How much bandwidth can GPIB provide? My program runs rather slow at only about ~1120 bps.

[Phredrick Dobbs]

Hey all,

My program once ran pretty responsively, but today my superior types had me add an additional 192 bits to be read per cycle from 2 instruments (96 bits each). Upon adding code to get these values, the program seems much more sluggish. It looks like it refreshes the data about 3 to 4 times a second, which comes to ~800 bps overall. (Assuming the 6 new values are 32 bit, and the old 4 all 8 bit)

Is this likely a bandwidth limitation, or some flaw in my code?

-Dobbs

[Ryan_M]

Depending on which GPIB controller you have and which instrument you are using, you can reach up to 1.5MB/s using standard 488 handshaking or 8MB/s using the HS488 protocol. I would suggest referring to your controller's specifications for more precise bentchmarking. You can find these specifications in the online catalog:
http://sine.ni.com/apps/we/nioc.vp?lang=US&pc=form&cid=1232

I might also suggest referring to your instrument's manual for information regarding its specified transer rates.

Ryan Mosley
National Instruments, Applications Engineer
http://www.ni.com/exchange

[Mark A. Hanning-Lee]

Ryan M wrote:
>
> Depending on which GPIB controller you have and which instrument you
> are using, you can reach up to 1.5MB/s using standard 488 handshaking
> or 8MB/s using the HS488 protocol. I would suggest referring to your
> controller's specifications for more precise bentchmarking. You can
> find these specifications in the online catalog:
> http://sine.ni.com/apps/we/nioc.vp?lang=US&pc=form&cid=1232
>
> I might also suggest referring to your instrument's manual for
> information regarding its specified transer rates.

E.g. in our experience, the Tektronix TDS 220 scope only transfers a
about a few KB/s! That was not exactly highlighted in the specs when we
bought ours a few years ago. They used a cheap GPIB chip.

The 3xx and other
scopes are better. See their Web site for transfer
rates.

Mark