Machine Vision

[Bart_Sky]

We have a flat panel connected with a NI 1424 framegrabber. The framegrabber is in a computer running on Windows 7 64-bit. In the past we used the imgGrab function of the NI-IMAQ functions.
Recently, to improve the speed of data acquisition, we started to use the imgSnap function. (http://zone.ni.com/reference/en-XX/help/370161G-01/imaqfr/imgsnap/) When we use this command, we noticed that some images were darker compared to the others. All images had the same exposure time. We looked in more detail to this problem and logged the time needed to take one frame. It turned out that for the normal, brighter images the time needed was around 1200 ms (1s exposure time and 200 ms read-out + overhead); for the darker images, the total time was around 250 ms (taken 200 ms read-out and overhead, the real exposure time is nearly 0s). We repeated the test on a Windows Vista 32-bit system and then we don’t have the problem. We took more than 8000 frames of 1s exposure time and never a frame was skipped. So probably, this is a 32-bit – 64-bit problem.

Did anyone experience the same problem and has anybody a solution?

Thanks,

Bart

[ThiCop]

Hello Bart,

Which drivers are you using?

Can you upload a MAX Report (http://digital.ni.com/public.nsf/allkb/271F252B4EF0A2E0862570E70056A1E4)?

Does the function return a 0 ("succes") after using it?

[Bart_Sky]

Hello Thierry,

The MAX report is in attachment. When doing imgsnap, we check if the function returns an errorcode, but always a 0 is returned.

FYI: We are using the Visual Studio 2010 C++ compiler to compile our software.

Thanks for the help,

Bart

[ThiCop]

Hello Bart,

I think I have found a probable cause for the issues you encounter.

According to these KB's the versions of NI-IMAQ and NI-IMAQdx you're using do not officially support Winddows 7 (32 or 64-bit):

http://digital.ni.com/public.nsf/allkb/E7725A08B0E152628625747900627A3E?OpenDocument

http://digital.ni.com/public.nsf/allkb/4A127014CA6E006B8625787A0066D329?OpenDocument

Can you download a newer version of these drivers that officially supports Windows 7?

[Bart_Sky]

Hello Thierry,

We updated the versions of NI-IMAQ and NI-IMAQdx. The new MAX report is in attachment. Still the problem exists.

Regards,

Bart

[Bart_Sky]

Hello Thierry,

We updated the versions of NI-IMAQ and NI-IMAQdx. The new MAX report is in attachment. Still the problem exists.

Regards,

Bart

[ThiCop]

Hello Bart,

Can you provide me with the MAX Reports of both pc's (working and non-working) and also with the code used on both pc's?

This will allow me to eliminate certain things.

Are you using exactly the same card (by moving it from one pc to the other) in both pc's or 2 seperate cards of the same type?

[Bart_Sky]

Hi Thierry,

In attachment, you will find the MAX report for the Windows Vista 32-bit system. The configuration for the Windows 7 64-bit system, you will find in one of the previous posts. We used the same NI card and detector in the both set-ups. We just swapped them from one computer to the other.

Regards,

Bart

[ThiCop]

Hello Bart,

The last MAX Report didn't really make me much wiser.

If you compare it to the previous one, then you can see that there is no card inserted or camera connected to it.

Therefor I cannot derive any of the settings/configurations that were done (at camera or framegrabber level) in MAX for the working Vista system.

Is there a way to provide me with this information?

[Bart_Sky]

Sorry Thierry, my mistake!

Now I attached the correct MAX report.

Regards,

Bart

[ThiCop]

Hello Bart,

I had a look at your MAX Reports and I saw the following differences:

1. Frame rate is 1.000 on the Vista system and 0.400 on the Windows 7 system
2. LUT: Binary Tesholds seem to be different between the Vista and the W7 system
3. Your card also seems to be recognized as part of a PXI-chassis on the Vista system (not on W7). Is this supposed by like this?
   I though you had a PCI-card in use?