Max throughput of FBUS PCI

[LeifS]

In what way do I get the best data rate reading tags from a couple of devices on the same H1 link, using the NI FBUS PCI card?

Today I've tried using the FFBus VIs an a simple sequential read of all preconfigured tags (approx 10 tags).

The devices is not configured using FBUS configurator or similar, just standard factory preloaded blocks (if this is important).

Is the FFBus VIs the best way or is the integrated OPC server faster, and if so, why is that?

Would I get any faster speed with multithreaded communication, e.g. a thread per tag, or does it get worse?

/Leif

[Vince Shen]

Hi,

The rate of FFLV and OPC should be similar. The fastest way is using C API which has less inter-process communication load.

By the way, what is your acceptable rate? Could you give me a rough number?

[LeifS]

Vince Shen wrote: 

By the way, what is your acceptable rate? Could you give me a rough number?

---

Well, today my update rate per tag is somewhere around 1-2 sec.

This is when I read them as mentioned above.

I'd like to get down to perhaps 0.1s for all ~10 tags. (Is that possible on a single H1 link or is it too much overhead in the communication protocol?)

Maybe the PCI-board or the devices need some adjustments?

They are all used with mostly factory default settings. (The few settings altered didn't change the data rate at all anyway but seemd to improve other things)

/Leif