Newsgroups: alt.folklore.computers Subject: Re: PCI and ISA References: Organization: Wizvax Communications, LLC From: multics@wizvax.wizvax.net (Richard Shetron) NNTP-Posting-Host: wizvax.wizvax.net Message-ID: <361c326d.0@news.wizvax.net> Date: 7 Oct 1998 23:33:01 -0500 X-Trace: 7 Oct 1998 23:33:01 -0500, wizvax.wizvax.net Lines: 31 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newspeer.monmouth.com!newsfeed.wizvax.net!news.wizvax.net!wizvax.wizvax.net!multics In article , Lorus wrote: >Can someone please help me! Seeing as I appear to be totally clueless >about hardware can anyone tell me the differences between PCI and ISA >slots. Thanks in advance >-- >Lorus Some basic information (may or may not be 100% accurate as there are some variations): ISA bus can have 8 bit or 16 bit slots. 8 bit cards can go into 16 bit, in general, 16 bit will not work in 8 bit. ISA bus runs at 6Mhz. I think the max transfer rate is 5MB/sec for 8bit slot, 10MB/sec for 16 bit. EISA bus, an enhanced ISA allowing EISA controller to get a 32 bit bus with a transfer rate of 20MB/sec. You can plug 8 or 16 bit ISA controllers into an EISA slot. VESA bus, an enhanced ISA allowing 32 bit transfers, but I think faster then EISA. Again, 8 or 16 bit ISA cards can work in a VESA slot. PCI bus is 32 bits and runs at 33Mhz allowing transfers rates up to 132MB/sec. PCI slots are not compatable with and of the others. PCI slots offere the best performance of the types. The PCI bus speed can vary dependong on the CPU clock speed, however 33Mhz is the most standard. ###### From: Terry Richards Newsgroups: alt.folklore.computers Subject: Re: PCI and ISA Date: Thu, 08 Oct 1998 09:42:24 -0400 Organization: Terry Richards Software Lines: 16 Message-ID: <361CC140.36E0@idt.net> References: <361c326d.0@news.wizvax.net> Reply-To: trs@idt.net NNTP-Posting-Host: ppp-36.ts-1.hp.idt.net Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 3.01C-IDT-v5 (WinNT; U) To: Richard Shetron Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newscore.univie.ac.at!btnet-peer!btnet!newshub.northeast.verio.net!news.idt.net!nntp.farm.idt.net!news Richard Shetron wrote: > > VESA bus, an enhanced ISA allowing 32 bit transfers, but I think faster > then EISA. Again, 8 or 16 bit ISA cards can work in a VESA slot. > IIRC, VESA was mechanically different and wouldn't accept ISA cards. Unfortunately, my last VESA M/B has gone to a better place(1) so I can't confirm this. (1) The Girl Scouts -- Terry Richards Terry Richards Software ###### From: "Hand&Bjafman-remove \"bantispame\" from my adress to send me a letter." Newsgroups: alt.folklore.computers References: Subject: Re: PCI and ISA Lines: 31 Organization: BartJim X-Newsreader: Microsoft Outlook Express 4.71.1712.3 X-MimeOLE: Produced By Microsoft MimeOLE V4.71.1712.3 Message-ID: Date: Thu, 08 Oct 1998 09:49:01 GMT NNTP-Posting-Host: 195.205.249.183 X-Complaints-To: abuse@tpsa.pl X-Trace: news.tpnet.pl 907840141 195.205.249.183 (Thu, 08 Oct 1998 11:49:01 MET DST) NNTP-Posting-Date: Thu, 08 Oct 1998 11:49:01 MET DST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.belnet.be!212.63.192.161.MISMATCH!newshub.bart.net!news.tele2.nl!newsfeed1.swip.net!swipnet!newsfeed.nyu.edu!wesley.videotron.net!Pollux.Teleglobe.net!orion.cst.tpsa.pl!news.tpnet.pl!not-for-mail Lorus napisał(a) w wiadomości: ... >Can someone please help me! Seeing as I appear to be totally clueless >about hardware can anyone tell me the differences between PCI and ISA >slots. Thanks in advance ISA slots are older. They are the longer ones. Mostly sound cards, modems, port cards, etc. must be placed in them. ISA slot is much slower then the PCI slot it's predicted that pretty soon it will disapear leaving us with AGP and PCI slots only. PCI slot is the shorter and faster one. Usually you place a video card, 3dfx card and some new sound&modem cards in it. PCI and ISA slots are usually put parallely together. In most cases it's 3 ISA and 4 PCI slots, but in older models it also can be 4 ISA 3 PCI. Some mother boards have 5 PCI and 3 ISA, etc. Just look for more than 4 parallel placed slots. If you have a P II processor you have to be careful not to insert ISA card into Slot 1 slot. You could ruin the motherboard as well as the card. ps: Don't miss up RAM/DRAM slots with PCI/ISA slots. RAM/DRAM slots are often placed parallely too. bartosz jan antoni frankowski (BJAF) (remove bantispame from my adress) ###### Newsgroups: alt.folklore.computers From: jwbirdsa@picarefy.picarefy.com (James W. Birdsall) Subject: Re: PCI and ISA Message-ID: <1998Oct8.165244.3659@picarefy.picarefy.com> Organization: Green Tiger Software References: <361c326d.0@news.wizvax.net> <361CC140.36E0@idt.net> Date: Thu, 8 Oct 1998 16:52:44 GMT Lines: 18 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newscore.univie.ac.at!btnet-peer!btnet!newsfeed.cwix.com!199.181.164.14!news.seanet.com!nntp.picarefy.com!picarefy!jwbirdsa In article <361CC140.36E0@idt.net> trs@idt.net writes: >Richard Shetron wrote: >> VESA bus, an enhanced ISA allowing 32 bit transfers, but I think faster >> then EISA. Again, 8 or 16 bit ISA cards can work in a VESA slot. > >IIRC, VESA was mechanically different and wouldn't accept ISA cards. >Unfortunately, my last VESA M/B has gone to a better place(1) so I can't >confirm this. I still have a couple. A VESA slot is just a standard 16-bit ISA slot with an additional connector, the same way that a 16-bit ISA slot is just an 8-bit ISA slot with an additional connector. -- James W. Birdsall http://www.picarefy.com/~jwbirdsa/ jwbirdsa@picarefy.com "For it is the doom of men that they forget." -- Merlin Get the Sun-2 Hardware Reference from ftp.picarefy.com:/pub/Sun-Hardware-Ref Sun-2 Hardware Reference Web Page: http://sun-www.picarefy.com/ ###### From: lucvdv@null.net (Luc Van der Veken) Newsgroups: alt.folklore.computers Subject: Re: PCI and ISA Date: Thu, 08 Oct 1998 19:33:54 GMT Organization: . Lines: 36 Message-ID: <36240d3c.6849277@news.uunet.be> References: <361c326d.0@news.wizvax.net> <361CC140.36E0@idt.net> NNTP-Posting-Host: pool02b-194-7-144-109.uunet.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: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newscore.univie.ac.at!btnet-peer!btnet!rill.news.pipex.net!pipex!rill.news.pipex.net!pipex!krypton.inbe.net!INbe.net!not-for-mail Also sprach Terry Richards on Thu, 08 Oct 1998 09:42:24 -0400 to alt.folklore.computers: > Richard Shetron wrote: > > > > > VESA bus, an enhanced ISA allowing 32 bit transfers, but I think faster > > then EISA. Again, 8 or 16 bit ISA cards can work in a VESA slot. > > > > IIRC, VESA was mechanically different and wouldn't accept ISA cards. > Unfortunately, my last VESA M/B has gone to a better place(1) so I can't > confirm this. > > (1) The Girl Scouts Just a second while I check - I've still got one under my bed ;) VESA has a third connector plus the two of a 16 bit ISA slot, so it will accept 8 and 16 bit ISA cards. Also, unless I was misinformed, VESA was actually the fastest of {ISA8, ISA16, EISA, VESA, PCI}; with bus clock frequencies up to 50 MHZ (depending on the processor speed, not fixed like the other ones - so on some systems it could actually be slower). If this is true, can someone explain why PCI won? (except for the obvious lower cost: PCI needs only 1 connector, about the same physical size as an 8-bit ISA one, versus ca. 3 times that length of connectors vor VESA). VHS versus Betamax again? -- To avoid having your dinner interrupted by one of those telemarketing calls, just leave your computer online. That way you'll only get spam. ###### From: korpela@islay.ssl.berkeley.edu (Eric J. Korpela) Newsgroups: alt.folklore.computers Subject: Re: PCI and ISA Date: 8 Oct 1998 21:45:08 GMT Organization: Cal Berkeley-- Space Sciences Lab Lines: 39 Message-ID: <6vjbp4$dfs$1@agate.berkeley.edu> References: <361c326d.0@news.wizvax.net> <361CC140.36E0@idt.net> <36240d3c.6849277@news.uunet.be> NNTP-Posting-Host: islay.ssl.berkeley.edu Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.belnet.be!newsfeed.wirehub.nl!btnet-peer!btnet!newsfeed.berkeley.edu!agate!islay.ssl.berkeley.edu!korpela In article <36240d3c.6849277@news.uunet.be>, Luc Van der Veken wrote: > >VESA has a third connector plus the two of a 16 bit ISA slot, so >it will accept 8 and 16 bit ISA cards. > >Also, unless I was misinformed, VESA was actually the fastest of >{ISA8, ISA16, EISA, VESA, PCI}; with bus clock frequencies up to >50 MHZ (depending on the processor speed, not fixed like the >other ones - so on some systems it could actually be slower). > >If this is true, can someone explain why PCI won? I'm assuming you mean VESA Local Bus. As opposed to VESA which is a consortium of video equipment manufacturers. The VESA Super VGA standard is still in use. The main reason to why PCI won over the VESA Local Bus is simple. The VESA local bus was tied to the 486 processor architecture. Signal (and therefore timings) were derrived directly from the 486. Designing a VLB chipset for the Pentium would have been a pain, and I'm not sure it was ever undertaken. (Anyone out there know if there was any Pentium chipset with VLB support?) There