Hacker News new | past | comments | ask | show | jobs | submit login

Oh, OK. I guess then I wouldn't say that upsets/bit/hour is an incorrect unit. (It's clearly what you want to know to calculate the chance of error for a given piece of RAM.) It's just that this parameter varies across time and manufacturers. Using the value from a particular model of RAM manufactured in 1988 is sure to lead to wrong conclusions.

Thanks.




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: