[Info-vax] DZQ11 on VAX/4000 Model 300

[Stephen Hoffman]

On 2019-01-15 23:18:00 +0000, David Wade said:

> On 15/01/2019 14:33, John E. Malmberg wrote:
>> On 1/15/2019 1:06 AM, Steven Schweda wrote:
>>> 
>>> http://bitsavers.trailing-edge.com/pdf/dec/vax/630/AX-GLBAA-MN_MicroVAX_II_630QB_Technical_Manual_Nov85.pdf 
>>> 
> 
> Does that apply to the BA400?

The Q22-class Q-bus itself doesn't work particularly differently across 
boxes, though different boxes do have different patterns of Q22/CD, 
Q22/Q22 and memory slots.
This is the serpentine stuff.
Know your box.

> But the cards go TU811,QZA,DZ11. Does this matter?

The only thing that matters is that ALL of the devices are included in 
the configuration.
The interrupt-happy devices and particularly the interrupt-happy 
terminal controllers are at the back of the bus, and the 
higher-priority devices at the front.
It was common for folks to omit a few devices from their configs, and 
that too generates a bogus configuration.
KZQSA presents a storage controller for each storage device present, IIRC.
KZQSA worked with specific CD-ROM devices, and specific tape devices.
KZQSA wasn't a particularly generic SCSI controller, though it did work 
for the devices that it was supported for.

> If order matters then the cards go TU81, QZA and DZ11. When I put them 
> into CONFIGURE on the vax I get different answers.

Put the DZ at the back of the bus.  Put the network at the front of the 
bus.  Otherwise and at the likely I/O loads here, the rest matters less.

> I was trying to avoid another box.

LAT was the late-term usual approach, as the serial controllers and 
particularly the DZ-class stuff was interrupt happy.
DEC was trying to get out of the morass of serial wiring and related.  
See the DECconnect information for details on that.
Most VAX console ports were poor choices for anything other than 
console usage.  Not a good idea to connect printers to console ports.


-- 
Pure Personal Opinion | HoffmanLabs LLC