MIO-16E-1 Output sample rate limited?

[cbaxter]

I have a program that I have used successfully on some old Pentium 2 computers. I recently purchased a new PCI-MIO-16E-1 board, and placed it into an AthlonXP 3000+ (2.1 GHz) computer (HP a700n). It seems that I have to limit my output sample rate to be less than 5kHz, where on the old systems the rate could be up to 1MHz. I am wondering if anyone has ran into the problem before or not?

I have tried the older MIO cards in this newer computer, and came with the same problematic result (no output).
I have also tried changing the output between IRQ and DMA modes (making sure the input channel is opposite).

[sergesl]

Hello Cbaxter,

Faster processor does not necessarily convert to faster transfer rate. How many other boards (Not necessarily NI's) do you have in your system? If you have other boards, does it make a difference if you take out the other boards? Also check if running less applications on your system has an effect.

Serges L.

[cbaxter]

I think you misunderstood exactly what I was saying. I can output a signal on a SLOW computer at a much faster rate than on the new, fast computer. The only other card in the system is a modem. No other software is running, as it is a clean install of WinXP Pro, NI-DAQ 7.3.1, and CVI6.0.

We've been toying with the card in other computers, and found that the problem does not happen on a Pentium 4 machine... So we're off to get us a new P4 machine, to see if that will be a solution... It's going to be a sad day if this is an AMD vs Intel issue with these NI boards.

[sergesl]

Hello CBaxter,

I know that a fast computer is supposed to operate at a faster rate, but it is not always the case. A pentium 2 might operate at a higher rate than a pentium 3 since the speed here is the processor speed and not necessarily the speed at which data can be moved on the PCI bus.
Please keep me posted on the status after you try it on the Intel machine.

Serges L.

[Jeremy_C]

cbaxter,
Can you give some additional details of the behavior you are experiencing on the AMD computer? It sounds like if you increase the update rate beyond a certain rate, then your program executes but you don't see a signal on your AO channels. Is this correct? Are you checking for errors in your program? If so, do you get any errors? It also sounds like you are performing AO and AI in the same application. Have you tried to run a simple AO example that ships with NI-DAQ? Are you using NI-DAQmx or Traditional NI-DAQ to communicate with your hardware?

[cbaxter]

Can you give some additional details of the behavior you are experiencing on the AMD computer? It sounds like if you increase the update rate beyond a certain rate, then your program executes but you don't see a signal on your AO channels. Is this correct?

Yes, Jeremy, this is correct. It seems that if we go above a 400kHz output sample rate, the signal either doesn't make it to the output buffer, or simply never gets placed on the AO of the board.

Are you checking for errors in your program? If so, do you get any errors? Are you using NI-DAQmx or Traditional NI-DAQ to communicate with your hardware?

Yes, we are checking for errors, and we get none. Using Traditional NI-DAQ, all responses are (0) as expected.

It also sounds like you are performing AO and AI in the same application. Have you tried to run a simple AO example that ships with NI-DAQ?

I was searching for an example to try when I though about trying a Pentium PC to see if the same thing happened. As far as I know, anytime our software has worked on any computer, I need not go back to trying samples, as they are typically more complex than what we end up with. We simply make a voltage array, and shoot the array to the board (After setting output rate).

[Jeremy_C]

The reason I asked about trying an example was because those examples that ship with NI-DAQ should work, and I could try to reproduce with the same example. However, there is an even simpler option. Route your AO0 to AI0 and open up a Traditional DAQ test panel for the E-1 in MAX. Select the AO tab and generate a sine wave at various update rates, all the way up to 1MS/s. Once you've started the sine wave, you should be able to flip back to that AI tab and see your signal being generated on AI0. Can you run this simple test to verify that the output quits updating at above 400kHz? Let me know what you find out and I can try to reproduce here. Thanks.