Subj: My Perspective on Pentium - AGS Path: math.fu-berlin.de!zib-berlin.de!news.th-darmstadt.de!terra.wiwi.uni-frankfurt.de!zeus.rbi.informatik.uni-frankfurt.de!news.dfn.de!Germany.EU.net!howland.reston.ans.net!ix.netcom.com!netnews From: rwirt@ix.netcom.com (Richard Wirt) Newsgroups: comp.sys.intel Subject: My Perspective on Pentium - AGS Date: 27 Nov 1994 19:31:21 GMT Organization: Netcom Lines: 102 Distribution: world Message-ID: <3bamq9$avt@ixnews1.ix.netcom.com> Andy Grove has asked me to post the following for him. Since it is the weekend and we are out of the office, I am posting from my home system. Richard Wirt Director SW Technology Intel Corp This is Andy Grove, president of Intel. I'd like to comment a bit on the conversations that have been taking place here. First of all, I am truly sorry for the anxiety created among you by our floating point issue. I read thru some of the postings and it's clear that many of you have done a lot of work around it and that some of you are very angry at us. Let me give you my perspective on what has happened here. The Pentium processor was introduced into the market in May of '93 after the most extensive testing program we at Intel have ever embarked on. Because this chip is three times as complex as the 486, and because it includes a number of improved floating point algorithms, we geared up to do an array of tests, validation, and verification that far exceeded anything we had ever done. So did many of our OEM customers. We held the introduction of the chip several months in order to give them more time to check out the chip and their systems. We worked extensively with many software companies to this end as well. We were very pleased with the result. We ramped the processor faster than any other in our history and encountered no significant problems in the user community. Not that the chip was perfect; no chip ever is. From time to time, we gathered up what problems we found and put into production a new "stepping" -- a new set of masks that incorporated whatever we corrected. Stepping N was better than stepping N minus 1, which was better than stepping N minus 2. After almost 25 years in the microprocessor business, I have come to the the conclusion that no microprocessor is ever perfect; they just come closer to perfection with each stepping. In the life of a typical microprocessor, we go thru half a dozen or more such steppings. Then, in the summer of '94, in the process of further testing (which continued thru all this time and continues today), we came upon the floating point error. We were puzzled as to why neither we nor anyone else had encountered this earlier. We started a separate project, including mathematicians and scientists who work for us in areas other than the Pentium processor group to examine the nature of the problem and its impact. This group concluded after months of work that (1) an error is only likely to occur at a frequency of the order of once in nine billion random floating point divides, and that (2) this many divides in all the programs they evaluated (which included many scientific programs) would require elapsed times of use that would be longer than the mean time to failure of the physical computer subsystems. In other words, the error rate a user might see due to the floating point problem would be swamped by other known computer failure mechanisms. This explained why nobody -- not us, not our OEM customers, not the software vendors we worked with and not the many individual users -- had run into it. As some of you may recall, we had encountered thornier problems with early versions of the 386 and 486, so we breathed a sigh of relief that with the Pentium processor we had found what turned out to be a problem of far lesser magnitude. We then incorporated the fix into the next stepping of both the 60 and 66 and the 75/90/100 MHz Pentium processor along with whatever else we were correcting in that next stepping. Then, last month Professor Nicely posted his observations about this problem and the hubbub started. Interestingly, I understand from press reports that Prof. Nicely was attempting to show that Pentium-based computers can do the jobs of big time supercomputers in numbers analyses. Many of you who posted comments are