We have an application that reads 24 bits of latched "event" information from a custom board, using a PC-DIO-96/PnP card. The application is written with LabWindows/CVI. Currently we are doing true latched I/O, meaning the LabWindows source code checks to see if data is latched, and then reads it.
The problem is that our custom hardware is MUCH faster at creating these 24-bit latched "events" than the CPU is at reading them. It's ok that we lose events, but it would be nice to get them faster. The obvious solution is DMA acquisition to take the CPU out of the loop.
So the question is, can we acquire 24 bits of information from this card using DMA? I realise that 3 ports is not a nice number
, so we could easily acquire 4 ports of information and just ignore the extra 8 bits.
I can't get a clear idea from the documentation whether or not this is possible from this particular board.