are, of course, other reason. VESA was basically designed as a video bus. Not much attention was paid to protocols for non-video devices. Signal path lengths and number of slots were limited with no option for multiple VLB busses. I don't think of this as a another case of VHS vs Betamax, but of a case of VHS versus analog video disk. The more versatile technology won. Eric -- Eric Korpela | An object at rest can never be korpela@ssl.berkeley.edu | stopped. Click for home page. ###### Path: chonsp.franklin.ch!usenet From: Neil Franklin Newsgroups: alt.folklore.computers Subject: Re: PCI and ISA Date: 08 Oct 1998 21:48:11 +0200 Organization: My own Private Self Lines: 61 Message-ID: References: <361c326d.0@news.wizvax.net> X-Newsreader: Gnus v5.3/Emacs 19.34 multics@wizvax.wizvax.net (Richard Shetron) writes: > > Some basic information (may or may not be 100% accurate And that on a.f.c :-) > ISA bus can have 8 bit or 16 bit slots. 8 bit cards can go into 16 bit, > in general, 16 bit will not work in 8 bit. ISA bus runs at 6Mhz. I think > the max transfer rate is 5MB/sec for 8bit slot, 10MB/sec for 16 bit. PC,XT Bus: introduced with IBM PC, unchanged in IBM PC/XT 8 bit, runs at 4.77 MHz, 4 clocks per data transfer -> 1.19 MByte/s. AT But: extended PC, introduced with IBM PC/AT 16 bit, runs at 6 (AT/02) or 8 MHZ (AT/03 and most clones), 2 clocks per data transfer -> 6 or 8 MByte/s ISA Bus: renaming of AT Bus because cloners don't like to call it AT after after IBM stopped making ATs. > EISA bus, an enhanced ISA allowing EISA controller to get a 32 bit > bus with a transfer rate of 20MB/sec. You can plug 8 or 16 bit ISA > controllers into an EISA slot. Microchannel (MCA): completely new, introduced with IBM PS/2 16 (286) or 32 bit (386 and above), runs at 10 MHz, 1 clock per transfer -> 20..40 MByte/s EISA: extended ISA, by an consortium as protest to MCA licensing conditions 32 bit, runs at 8 MHZ, 1 clock per transfer -> 32 MByte/s > VESA bus, an enhanced ISA allowing 32 bit transfers, but I think faster > then EISA. Again, 8 or 16 bit ISA cards can work in a VESA slot. EISA: extended ISA, by annother consortium because EISA too expensive 32 bit, runs at 25..50 MHz (processor external clock), 1 clock per transfer -> 100..200 MByte/s > PCI bus is 32 bits and runs at 33Mhz allowing transfers rates up > to 132MB/sec. PCI slots are not compatable with and of the others. > PCI slots offere the best performance of the types. The PCI bus speed can > vary dependong on the CPU clock speed, however 33Mhz is the most > standard. PCI: completely new, by Intel, bus mastering on all slots 32 or 64 bit, 33 (proc 33..50) or 66 MHz (proc 66..100) 1 clock per transfer -> 132..528 MByte/s Bus Politics: something horrible that harms users. -- *** New home Addresses Mail and Web *** home: neil@franklin.ch.remove http://neil.franklin.ch/ work: franklin@arch.ethz.ch.remove http://caad.arch.ethz.ch/~franklin/ Microsoft is Software Communism, Fight for GNU Freedom! ###### From: `David O'Bedlam' Subject: Re: PCI and ISA Newsgroups: alt.folklore.computers References: <361c326d.0@news.wizvax.net> <361CC140.36E0@idt.net> <36240d3c.6849277@news.uunet.be> Organization: Gehenna-By-The-Sea Lines: 14 Message-ID: NNTP-Posting-Host: 207.201.34.8 X-Trace: news6.ispnews.com 908195531 207.201.34.8 (Mon, 12 Oct 1998 08:32:11 EDT) NNTP-Posting-Date: Mon, 12 Oct 1998 08:32:11 EDT Date: Mon, 12 Oct 1998 12:32:11 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.belnet.be!newsgate.cistron.nl!het.net!news-feed.inet.tele.dk!bofh.vszbr.cz!nntp.abs.net!hub1.ispnews.com!news6.ispnews.com.POSTED!not-for-mail Luc Van der Veken wrote: > Also sprach Terry Richards on Thu, 08 Oct 1998 >> (1) The Girl Scouts > Just a second while I check - I've still got one under my bed ;) A Girl Scout? It's a bed on stilts, right? David -- "Say you have a flock of sheep." -- Jonah Thomas ###### From: Bruce Lin Newsgroups: alt.folklore.computers Subject: Re: PCI and ISA Date: 12 Oct 1998 01:26:06 GMT Organization: Princeton University Lines: 24 Sender: Bruce Lin Message-ID: <6vrlre$9tj$1@cnn.Princeton.EDU> References: <361c326d.0@news.wizvax.net> <361CC140.36E0@idt.net> <36240d3c.6849277@news.uunet.be> <6vjbp4$dfs$1@agate.berkeley.edu> NNTP-Posting-Host: phoenix.princeton.edu User-Agent: tin/pre-1.4-980818 ("Laura") (UNIX) (SunOS/5.6 (sun4u)) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newshub.northeast.verio.net!news3.cac.psu.edu!cnn.Princeton.EDU!not-for-mail There definitely were Pentium-class VLB motherboards. I believe they were only P-60 and P-66 5 volt