Re: get-real-time bug?

On May 25, 2005, at 6:42 PM, James M Grimes wrote:

> I haven't seen  a reply to this yet, so I'll put my two cents in,  
> and the real experts can correct me if I'm wrong.
>
> "get-real-time" returns an integer, or at least an integral value.   
> The manual says:
>
> get-real-time
>     Returns the number of seconds since January 1st, 1970 with  
> microsecond accuracy.
>
> which could imply a float return.  However, the "microsecond  
> accuracy" only means that it will flip to the next integer with  
> million to one accuracy.  I'm not sure what that means as far as  
> accuracy over longer durations, and obviously your computer would  
> have to be synched to world time to at least that accuracy to get  
> absolute time (but hey, we're running simulations, no?).

This explanation would allow me to cover up for a bug in my code, but  
alas, it is a bug -- get-real-time should be returning time with  
microsecond precision (the word I should have used instead of  
"accuracy", I suppose).

The problem has been fixed for the next release.

- jon klein

>> I'm not sure if this is a problem somewhere in my
>> code, but while "get-time" seems to update itself
>> properly, "get-real-time" seems to update the returned
>> value only once a second.

_______________________________________________
breve mailing list
[email protected]
http://lists.spiderland.org/mailman/listinfo/breve