Subject: From: Date:

Re: [ECS] Security-IFC-A LEC1964.aol.com Fri, 10 Jul 1998 00:29:37 EDT

---

Ingo, thanks for the info.
      It's a viable option and one I will use if necessary.

     Hopefully, I won't be trying the patience of too many if I add some
additional comments. I have used similar programs at work for some years. One
thing that I had noticed that if the interval of the calls were the same the
amount of error was generally the same. That is, if we called daily we
corrected for a X seconds always positive or always negative depending on the
particular PC. I think this is to be expected with hardware design in this
area of most motherboards. So, as you might expect, we corrected for X seconds
manually each day and extended the need for access to the NBS. We saved on a
few phone calls and kept pretty good accuracy. It just didn"t seem logical to
call on something you already knew.
     Also, error will be accumulative. If the intervals between calls is seven
days you will have 7X seconds of error on day seven. If I adjust daily my
accumulative error on day seven should be one or two seconds which is worst
case for the entire week.
     There are two or three other was ways to correct for CMOS clock error and
I will do a little experimenting. I will probably go with something simple and
gets the job done.
     Hopefully, I'm not totally off base here.

Thanks to everyone,

Larry C.
      

---