True, but the range and station count wasn't our primary concern as we viewed it as a kind of extended range bus.

It was a big deal to us because of the RH slot that got freed up; it you had a ÇI and an NI, that was three out of RH eight slots before you thought about other peripherals.  There was also the cost savings involved with not having to buy an NI plus the memory saved by not having the NI drivers in the monitor.

I don't remember how much we cared about the money.  I mean, as compared to what everything else cost.   What we really cared about was that extra RH slot and the memory.  Particularly the memory, as part of it was dedicated resident/non-swappable.  Even with 3.75KW, we just never had enough memory for users.  We were always looking around how to save memory.

That's one thing I like about having a hobbyist machine; I don't have to care about memory, so some of my solutions are easier and faster because I don't get hung up about the space/time tradeoff.

For example, when Kermit-20 has to put parity on a packet (which it typically doesn't), it had to call a routine to compute the parity for each and every single character.  There is a discussion in SYSREF about how to compute parity and what the tradeoffs were.  They're interesting, particularly when learning about the PDP-10's ISA--in short, some fine code.  I used none of them and replaced code with a halfword lookup table and a single EXTEND instruction, MOVST.

Boom.  I cut the processing time down by 1/3 and I didn't even have to do anything nearly as clever as SYSREF.

I have run out of swap space, which appears to be limited to 20,000 pages.  Don't do that.  You'll pretty much hang solid.  I have looked at the issue and (sort of) remember that the current limit is 20,000, but I think it could be expanded to 30,000 pages.

I wouldn't swear to it, but when we were a field test for 6.x, there was some internal documentation on DECnet over the CI.  I think there is still some kind of write up on it, somewhere.
On 5/17/24 11:19 AM, Paul Koning wrote:
It seems quite reasonable except for the fact that CI has such a limited range and station count.  It's just that it wasn't specified in any DNA spec.

	paul