[ntp:questions] Trying to use Dimension 4 time keeper
Brian.Inglis at SystematicSw.ab.ca
Mon Sep 9 03:52:50 UTC 2013
On 2013-09-08 11:51, W. eWatson wrote:
> On 9/7/2013 9:16 PM, David Taylor wrote:
>> On 07/09/2013 22:38, W. eWatson wrote:
>>> I'm told that Dimension 4 is an accurate time keeper.
>>> <http://www.thinkman.com/dimension4/> I downloaded it and installed it.
>>> It's freeware, but would appreciate a $10 donation. Oddly it comes w/o
>>> any description to use it. Am I missing something? I'm running it on my
>>> XP machine. Nothing is said about an OS.
>>> Out of curiosity I installed it on this Win7 PC. It acts like my XP
>>> install. That is, it has no useful help, what to do is not obvious.
D4 was used with dialup Internet on Win9x.
You set it up to run when you connected to the Internet.
You could add NTP servers chosen from the publicly available Stratum 2 servers
on NTP sites.
D4 uses legacy Windows .hlp.
You can download legacy WinHelp for XP and 7 from MS if you need to see that.
>> There are instructions for installing NTP here:
> That's a daunting install process. See my comments about Meinberg below.
>> Why would you want to install NTP rather than Dimension 4, or the
>> Windows own time service? See:
> My XP machine quickly loses accuracy. BTW, it needs to be pretty accurate, since
> I have a science application that runs on it 7/24. The XP is on a LAN of three
> PCs. Two are on Win7 in my den. The XP PC is in a building 120' from where I am
> now. The XP PC Data from the app is uploaded every day to a distant location,
> and saved in a DB. 1,000 miles from me here in N. Calif.
> I tried Meinberg for quite some time, but it flops fairly often. As I recall, my
> Meinberg software was not nearly as involved as shown in the link above. Given
> the XP is outside, I don't often go out there to check the time against an
> atomic clock. Consequently the data becomes erroneous.
You may want to try the latest stable NTP release 4.2.6p5 from Meinberg, for
ease of installation, on all three of your systems.
With stable connections and remote NTP servers it should get your time offset
within 100ms in a few hours, and within a few ms (or better depending on your
network and sources) over a period of about three weeks, and the drift within
PPB of your hardware clock over that period.
If you are in the US, you may want to select some servers from
http://www.ustiming.org/?page_id=458 or http://tf.nist.gov/tf-cgi/servers.cgi
and keep the pool servers in ntp.conf as backup in case some of your selections
pool CC.pool.ntp.org iburst minpoll 4 maxpoll 10 # CC country domain e.g us
Add parameters minpoll 4 maxpoll 10 after all your remote server iburst
parameter to allow the offset to quickly converge then automatically reduce the
If your systems have DNS names, static IP addresses, or a small range of LAN
DHCP addresses (typically 192.168.1.n) you can add restrict statements after
those already in your conf to allow you to monitor them all from any of them
(using ntpq and ntpdc) e.g. after usual:
restrict default limited kod nomodify notrap ...
restrict ::1 # IPV6 localhost6
# 192.168.1.1 usually router, DHCP server, default gateway, DNS server address
Once you have all your systems reliably running NTP, and if they have fairly
static IP addresses, you could select different servers in each conf and define
them as peers of each other (comment one out on each server):
peer 192.168.1.2 iburst minpoll 4 maxpoll 6
peer 192.168.1.3 iburst minpoll 4 maxpoll 6
peer 192.168.1.4 iburst minpoll 4 maxpoll 6
More information about the questions