Regeneration to an offset

[Steve_Block]

I would like to use code which is similar to Digital - Continuous Output.vi from the examples to run a repeating pattern, but to start at an offset the second time around with the full pattern running the first time. I want to do this completely with hardware timing(not sw control) so there are no glitches as the transition. Is it possible to do this?

[Minions]

You can use one of the numerous FPGA based NI devices for hardware timed outputs.

[Steve_Block]

I see that I can create an "Every N Samples" transferred from buffer and it allows me to write additional samples(I'm not yet sure if that is erasing the existing samples). I have not been able to figure out how to acquire the buffer size so I can continuously feed samples into it and have it act like a FIFO. Do any such examples exist?

[Minions]

The FPGA hardware does have the capability to create FIFO's and other memory devices.  Start here to get a better understanding of how the data transfers between hardware and PC work. http://zone.ni.com/reference/en-XX/help/371599N-01/TOC16.htm

[Steve_Block]

I'm not using Labview FPGA. I'm using a 9401 in a 9178 chassis with DaqMx.

[Minions]

Sounded like before that you were asking how to do something more in the planning phase.  This is the first point that you mentioned any hardware.

[Minions]

The NI cDAQ-9178 talks about a 127 sample FIFO per slot.  While the 9401 can sample one input every 100ns Max.  What are your sampling frequencies and number of inputs being sampled?  Can you guarantee that you will not overrun the buffer with those plans?

[Steve_Block]

I really hope 127 samples is not all the device has. Did you mean 127k or 127M? You can never guarantee you don't overrun a buffer, however, you can monitor it. What I would like to know regardless of my sampling frequency is how to properly load and monitor it so I can control how fast I load data into it. Do any examples exist?

[Steve_Block]

Looks like the 2047 might not be enough for this application but potentially others if such an example exists. Is there a way to configure the regeneration to jump back to a sample other than the first sample?

From: http://www.ni.com/pdf/manuals/374046a.pdf

Waveform generation (DO) FIFO

.S..l.o..t.s.. .1.. .t.o.. .4...................................................2,047 samples

.S..l.o..t.s.. .5.. .t.o.. .8...................................................1,023 samples