two issues. first DEC subsetted the modem control lines so running modems - particularly when you wanted hardware flow control like the trailblazers - did not work. second as others pointed out there the buffering was mostly lacking and the interrupt load was terrible on the OS if DEC had used different and better USARTs this would not have been as bad. When it came out there were other options for the HW guys but IMO it was an example of tunnel vision - plus HW guys not really understanding the SW implications of the choice. one of the guys behind the DZ would later become a good friend of mine. I realized Dave never looked at RS-232 the same way SW people did. To Dave the DZ was great because it was two boards to do what he thought was the same thing as a DH - ie cheaper / less power / smaller etc. But the saving was a HW one and not what we needed. that said DEC sold a lot of them on VMS systems. it was us UNIX guys that switched to Able's board as soon as we could. Cheaper and better functionality for the OS. ken once told me he always loved us Unix guys because we got it. he also listened too us and built what we needed when he could. I tried to get him to put the DH/DM on a AT form factor but he said he could not make money at it. Eventually the rocket port guys built and ASIC and pretty much did it Clem

On Fri, Nov 28, 2014 at 06:45:38AM +1100, Dave Horsfall wrote: This problem is old as the hills, at SGI we did similar stuff for networking cards. We could run something like 8 HIPPI cards at full bandwidth with some 200mhz MIPS cpus. Lots of work in the networking stack to go get the next packet if it was there after we did this one.

On Sun, 30 Nov 2014, Warren Toomey wrote: That's the one! Bless you :-) It'll take a while to work through it again, and I can see that I was wrong about the use of the buffer pool (our printer drivers certainly used them, along with the Calcomp and the Versatec). but I can see that it does funky things otherwise. -- Dave Horsfall DTM (VK2KFU) "Bliss is a MacBook with a FreeBSD server." http://www.horsfall.org/spam.html (and check the home page whilst you're there)

On Sun, 30 Nov 2014, Warren Toomey wrote: Hadn't realised that there were two different *fast* versions, but I do recall using the KW-11P gave blazing performance. Well, when you had the 11/70 loaded with *seven* DZ-11s, something had to be done... Warren, would the old CSU40/CSU60 stuff be around? I pretty much ran that place until I left; a 9-track tape followed me home, but even if I knew where it was now it ain't gonna be readable after over 30 years... If I could figure out how to search by file name, I'd look for, oh, "ei.c" (the UT200 driver that I wrote), "lv.c" (Versatec LV-11 driver), and "xy.c" (which would be the Calcomp plotter). -- Dave Horsfall DTM (VK2KFU) "Bliss is a MacBook with a FreeBSD server." http://www.horsfall.org/spam.html (and check the home page whilst you're there)