Prev: CFP: Informatics 2010: new date - until 15 March 2010
Next: Call for papers: HPCS-10, USA, July 2010
From: Del Cecchi on 14 Mar 2010 16:10 "Stephen Fuld" <SFuld(a)alumni.cmu.edu.invalid> wrote in message news:hnj8k4$tsd$1(a)news.eternal-september.org... > On 3/14/2010 1:46 AM, Terje Mathisen wrote: >> Stephen Fuld wrote: >>> On 3/11/2010 10:15 AM, Terje Mathisen wrote: >>>> The 8086 model did have a 16-bit bus as well as an 8-byte >>>> instruction >>>> buffer, which meant that at least during those instructions that >>>> didn't >>>> load or store anything, the bus could be used to load one or two >>>> upcoming opcodes. >>> >>> Yes, and IIRC, the 8088 has a slightly larger instruction buffer >>> as a >>> counter to the smaller bus width. Again, an intelligent response >>> to what >>> was available and needed. >>> >> The opposite: >> >> The 8088, as I showed above, had almost zero opportunity to ever >> load >> anything into that prefetch buffer, so they _reduced_ the size from >> 8 to >> 6 bytes. > > Thanks to Tim and Terje for correcting my recollections. But > despite my error, I still maintain the original point, that these > decisions were made using good engineering judgment and not out of > ignorance of caches, or certainly of different bus widths. > It is either amusing or annoying the way that some of the posters here sometimes talk as if designers for major companies were ignorant fools when they made design decisions.
From: nmm1 on 14 Mar 2010 18:26 In article <804u0fF75pU1(a)mid.individual.net>, Del Cecchi <delcecchi(a)gmail.com> wrote: >"Stephen Fuld" <SFuld(a)alumni.cmu.edu.invalid> wrote in message >news:hnj8k4$tsd$1(a)news.eternal-september.org... >> >> Thanks to Tim and Terje for correcting my recollections. But >> despite my error, I still maintain the original point, that these >> decisions were made using good engineering judgment and not out of >> ignorance of caches, or certainly of different bus widths. >> >It is either amusing or annoying the way that some of the posters here >sometimes talk as if designers for major companies were ignorant fools >when they made design decisions. There have been such incidents! Normally because some dogmatists have got dominance and have simply denied the inconvenient facts that were pointed out to them (firstly by their more pragmatic colleagues and, later, by the real world). But I agree that it's far rarer than is often made out. Regards, Nick Maclaren.
From: Tim McCaffrey on 14 Mar 2010 23:49 In article <804u0fF75pU1(a)mid.individual.net>, delcecchi(a)gmail.com says... > >It is either amusing or annoying the way that some of the posters here >sometimes talk as if designers for major companies were ignorant fools >when they made design decisions. > > Oh, golly, I could go on for some time about questionable design decisions in both the 8088 and the IBM PC. Things that were either obvious at the time (as I recall), or it seems like they should have been obvious goofs. There are even contemporary examples were other companies didn't make those mistakes, so it isn't all 20/20 hindsight. - Tim
From: kenney on 15 Mar 2010 00:39 In article <hnhiqf$gap$1(a)news.eternal-september.org>, SFuld(a)alumni.cmu.edu.invalid (Stephen Fuld) wrote: > The reason they gave > was that system cost would have been too high with a 15 bit bus. The development of 16 bit I/O and buffer chips had badly lagged processor development. Using an 8 bit data bus meant that existing and much cheaper of the shelf components could be used. IIRC the only new chip in the PC was the 8088 all the others were in common use. Still there was also the rumour that an 8 bit bus was used to make sure the machine was slower than IBM mini-computers. Ken Young
From: Del Cecchi on 15 Mar 2010 10:43 "Tim McCaffrey" <timcaffrey(a)aol.com> wrote in message news:hnkanh$etu$1(a)USTR-NEWS.TR.UNISYS.COM... > In article <804u0fF75pU1(a)mid.individual.net>, delcecchi(a)gmail.com > says... >> > >>It is either amusing or annoying the way that some of the posters >>here >>sometimes talk as if designers for major companies were ignorant >>fools >>when they made design decisions. >> >> > > Oh, golly, I could go on for some time about questionable design > decisions in both the 8088 and the IBM PC. Things that were either > obvious at the time (as I recall), or it seems like they should have > been obvious goofs. > > There are even contemporary examples were other companies didn't > make > those mistakes, so it isn't all 20/20 hindsight. > > - Tim > Absolutely true that, as the politicians say, mistakes have been made, in design, architecture, and marketing by all kinds of people and companies. You should have seen the internal IBM forums in the early days of the PC, not to mention wonders like the PCjr. Sometimes those errors were due to lack of vision and foresight and some times they were just errors of commission or omission. And there were occasional outright screwups. But sometimes things that were legitmate design decisions driven by technical or economic factors are criticized as if they were made by dough heads when they were in fact made by intelligent people after evaluating the alternatives. Other companies and folks might have made different choices but that doesn't mean any of the choices are necessarily stupid or evil. del
First
|
Prev
|
Next
|
Last
Pages: 1 2 3 4 5 Prev: CFP: Informatics 2010: new date - until 15 March 2010 Next: Call for papers: HPCS-10, USA, July 2010 |