Обсуждение: Timestamp precision in Windows and Linux
Hi,
The page http://www.postgresql.org/docs/7.2/static/datatype-datetime.html mentions that the resolution of all time and timestamp data types is 1 microsecond. I have an application that runs on both a Windows (XP with SP2) machine and a Linux (SUSE 10.2) machine. I saw that on postgres enterprisedb 8.3 installed on both these machines, the default timestamp precision on the former is upto a millisecond and on the latter it is 1 microsecond.
My curiosity is : is this a universal phenomenon ie a basic issue with Windows? Or could there be some hardware or architectural differences or something else...
And my problem is: is there any way to enforce a higher precision in Windows? Because my application badly needs it.
Please help / guide.
Thanks a million,
Shruthi
The page http://www.postgresql.org/docs/7.2/static/datatype-datetime.html mentions that the resolution of all time and timestamp data types is 1 microsecond. I have an application that runs on both a Windows (XP with SP2) machine and a Linux (SUSE 10.2) machine. I saw that on postgres enterprisedb 8.3 installed on both these machines, the default timestamp precision on the former is upto a millisecond and on the latter it is 1 microsecond.
My curiosity is : is this a universal phenomenon ie a basic issue with Windows? Or could there be some hardware or architectural differences or something else...
And my problem is: is there any way to enforce a higher precision in Windows? Because my application badly needs it.
Please help / guide.
Thanks a million,
Shruthi
Shruthi A <shruthi.iisc@gmail.com> writes: > The page http://www.postgresql.org/docs/7.2/static/datatype-datetime.htmlmentions > that the resolution of all time and timestamp data types is 1 > microsecond. I have an application that runs on both a Windows (XP with > SP2) machine and a Linux (SUSE 10.2) machine. I saw that on postgres > enterprisedb 8.3 installed on both these machines, the default timestamp > precision on the former is upto a millisecond and on the latter it is 1 > microsecond. I suppose what you're really asking about is not the precision of the datatype but the precision of now() readings. You're out of luck --- Windows just doesn't expose a call to get the wall clock time to better than 1 msec. Keep in mind that whatever the Linux machine is returning might be largely fantasy in the low-order bits, too. regards, tom lane