NI-488.2 Version 2.0 vs 1.70

[x@no.mail]

In my application I am calling ibwait(board,0) to get the current ibsta
value. I need to test the END bit to see if my ibrda() call is finished. I
call ibwait as often as I can in the program (lots of data coming in). I am
using System Monitor on Win98se to watch memory usage. When I use NI-488.2
Version 2.0, I can see Allocated Memory increase very quickly (by 1M every
5 seconds). If I use version 1.7, this does not happen. Any ideas? Is
memory leakage a known issue with the 2.0 driver? I prefer the 2.0 driver
because it seems quicker, but the memory issue will kill any chance of being
able to use it.

Thanks

[Geneva L.]

Hello,

Is this the same setup (PC, OS, and application) that you had with the "Multithread problem with ibstop" posting?

http://exchange.ni.com/servlet/ProcessRequest?RHIVEID=101&RPAGEID=135&HOID=5065000000080000006F650000&UCATEGORY_0=_26_%24_13_&UCATEGORY_S=0

I just want to make sure I have a good picture of your setup. Also, which GPIB board are you using?

Can you attach an NI Spy capture for when this happens? You can log a capture using the Spy>>Options>>Log to File menu item.

Regards,
Geneva L.
GPIB/Serial Product Support Engineer
National Instruments
http://www.ni.com/ask

[DiegoF]

Hi,

I ran a test using the code attached below which just calls ibwait over and over again. I ran it for an hour and a half and the memory used remained at 4196K according to the Windows 2000 task manager.

At what rate does the memory used increases in your system?.

I'm using NI-488.2 2.0 with Windows 2000 (SP2) and the USB-GPIB-A.

Check out the code and see if there's anything to add or if it does increase memory usage in your system.

DiegoF

[x@no.mail]

In order to run your code I had to reinstall 2.0. I did this over 1.7 (I
did not uninstall 1.7 first). When I did this your example worked fine and
so did my program. So for now I can't even duplicate it. Thanks for your
reply and for your investment of time into my problem.

FYI I'm using Win98se, PCI-GPIB. When it happened memory usage was
increasing at the rate of 1M per 5 seconds.


"DiegoF" wrote in message
news:50650000000500000034CD0000-1042324653000@exchange.ni.com...
> Hi,
>
> I ran a test using the code attached below which just calls ibwait
> over and over again. I ran it for an hour and a half and the memory
> used remained at 4196K according to the Windows 2000 task manager.
>
> At what rate does the memory used
increases in your system?.
>
> I'm using NI-488.2 2.0 with Windows 2000 (SP2) and the USB-GPIB-A.
>
> Check out the code and see if there's anything to add or if it does
> increase memory usage in your system.
>
> DiegoF

[x@no.mail]

Yes, it's the same. I got around the multithread problem (I hope) by making
sure that all my accesses of the GPIB are done via a single thread. I'm not
happy about this because it's a bit of a 'hack' of the program. I'm using
a PCI-GPIB. Thank you for reply and concern over my previous problem.
My reply to Diego's post explains where I am with my current problem - so I
won't be sending an NI Spy capture. Thanks again.


"Geneva L." wrote in message
news:5065000000050000002FCD0000-1042324653000@exchange.ni.com...
> Hello,
>
> Is this the same setup (PC, OS, and application) that you had with the
> "Multithread problem with ibstop" posting?
>
>
http://exchange.ni.com/servlet/ProcessRequest?RHIVEID=101&RPAGEID=135&HOID=5
065000000080000
006F650000&UCATEGORY_0=_26_%24_13_&UCATEGORY_S=0
>
> I just want to make sure I have a good picture of your setup. Also,
> which GPIB board are you using?
>
> Can you attach an NI Spy capture for when this happens? You can log a
> capture using the Spy>>Options>>Log to File menu item.
>
> Regards,
> Geneva L.
> GPIB/Serial Product Support Engineer
> National Instruments
> http://www.ni.com/ask