Thanks for the quick reply!
The exact error message I am getting is: Error 42 occurred at RLP Invoke Node, Possible reason(s):
The DMA buffer overflowed because data was not read from the buffer as fast as the DMA channel wrote to the buffer.
I tried increasing the input buffer first as per your first suggestion and that made it possible to run at 2 kHz with out getting this error and to run at 5 kHz for about 10 seconds. This is about twice as good as before. I increased the number of samples in the buffer untill I got an error saying the buffer was to large(30,000). I hope this is what you intended.
I also added a "case structure" to the producer loop that flushes the queue if it has more then 3 values in it. I did this because I hoped it would keep the data read by the consumer loop reasonably fresh and it would keep the queue from growing to large. I don't know if it was necessary or if it makes the problem worse but it didn't seem like it would cost to much performance.
Next I changed the dmx base read.vi setting to taking multiple samples in multiple channels as per your last suggestion. Right now I am running 5000 samples at 20 kHz and it has ran apparently without erroring(I stopped it to run other tests). I ran it at 1000 samples at 20 kHz and it errored after about a minute and at 100 samples at 20 kHz it errored more quickly (15 seconds about... these are just guesses I don't have a stop watch with me and I am doing other things at the same time so my sense of time could be off). I just tried 5000 samples at 80 kHz and it errors after about a minute or two.
It seems to work pretty well so thank you! I havn't written the data analysis part of the program yet so that might slow things down but at least it is running on the order of what the final product has to run at. However, if you think it should be running faster I would be happy to hear any additional suggestions.
Thank You Again,
Craig
