[ntp:questions] Re: Initial Setting

mgrd mgrd at gmx.net
Mon Apr 25 22:29:57 UTC 2005


David Schwartz wrote:
> "mgrd" <mgrd at gmx.net> wrote in message 
>>>There's also the problem that RFC868 is internally inconsistent and not 
>>>sufficient to document a usable time protocol.
> 
> 
>>Hmm, I just read RFC868.
>>I can't get the inconsistency. Could you give me a hint?
>>But the insufficiency is obvious since it's no very detailed.
> 
> 
> "The most important thing is, the time is the number of seconds since 00:00 
> (midnight) 1 January 1900 GMT, such that the time 1 is 12:00:01 am on 1 
> January 1900 GMT; this base will serve until the year 2036."
> 
>     A 32-bit signed value can represent rougly plus or minus 2^31 seconds, 
> which is about 68 years. This would mean that this protocol can represent 
> times from about 1832 to 1968. The examples include:
> 
> "2,629,584,000 corresponds to 00:00 1 May 1983 GMT", and
> "-1,297,728,000 corresponds to 00:00 17 Nov 1858 GMT"
> 
>     Do you know any way for a 32-bit signed integer to get to 2.6 billion?

Thanks for pointing it out!

-- 
reply to usenet only



More information about the questions mailing list