From: l3q90@jupiter.sun.csd.unb.ca (Warnica) Newsgroups: alt.folklore.computers Subject: HCF (was: Re: history of computer input devices) Date: 16 Dec 1997 15:55:58 GMT Organization: University of New Brunswick, Fredericton, NB, Canada Lines: 29 Message-ID: <6768ae$mev@sol.sun.csd.unb.ca> References: <01bd06e3$e0e5eeb0$d6119582@sirius> <34915C5E.69AD@azstarnet.com> <349437DC.19B9@goldrush.com> <673oli$569@sol.sun.csd.unb.ca> <6742c2$rg0$1@nntp2.ba.best.com> NNTP-Posting-Host: sol-alt1.unb.ca Aproved: god@hell.org (root) X-Newsreader: TIN [version 1.2 PL2] Path: ccw.ch!aetna.dolphins.ch!news.planetc.com!leto.ou.edu!hammer.uoregon.edu!vixen.cso.uiuc.edu!logbridge.uoregon.edu!newsfeed.direct.ca!torn!garnet.nbnet.nb.ca!news.unb.ca!sol!l3q90 George Byrd (george@apan.org.NO_UCE) wrote: : Speaking of "Re: history of computer input devices" : in on 15 Dec 1997 17:16:34 GMT, : said: : >I guess the two (gun & bullet) together would be a potential universial : >HCF instruction :) [SNIP] : Anybody have other entry points for the instruction? The Jargon File is unusualy non-helpful on this one - it points out that the IBM 360 has one, but not what it was, or what it did. Thr MC6800 had a HCF that caused lines to togglr sdtate quickly, potentialy causing the solder traces to burn up. (Im dating myself, but this was slightly before my time) I do know on at least one C= PET model, you could POKE something to some adderess and a couple of IO ports (described to me this way) would fight over some lines, burning our the ports. (and things attached to them?) Along the same lines would be the ASR-33's EOU (End Of User) charactor. -- Jeff Warnica jeff.warnica@unb.ca l3q90@unb.ca warnica@unb.ca From the University Of New Brunswickas really crappy NFSd fortune file: Elevators smell different to midgets ###### Newsgroups: alt.folklore.computers From: n1ist@netcom.com (Michael L. Ardai) Subject: Re: HCF (was: Re: history of computer input devices) Message-ID: Organization: Utopia Planetia Shipyards - Mars References: <673oli$569@sol.sun.csd.unb.ca> <6742c2$rg0$1@nntp2.ba.best.com> <6768ae$mev@sol.sun.csd.unb.ca> Date: Tue, 16 Dec 1997 16:35:30 GMT Lines: 23 Sender: n1ist@netcom5.netcom.com Path: ccw.ch!aetna.dolphins.ch!news.planetc.com!wilbur.ohww.norman.ok.us!newsfeed.kornet.nm.kr!howland.erols.net!ix.netcom.com!n1ist In article <6768ae$mev@sol.sun.csd.unb.ca> l3q90@jupiter.sun.csd.unb.ca (Warnica) writes: -George Byrd (george@apan.org.NO_UCE) wrote: -The Jargon File is unusualy non-helpful on this one - it points out that -the IBM 360 has one, but not what it was, or what it did. Thr MC6800 had a -HCF that caused lines to togglr sdtate quickly, potentialy causing the -solder traces to burn up. Nope. It just cycled the address lines. You aren't going to burn traces with address/data line drivers... -I do know on at least one C= PET model, you could POKE something to some -adderess and a couple of IO ports (described to me this way) would fight -over some lines, burning our the ports. (and things attached to them?) Actually, it changed the horizontal timing. On some PETs it speeded up the system. On others, it toasted the monitor. You could do the same thing with the original mono screen on the PC. /mike -- \|/ Michael L. Ardai N1IST n1ist@netcom.com \|/ -*- === Boston Amateur Radio Club: http://www.barc.org/barc === -*- /|\ or send "info barc-list" to listserv@majordomonetcom.com /|\ ###### From: tsw@cagent.com (Tom Watson) Newsgroups: alt.folklore.computers Subject: Re: HCF (was: Re: history of computer input devices) Date: Tue, 16 Dec 1997 16:44:09 -0800 Organization: CagEnt, Inc. Lines: 33 Message-ID: References: <01bd06e3$e0e5eeb0$d6119582@sirius> <34915C5E.69AD@azstarnet.com> <349437DC.19B9@goldrush.com> <673oli$569@sol.sun.csd.unb.ca> <6742c2$rg0$1@nntp2.ba.best.com> <3497e073.7243092@news.innet.be> NNTP-Posting-Host: alfred.cagent.com Cache-Post-Path: alfred.cagent.com!unknown@cypher.cagent.com Path: ccw.ch!aetna.dolphins.ch!news.planetc.com!leto.ou.edu!hammer.uoregon.edu!vixen.cso.uiuc.edu!howland.erols.net!news.maxwell.syr.edu!xfer.kren.nm.kr!newshub1.home.com!news.home.com!news.rdc1.sfba.home.com!cypher.cagent.com!user