motherboards, though. The Nx586 came in VLB varieties, too, if I remember correctly. Here's a question: is a PCI hard drive controller inherently faster than a VLB hard drive controller? Because I remember running my IBM 2.1 GB hard drive at 40 MHz on my 486 VLB system, and that was "okay" in terms of meeting specs (and not getting errors), but running my PCI bus at 37.5 or 41.7 MHz is perceived as overclocking my hard drive. So ... was I exceeding specs with 40 MHz VLB, or is there something inherently different about PCI that makes it more sensitive to clock speeds? Or perhaps is it dependent on the controller itself rather than the bus? Bruce P.S. Aside: I thought I read somewhere that the specs required that only one VLB card is allowed to run at 50 MHz. -- center for energy and environmental studies, princeton university fuel cell scooters : http://www.princeton.edu/~brucelin/thesis ###### From: Simon Lyall Newsgroups: alt.folklore.computers Subject: Re: PCI and ISA Date: 12 Oct 1998 02:22:15 GMT Organization: Darkmere Private access Internet, Auckland, NZ Lines: 12 Message-ID: <6vrp4n$885@darkmere.gen.nz> References: <361c326d.0@news.wizvax.net> NNTP-Posting-Host: p32-max39.akl.ihug.co.nz User-Agent: tin/pre-1.4-980226 (UNIX) (Linux/2.0.27 (i486)) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.belnet.be!newsgate.cistron.nl!het.net!news-feed.inet.tele.dk!bofh.vszbr.cz!newspeer.monmouth.com!newsfeed.cwix.com!203.97.37.7!newsfeed.clear.net.nz!ihug.co.nz!darkmere.gen.nz!darkmere.gen.nz!simon Neil Franklin wrote: [ snipped ] So how does AGP fit in? . I went to buy a new PC a few weeks back and suddenly found all the video cards were using it. Never even heard of it before then. -- Simon J. Lyall | Very Busy | Mail: simon@darkmere.gen.nz "Inside me Im Screaming, Nobody pays any attention." | MT. PGP Keyprint - F3 41 25 CB A7 6E FE BC 75 4E 27 E5 BF CB A7 93 ###### Path: chonsp.franklin.ch!usenet From: Neil Franklin Newsgroups: alt.folklore.computers Subject: Re: PCI and ISA Date: 12 Oct 1998 21:56:16 +0200 Organization: My own Private Self Lines: 15 Sender: neil@chonsp.franklin.ch Message-ID: References: <361c326d.0@news.wizvax.net> <361CC140.36E0@idt.net> <36240d3c.6849277@news.uunet.be> <6vjbp4$dfs$1@agate.berkeley.edu> <6vrlre$9tj$1@cnn.Princeton.EDU> X-Newsreader: Gnus v5.3/Emacs 19.34 Bruce Lin writes: > > P.S. Aside: I thought I read somewhere that the specs required that > only one VLB card is allowed to run at 50 MHz. The official VESA VLB specs went up to 40MHz (max 2 cards), with an lower 33MHz option where they allowed 3 cards. This was inofficially extended to 50MHz, limited to 1 card. -- *** New home Addresses Mail and Web *** home: neil@franklin.ch.remove http://neil.franklin.ch/ work: franklin@arch.ethz.ch.remove http://caad.arch.ethz.ch/~franklin/ Microsoft is Software Communism, Fight for GNU Freedom! ###### Path: chonsp.franklin.ch!usenet From: Neil Franklin Newsgroups: alt.folklore.computers Subject: Re: PCI and ISA Date: 12 Oct 1998 22:00:08 +0200 Organization: My own Private Self Lines: 20 Sender: neil@chonsp.franklin.ch Message-ID: References: <361c326d.0@news.wizvax.net> <6vrp4n$885@darkmere.gen.nz> X-Newsreader: Gnus v5.3/Emacs 19.34 Simon Lyall writes: > > So how does AGP fit in? . I went to buy a new PC a few weeks back and > suddenly found all the video cards were using it. Never even heard of it > before then. AGP is an extension to PCI (has AFAIK the full bus protocol) but limited to one fast card (less buffer delay, but so less power on the signal lines). Intel has been pushing it for about 1 year, to make life difficult for the cloners. Neil "upgraded from ISA to PCI video only 1/2 year ago" Franklin -- *** New home Addresses Mail and Web *** home: neil@franklin.ch.remove http://neil.franklin.ch/ work: franklin@arch.ethz.ch.remove http://caad.arch.ethz.ch/~franklin/ Microsoft is Software Communism, Fight for GNU Freedom! ###### From: skb@xmission.removethis.com (Scott Brown) Newsgroups: alt.folklore.computers Subject: Re: PCI and ISA Date: Tue, 13 Oct 1998 04:58:28 GMT Organization: (none) Lines: 22 Message-ID: <3622cf73.424586589@news.xmission.com> References: <361c326d.0@news.wizvax.net> <6vrp4n$885@darkmere.gen.nz> NNTP-Posting-Host: slc356h.modem.xmission.com X-Trace: news.xmission.com 908254524 3139 166.70.2.102 (13 Oct 1998 04:55:24 GMT) X-Complaints-To: abuse@xmission.com NNTP-Posting-Date: 13 Oct 1998 04:55:24 GMT X-Newsreader: Forte Free Agent 1.11/32.235 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.belnet.be!news-raspail.gip.net!news.gsl.net!gip.net!oleane!news.maxwell.syr.edu!xmission!nnrp.xmission!not-for-mail