Prologix wrote: > > One solution is to turn read-after-write off (++auto 0) before connecting > the adapter to 3478A, and then use ++read command to read one measurement at > a time. Please see the manual (www.prologix.biz) for ++read command options. > I nearly always turn off auto mode on the prologix. I've run into a number of slightly flaky problems when the prologix guesses to turn on the instrument to talk. I've never noticed a performance penalty because I need to issue a "++read" command to the prologix whenever I want to read something back. I think the prologix device is great. I'm a linux guy, I haven't used windows in many years. NI has never really understood open source software. It was always a pain to have a reliable driver for the NI GPIB PCI cards, so I was quite happy to find something like the Prologix device that is reliable and has drivers that are totally open source and distributed as part of the linux kernel. I have a slightly off-topic question about reads. When issuing a "++read eoi", is it possible to tell whether an EOI was actually returned by the instrument or whether the read was terminated by timeout or block size limitation? jeff

Hello Jeff, Thanks for the compliments. Much appreciated. You may configure the Prologix adapter (using ++eot_enable and ++eot_char) to send (append, really) a user-specified character to USB ouput when it detects EOI. By checking for the character you can determine if EOI was asserted. Regards, Abdul -----Original Message----- From: time-nuts-bounces@febo.com [mailto:time-nuts-bounces@febo.com] On Behalf Of Jeff Mock I have a slightly off-topic question about reads. When issuing a "++read eoi", is it possible to tell whether an EOI was actually returned by the instrument or whether the read was terminated by timeout or block size limitation? jeff