Copied back to the main list?. OK on the L66 machines I worked on we always kept the doors closed, there was an application, can?t remember its name, we ran on a VDU by the system console that displayed the Job Queues, State of Active Jobs, CPU utilization etc. There was also a MIPS meter on the console. They were very modular, and as the store was in the SCU?s on a multi-cpu system you also needed cross coupled cache cables to invalidate the CPU cache when the other CPU overwrote a word in main sore that it also had cached. Interleave was also interesting, so you could configure interleave memory word by word so alternate accesses went to the ?other? SCU. A guy we had in at NERC Bidston (was www.pol.ac.uk <http://www.pol.ac.uk> ), Vince Martin I think his name was who had worked in the Honeywell performance labs, I believe in Scottsville. Arizona said this would give much better performance. He also said the big bottle neck on the L66 was memory bandwidth. With the fast 6250 BPI tapes he said the tape drives could drive the memory flat out, locking the CPU out?. Dave From: Charles Anthony [mailto:charles.unix.pro at gmail.com] Sent: 12 March 2016 15:27 To: Dave G4UGM <dave.g4ugm at gmail.com> Cc: Ed Sharpe <COURYHOUSE at aol.com>; jws at jwsss.com; spacewar at gmail.com; jwsmail at jwsss.com; General Discussion: On-Topic and Off-Topic Posts <cctalk at classiccmp.org>; Kevin Monceaux <Kevin at rawfeddogs.net>; healyzh at aracnet.com; couryhouse.smecc at gmail.com Subject: Re: Honneywell multics? from panels. the inline phots in this message folks -smecc On Sat, Mar 12, 2016 at 6:44 AM, Dave G4UGM <dave.g4ugm at gmail.com <mailto:dave.g4ugm at gmail.com> > wrote: The panels would be pretty much un-used Unlike 360 panels these were hidden behind doors for most of the time. Assuming the work the same on a Multics box as on a regular L66/DPS box the only time they were really used was if you split a 2 x CPU system into 2 x 1 CPU system, or changed the memory configuration from interleaved to non-interleaved. Pretty sure you could IPL from the console. [My understanding; I wasn't there] The display panels changed over time; later models had a minicomputer with management software instead of the big panels. Early models: The CPU cabinet had two doors, with the panels on the inside of the door, so that opening the door swung the panel out. (There were probably additional panels exposed, but I'm not sure.) One the panels was the CPU status, showing the contents of the registers (A, Q, PPR, TSR, Xn, PARn), the major CPU states (fetch, execute, interrupt, fault), and major status bits (Append Unit, Operation Unit, Decimal Unit). I understand it was common practice to leave that door open for it visual appeal and highly visible 'things are running' status (including the 'idle crawl' a distinctive pattern display when Multics was idle. The panels labeled SCU are System Control Units; they were separate cabinets. The SCU's contained the memory and handled all communication between CPUs and between CPUs and IO controllers. The panels labeled IOM are Input Output Managers; they connected the SCUs to peripheral devices; also sometimes 'IOP' (Input Output Processor). The enormous number of configuration switches is due to the extreme modularity of the system. Core memory was in the SCUs, not the CPUs; each SCU could have a varying amount of memory, in up to 4 banks of varying sizes. Each bank could taken out of service, so defining the addressing of the memory in each SCU was complex. Each CPU had to have a matching memory configuration panel so that it knew which SCU to talk to access a particular memory location. Each CPU had to be cabled to each SCU. Since the IOMs did DMA, the memory configuration panels are duplicated there as well, and each IOM cabled to each SCU. The only display panel I would want to utilize is the CPU status; the other panels display information which the emulator does not necessarily have and do not have the 'wow' factor. -- Charles