Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!nntp.coast.net!oleane!jussieu.fr!math.ohio-state.edu!howland.reston.ans.net!world1.bawave.com!news.clark.net!mr.net!newshub.tc.umn.edu!zib-berlin.de!news.tu-chemnitz.de!irz401!uriah.heep!news From: j@uriah.heep.sax.de (J Wunsch) Newsgroups: comp.unix.bsd.freebsd.misc Subject: Re: Data corruption on an ASUS P/I-P55TP4N motherboard (summary) Date: 26 May 1996 21:43:24 GMT Organization: Private BSD site, Dresden Lines: 32 Message-ID: <4oaj9s$3ac@uriah.heep.sax.de> References: <4mvvst$5jf@agate.berkeley.edu> <4n3s10$b2h@news.csie.nctu.edu.tw> <4ncg6e$ol1@vidar.diku.dk> <4nt94b$8me@agate.berkeley.edu> Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) NNTP-Posting-Host: localhost.heep.sax.de Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit X-Newsreader: knews 0.9.6 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E dan@math.berkeley.edu (Dan Strick) wrote: > I received several email responses and there were several followups. > Nobody knew of any general problems with the ASUS P/I-P55TP4N motherboard. Btw., there have been several reports about other problems with these boards (Bruce Evans did a great job in investigating it), but they were apparently unrelated to your problem. The biggest offender on this (and not only this) board is the UMC 8669F multi-IO chip. It features all the ``standard IO'' tasks, but at least its builtin 16550 UARTs and FDC are horribly broken. > There is an important moral to this story. Tell your motherboard vendors > that you will PAY EXTRA for error checking in main memory, external cache, > and I/O busses. Tell them that you are willing to BUY FROM SOMEONE ELSE > to get these features. Let them know that you care. Only then will they > begin to produce reliable products. Luckily enough, the successor of this chipset is now available. I'm now happily running a Tr*ton-II based board (P/I-P55T2P4). Finally, it supports parity SIMMs again, and at an expense of 10 ... 15 % of your memory bandwidth (according to Rodney Grimes), you even get ECC memory if you're using parity SIMMs. (ECC: you've got 64+8 bits of memory bus, thus you can correct 1-bit errors, and detect 2-bit errors.) -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-)