On 12 Oct 1998 22:00:08 +0200, Neil Franklin wrote: >AGP is an extension to PCI (has AFAIK the full bus protocol) but >limited to one fast card (less buffer delay, but so less power on the >signal lines). > >Intel has been pushing it for about 1 year, to make life difficult for >the cloners. Never mind that Intel *is* a cloner. Lest we forget that it was Big Blue who started this shindig, remember that under all the AGP, PCI, VLB, 500MHz Slot-this and Socket-that, there lives an 8088 PC. As far as I can tell, AGP is just a cheap way to build mediocre 3D accelerator boards. Coming from Intel, this is not a surprise. >Neil "upgraded from ISA to PCI video only 1/2 year ago" Franklin Lucky dog. I had to dump a rather nice VLB video card when I got an ISA+PCI mainboard. ###### From: lucvdv@null.net (Luc Van der Veken) Newsgroups: alt.folklore.computers Subject: Re: PCI and ISA Date: Tue, 13 Oct 1998 17:08:24 GMT Organization: . Lines: 15 Message-ID: <36238856.1488104@news.uunet.be> References: <361c326d.0@news.wizvax.net> <6vrp4n$885@darkmere.gen.nz> <3622cf73.424586589@news.xmission.com> NNTP-Posting-Host: pool02b-194-7-147-188.uunet.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: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.belnet.be!newsgate.cistron.nl!het.net!news-feed.inet.tele.dk!bofh.vszbr.cz!rill.news.pipex.net!pipex!krypton.inbe.net!INbe.net!not-for-mail Also sprach skb@xmission.removethis.com (Scott Brown) on Tue, 13 Oct 1998 04:58:28 GMT to alt.folklore.computers: > Never mind that Intel *is* a cloner. Lest we forget that it was Big > Blue who started this shindig, remember that under all the AGP, PCI, > VLB, 500MHz Slot-this and Socket-that, there lives an 8088 PC. And an 8088 was (and is) Intel, if you ask me (or were you referring to motherboards?) > Lucky dog. I had to dump a rather nice VLB video card when I got an > ISA+PCI mainboard. Same story here (except that here it's ISA+PCI+AGP). ###### From: skb@xmission.removethis.com (Scott Brown) Newsgroups: alt.folklore.computers Subject: Re: PCI and ISA Date: Tue, 13 Oct 1998 20:43:13 GMT Organization: (none) Lines: 17 Message-ID: <3623ba06.484646520@news.xmission.com> References: <361c326d.0@news.wizvax.net> <6vrp4n$885@darkmere.gen.nz> <3622cf73.424586589@news.xmission.com> <36238856.1488104@news.uunet.be> NNTP-Posting-Host: slc356h.modem.xmission.com X-Trace: news.xmission.com 908311205 13826 166.70.2.102 (13 Oct 1998 20:40:05 GMT) X-Complaints-To: abuse@xmission.com NNTP-Posting-Date: 13 Oct 1998 20:40:05 GMT X-Newsreader: Forte Free Agent 1.11/32.235 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeed.uk.ibm.net!ibm.net!logbridge.uoregon.edu!xmission!nnrp.xmission!not-for-mail On Tue, 13 Oct 1998 17:08:24 GMT, lucvdv@null.net (Luc Van der Veken) wrote: >Also sprach skb@xmission.removethis.com (Scott Brown) on Tue, 13 >Oct 1998 04:58:28 GMT to alt.folklore.computers: > >> Never mind that Intel *is* a cloner. Lest we forget that it was Big >> Blue who started this shindig, remember that under all the AGP, PCI, >> VLB, 500MHz Slot-this and Socket-that, there lives an 8088 PC. > >And an 8088 was (and is) Intel, if you ask me (or were you >referring to motherboards?) IBM PC microcomputers. Intel builds clone mainboards, because only IBM can build genuine PC mainboards. AMD builds clones of Intel's CPUs, as does IBM, I think. ###### From: bill_h Newsgroups: alt.folklore.computers Subject: Re: PCI and ISA Date: Tue, 13 Oct 1998 23:06:48 -0700 Organization: http://www.supernews.com, The World's Usenet: Discussions Start Here Lines: 13 Message-ID: <36243F78.6F24@azstarnet.com> References: <361c326d.0@news.wizvax.net> <6vrp4n$885@darkmere.gen.nz> <3622cf73.424586589@news.xmission.com> <36238856.1488104@news.uunet.be> Reply-To: bill_h@azstarnet.com NNTP-Posting-Host: 169.197.40.52 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: 908345335 W4NEQTDLS2834A9C5C usenet80.supernews.com X-Complaints-To: newsabuse@supernews.com X-Mailer: Mozilla 3.04Gold (Win16; I) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.belnet.be!newsgate.cistron.nl!het.net!newsfeed.wirehub.nl!btnet-peer!btnet!newsfeed.corridex.com!Supernews73!supernews.com!Supernews69!not-for-mail Neil Franklin wrote: > ...... IBM with the PC actually for the first time made an > computer with an third party off the shelf CPU. > > Introduction dates: 8086: 1977, 8088: 1978, IBM PC: 1981 Didn't IBM have a license to MAKE THEIR OWN 8086 chips, before the PC was ever built, for among other things the Displaymaster? Bill Tucson ###### Path: chonsp.franklin.ch!usenet From: Neil Franklin Newsgroups: alt.folklore.computers Subject: Re: PCI and ISA Date: 14 Oct 1998 00:09:02 +0200 Organization: My own Private Self Lines: 34 Sender: neil@chonsp.franklin.ch Message-ID: References: <361c326d.0@news.wizvax.net> <6vrp4n$885@darkmere.gen.nz> <3622cf73.424586589@news.xmission.com> <36238856.1488104@news.uunet.be> X-Newsreader: Gnus v5.3/Emacs 19.34 lucvdv@null.net (Luc Van der Veken) writes: > > Also sprach skb@xmission.removethis.com (Scott Brown) on Tue, 13 > Oct 1998 04:58:28 GMT to alt.folklore.computers: > > > Never mind that Intel *is* a cloner. Lest we forget that it was Big > > Blue who started this shindig, remember that under all the AGP, PCI, > > VLB, 500MHz Slot-this and Socket-that, there lives an 8088 PC. > > And an 8088 was (and is) Intel, if you ask me You are correct. 8086/88 are pure Intel designs, descendant form their 8080/85 design. IBM with the PC actually for the first time made an computer with an third party off the shelf CPU. Introduction dates: 8086: 1977, 8088: 1978, IBM PC: 1981 > > >Neil "upgraded from ISA to PCI video only 1/2 year ago" Franklin > > Lucky dog. I had to dump a rather nice VLB video card when I got an > > ISA+PCI mainboard. > > Same story here (except that here it's ISA+PCI+AGP). Then I must be very lucky, as this is actually an VLB(1)+ISA(3)+PCI(3) motherboard (486). I just never had an VLB card to put in it. -- *** New home Addresses Mail and Web *** home: neil@franklin.ch.remove http://neil.franklin.ch/ work: franklin@arch.ethz.ch.remove http://caad.arch.ethz.ch/~franklin/ Microsoft is Software Communism, Fight for GNU Freedom! ###### From: lucvdv@null.net (Luc Van der Veken) Newsgroups: alt.folklore.computers Subject: Re: PCI and ISA Date: Wed, 14 Oct 1998 18:50:19 GMT Organization: . Lines: 9 Message-ID: <3626ea9c.4538284@news.uunet.be> References: <361c326d.0@news.wizvax.net> <6vrp4n$885@darkmere.gen.nz> <3622cf73.424586589@news.xmission.com> <36238856.1488104@news.uunet.be> NNTP-Posting-Host: pool02b-194-7-148-75.uunet.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: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.belnet.be!news-raspail.gip.net!news-peer.gip.net!news.gsl.net!gip.net!news.maxwell.syr.edu!rill.news.pipex.net!pipex!rill.news.pipex.net!pipex!krypton.inbe.net!INbe.net!not-for-mail Also sprach Neil Franklin on 14 Oct 1998 00:09:02 +0200 to alt.folklore.computers: > 8086/88 are pure Intel designs, descendant form their > 8080/85 design. A scary thought just came up - my P2 is a descendant from the 4004 ;) ###### From: cjt&trefoil Newsgroups: alt.folklore.computers Subject: Re: PCI and ISA Date: Wed, 14 Oct 1998 20:45:21 -0500 Organization: Prodigy Services Corp Lines: 17 Message-ID: <362553B1.13B8@prodigy.net> References: <361c326d.0@news.wizvax.net> <6vrp4n$885@darkmere.gen.nz> <3622cf73.424586589@news.xmission.com> <36238856.1488104@news.uunet.be> <36243F78.6F24@azstarnet.com> Reply-To: cheljuba@prodigy.net NNTP-Posting-Host: dllsb203-01.splitrock.net Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Post-Time: 15 Oct 1998 01:46:52 GMT X-Auth-User: 002709921/dc8931acea349e9b X-Problems-To: abuse@prodigy.net X-Mailer: Mozilla 3.04 (Win95; U) To: bill_h@azstarnet.com Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!surfnet.nl!newsfeed.wirehub.nl!btnet-peer!btnet!newsfeed.cwix.com!4.1.16.34!cpk-news-hub1.bbnplanet.com!news.bbnplanet.com!portc02.blue.aol.com!newscon02!prodigy.com!not-for-mail bill_h wrote: > > Neil Franklin wrote: > > > ...... IBM with the PC actually for the first time made an > > computer with an third party off the shelf CPU. > > > > Introduction dates: 8086: 1977, 8088: 1978, IBM PC: 1981 > > Didn't IBM have a license to MAKE THEIR OWN 8086 chips, before > the PC was ever built, for among other things the Displaymaster? > > Bill > Tucson I also think that at one time IBM owned a significant fraction of Intel's stock. ###### Newsgroups: alt.folklore.computers From: "Ralph Wade Phillips" Subject: Re: PCI and ISA X-Nntp-Posting-Host: zs150050.shrv.bna.boeing.com Message-ID: X-Mimeole: Produced By Microsoft MimeOLE V4.72.2106.4 Lines: 22 Sender: nntp@news.boeing.com (Boeing NNTP News Access) Organization: The Boeing Company X-Newsreader: Microsoft Outlook Express 4.72.2106.4 References: <361c326d.0@news.wizvax.net> <6vrp4n$885@darkmere.gen.nz> <3622cf73.424586589@news.xmission.com> <36238856.1488104@news.uunet.be> <3626ea9c.4538284@news.uunet.be> Date: Thu, 15 Oct 1998 12:38:18 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newscore.univie.ac.at!europa.clark.net!208.134.241.18!newsfeed.cwix.com!192.220.250.21!netnews1.nw.verio.net!netnews.nwnet.net!xyzzy!not-for-mail