On Fri, 19 Jan 2007 11:03:20 -0600, Eric Chevalier wrote: > >Come on, folks; let's have a little perspective on this issue! Perhaps >you can't find laptops with ECC memory because the non-ECC memory is >completely reliable for all _practical_ purposes? Am I running a risk >of data corruption on my laptop because of some random memory error >that goes undetected? Sure.
Perhaps it's because it's more expensive and not enough people are willing to pay extra for it. The best clue that I can think of is to look at your mainframe and find out how many memory errors it has corrected this month. If you're on good enough terms with your hardware guy, maybe he'll tell you. It's hard to find. They are not reported to EREP unless there is a _huge_ number of them. > >My employer's shiny new Z9 processor doesn't have to worry about that >particular risk! :-) > IIRC, modern processors have beefed up the ECC form the old single bit correction and duoble bit detection to include double bit correction. Do you really think that would have been necessary if modern memories are as reliable as you say? -- Tom Marchant ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
