Possibly off topic - Jitter on Ethernet over power adapters
I'm not sure if this is the best place to ask the question, but does anyone
have experience of using Ethernet over power line adapters? I have an
outside office, and my router is in the house plugged into the phone master
socket. I have used two Ethernet over power adapters, one at the router and
one in the office here to get internet access. The output of the adapter
then goes to a multi-port hub to give me Ethernet to all my office devices
including two Meinberg NTP servers.
I've noticed large jitter readings on Meinberg's NTP monitor program. Can
be as low as 2ms, but much higher (50mS +), and at this point NTP goes
haywire.
Not sure if it is the physical set up or something else.
Any comments appreciated.
Thanks.
Rob
From: Rob Kimberley
[]
I'm not sure if this is the best place to ask the question, but does anyone
have experience of using Ethernet over power line adapters? I have an
outside office, and my router is in the house plugged into the phone master
socket. I have used two Ethernet over power adapters, one at the router and
one in the office here to get internet access. The output of the adapter
then goes to a multi-port hub to give me Ethernet to all my office devices
including two Meinberg NTP servers.
I've noticed large jitter readings on Meinberg's NTP monitor program. Can
be as low as 2ms, but much higher (50mS +), and at this point NTP goes
haywire.
Not sure if it is the physical set up or something else.
Any comments appreciated.
Thanks.
Rob
---=
Rob,
It's not quite clear which direction you are measuring. I take it your
Meinberg servers are "perfect" in NTP terms, and you are monitoring from the
house? Or vice-versa? Anyway, my first guess is that jitter might be not
dissimilar to Wi-Fi, in which case my lightly-loaded Wi-Fi results might be
a starting point:
http://www.satsignal.eu/mrtg/performance_ntp_wifi.php
Note the improvement with Windows-8 and the latest NTP (top graph, PC
Bergen), and the others are somewhat variable.
Cheers,
David
SatSignal Software - Quality software written to your requirements
Web: http://www.satsignal.eu
Email: david-taylor@blueyonder.co.uk
THose power over Ethernet devices work with analog signals and don't
evn look at the data packets. All they do is place a DC bias on the
twisted pair. Ethernet is always transformer coupled so your
routers, switches and computers never see DC.
What is your NTP server using for a reference clock? I'd suspect that
is the problem. If the reference is an Internet pool server than a
few mS is about what you should expect. If using GPS then look to
see if you have a good signal from enough satellites.
But those POE boxes don't mess with the data packets, or at least the
are not designed to do that. If one is broken it could be adding
noise to the line. Broken hardware can do "anything".
On Sun, Feb 10, 2013 at 3:18 AM, Rob Kimberley
robkimberley@btinternet.com wrote:
I'm not sure if this is the best place to ask the question, but does anyone
have experience of using Ethernet over power line adapters? I have an
outside office, and my router is in the house plugged into the phone master
socket. I have used two Ethernet over power adapters, one at the router and
one in the office here to get internet access. The output of the adapter
then goes to a multi-port hub to give me Ethernet to all my office devices
including two Meinberg NTP servers.
I've noticed large jitter readings on Meinberg's NTP monitor program. Can
be as low as 2ms, but much higher (50mS +), and at this point NTP goes
haywire.
Not sure if it is the physical set up or something else.
Any comments appreciated.
Thanks.
Rob
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
--
Chris Albertson
Redondo Beach, California
I'm using Meinberg GPS NTP Servers. They are working fine and no problem
with SVs.
Still doing some tests. Will report back once I have anything extra to add.
Thanks.
Rob
-----Original Message-----
From: time-nuts-bounces@febo.com [mailto:time-nuts-bounces@febo.com] On
Behalf Of Chris Albertson
Sent: 10 February 2013 16:15
To: Discussion of precise time and frequency measurement
Subject: Re: [time-nuts] Possibly off topic - Jitter on Ethernet over power
adapters
THose power over Ethernet devices work with analog signals and don't
evn look at the data packets. All they do is place a DC bias on the
twisted pair. Ethernet is always transformer coupled so your
routers, switches and computers never see DC.
What is your NTP server using for a reference clock? I'd suspect that
is the problem. If the reference is an Internet pool server than a
few mS is about what you should expect. If using GPS then look to
see if you have a good signal from enough satellites.
But those POE boxes don't mess with the data packets, or at least the
are not designed to do that. If one is broken it could be adding
noise to the line. Broken hardware can do "anything".
On Sun, Feb 10, 2013 at 3:18 AM, Rob Kimberley robkimberley@btinternet.com
wrote:
I'm not sure if this is the best place to ask the question, but does
anyone have experience of using Ethernet over power line adapters? I
have an outside office, and my router is in the house plugged into the
phone master socket. I have used two Ethernet over power adapters, one
at the router and one in the office here to get internet access. The
output of the adapter then goes to a multi-port hub to give me
Ethernet to all my office devices including two Meinberg NTP servers.
I've noticed large jitter readings on Meinberg's NTP monitor program.
Can be as low as 2ms, but much higher (50mS +), and at this point NTP
goes haywire.
Not sure if it is the physical set up or something else.
Any comments appreciated.
Thanks.
Rob
time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
--
Chris Albertson
Redondo Beach, California
time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
From: Chris Albertson
[]
THose power over Ethernet devices work with analog signals and don't
evn look at the data packets. All they do is place a DC bias on the
twisted pair. Ethernet is always transformer coupled so your
routers, switches and computers never see DC.
[]
Chris - I think the query was about power-line networking, not PoE.
David
SatSignal Software - Quality software written to your requirements
Web: http://www.satsignal.eu
Email: david-taylor@blueyonder.co.uk
Rob,
One of the common characteristics of power lines is noise.
Seems to me that bursts of noise could interrupt the Ethernet signal,
causing retries of the transmission.
Now, I'm only familiar with SNTP, which uses UDP messages (User Datagram
Protocol). The more familiar TCP will retry messages within the protocol,
but UDP does not. It relies on the application to retry. Perhaps the
Ethernet to power line boxes are smart enough to retry after noise bursts.
Can you monitor your power line for noise? Do you see any pattern to the
50 millisecond delays?
Is there a way to tell NTP to discard the data and retry if the round trip
time is unreasonable?
Can you run a fiber optic pair to your office? How about internet access
with a cell phone?
Hope there was something useful in all that.
Bill Hawkins
P.S. The Meinberg article at their site says that NTP and SNTP both use
TCP/IP. I know that SNTP uses UDP/IP, so perhaps they are confused. TCP
(Transmission Control Protocol) requires a request/confirm / indication/
/response handshake using those protocol primitives. UDP simply sends a
message in the hope that it will get to the receiver. When it works, you
get good time data, but you get garbage when it doesn't. The TCP handshake
would be coumterproductive for NTP.
Here is the original from Rob Kimberly, for those with short memories:
"I'm not sure if this is the best place to ask the question, but does anyone
have experience of using Ethernet over power line adapters? I have an
outside office, and my router is in the house plugged into the phone master
socket. I have used two Ethernet over power adapters, one at the router and
one in the office here to get internet access. The output of the adapter
then goes to a multi-port hub to give me Ethernet to all my office devices
including two Meinberg NTP servers.
I've noticed large jitter readings on Meinberg's NTP monitor program. Can
be as low as 2ms, but much higher (50mS +), and at this point NTP goes
haywire.
Not sure if it is the physical set up or something else.
Any comments appreciated.
Thanks.
Rob"
The poster is asking about ethernet over power line and not power over
ethernet. As you point out, the later should have zero effect on
ethernet latency.
There are several ethernet over power standards. Latency will include
a bridge in each adapter, the effects of a noisy uncontrolled AC power
line when ARQ (automatic repeat query) is used, and the TDMA or CSMA
media access control system.
I suspect varying power line conditions will have the largest effect
because any jitter from the media access control system will be
multiplied by ARQ.
On Sun, 10 Feb 2013 08:14:38 -0800, Chris Albertson
albertson.chris@gmail.com wrote:
THose power over Ethernet devices work with analog signals and don't
evn look at the data packets. All they do is place a DC bias on the
twisted pair. Ethernet is always transformer coupled so your
routers, switches and computers never see DC.
What is your NTP server using for a reference clock? I'd suspect that
is the problem. If the reference is an Internet pool server than a
few mS is about what you should expect. If using GPS then look to
see if you have a good signal from enough satellites.
But those POE boxes don't mess with the data packets, or at least the
are not designed to do that. If one is broken it could be adding
noise to the line. Broken hardware can do "anything".