evidently also involved in pretty heavy duty mathematical work. That gets us to the present time and what we do about all this. We would like to find all users of the Pentium processor who are engaged in work involving heavy duty scientific/floating point calculations and resolve their problem in the most appropriate fashion including, if necessary, by replacing their chips with new ones. We don't know how to set precise rules on this so we decided to do it thru individual discussions between each of you and a technically trained Intel person. We set up 800# lines for that purpose. It is going to take us time to work thru the calls we are getting, but we will work thru them. I would like to ask for your patience here. Meanwhile, please don't be concerned that the passing of time will deprive you of the opportunity to get your problem resolved -- we will stand behind these chips for the life of your computer. Sorry to be so long-winded -- and again please accept my apologies for the situation. We appreciate your interest in the Pentium processor, and we remain dedicated to bringing it as close to perfection as possible. I will monitor your communications in the future -- forgive me if I can't answer each of you individually. Andy Grove -------------------------------------------------------------------------------- Subject: Re: *** The President of Intel responds... Newsgroups: comp.sys.intel Path: math.fu-berlin.de!zib-berlin.de!fauern!xlink.net!howland.reston.ans.net!gatech!psuvax1!rutgers!att-out!oucsace!bobcat!venkat From: venkat@bobcat.ent.ohiou.edu (Venkatesh Thirumalai) Subject: Re: *** The President of Intel responds... Message-ID: Sender: news@oucsace.cs.ohiou.edu (news account) Organization: Ohio University, College of Engineerng & Technology References: <3baumb$14c@carbon.cudenver.edu> Date: Mon, 28 Nov 1994 01:14:07 GMT Lines: 136 First of all to be CREDIBLE, please repost this from a machine in the Intel domain. If the problem is fixed as claimed by you (Intel) , release the codes on the cpus, after which the buggy Pentiums are no longer being produced. In all of that long winded post, Mr. Grove , could have included some substance. HOPING you guys would post that important information, here as well. If Intel is going to decide on a case by case basis, is that not going to lead to subjective judgements, (usually decided by you guys, in YOUR FAVOR) to save $$. Do you think this is ethical and FAIR.?? Venkat (Ph # 614.592.5357) (Fax # 614.592.6299) -------------------------------------------------------------------------------- Subject: Re: PENTIUM REPLACEMENT Path: math.fu-berlin.de!zib-berlin.de!news.th-darmstadt.de!fauern!xlink.net!howland.reston.ans.net!pipex!uunet!news.uiowa.edu!icaen!macpd From: macpd@icaen.uiowa.edu (Mac Administration) Newsgroups: comp.sys.intel Subject: Re: PENTIUM REPLACEMENT Date: 28 Nov 1994 06:52:25 GMT Organization: Iowa Computer Aided Engineering Network, University of Iowa Lines: 35 Message-ID: <3bbun9$l05@news.icaen.uiowa.edu> References: <3baq3e$5rm@jabba.cybernetics.net> <3bau6u$2r1@news.service.uci.edu> Trangdaithi Hoang (thoang@orion.oac.uci.edu) wrote: : Has ANYONE passed Intel's review board for a replacement Pentium CPU and : has gotten it? It looks to me like the only person who's gotten a : replacement Pentium without the FDIV bug is Prof. Nicely, one of the first : to discover the bug and now employed by Intel. I'm starting to wonder. : 1. No one has reported a Pentium without the FDIV bug. : 2. No one has been approved for or has gotten a replacement Pentium : without the FDIV bug other than Prof. Nicely. : Does the fixed Pentium without the FDIV bug even exist? Coud this be the : reason why Intel hasn't replaced anyone's Pentium and can't sell you : one right now because it doesn't exist? : -- : Prof. Nicely, : Could you be so kind as to post the markings on your fixed Pentium : professor(s) so that everyone wishing to purchase a Pentium system know : what to look for. : Thank You The "Pentium" Prof. Nicely has may be a board level prototype of the next Pentium chip, or other simulation. The following info from another note may be of interest: From: andrew@srsunc.shlrc.mq.edu.au (Andrew McVeigh) Newsgroups: comp.sys.intel Subject: Re: Intel Bug Fix Delay for Good Reason Message-ID: [unrelated lines before this one removed] I called the Australian distibutor for Intel in Australia and was told #### that the bug had not been corrected yet and would only be corrected in #### chips produced after March next year... #### [following lines removed for brevity] -------------------------------------------------------------------------------- Newsgroups: comp.sys.intel Path: math.fu-berlin.de!zib-berlin.de!news.th-darmstadt.de!terra.wiwi.uni-frankfurt.de!zeus.rbi.informatik.uni-frankfurt.de!news.dfn.de!Germany.EU.net!howland.reston.ans.net!ix.netcom.com!netcomsv!zodiac!szh From: szh@zcon.com (Syed Zaeem Hosain) Subject: Maybe it's gonna go down (was: Re: Intel stock goes up!) Message-ID: <1994Nov28.102447.9487@zcon.com> Sender: szh@zcon.com (Syed Zaeem Hosain) Reply-To: szh@zcon.com Organization: Z Consulting Group References: <3b5bqg$pom@news.bridge.com> Date: Mon, 28 Nov 1994 10:24:47 GMT Lines: 58 In article pom@news.bridge.com, thogard@bridge.com (Tim Hogard) writes: >Ramaswamy Krishnan (kg@thor.ece.uc.edu) wrote: >: In article <1994Nov22.133304.15463@mercury.ncat.edu>, >: wrote: >: >Despite the buggy pentiums there was a strong buy recommendation for intel >: >and it's stock went up yesterday....... > >: Of cousre... NYSE also uses the buggy chips :) > >Has anyone noticed that if you add the number of say the NYSE volumes for >say Intel (INTC), IBM, Microsoft (MSFT) and divide it by the sum of the >volumes of AMD, Cyrix (CYRX) on a Pentium, you might get the wrong answer. > >Every try to do a market value calculation with bad doubles? The typical >volumes in certain countries are in the billions. Take that times fractions >of a cent and then divide by a 7 digit currency conversion and you could >come out real wrong. > >Intel claims that this won't effect normal users, but it might effect those >taht have invested in them that use their CPUs. As a side note, I called up my stock broker at home today (Sunday). We both decided that it was time to get out of Intel stock for a while - just in case! So, when the NYSE opens tomorrow morning, we are gonna sell *all* the Intel stock that we both happen to have, at whatever the market value is at that time. I think he is also planning to contact his other clients and tell them to do the same. I have worked for semiconductor companies before, and have been in a few situations where the company had to swallow returns due to chip problems. It was always done by telling the customers about problems when we found them out, and replacing the chips. Yeah, one time, it caused earnings to be half what was expected for a couple of quarters, and the stock took the hit. But that was the right thing to have done in each and every case. Personally, I may stay out of Intel stock for a long, long time. I would have been willing to hang in there, but I find their attitude about replacing faulty chips totally incomprehensible and unacceptable for a semiconductor company. If this is the way they choose to operate, very unlike other semiconductor companies in the market, I can choose not to deal with them anymore. I was about to buy a Pentium 90 or 100 MHz motherboard to update my 486/33 system and for testing some CAD software for one of my clients (who were going to get some Pentium systems for schematic capture and simulation, but I will be recommending Sun instead). For my own system, I think I will spend the money on upgrading my Sun workstation instead of squandering it on a Pentium motherboard. -- ------------------------------------------------------------------------- | Syed Zaeem Hosain P. O. Box 610097 (408) 441-7021 | | Z Consulting Group San Jose, CA 95161 szh@zcon.com | ------------------------------------------------------------------------- -------------------------------------------------------------------------------- Subject: Re: *** The President of Intel responds... Path: math.fu-berlin.de!zib-berlin.de!news.th-darmstadt.de!terra.wiwi.uni-frankfurt.de!zeus.rbi.informatik.uni-frankfurt.de!news.dfn.de!swiss.ans.net!howland.reston.ans.net!gatech!swrinde!ringer.cs.utsa.edu!lonestar.jpl.utsa.edu!yfraiman From: "Yaron . Fraiman " Newsgroups: comp.sys.intel Subject: Re: *** The President of Intel responds... Date: Mon, 28 