Multifunction DAQ
what is the bottleneck (error 200279)
I have been having this problem recently since the IT department decided to "upgrade" my computer from a 2.8 Ghz to a dual 550Mhz (against my recomendations
). I am running the m series card pci6221 acquiring 7 analog and 1 encoder. The VI will progressively fill up the buffer no matter what i do, change buffer size, read all avail, read more often etc. However if I exclude the encoder the VI will run a maximum of 2000 hz. If this error is produced after the adc conversion, then the PCI bus seems like the culprit but i do not believe it. My processor, even though a dinosoar, is not taxed 6-10%, i also have 1Gb ram. How can this be?
). I am running the m series card pci6221 acquiring 7 analog and 1 encoder. The VI will progressively fill up the buffer no matter what i do, change buffer size, read all avail, read more often etc. However if I exclude the encoder the VI will run a maximum of 2000 hz. If this error is produced after the adc conversion, then the PCI bus seems like the culprit but i do not believe it. My processor, even though a dinosoar, is not taxed 6-10%, i also have 1Gb ram. How can this be? Windows XP, Latest nidaq, sc2345 carrier
Personally i believe the dinosoar computer to be the problem, but i need evidence. And if i have 2 or 3 DMA channels, why would the encoder have any effect since it would go through a seperate conversion on the daq card independant of analog.
BTW i did read previous boards and did not see an answer that suited me. my application is a state machine when left alone just reads data and in this state it still hangs up.
I appreciate any help or tips on the subject.
Chris
Abhinav Thank you for the reply.
Yes, the available samples per channel never stops increasing. I have even tried to set the #of samples equal to the sampling rate at various rates (500Hz-1000Hz) which would mean the read function will not be called until it has received the number of samples (1 second). Good for the processor bad for a program. I timed the program at 312ms which i thought was slow, but it would also mean that i should be able to run the program at around 1000 Hz and 300 samples. Also I thought that when the buffer was filling up that it was because the computer cannot catch up to the acquisition. But this implies that I would see the processor begging for mercy which it is not, according to the task manager. Maybe my understanding is wrong- please tell me if it is.
I wish i had the VI analyser add on. Not that it would help me but I would feel better.
Could a processor in a dual system be "hurting" or "running on E" and not show it. Could it be an SCC carrier limitation or a DAQ Limitation. I do not know about the carrier, but the DAQ is the one doing its job filling up the buffer right? GOT to be a processor. What else does that leave me with, Buss speed. Maybe video card issues. Follow me on this for a second. Maybe the screen update with the graph is too much for the video to process through. Now i have no idea how video works with no card, but i thought it was feed video ram on the motherboard and processed by a separate chip on the board, in which case the task manager would not detect a processor lag because it is lagging on video???? Man do I sound madd or what. Funny farm here I come!
Chris
