At one time DEC left it up to the branches whether or not to maintain non-DEC parts. The switch would help by disabling the device while running diags and doing other maintenance. Nobody wanted to be liable if a non-DEC board had to be pulled.
Paul On Mon, Jul 23, 2018 at 11:57 PM, Paul Birkel via cctalk < cctalk@classiccmp.org> wrote: > -----Original Message----- > From: Paul Birkel [mailto:pbir...@gmail.com] > Sent: Monday, July 23, 2018 2:36 AM > To: General Discussion: On-Topic and Off-Topic Posts > Subject: RE: Strange third party board in PDP-11/45 > > ... > > From the ABLE marketing literature: > > CACHE/ 45 (CACHE BUFFER MEMORY) INSTALLS IN: PDP-11/45, -11/50 and -11/55 > CAPACITY: 2048 byte (1 K word). > ENHANCEMENT FACTOR: Run time reductions to 50% (100% speed improvement) are > achievable. > CACHE PARITY: Automatically goes off-line in event of any data error. > RANGE SELECTION: User may optimize hit ratio by upper/lower limit switch > settings. > SPECIAL FEATURE: Cache/ 45 can be enabled via software or console switches. > > Presumably that's what Mattis has in-hand. > > ----- > > https://books.google.com/books?id=hYD3sny2NzEC&pg=PA23& > lpg=PA23&dq=ABLE+Comp > uter+technology+ACT+DEC+%22CACHE/45%22&source=bl&ots= > b15iACJbMd&sig=oLMrJMn2 > qEFmxiKSMIhpLF5qYnk&hl=en&sa=X&ved=2ahUKEwjcyvzi87TcAhVOw1kKHdc5A > 8kQ6AEwAHoE > CAIQAQ#v=onepage&q=ABLE%20Computer%20technology%20ACT% > 20DEC%20%22CACHE%2F45% > 22&f=false > > Computerworld, Page 23, July 26, 1976 > > ACT Has PDP-11/45 Buffer > SANTA ANA, Calif. - Able Computer Technology (ACT) has a 2K-byte cache > buffer for use with the Digital Equipment Corp. POP-11/45. > The Cache/45 is contained on a single printed circuit board that plugs into > the > system's chassis. > Buffer control is provided over every core memory address location on the > Unibus. A switch within the memory buffer permits a choice of either > on-line > or off-line operation, the company said. > The buffer costs $7 ,000. ACT is at I 538-E Chestnut St., Santa Ana, Calif. > 92705. > > (Somewhat strange use of the expressions "on-line" and "off-line" > operation, > IMO.) > > I presume that the reason for "single printed circuit board" is that only a > single "normal" slot is occupied. > > paul > >