There is a chance that that new behaviour in LabVIEW 8.5 is in preparation of something that might change the refnum entirely and that might make the workaround we developed for LuaVIEW in LabVIEW 8.5 invalid again in a future LabVIEW version. Rolf Kalbermatter

QUOTE(PeterB @ Jan 28 2008, 05:43 AM) It means that the event structure might be serving an UI event at the moment the user event arrives and hence not be able to react immediately. A Read Queue waiting for an event to arrive will return instantly or at least as fast as possible, considering the contstraints of available CPU power and load only. Rolf Kalbemratter

QUOTE(macbeth @ Jan 21 2008, 03:52 PM) Why not show your VIs? Rolf Kalbermatter

QUOTE(crelf @ Jan 17 2008, 08:36 AM) This dialog should only be triggered when upgrading a pre 8.0 VI. So I'm afraid it is probably more a question of a corrupted VI somehow. For some reasons LabVIEW 8.5 does believe it is a pre 8.0 VI and applies that check but then stumbles over something else that is inconsistent in respect to this assumption. A bug? Well strictly speaking the bug happened either when saving that VI in 8.2 or the VI was corrupted somehow in such a way that 8.2 did not have trouble to read it but 8.5 has. It would be of course fine if LabVIEW 8.5 would just inform you of the inconsistency and continue instead of hanging, but there are miriads of things that can be inconsistent and anticipating them all is not very easy and checking them all is even less practical. I do think that technical support would be interested in this VI to be attached to the CAR so that LabVIEW development may look into the kind of corruption and maybe apply an extra check in the VI loader. Rolf Kalbermatter

QUOTE(silmaril @ Jan 10 2008, 08:26 AM) It may be that you have fiddled with the default fonts in your development environment. Check your LabVIEW ini file for entries like FPFont=, BDFont=, AppFont=, DlgFont= etc. When you change the font settings while you have some control or text setting selected the change applies to that control or text. Without anything selected the change applies globally and will also be written to the INI file if it is not the default setting. An application has its own INI file and by default no Font settings are in there which will cause LabVIEW to use default Font settings. The LabVIEW 8.5 Application Builder does copy some of the INI settings from LabVIEW.ini into an application specific INI File unless you disable that by specifying your own INI file in the build. But I'm not sure if Fonts happen to be among them, as I almost always use customer specific Font settings for Buiild applications. So no, I'm sure it is not a bug, but yes I do think the Font settings and default Font configuration could use some extra explanation somewhere in a prominent spot in the online help. And the Font configuration dialog could be more intuitive. The checkboxes for Panel Default and Daigramm Default are a bit hard to understand in respect to how they influence the INI file and future changes to a VI. Rolf Kalbermatter

QUOTE(ben5248 @ Jan 24 2008, 07:53 AM) Probably just a change in marketing. Those modules were introduced with cRIO but in the meantime are also usable with the new USBDAQ modules NI sells. So I guess they decided to drop the cRIO in the labeling to not confuse users. Rolf Kalbermatter

QUOTE(ashwin @ Jan 27 2008, 04:56 AM) Forget the queues for a moment. You want to call those DLL functions in a loop or something and test it's return value or it's returned parameter until it gives you the value to tell you it is ready. Then you call the function to retrieve the data or do whatever else you want to do. There is no way to use LabVIEW queues or such directly with the DLL. Once you have the DLL calls working properly you can of course develop a VI library that could give you a queue based interface. But I do not think that would have real advantages over a more direct VI library approach directly. Rolf Kalbermatter

QUOTE(PeterB @ Jan 27 2008, 06:34 PM) But they execute in a structure that also handles all the normal UI events that WILL switch to the UI thread for sure. And also might contain a timeout case that will also snoop some time for handling. ONe extra point. Queues are very specifically optimized to NOT copy data at all but instead use the reference to the data instead. User events might not be able to use the same optimization in all cases or even at all, especially if you use multiple event structures to receive the same event. Rolf Kalbermatter

QUOTE(ashwin @ Jan 26 2008, 09:32 AM) You can't. For a driver to inform you of something it must implement some kind of event notification. For DLLs there are two possible solutions, they can either allow you to install a callback function that will be called when an event occurres or they can use OS events to inform the caller about such an event. In any case these are things that are not directly translateable into LabVIEW events so you will basically have to write a wrapper DLL that calls your Device DLL with the event notification method that your driver uses and translates it into something more LabVIEW friendly such as a user event or occurrence. C programming knowledge will be required for sure to do that. Rolf Kalbermatter

QUOTE(Yen @ Jan 25 2008, 06:52 AM) Yen you are right here, but there needs to be some extra clarification since a LabVIEW refnum is a memory resource too, and I have trouble to believe that .Net will have any influence on the life time of that refnum itself, even if .Net has a very smart Garbage Collector. It may invalidate (but probably won't since the LabVIEW refnum holds a reference to the .Net object) the .Net resource that was referenced by the refnum, (and possibly though not for sure, the Not A Refnum node may be able to detect that too), but it most likely will not dispose of the LabVIEW refnum before the Top Level hierarchy that created it goes idle. So Your concerns are fully valid and Big Hanks claims are likely to be not the full story with using .Net in LabVIEW. Rolf Kalbermatter