Nov 1994 12:29:51 -0600 Organization: The University of Texas at San Antonio Lines: 8 Message-ID: References: <3baumb$14c@carbon.cudenver.edu> all nice and well but one question is still un-answered: why once the "flaw" was detected intel kept selling these "flawed" processors to innocent customers who now have to be humiliated when asking for what they paid for? thank you yfraiman@lonestar.jpl.utsa.edu -------------------------------------------------------------------------------- Subject: Re: Morality vs. reality. Newsgroups: comp.sys.intel Path: math.fu-berlin.de!zib-berlin.de!news.th-darmstadt.de!terra.wiwi.uni-frankfurt.de!zeus.rbi.informatik.uni-frankfurt.de!news.dfn.de!Germany.EU.net!howland.reston.ans.net!news.sprintlink.net!EU.net!uunet!hearst.acc.Virginia.EDU!murdoch!delton.psyc.virg From: smb3u@delton.psyc.virginia.edu (Steven M. Boker) Subject: Re: Morality vs. reality (was Re: P24T chip: When? When? When?) Message-ID: Sender: usenet@murdoch.acc.Virginia.EDU Organization: University of Virginia, Department of Psychology References: <3b8sau$kpt@crl6.crl.com> <3baaau$js8@falcon.ccs.uwo.ca> Date: Mon, 28 Nov 1994 17:07:47 GMT Lines: 49 [ snipped ] As a consultant I talk to many different kinds of users. One was just in my office today. She had heard about the Pentium bug and was wondering what a Pentium was. I told her it was a chip made by Intel which was in most of the IBM-PC compatible computers. She said, "Oh, its a problem with all IBM computers." I said no, only computers with an Intel Pentium chip in them. She said, "Oh, then I should avoid all those computers with the Intel Inside sticker on them." She then walked out of my office. This scene is being replayed around the country. The vast majority of the buying public will never know (or care to know) that a mantissa isn't some kind of praying insect. But they will learn to avoid the "Intel Inside" sticker. If Intel waits until the intangibles are outweighing the cost of replacement, it will be far, far too late. You must take into account the resistance against good news. Bad news travels fast, good news mostly doesn't sell newspapers. It will take massive amounts of advertising dollars to repair the damage that is occurring at a frightening rate. And Intel is taking the old Nixonian stonewall stance. Its a sad posture, and a sad day for Intel stockholders. Steve -- #====#====#====#====#====#====#====#====#====#====#====#====#====# # Steven M. Boker # "Two's bifurcation # # boker@virginia.edu # but three's chaotic" # #====#====#====#====#====#====#====#====#====#====#====#====#====# -------------------------------------------------------------------------------- Path: math.fu-berlin.de!zib-berlin.de!news.th-darmstadt.de!terra.wiwi.uni-frankfurt.de!zeus.rbi.informatik.uni-frankfurt.de!news.dfn.de!Germany.EU.net!EU.net!howland.reston.ans.net!swrinde!pipex!uunet!mozz.unh.edu!christa.unh.edu!ces From: ces@christa.unh.edu (Christan E Schoenfeld) Newsgroups: comp.sys.intel Subject: Re: FDIV bug in Pentium chips! Date: 28 Nov 1994 17:01:04 GMT Organization: University of New Hampshire - Durham, NH Lines: 18 Message-ID: <3bd2cg$3ig@mozz.unh.edu> References: <39vh32$jgj@nntpd.lkg.dec.com> <3a22p5$b0b@abyss.West.Sun.COM> <3a8na6$f40@hpbab.wv.mentorg.COM> <1994Nov21.065616.19455@cs.cornell.edu> I just wanted to add a little anecdote to this thread for all those people who are arguing that this is not a critical bug: Remember about a year or two ago when the shuttle astronauts were having a hell of a time lassoing a satellite into the cargo bay? That was entirely due to the fact that one of the computer programmers had inadvertantly used a float when he should have used a double. The bug was caught and fixed, but if they had been using a Pentium, and the FDIV occured, there would have not been any code errors to find, and they'd still be trying to grab that thing today ;-). -- Chris Schoenfeld | "If you have the means, I highly Communications/Political Science Senior | recommend it." University of New Hampshire ces@unh.edu | -Ferris Beuler -------------------------------------------------------------------------------- "I n t e l i n s i d e, b u t c a n i t d i v i d e ?" --------------------------------------------------------------------------------