Sender: eric@ruckus.brouhaha.com From: Eric Smith Newsgroups: alt.sys.pdp10 Subject: Massbus (was Re: I have a KS10 now! ^_^) References: <8vtt5l$gd1$6@bob.news.rcn.net> X-Disclaimer: Everything I write is false. Organization: Eric Conspiracy Secret Labs X-Eric-Conspiracy: There is no conspiracy. Date: 28 Nov 2000 11:31:38 -0800 Message-ID: Lines: 38 X-Newsreader: Gnus v5.7/Emacs 20.7 NNTP-Posting-Host: ruckus.brouhaha.com X-Trace: 28 Nov 2000 11:32:55 -0800, ruckus.brouhaha.com Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.nextra.ch!news1.sunrise.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!nntp.primenet.com!nntp.gblx.net!enews.sgi.com!news.sgi.com!news.spies.com!ruckus.brouhaha.com Xref: chonsp.franklin.ch alt.sys.pdp10:1362 jmfbahciv@aol.com writes: > You've got -11s and no Massbus stuff? Why do you need a Massbus > for a KS (or have I really got my hardware jargon brain farted all > over the place?). You're probably thinking of Unibus, which is the PDP-11 bus. Massbus was the host-independent disk/tape interface in common use on high-end DEC gear from 1973 (?) to the mid 1980s. It was later replaced by other interconnects such as CI, SDI, STI, DSSI, and SCSI. Massbus has an 18-bit data path, a 16-bit control/status path, and a bunch of handshaking and control signals. There are a total of 60 differential pairs to a Massbus. Internal Massbus cabling uses three 40-conductor flat cables with the ubiquitous 40-bit Berg sockets. External Massbus cabling uses twisted pairs, with connectors consisting of three of the Berg sockets in parallel in a connector shell. The data path and control/status path can be used simultaneously; once a data transfer is set up, the the host can use the control/status path to monitor the drive, seek on other drives, or prepare another drive for a data transfer. Note that although there is no electrical problem with putting disks and tapes on the same Massbus, in practice none of the host software supports it. So most systems have at least two Massbus channels. The KA10 and KI10 use an external RH10 Massbus channel. The KL10 can use either the RH10 or the internal RH20 Massbus channel. The KS10 uses a slightly modified form of the RH11 Massbus adapter as used on the PDP-11 Unibus. VAXen use the RH780 and RH750 Massbus adapters. Commonly used Massbus drives include the RP04, RP05, RP06, RP07, RM03, RM05, and RS04 disks, and the TE16, TU45, TU77, TU78, and TU79 tapes. ###### Path: chonsp.franklin.ch!not-for-mail From: Neil Franklin Newsgroups: alt.sys.pdp10 Subject: Re: Massbus (was Re: I have a KS10 now! ^_^) Date: 29 Nov 2000 01:32:41 +0100 Organization: My own Private Self Lines: 36 Message-ID: <6un1ejhb2u.fsf@chonsp.franklin.ch> References: <8vtt5l$gd1$6@bob.news.rcn.net> NNTP-Posting-Host: chonsp.franklin.ch X-Trace: chonsp.franklin.ch 975457961 969 10.0.3.2 (29 Nov 2000 00:32:41 GMT) X-Complaints-To: news@chonsp.franklin.ch NNTP-Posting-Date: 29 Nov 2000 00:32:41 GMT X-Newsreader: Gnus v5.7/Emacs 20.4 Xref: chonsp.franklin.ch alt.sys.pdp10:1369 Eric Smith writes: > The KA10 and KI10 use an external RH10 Massbus channel. > > The KL10 can use either the RH10 or the internal RH20 Massbus channel. > > The KS10 uses a slightly modified form of the RH11 Massbus adapter > as used on the PDP-11 Unibus. Now that is interesting. In the PDP-11 FAQ there is this: --------- 11/70. [snip] Note that MASSBUS (RH70's) adapters attached directly to memory were unique to the 11/70 -- MBAs only otherwise appeared on different series (PDP-10, VAX etc). Unibus systems could use MASSBUS peripherals, but the controllers (RH11's) spoke over the Unibus and lost most of the speed advantage. While this machine was not the fastest -11 (that title belongs to the bipolar memory equipped 11/55), no other -11 could approach the 11/70 in terms of I/O throughput over its multiple massbus/memory bus architecture. ------- So they say that Unibus Massbus adapters were inferior in an PDP11. Now you write that the KS10 switched to using them. Did they do something to them to make them faster? Or did the KS10 take an disk throughput speed hit for this? -- Neil Franklin, neil@franklin.ch.remove http://neil.franklin.ch/ Nerd, Geek, Hacker, Unix Guru, Sysadmin, Roleplayer, LARPer, Mystic ###### Newsgroups: alt.sys.pdp10 From: Daniel Seagraves Subject: Re: Massbus (was Re: I have a KS10 now! ^_^) In-Reply-To: <6un1ejhb2u.fsf@chonsp.franklin.ch> Message-ID: References: <8vtt5l$gd1$6@bob.news.rcn.net> <6un1ejhb2u.fsf@chonsp.franklin.ch> Approved: Why bother? MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Lines: 24 Date: Tue, 28 Nov 2000 19:00:46 -0600 NNTP-Posting-Host: 198.199.189.6 X-Trace: newsfeed.slurp.net 975458732 198.199.189.6 (Tue, 28 Nov 2000 18:45:32 CDT) NNTP-Posting-Date: Tue, 28 Nov 2000 18:45:32 CDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.nextra.ch!news1.sunrise.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.slurp.net!bony.umtec.com!root Xref: chonsp.franklin.ch alt.sys.pdp10:1382 On 29 Nov 2000, Neil Franklin wrote: > So they say that Unibus Massbus adapters were inferior in an PDP11. Not surprised. > Now you write that the KS10 switched to using them. Did they do > something to them to make them faster? Or did the KS10 take an disk > throughput speed hit for this? I doubt it. The disk RH11 is in the KS logic cardcage, so it might be special, but the tape RH11 is in the UNIBUS box, which is a standard PDP-11 type Unibus (with the exception that it uses the parity bits as data bits). "Confuse, annoy, and DEE-STROY!" -- Jet Wolf | "Nothing Happens." -- ADVENT "You'd be surprised what you can live through..." -- Anonymous "...A man can pass his family and his name down through his sons, but it's his honour that gets passed through his daughters. He can see the best and worst of life in his girls. A daughter is something far too precious, and he'll do anything to protect her." -- Reichsfuehrer Siegfried Koenig, _Matrose_Mond_, David Oliver ###### Message-ID: <3A246885.1B478B90@bellatlantic.net> From: hg/jb Reply-To: shsrms@bellatlantic.net X-Mailer: Mozilla 4.75 [en]C-CCK-MCD BA45DSL (WinNT; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.sys.pdp10 Subject: Re: Massbus (was Re: I have a KS10 now! ^_^) References: <8vtt5l$gd1$6@bob.news.rcn.net> <6un1ejhb2u.fsf@chonsp.franklin.ch> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 50 Date: Wed, 29 Nov 2000 02:16:25 GMT NNTP-Posting-Host: 138.88.37.194 X-Complaints-To: newsadmin@bellatlantic.net X-Trace: typhoon2.ba-dsg.net 975464185 138.88.37.194 (Tue, 28 Nov 2000 21:16:25 EST) NNTP-Posting-Date: Tue, 28 Nov 2000 21:16:25 EST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!newsfeed.skycache.com!Cidera!cyclone1.ba-dsg.net!typhoon2.ba-dsg.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:1377 Neil Franklin wrote: > > Eric Smith writes: > > > The KA10 and KI10 use an external RH10 Massbus channel. > > > > The KL10 can use either the RH10 or the internal RH20 Massbus channel. > > > > The KS10 uses a slightly modified form of the RH11 Massbus adapter > > as used on the PDP-11 Unibus. > > Now that is interesting. In the PDP-11 FAQ there is this: > > --------- > 11/70. > > [snip] > > Note that MASSBUS (RH70's) adapters attached directly to memory were > unique to the 11/70 -- MBAs only otherwise appeared on different > series (PDP-10, VAX etc). Unibus systems could use MASSBUS > peripherals, but the controllers (RH11's) spoke over the Unibus and > lost most of the speed advantage. While this machine was not the > fastest -11 (that title belongs to the bipolar memory equipped 11/55), > no other -11 could approach the 11/70 in terms of I/O throughput over > its multiple massbus/memory bus architecture. The 11/70 memory box was a little different. I don't recall who designed it but it was related to the 45/50/55 memory box BUT capitalized on changes in technology between the two machines and I think I recall a difference in the cache.. Of the two machines (45/50/55 and the 70), the 70 Mbox and cache were tuned to allow more balanced I/O - meaning DMA or NPR xfrs to or from memory were more sustainable on the 70. bob > ------- > > So they say that Unibus Massbus adapters were inferior in an PDP11. > > Now you write that the KS10 switched to using them. Did they do > something to them to make them faster? Or did the KS10 take an disk > throughput speed hit for this? > > -- > Neil Franklin, neil@franklin.ch.remove http://neil.franklin.ch/ > Nerd, Geek, Hacker, Unix Guru, Sysadmin, Roleplayer, LARPer, Mystic ###### From: wilson@dbit.com (John Wilson) Newsgroups: alt.sys.pdp10 Subject: Re: Massbus (was Re: I have a KS10 now! ^_^) Date: 28 Nov 2000 23:02:58 -0500 Organization: D Bit, Troy, NY Lines: 24 Message-ID: <901v5i$agv$1@dbit.dbit.com> References: <8vtt5l$gd1$6@bob.news.rcn.net> <6un1ejhb2u.fsf@chonsp.franklin.ch> NNTP-Posting-Host: dbit.dbit.com X-Original-NNTP-Posting-Host: dbit.dbit.com X-Trace: 28 Nov 2000 23:03:52 -0500, dbit.dbit.com XPident: wilson X-Original-NNTP-Posting-Host: 199.181.141.3 XPident: Unknown Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.nextra.ch!news1.sunrise.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.mathworks.com!news.mv.net!newspeer.phoen-x.net!news.wizvax.net!dbit.com!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:1370 In article <6un1ejhb2u.fsf@chonsp.franklin.ch>, Neil Franklin wrote: >Now you write that the KS10 switched to using them. Did they do >something to them to make them faster? Or did the KS10 take an disk >throughput speed hit for this? Yes and yes. The RH11C is jumpered as a "bus hog" with some ridiculous burst length, which is the only way it could keep up with the RM03 and RM80 (which are officially not supported on the RH11AB since there aren't enough 3341s to buffer a whole sector, or something like that). The "disk" RH11C has to be the only thing on the "disk" Unibus to pull this off. But it's still pretty weak, it would have been much cleaner if DEC had built an "RH2020" or whatever, i.e. an MBA which sits directly on the KS10 backplane bus, so it could have been faster, plus it wouldn't have had to get mired in the Unibus map ugliness. But obviously DEC didn't feel it was worth the trouble for a low-end system. So you get a pretty long chain for e.g. RM80 disk I/O on a KS: backplane bus => UBA => RH11C => RM adapter => SMD personality module => R80 Each step of the way probably burns another 50 watts, too ... John Wilson D Bit ###### From: inwap@best.com (Joe Smith) Newsgroups: alt.sys.pdp10 Subject: Re: Massbus (was Re: I have a KS10 now! ^_^) Date: 4 Dec 2000 19:44:50 GMT Organization: Chez Inwap Lines: 29 Message-ID: <90gs7i$2jml$1@nntp1.ba.best.com> References: <6un1ejhb2u.fsf@chonsp.franklin.ch> <901v5i$agv$1@dbit.dbit.com> NNTP-Posting-Host: shell3.ba.best.com X-Trace: nntp1.ba.best.com 975959090 85717 206.184.139.134 (4 Dec 2000 19:44:50 GMT) X-Complaints-To: abuse@best.com NNTP-Posting-Date: 4 Dec 2000 19:44:50 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.nextra.ch!news1.sunrise.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!news1.best.com!nntp1.ba.best.com!inwap Xref: chonsp.franklin.ch alt.sys.pdp10:1451 In article <901v5i$agv$1@dbit.dbit.com>, John Wilson wrote: >In article <6un1ejhb2u.fsf@chonsp.franklin.ch>, >Neil Franklin wrote: >>Now you write that the KS10 switched to using them. Did they do >>something to them to make them faster? Or did the KS10 take an disk >>throughput speed hit for this? > >Yes and yes. The RH11C is jumpered as a "bus hog" with some ridiculous burst >length, which is the only way it could keep up with the RM03 and RM80 Yep. That was fine for handling a single I/O request: consecutive disk sectors going to a single chunk of contiguous memory. But for buffered I/O, in which the actual buffers were separated by three-word buffer headers, a single I/O request could only do one block at a time. And in the amount of time it took the KS to respond to the interrupt and initiation a new I/O request for the next block, the disk rotated enough so that it just missed the next sector on the track. In other words, one block per revolution (60 blocks per second) when using buffered I/O on the KS. That is one of the reasons for "SET BLOCKSIZE 4" to declare the default buffer size to be 4 blocks for your particular job. I don't know if anyone experimented with changing the disk formatter to use interleaved sectors instead of consecutive sectors when formatting the disk track. -Joe -- See http://www.inwap.com/ for PDP-10 and "ReBoot" pages.