Jump to content

didierj

Members
  • Posts

    363
  • Joined

  • Last visited

    Never

Posts posted by didierj

  1. I am under the impression that if you don't close the file and you're writing to it, that it might be cached and then you would lose status messages that should have been saved to disk, if the machine were to go down.

    2303[/snapback]

    In the palette "File I/O">>"Advanced File Functions" there is a function named "Flush File". It is probably an "antique" function, when computers weren't that fast and writing to disk stopped every other execution (including the mouse) on your computer.

    Open once the file and keep the reference open as long as your app runs. Write to the file as wished and every time after a write add "Flush File".

    Didier

  2. Is there a way I can name the file with a timestamp similar to the options in write LVM DAQ assistant?

    2296[/snapback]

    In the "Time & Dialog" palette you will find a function that is named "Format Date/Time String". Use it to format the date/time the way it fits for your filename:

    post-253-1097729468.png?width=400

    Didier

  3. But for us Europeans it's no problemĀ  :D

    2288[/snapback]

    It is a problem!

    I don't want to imagine how angry my wife would get, if I hang evening long on the computer chatting around and giving her (and soon my baby) the cold shoulder... :(

  4. When you offer no option for the innocent or opressed to defend themselves, then fairness is lost and bullies win.

    The joke of the poor-innocent-back-yard-researcher-that-has-to-be-protected was something for the beginning of last century, antiquated nowadays.

    Patents is a billion dollar market. When a firm is sold, the counts of patents this firm holds is an important factor on the selling price.

    When I hear that such stuff gets patented like: "mouse click on a window scrollbar" or "sperma treated with some hormone to enhance reproductivity" then I get serious problems concerning the idea of patents. :thumbdown:

    I prefer the way of view, that when others take your Idea, then it's a greate one, otherwise it's bullshit. :thumbup:

    Free market means that you have to be better than your concurence, not that you have to kill them by kicking in the ######, this is more wild-west.

    ...just my $0.02 opinion.

    Didier

  5. I wouldn't mind having the scripting possiblities as a tool-box. I probably would even be one of the first bying it. It extends LV so much and one can create help tools for easier programming.

    I even doupt there might be some legal problems. The ability to write code from code would mean that one could build some new programming language from LV and distribute it. NI wouldn't be glad seing that people program LV without bying LV development suite...

    Didier

  6. If I interpret this chart correct, the biggest traffic are after lunch and in the afternoon of the american east-coast.

    Considering that in Europe (switzerland + germany at least) we have GMT+1. This would mean, best lounge times would be (for me) 8 p.m. and later. :blink:

    Uh, usually at this time I'm at home enjoying a tea and dessert, and not reading/writing LAVA forums. Others would even say they enjoy a movie on "home cinema", but since I have no tv-set...

    My best time for joining a live chat would be in my early morning hours (7-9 a.m. GMT+1). Afterwards it's getting hectic in the office.

  7. I don't know if in LV7.1 it's possible, otherwise, count the samples acquired and if needed create a new file:

    (10Megs) / (5channels x 4bytes per value) = 524288samples / 655536samples per second = 8 seconds

    Means: after 524288 (=2^9) samples or 8 seconds you'll have to create a new file.

    Best add an index to the file or it's extension (in windoze) and wire it to the Write LVM file.vi

    Didier

  8. Not quite a code challenge, I'v already done something like that, with quite some statistics (haven't the code anymore present, must have lost when I last changed job :( )

    post-253-1097558950.png?width=400

    You'll still have to wire them into your code, but less complicated than with sequences. If you make the icon footprint small (e.g. a small clock, 8x8pixel), it won't disturb your code too much.

    Didier

  9. I'v just seen we have a full chat room... no more "trial version" :thumbup: :thumbup:

    Maybe we can negotiate a fixed chat time.

    What do you think?

    This could be a bit difficult, since we have members from America, Europe, Asia, Australia, Africa(?), all with different timezones.

    I'm not soooo used to chat-rooms, but I think best would be to log in and wait, or send a PM (since we see who is online).

    is the chat-room already considered as an alternative for the LAVA-meetings?

  10. It has been suggested that the splashscreen loads the main application. It is not necessary since if both splashscreen and main VI are TLVIs, they'll load and run simultaneously at startup. If the splash screen is lightweight and opens quickly, it will also prevent the "loading..." screen to appear when the main application takes too long to load and open its FP.

    That's the way I always did it... load+open splashscreen, load main app dynamically, end splashscreen.

    This is a much easier method. :thumbup:

    Another question rises up there: how do you know when the splashscreen has to close? Is it just a timed loop, or do you have any sync (occurence or something else)?

  11. Soon they only want 'one button' to do the whole thing

    Like general computer rules...

    In win 3.1 every setting was within 1-3 mouse-click. in XP it's withing 6-9 clicks :thumbdown:

    Reminds me the "aunt Olga"-test of a german computer magazine. An XT (computer generation of somewhere in 1980) won in time against a Pentium IV in sending an email to "aunt Olga", including boot, connecting to the internet, write mail, send, shutdown. :thumbup:

  12. If you are afraid of loosing data over the net, then you'll have to add an acknowledge:

    1. Send the data to each client

    2. each client calculates a checksum from the received data, e.g. CRC16

    3. the clients sends the checksum back

    4. if the server gets a wrong or no checksum back from one client, he sends the data again to this client.

    This would also mean that each connection is handled (in the server) separately, so your server exactly knows which client doesn't work correctly. A broadcast won't work.

    Didier

  13. Hi Sarah,

    If you have pasted your sub-vi into your diagram, then the program flow waits until your sub-vi finished execution (except operations that have parallel dataflows). Your sub-vi executes all the code inside itself until it returns. As the vi is loaded at program-begin in such a situation, there shouldn't be any exception in the sort of exit before loading.

    I just see two possibilities:

    1. Your sub-vi is not in the main (user interface) loop. There the loop and your sub-vi executes in parallel and your sub-vi is run just in the beginning of programm execution.

    2. You call the sub-vi (or a sub-vi in the sub-vi :wacko: ) with vi-server. Then you must be sure to have the method "run vi" with its property "wait until done" set to true.

    Didier

×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.