On Sun, Feb 10, 2013 at 3:18 AM, Rob Kimberley
robkimberley@btinternet.com wrote:
I'm not sure if this is the best place to ask the question, but does anyone
have experience of using Ethernet over power line adapters? I have an
outside office, and my router is in the house plugged into the phone master
socket. I have used two Ethernet over power adapters, one at the router and
one in the office here to get internet access. The output of the adapter
then goes to a multi-port hub to give me Ethernet to all my office devices
including two Meinberg NTP servers.
I've noticed large jitter readings on Meinberg's NTP monitor program. Can
be as low as 2ms, but much higher (50mS +), and at this point NTP goes
haywire.
Not sure if it is the physical set up or something else.
Any comments appreciated.
Thanks.
Rob
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
My gut feeling would be that overall noise/power/length of run etc is
going to be a significant factor, too - ie, longer runs or noisier
power environments will have an impact.
As with all things sensitive, it's best to isolate things - I have yet
to see any ethernet over powerline installations that couldn't be done
with 30m of category 5 cable and 10 minutes with a drill (or lots of
internal cable pins). Given that, my instinct would be to replace the
ethernet over power boxes with real cables. Another upshot: You'll
make any radio hams nearby very happy.
Cheers,
James
On 10/02/2013 19:00, David wrote:
The poster is asking about ethernet over power line and not power
over ethernet. As you point out, the later should have zero effect
on ethernet latency.
There are several ethernet over power standards. Latency will
include a bridge in each adapter, the effects of a noisy
uncontrolled AC power line when ARQ (automatic repeat query) is
used, and the TDMA or CSMA media access control system.
I suspect varying power line conditions will have the largest
effect because any jitter from the media access control system will
be multiplied by ARQ.
On Sun, 10 Feb 2013 08:14:38 -0800, Chris Albertson
albertson.chris@gmail.com wrote:
THose power over Ethernet devices work with analog signals and
don't evn look at the data packets. All they do is place a DC
bias on the twisted pair. Ethernet is always transformer
coupled so your routers, switches and computers never see DC.
What is your NTP server using for a reference clock? I'd suspect
that is the problem. If the reference is an Internet pool
server than a few mS is about what you should expect. If using
GPS then look to see if you have a good signal from enough
satellites.
But those POE boxes don't mess with the data packets, or at least
the are not designed to do that. If one is broken it could be
adding noise to the line. Broken hardware can do "anything".
On Sun, Feb 10, 2013 at 3:18 AM, Rob Kimberley
robkimberley@btinternet.com wrote:
I'm not sure if this is the best place to ask the question, but
does anyone have experience of using Ethernet over power line
adapters? I have an outside office, and my router is in the
house plugged into the phone master socket. I have used two
Ethernet over power adapters, one at the router and one in the
office here to get internet access. The output of the adapter
then goes to a multi-port hub to give me Ethernet to all my
office devices including two Meinberg NTP servers.
I've noticed large jitter readings on Meinberg's NTP monitor
program. Can be as low as 2ms, but much higher (50mS +), and
at this point NTP goes haywire.
Not sure if it is the physical set up or something else.
Any comments appreciated.
Thanks.
Rob
_______________________________________________ time-nuts mailing
list -- time-nuts@febo.com To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow
the instructions there.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.17 (MingW32)
iEYEARECAAYFAlEX9ncACgkQ22kkGnnJQAwcRACcC+nnF/iN4sPZ3S25FX4y7WRS
VSEAn2fHeSA/4Ri6ZjJgdUek/y6xizGC
=wM2v
-----END PGP SIGNATURE-----
Hi Bill,
On 02/10/2013 07:44 PM, Bill Hawkins wrote:
P.S. The Meinberg article at their site says that NTP and SNTP both use
TCP/IP. I know that SNTP uses UDP/IP, so perhaps they are confused. TCP
(Transmission Control Protocol) requires a request/confirm / indication/
/response handshake using those protocol primitives. UDP simply sends a
message in the hope that it will get to the receiver. When it works, you
get good time data, but you get garbage when it doesn't. The TCP handshake
would be coumterproductive for NTP.
You should read "TCP/IP" as "Internet Protocols" (notice plural form
here). It points to the stack of protocols, which includes stuff like
ARP etc without it also being mentioned. It is common for dual uses like
that. UDP comes in by association in the TCP/IP suite of protocols.
NTP and SNTP uses UDP.
Cheers,
Magnus
On Sun, 10 Feb 2013 12:44:57 -0600, "Bill Hawkins" bill@iaxs.net
wrote:
Rob,
One of the common characteristics of power lines is noise.