In article <3497e073.7243092@news.innet.be>, lucvdv@null.net (Luc Van der Veken) wrote: > > It also says that the 68000 was the first processor to actually > implement an instruction with this effect (it could burn the bus in > some configurations). I doubt if the mnemonic was HCF, however. You are off a digit. It was the 6800 that had such an instruction. It was a variant of the 'JMP' (jump) instruction that "jumped to the accumulator". The opcodes are 0x4E and 0x5E (for the A and B registers respectively). While not used in "normal" operations, they are convient for bringing up new designs. One could wire the data buss to either of the codes, and reset the processor. Upon reset, the processor reads out locations 0xFFFE and 0xFFFF for the location of the first instruction to execute (it gets 0x4E 0x4E), then it sets the address buss to that location and reads an instruction from there, getting 0x4E. This causes the address buss to increment and cycle thru, reading every location on the buss. With a scope, properly attached, quite helpful. Shows up decodes and the like for the various I/O, RAM, ROM devices. Might even show up conflicts, or a mal formed decode PAL. The modern equivalent is some silly Pentium instruction that may do something similar internally. It wasn't decoded as a special case, and fell thru the cracks. Here we are over 20 years later, and lessons are never learned. Probably accounts for segmented stuff as well..... As for mnemonic, there was none. Later data books put a '*' next to the op codes and mentioned what happened. -- tsw@cagent.com (Home: tsw@johana.com) Please forward spam to: annagram@hr.house.gov (my Congressman), I do. ###### From: lucvdv@null.net (Luc Van der Veken) Newsgroups: alt.folklore.computers Subject: Re: HCF (was: Re: history of computer input devices) Date: Tue, 16 Dec 1997 22:53:12 GMT Organization: . Lines: 38 Message-ID: <3498e24a.7713556@news.innet.be> References: <01bd06e3$e0e5eeb0$d6119582@sirius> <34915C5E.69AD@azstarnet.com> <349437DC.19B9@goldrush.com> <673oli$569@sol.sun.csd.unb.ca> <6742c2$rg0$1@nntp2.ba.best.com> <6768ae$mev@sol.sun.csd.unb.ca> NNTP-Posting-Host: pmpool053-44.innet.be Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Newsreader: Forte Agent 1.5/32.451 X-No-Archive: yes Path: ccw.ch!aetna.dolphins.ch!news.planetc.com!leto.ou.edu!hammer.uoregon.edu!logbridge.uoregon.edu!news-peer.gip.net!news.gsl.net!gip.net!news-peer.sprintlink.net!news-peer-east.sprintlink.net!news.sprintlink.net!Sprint!rill.news.pipex.net!pipex!uunetukout!warm.news.pipex.net!pipex!krypton.inbe.net!INbe.net!not-for-mail [sorry - missed this one and my other post was on the road already. Some1 beat me] On 16 Dec 1997 15:55:58 GMT, l3q90@jupiter.sun.csd.unb.ca (Warnica) told the world, or rather subsection alt.folklore.computers of it, that: > The Jargon File is unusualy non-helpful on this one - it points out that > the IBM 360 has one, but not what it was, or what it did. That's not how I interpreted it. I think it means there was talk about a HCF instruction that far back. > Thr MC6800 had a > HCF that caused lines to togglr sdtate quickly, potentialy causing the > solder traces to burn up. Not exactly the solder traces (you would need amps of current for that, for a prolonged period), but probably the bus drivers (outputs). > (Im dating myself, but this was slightly before my time) > > I do know on at least one C= PET model, you could POKE something to some > adderess and a couple of IO ports (described to me this way) would fight > over some lines, burning our the ports. (and things attached to them?) > > Along the same lines would be the ASR-33's EOU (End Of User) charactor. I once saw an entire list of instructions in this category. The only one I remember right away is RBT - Rewind and Break Tape. Many will have seen it already, but if I find it I just might post a copy for the newbies (and for the rest, to refresh their memories). -- The address in the "from" field is a real address, used as a spambox. Mail there may be read, or it may not. If you want to be sure, replace the domain by innet.be ###### From: lucvdv@null.net (Luc Van der Veken) Newsgroups: alt.folklore.computers Subject: Re: HCF (was: Re: history of computer input devices) Date: Tue, 16 Dec 1997 23:38:01 GMT Organization: . Lines: 17 Message-ID: <34a20cb3.18572630@news.innet.be> References: <01bd06e3$e0e5eeb0$d6119582@sirius> <34915C5E.69AD@azstarnet.com> <349437DC.19B9@goldrush.com> <673oli$569@sol.sun.csd.unb.ca> <6742c2$rg0$1@nntp2.ba.best.com> <6768ae$mev@sol.sun.csd.unb.ca> <3498e24a.7713556@news.innet.be> NNTP-Posting-Host: pmpool053-47.innet.be Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Newsreader: Forte Agent 1.5/32.451 X-No-Archive: yes Path: ccw.ch!aetna.dolphins.ch!news.planetc.com!leto.ou.edu!hammer.uoregon.edu!logbridge.uoregon.edu!newsfeed.ecrc.net!rill.news.pipex.net!pipex!uunetukout!warm.news.pipex.net!pipex!krypton.inbe.net!INbe.net!not-for-mail > Many will have seen it already, but if I find it I just might post a > copy for the newbies (and for the rest, to refresh their memories). I did find this fine example of the use of the HCF instruction (it's all the way at the end): http://comedy.clari.net/rhf/jokes/91q3/strfry.html The rest can be found at http://www.pvv.unit.no/~steinl/vitser/aseembly.html (and that's not a typo, at least not mine). The one I miss the most, is RPM (Read Programmer's Mind). -- The address in the "from" field is a real address, used as a spambox. Mail there may be read, or it may not. If you want to be sure, replace the domain by innet.be ###### From: lucvdv@null.net (Luc Van der Veken) Newsgroups: alt.folklore.computers Subject: Re: HCF (was: Re: history of computer input devices) Date: Wed, 17 Dec 1997 23:53:56 GMT Organization: . Lines: 60 Message-ID: <349c6401.5945634@news.innet.be> References: <01bd06e3$e0e5eeb0$d6119582@sirius> <34915C5E.69AD@azstarnet.com> <349437DC.19B9@goldrush.com> <673oli$569@sol.sun.csd.unb.ca> <6742c2$rg0$1@nntp2.ba.best.com> <3497e073.7243092@news.innet.be> NNTP-Posting-Host: pmpool053-51.innet.be Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Newsreader: Forte Agent 1.5/32.451 X-No-Archive: yes Path: ccw.ch!aetna.dolphins.ch!news.planetc.com!leto.ou.edu!hammer.uoregon.edu!xfer.kren.nm.kr!news-peer.gip.net!news.gsl.net!gip.net!news-peer.sprintlink.net!news-peer-east.sprintlink.net!news.sprintlink.net!Sprint!rill.news.pipex.net!pipex!uunetukout!warm.news.pipex.net!pipex!krypton.inbe.net!INbe.net!not-for-mail On Tue, 16 Dec 1997 16:44:09 -0800, tsw@cagent.com (Tom Watson) told the world, or rather subsection alt.folklore.computers of it, that: > You are off a digit. It was the 6800 that had such an instruction. It > was a variant of the 'JMP' (jump) instruction that "jumped to the > accumulator". The opcodes are 0x4E and 0x5E (for the A and B registers > respectively). While not used in "normal" operations, they are convient > for bringing up new designs. One could wire the data buss to either of > the codes, and reset the processor. Upon reset, the processor reads out > locations 0xFFFE and 0xFFFF for the location of the first instruction to > execute (it gets 0x4E 0x4E), then it sets the address buss to that > location and reads an instruction from there, getting 0x4E. This causes > the address buss to increment and cycle thru, reading every location on > the buss. With a scope, properly attached, quite helpful. Shows up > decodes and the like for the various I/O, RAM, ROM devices. Might even > show up conflicts, or a mal formed decode PAL. Not exactly what I would call a "Halt & Catch Fire" instruction. The one on the 68k would toggle a number of lines on and off as fast as it could, effectively burning the bus in some systems (at least, that's what I read). An instruction almost like you describe here existed (exists) on the 8085 also: opcode FF is a restart (like they used to call software interrupts), causing the PSW and PC to be pushed on the stack, whereafter the next instruction was read from address 003C. Normally there would be another instruction there, but if that was FF again, it would push the PSW & PC on the next locations of the stack, etc. - making the processor write to any memory location in it's addess