Hi there! Luc Van der Veken wrote in message <3626ea9c.4538284@news.uunet.be>... >Also sprach Neil Franklin on 14 Oct >1998 00:09:02 +0200 to alt.folklore.computers: > >> 8086/88 are pure Intel designs, descendant form their >> 8080/85 design. > >A scary thought just came up - my P2 is a descendant from the >4004 ;) > You just now realized that? But, the calculator the 4004 was designed for at least DID accurate math, within its limits of accuracy. Can't always say the same for the Pentium and its followers ... References: <3626ea9c.4538284@news.uunet.be> NNTP-Posting-Host: man-134.dialup.zetnet.co.uk X-Trace: irk.zetnet.co.uk 908490538 17408 194.247.40.171 (15 Oct 1998 22:28:58 GMT) NNTP-Posting-Date: 15 Oct 1998 22:28:58 GMT X-Everything: Net-Tamer V 1.08X Lines: 20 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.belnet.be!newsfeed.wirehub.nl!btnet-peer!btnet!dispose.news.demon.net!demon!peer.news.zetnet.net!zetnet.co.uk!not-for-mail On 1998-10-14 lucvdv@null.net(LucVanderVeken) said: :Also sprach Neil Franklin on 14 Oct :1998 00:09:02 +0200 to alt.folklore.computers: :> 8086/88 are pure Intel designs, descendant form their :> 8080/85 design. :A scary thought just came up - my P2 is a descendant from the :4004 ;) Worse - I wouldn't be surprised if you could still compile 4004 code to run on it with the right compiler... certainly, there's still a 1-to-1 mapping between 8008 instructions and x86 instructions, so you can still use all that legacy 8008 code... ;> -- Communa (together) we remember... we'll see you falling you know soft spoken changes nothing to sing within her... Net-Tamer V 1.08X - Test Drive ###### From: ftit@engin.umich.edu (Sergej Roytman) Newsgroups: alt.folklore.computers Subject: Re: PCI and ISA Date: 16 Oct 1998 17:46:22 GMT Organization: University of Michigan, College of Engineering Lines: 19 Message-ID: <7080pe$gfn@srvr1.engin.umich.edu> References: <36238856.1488104@news.uunet.be> <3626ea9c.4538284@news.uunet.be> NNTP-Posting-Host: price.ummu.umich.edu Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.belnet.be!212.63.192.161.MISMATCH!newshub.bart.net!news.tele2.nl!newsfeed1.swip.net!swipnet!news-feed.inet.tele.dk!bofh.vszbr.cz!news.eecs.umich.edu!srvr1.engin.umich.edu!ftit In article <3626ea9c.4538284@news.uunet.be>, Luc Van der Veken wrote: >Also sprach Neil Franklin on 14 Oct >1998 00:09:02 +0200 to alt.folklore.computers: >> 8086/88 are pure Intel designs, descendant form their >> 8080/85 design. >A scary thought just came up - my P2 is a descendant from the >4004 ;) You gots it. A few years ago in one of my first grad-school classes, the prof was talking about backward compatability, and how if you were to look at a Pentium (brand new back then) you could still see an 8080. One of the students immediately came back with, "You wouldn't have to look very hard!" The class erupted into laughter. -- Sergej Roytman ###### From: lucvdv@null.net (Luc Van der Veken) Newsgroups: alt.folklore.computers Subject: Re: PCI and ISA Date: Fri, 16 Oct 1998 20:46:04 GMT Organization: . Lines: 43 Message-ID: <3628a431.4250972@news.uunet.be> References: <3626ea9c.4538284@news.uunet.be> <705sva$h00$1@irk.zetnet.co.uk> NNTP-Posting-Host: pool02b-194-7-144-22.uunet.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: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!howland.erols.net!rill.news.pipex.net!pipex!join.news.pipex.net!pipex!krypton.inbe.net!INbe.net!not-for-mail Also sprach lisard@zetnet.co.uk on 15 Oct 1998 22:28:58 GMT to alt.folklore.computers: > > > On 1998-10-14 lucvdv@null.net(LucVanderVeken) said: > :A scary thought just came up - my P2 is a descendant from the > :4004 ;) > Worse - I wouldn't be surprised if you could still compile 4004 code to > run on it with the right compiler... certainly, there's still a 1-to-1 > mapping between 8008 instructions and x86 instructions, so you can still > use all that legacy 8008 code... ;> That's funny - I remember checking (in my PC-XT days) if the 8080 or '85 opcodes matched the 8086 ones, and I thought they didn't. OTOH there _was_ something like a NEC V20-V30-V40 (hope I got the letter right): at least one of them had a Z80 emulation mode (and was 80186 compatible 'normally' IIRC), so that one wouldn't barf on 8080 code. There even existed a Z-80 CP/M emulator for it that ran under DOS (or should I say a CP/M emulator under CP/M :o) Now I get it - you said "with the right compiler." One being a forefather of the other, the 4004's instructions