Seems to me that bursts of noise could interrupt the Ethernet signal,
causing retries of the transmission.
Now, I'm only familiar with SNTP, which uses UDP messages (User Datagram
Protocol). The more familiar TCP will retry messages within the protocol,
but UDP does not. It relies on the application to retry. Perhaps the
Ethernet to power line boxes are smart enough to retry after noise bursts.
The power line is definitely a hostile environment for data
transmission. The recent ethernet over power line standards use ARQ
(Automatic Repeat Query) as a backup for the FEC (forward error
correction) so when data cannot be recovered at the receiver, it will
be resent.
While the adapters will certainly resent corrupted TCP data, I do not
know if they do the same for UDP data. The marketing makes noise
about special support for streaming content but who knows what that
means.
The standards also include both TDMA and CSMA in some combination
because they are half duplex. Those by themselves will add latency
and jitter.
On Sun, Feb 10, 2013 at 3:49 PM, David davidwhess@gmail.com wrote:
On Sun, 10 Feb 2013 12:44:57 -0600, "Bill Hawkins" bill@iaxs.net
wrote:
Rob,
One of the common characteristics of power lines is noise.
Seems to me that bursts of noise could interrupt the Ethernet signal,
causing retries of the transmission.
It is unlikely to add much noise. The PoE device only puts a DC bias
on the twisted pair. The data signal is differential. It is
transformer couple to is pretty much is immune to common mode noise.
So even iif the DC bias was noisy I don't thing it would matter.
Chris Albertson
Redondo Beach, California
It is unlikely to add much noise. The PoE device only puts a DC bias
on the twisted pair. The data signal is differential. It is
transformer couple to is pretty much is immune to common mode noise.
So even iif the DC bias was noisy I don't thing it would matter.
Chris Albertson
Redondo Beach, California
---===
Chris,
This isn't PoE we're discussing. It's data over "mains wiring". 110 V AC
to you, 240 V AC to me. Homeplug.
Cheers,
David
SatSignal Software - Quality software written to your requirements
Web: http://www.satsignal.eu
Email: david-taylor@blueyonder.co.uk
Ahem!
Ethernet over powerline!
NOT PoE!
The various forms of ether over power are (for practical purposes) a
wireless ethernet protocol coupled into the AC wiring.
And yes, it is noisy timing wise, for all the same reason that a
simplex/shared variable rate 802.11 system is.
On 2/10/13 9:15 PM, David J Taylor wrote:
It is unlikely to add much noise. The PoE device only puts a DC bias
on the twisted pair. The data signal is differential. It is
transformer couple to is pretty much is immune to common mode noise.
So even iif the DC bias was noisy I don't thing it would matter.
Chris Albertson
Redondo Beach, California
---===
Chris,
This isn't PoE we're discussing. It's data over "mains wiring". 110 V AC
to you, 240 V AC to me. Homeplug.
Cheers,
David
All,
Strangely today the jitter numbers seem to be behaving themselves! Nothing
has been done to the set up.
Thanks to everyone for their comments. I'm looking at putting a direct
Ethernet cable in here to see what the difference might make.
Attached is a picture of latest NTP Monitor readout. Bottom two devices are
the Meinberg LanTimes. The others are pool servers.
Rob
-----Original Message-----
From: time-nuts-bounces@febo.com [mailto:time-nuts-bounces@febo.com] On
Behalf Of David
Sent: 10 February 2013 23:49
To: Discussion of precise time and frequency measurement
Subject: Re: [time-nuts] Possibly off topic - Jitter on Ethernet over power
adapters
On Sun, 10 Feb 2013 12:44:57 -0600, "Bill Hawkins" bill@iaxs.net
wrote:
Rob,
One of the common characteristics of power lines is noise.
Seems to me that bursts of noise could interrupt the Ethernet signal,
causing retries of the transmission.
Now, I'm only familiar with SNTP, which uses UDP messages (User
Datagram Protocol). The more familiar TCP will retry messages within
the protocol, but UDP does not. It relies on the application to retry.
Perhaps the Ethernet to power line boxes are smart enough to retry after
noise bursts.
The power line is definitely a hostile environment for data transmission.
The recent ethernet over power line standards use ARQ (Automatic Repeat
Query) as a backup for the FEC (forward error
correction) so when data cannot be recovered at the receiver, it will be
resent.
While the adapters will certainly resent corrupted TCP data, I do not know
if they do the same for UDP data. The marketing makes noise about special
support for streaming content but who knows what that means.