range in a real minimum of time. Now if there's nothing at address 3C (no rom, no ram, no outputs enabled at all), guess what the processor sees there... FF (also depends on the design a bit). I used to exploit this "feature" to clear battery-backed CMOS RAMs in our gaming machines that ran on an 8085 in those days: power off, remove eprom from socket, power on, power off again, put eprom back, done: RAM filled with trash. > The modern equivalent is some silly Pentium instruction that may do > something similar internally. It wasn't decoded as a special case, and > fell thru the cracks. Here we are over 20 years later, and lessons are > never learned. Probably accounts for segmented stuff as well..... Some undocumented opcodes like that existed in the Z80 also, but they were quite useful there. They made it possible to do anything you could do with the HL register with the IX and IY registers too, by just prefixing the instruction with a certain byte value, so you actually had 3 sets of HL registers. Even though the prefixes were only documented for a few instructions, they actually worked for all of them, but possibly not in all brands of Z80's (just about every chip manufacturer there was must have made them at a certain moment.) -- The address in the "from" field is a real address, used as a spambox. Mail there may be read, or it may not. If you want to be sure, replace the domain by innet.be ###### From: William.Hamblen@nashville.com (William Hamblen) Newsgroups: alt.folklore.computers Subject: Re: HCF (was: Re: history of computer input devices) Date: Thu, 18 Dec 1997 01:43:04 GMT Organization: Utterly Disorganized Lines: 11 Message-ID: <349976c3.2299363@news.nashville.com> References: <01bd06e3$e0e5eeb0$d6119582@sirius> <34915C5E.69AD@azstarnet.com> <349437DC.19B9@goldrush.com> <673oli$569@sol.sun.csd.unb.ca> <6742c2$rg0$1@nntp2.ba.best.com> <6768ae$mev@sol.sun.csd.unb.ca> Reply-To: William.Hamblen@nashville.com NNTP-Posting-Host: 14706@207.65.181.138 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Newsreader: Forte Agent 1.5/32.452 Path: ccw.ch!aetna.dolphins.ch!news.planetc.com!atl-news-feed1.bbnplanet.com!cpk-news-hub1.bbnplanet.com!news.bbnplanet.com!newsfeed.direct.ca!Supernews60!supernews.com!Supernews69!not-for-mail On 16 Dec 1997 15:55:58 GMT, l3q90@jupiter.sun.csd.unb.ca (Warnica) wrote: >I do know on at least one C= PET model, you could POKE something to some >adderess and a couple of IO ports (described to me this way) would fight >over some lines, burning our the ports. (and things attached to them?) The Commodore Pet uses a programmable CRT controller and you can poke values into it that will destroy the monitor. This is not so different from what you can do with IBM clone video cards and monitors. ###### Path: ccw.ch!usenet From: Neil Franklin (reply remove nospam.) Newsgroups: alt.folklore.computers Subject: Re: HCF (was: Re: history of computer input devices) Date: 20 Dec 1997 04:03:27 +0100 Organization: My own Private Self Lines: 17 Message-ID: <1zz8emww.fsf@chonsp.franklin.lugs.ch> References: <01bd06e3$e0e5eeb0$d6119582@sirius> <34915C5E.69AD@azstarnet.com> <349437DC.19B9@goldrush.com> <673oli$569@sol.sun.csd.unb.ca> <6742c2$rg0$1@nntp2.ba.best.com> <6768ae$mev@sol.sun.csd.unb.ca> <3498e24a.7713556@news.innet.be> <34a20cb3.18572630@news.innet.be> X-Newsreader: Gnus v5.3/Emacs 19.34 lucvdv@null.net (Luc Van der Veken) wrote: >The rest can be found at > http://www.pvv.unit.no/~steinl/vitser/aseembly.html >(and that's not a typo, at least not mine). >The one I miss the most, is RPM (Read Programmer's Mind). That is a fairly incomplete copy of the canonical opcode list. Here a lot larger copy: http://www.ccw.ch/Neil.Franklin/Jokes_and_Fun/Canon_Opcode And if you're asking: I haven't got the URL of the original author. -- Neil.Franklin@ccw.ch, http://www.ccw.ch/Neil.Franklin/ for Geek Code, Papernet, Voicenet, PGP public key see http: Any computer, that is running optimally, is outdated; including my Cx486 ###### canonical opcode list moved to: http://neil.franklin.ch/Jokes_and_Fun/Canon_Opcode home page generally moved to: http://neil.franklin.ch/