probably all have close counterparts in the Pentium set. I wonder for how many other processors of completely different origin your statement would come close: for example, and for as much as I remember of it, it looks to me as if there was more resemblance between the 8080's instruction set and the PDP-11, than between the PDP-8 and the 11 (and a lot more than between the 11 and a 6502 or so.) (must admit that I only tried to use PDP11 assembler a few times, and I only *saw* PDP8 assembler once or twice). What I mean is: I think it would be a lot easier to port an 8080 assembler routine to the PDP11 than (to name another uP) a 6809 one - but the 11 knew tricks an 8080 didn't even dream of, so the opposite would be all but obvious (like JMP @R4+, if I've got the syntax right). ###### From: lisard@zetnet.co.uk Newsgroups: alt.folklore.computers Subject: Re: PCI and ISA Date: 17 Oct 1998 03:32:13 GMT Lines: 61 Message-ID: <70933t$lit$8@irk.zetnet.co.uk> References: <3628a431.4250972@news.uunet.be> NNTP-Posting-Host: man-111.dialup.zetnet.co.uk X-Trace: irk.zetnet.co.uk 908595133 22109 194.247.40.142 (17 Oct 1998 03:32:13 GMT) NNTP-Posting-Date: 17 Oct 1998 03:32:13 GMT X-Everything: Net-Tamer V 1.08X Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.belnet.be!news-raspail.gip.net!news.gsl.net!gip.net!newsfeed.ecrc.net!newsfeed.nacamar.de!demeter.clara.net!news.clara.net!peer.news.zetnet.net!zetnet.co.uk!not-for-mail On 1998-10-16 lucvdv@null.net(LucVanderVeken) said: :> Worse - I wouldn't be surprised if you could still compile 4004 :>code to run on it with the right compiler... certainly, there's :>still a 1-to-1 mapping between 8008 instructions and x86 :>instructions, so you can still use all that legacy 8008 code... ;> :That's funny - I remember checking (in my PC-XT days) if the 8080 :or '85 opcodes matched the 8086 ones, and I thought they didn't. No, they didn't. But at the source level, near as dammit every 8080 instruction had an 8086 instruction that it mapped to directly. The major exceptions were the conditional call and return instructions. :OTOH there _was_ something like a NEC V20-V30-V40 (hope I got the :letter right): at least one of them had a Z80 emulation mode (and :was 80186 compatible 'normally' IIRC), so that one wouldn't barf :on 8080 code. There even existed a Z-80 CP/M emulator for it that :ran under DOS (or should I say a CP/M emulator under CP/M :o) V20 and V30 had 8080 emulation; V40 may have had Z80, I never saw one. :much as I remember of it, it looks to me as if there was more :resemblance between the 8080's instruction set and the PDP-11, :than between the PDP-8 and the 11 (and a lot more than between :the 11 and a 6502 or so.) I've never seen anything that pushed as far along the memory-to-memory concept as the PDP-11. Even in the M68k, which was about as close as anyone ever got without breaking copyright in the MPU world, one of the operands to the arithmetic instructions had to be a register; that wasn't the case with the PDP-11. But what I'm trying to say is that some of the things you could do on a PDP-11 would be impossible to express in a single instruction or set of instructions on an 8080, and many others would be far too verbose. The PDP-8 seems fairly similar to the other PDP machines (10, 7, etc) from what (very little) I've seen of them (a good look through the emulators on DECUS - btw, if anyone in the UK has a PDP-anything or a Nova going spare, please give us a shout...). But they also seem to have fairly heavily influenced the design of the 6800 (and hence the 6502). :What I mean is: I think it would be a lot easier to port an 8080 :assembler routine to the PDP11 than (to name another uP) a 6809 :one - but the 11 knew tricks an 8080 didn't even dream of, so the :opposite would be all but obvious (like JMP @R4+, if I've got the :syntax right). I'm not so sure. I'd venture that the 6809 would be the easier port, simply because of the generality of its addressing modes and registers. The 8080 was overwrought with an awful lot of special-purposeness, and Intel never did figure out why this is a silly idea. But porting assembly between processors is a tiny fraction of the battle - all interface code would always need a complete redesign, for a start. The strength of the 8086 was that you could pretty much drop it in the same bus as an 8080 (well, 8085 certainly) and have it carry on functioning. Just redesign a little glue, recompile your program, and go. -- Communa (together) we remember... we'll see you falling you know soft spoken changes nothing to sing within her...