The standards also include both TDMA and CSMA in some combination because
they are half duplex. Those by themselves will add latency and jitter.
time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
All,
Strangely today the jitter numbers seem to be behaving themselves! Nothing
has been done to the set up.
Thanks to everyone for their comments. I'm looking at putting a direct
Ethernet cable in here to see what the difference might make.
Attached is a picture of latest NTP Monitor readout. Bottom two devices are
the Meinberg LanTimes. The others are pool servers.
Rob
---==
Rob,
Two comments from your screen-shot.
-
NTP version 4.2.4p8 is now rather old. You may well find better
performance using a more recent version, and especially with the most recent
development versions on Windows-7 and Windows-Vista. There are some upgrade
notes on my Web site. -
with a local stratum-1 server, you can get better performance using a more
frequent poll i.e. limit maxpoll. For my PCs I typically have (with NTP
4.2.6 or later):
Local stratum-1 LAN-based servers
server 192.168.0.3 iburst minpoll 5 maxpoll 5 prefer # Pixie
server 192.168.0.2 iburst minpoll 5 maxpoll 5 # Feenix
External servers from the UK pool
pool uk.pool.ntp.org minpoll 10
For measurement, you may want to let the poll interval drift to 1024s, but
for lower offsets keeping poll to 32s (maxpoll 5) gives better results.
My apologies if you were already aware of this.
Cheers,
David
SatSignal Software - Quality software written to your requirements
Web: http://www.satsignal.eu
Email: david-taylor@blueyonder.co.uk
-----Original Message-----
From: time-nuts-bounces@febo.com [mailto:time-nuts-bounces@febo.com] On
Behalf Of David J Taylor
Sent: 11 February 2013 16:13
To: 'Discussion of precise time and frequency measurement'
Subject: Re: [time-nuts] Possibly off topic - Jitter on Ethernet overpower
adapters
All,
Strangely today the jitter numbers seem to be behaving themselves! Nothing
has been done to the set up.
Thanks to everyone for their comments. I'm looking at putting a direct
Ethernet cable in here to see what the difference might make.
Attached is a picture of latest NTP Monitor readout. Bottom two devices are
the Meinberg LanTimes. The others are pool servers.
Rob
---==
Rob,
Two comments from your screen-shot.
-
NTP version 4.2.4p8 is now rather old. You may well find better
performance using a more recent version, and especially with the most recent
development versions on Windows-7 and Windows-Vista. There are some upgrade
notes on my Web site. -
with a local stratum-1 server, you can get better performance using a more
frequent poll i.e. limit maxpoll. For my PCs I typically have (with NTP
4.2.6 or later):
Local stratum-1 LAN-based servers
server 192.168.0.3 iburst minpoll 5 maxpoll 5 prefer # Pixie
server 192.168.0.2 iburst minpoll 5 maxpoll 5 # Feenix
External servers from the UK pool
pool uk.pool.ntp.org minpoll 10
For measurement, you may want to let the poll interval drift to 1024s, but
for lower offsets keeping poll to 32s (maxpoll 5) gives better results.
My apologies if you were already aware of this.
Cheers,
David
David,
Many thanks for that. I do have a copy of 6.2.6 which I have tried, but the
jitter numbers appeared higher. I will try again and also change the minpoll
settings as suggested.
The settings shown were the default settings from Meinberg and I haven't
tweaked them.
Will see what I get.
Rob
On 2/10/2013 6:04 PM, Magnus Danielson wrote:
You should read "TCP/IP" as "Internet Protocols" (notice plural form
here). It points to the stack of protocols,
Actually, no. IP is Internet Protocol, singular, and is the L3 (mostly -
IP predates the ISO/OSI model layers, so IP suite protocols don't map
exactly) protocol upon which both TCP and UDP are built. It's defined by
RFC 791.
TCP/IP, simply because those are the most commonly used protocols in the
suite.
Thanks for the details, Mike.
I read TCP/IP as TCP over IP, because TCP can be used with any
data link layer that doesn't guarantee delivery.
OTOH, the first book I read about the Internet protocols was titled
"TCP/IP" so there is a tendency to lump then together. I've read that
the developers of IP were going to stop at the data link layer, and
then decided that it wasn't a good idea to let each user develop their
own scheme for verifying message delivery. TCP introduced ports so
that one IP address could have 64000 sub-addresses. UDP, while not
verified, does use ports.
