[ntp:questions] New Version of NTP Installer for Windows from Meinberg

Danny Mayer mayer at ntp.org
Wed Jun 3 11:37:51 UTC 2009

David J Taylor wrote:
> Dave Hart wrote:
>> David J Taylor:
>>> Does this new release include the improved interpolation developed
>>> by Dave Hart, and the Windows Vista/7 optimisations?
>> No, those are only in 4.2.5p162 and later.  See
>> http://bugs.ntp.org/216 Very soon a release candidate cycle for 4.2.6
>> should start, so it shouldn't be terribly long before a -stable NTP
>> release includes the new Windows interpolation code.  4.2.4p7 does
>> have substantial improvements vs 4.2.4p6 in the Windows serial code
>> used by, for example, the NMEA reference clock driver.  However,
>> lacking the 216 fix, it's not really a good choice for any upstream
>> server duty, though it's great as a client assuming better than 1
>> millisecond accuracy is unimportant.
> Thanks, Dave.  Looking forward to 4.2.6!
>>> You may know that I discovered a problem with Windows 2000 afer bug
>>> fix 1149. Is the Windows 2000 fix 1191 included?
>> Yes it is.  Heiko also identified and fixed in the Meinberg release a
>> problem building current NTP (-stable and -dev) using the now-ancient
>> Visual C++ 6 compiler.  Anyone attempting to build from source using
>> VC6 should see http://bugs.ntp.org/1196 for the solution.  I would
>> appreciate contact from anyone building NTP using VC6, as I believe
>> only Meinberg and possibly Danny Mayer are still using it with NTP.  I
>> hope to see support for VC6 dropped before too much longer, as it
>> requires some ugliness in the code to work around half-implemented 64-
>> bit scalar support.
>> Cheers,
>> Dave Hart
> That's good to hear.
> If it were possible to keep Visual Studio 2005 support that would be 
> "nice", as it might allow me to try building NTP here once again.  Not 
> eseential, though.

I use VS 2005 for most of the work but I make sure that we can still
support VC 6.0. I'm not sure what Dave means by half-implemented 64-bit
scalar support, so maybe he can elaborate.


This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.

More information about the questions mailing list