Maximum rate for NI PCI 6259 different than NI device specifications

[RobinGeorgeKoshy]

The documented maximum sample rate of NI PCI 6259 is 2.86M samples/s but DAQmxGetSampClkMaxRate returns 1M samples/s.

On setting the sample rate to 2.86M samples/s via DAQmxCfgSampClkTiming throws warning

200040: Warning: Sample clock rate specified is so high that it violates the settling time requirements for the generation.
Reduce the sample clock rate, or the accuracy of the generated signal might be compromised.

Attached: Reproduction code in Source.zip

Specification Sheet: https://www.ni.com/pdf/manuals/375216c.pdf

[Kevin_Price]

Just to add another data point.

I saw the same thing here with LabVIEW 18 and DAQmx 18.6.   A timing property node reported SampClk.MaxRate as 1.0 MHz and the task gave the same 200040 warning when I tried to set the rate higher anyway.

FWIW, when I used a *device* property node rather than a "timing" property node to query for AO.MaxRate, it output the documented 2.86 MHz.   (Rounded -- technically a timing property node reported a timebase divisor of 7, so the task apparently derives its sample clock from the internal 20 MHz timebase.)

I further explored what I got from an AI task.  That worked correctly, giving results expected from the spec sheet.  A 1-channel task reported SampClk.MaxRate as 1.25 MHz while a 2-channel tax reported 500 kHz (corresponding to the 1 MHz *aggregate* in the specs.)   The device property node had separate properties to query for single channel and for multi-channel.  Single channel again gave 1.25 MHz while multi channel reported the *aggregate* rate of 1.0 MHz.  The online help for the property makes it clear that it's *meant to* report the aggregate rate, so no real discrepancy there.

Looks to me like a bug or a DAQmx database mistake.  During task config, it must consult the "timing node" property to compare the requested rate to (what it thinks is) the device's max rate and then issues the observed warning.

What DAQmx version do you have on your end?

-Kevin P

[RobinGeorgeKoshy]

Thanks for the reply Kevin.

I tried with NI DAQmx 20.0.Also, NI PCI 6363 has same spec and it works fine with timing info.

[RobinGeorgeKoshy]

How can you ever sample at 2.86 MHz from PCI 6259 for AO? 

[Kevin_Price]

I'd say give it a try.  I think there's a good chance this is a *benign* bug in that DAQmx may (mistakenly) issue a warning, but will nevertheless configure the sample rate you ask for.  When I request a 2.86 MHz sample rate, yes there's a warning, but if I then query a timing property node for the actual sample rate, it tells me 2.86 MHz.   (At least on a simulated device.)

-Kevin P

[RobinGeorgeKoshy]

Yes you are right. Thanks for the insight.