Then there all of those other application layer protocols designed to
work with IP, such as ARP, DHCP, NTP, SNMP, and SNTP. Each has one or
more of the 256 ports reserved for use with IP by the designers. Mills
chose port 123, probably because it suited his sense of humor.
Bill Hawkins
-----Original Message-----
From: Mike S
Sent: Tuesday, February 12, 2013 1:20 PM
On 2/10/2013 6:04 PM, Magnus Danielson wrote:
You should read "TCP/IP" as "Internet Protocols" (notice plural form
here). It points to the stack of protocols,
Actually, no. IP is Internet Protocol, singular, and is the L3 (mostly -
IP predates the ISO/OSI model layers, so IP suite protocols don't map
exactly) protocol upon which both TCP and UDP are built. It's defined by
RFC 791.
TCP/IP, simply because those are the most commonly used protocols in the
suite.
David,
Thanks for posting that. I'm currently doing some testing over wifi
links myself, and found that page very useful. You do a really good job
documenting your experiences with GPS-based NTP refclocks, and I
appreciate all the hard work.
I just wanted to ask though, are you compiling your own NTP now or what?
http://www.davehart.net/ntp/win/x86.please.see.http.support.ntp.org-people-hart-win-x86/
but then under http://support.ntp.org/people/hart/win/x86/
... most recent seems to be ntp-4.2.7p310
^Basically, you were previously documenting use of dave hart's builds
(overlaid over a meinberg ntp install or otherwise)
Sorry rob, I don't have any experience with powerline adapters, but I'm
treating your experiences (which don't seem to be promising) as a data
point showing that they're no better for timing than using wifi...
I'm getting high & unpredictable jitter with NTP over wifi as well
(compared to cat 6 RJ-45 crossover cable directly between NTP servers)
1-20ms jitter for 5ghz band, 802.11n connection running with bitrate
manually limited to 6mbit/s
5-70ms jitter for the 2.4ghz band, 802.11g connection running with
bitrate auto-negotiation (up to 54mbit/s)
... My best case scenario for NTP jitter is about 0-5ms between a
stratum 1 and a stratum 2 server directly connected via gigabit ethernet
crossover (and the stratum 1 itself with a connected refclock seems to
be at a baseline of 0-1ms most of the time, and rarely higher than 2ms)
Those are just rough estimates based on casual observation, and I
haven't done any long-term measurements yet like David Taylor's work...
I'm "getting there" though it's coming slowly because of my trouble with
learning curve in this area.
--Sarah
P.S. I renamed this post. The title "Possibly off topic - Jitter on
Ethernet over poweradapters" seemed silly.
Rob,
It's not quite clear which direction you are measuring. I take it your
Meinberg servers are "perfect" in NTP terms, and you are monitoring from
the house? Or vice-versa? Anyway, my first guess is that jitter might
be not dissimilar to Wi-Fi, in which case my lightly-loaded Wi-Fi
results might be a starting point:
http://www.satsignal.eu/mrtg/performance_ntp_wifi.php
Note the improvement with Windows-8 and the latest NTP (top graph, PC
Bergen), and the others are somewhat variable.
Cheers,
David
David,
[]
I just wanted to ask though, are you compiling your own NTP now or what?
http://www.davehart.net/ntp/win/x86.please.see.http.support.ntp.org-people-hart-win-x86/
but then under http://support.ntp.org/people/hart/win/x86/
... most recent seems to be ntp-4.2.7p310
^Basically, you were previously documenting use of dave hart's builds
(overlaid over a meinberg ntp install or otherwise)
[]
--Sarah
Yes, Sarah, with initial help from Dave Hart and later help from more
C-fluent friends on the various groups, I have managed to compile NTP from
the available source code - details here:
http://www.satsignal.eu/ntp/setup.html#build
To try and offer something in return, I make my recent builds available
here:
http://www.satsignal.eu/ntp/x86/
although they won't work on Windows-2000 and earlier due to limitations on
the MS Visual Studio 2010 Express I am using. You can overlay my ntp*.exe
over a Meinberg install. For Windows-Vista and Windows-7 there have been
some significant improvements in the most recent builds, and Windows-8 can
provide even more improvements, nearer to Linux and FreeBSD than previous
versions of Windows, but still not as good.
Dave Hart has been involved in a house move recently, so he has been rather
quiet.
Cheers,
David
SatSignal Software - Quality software written to your requirements
Web: http://www.satsignal.eu
Email: david-taylor@blueyonder.co.uk