Date
1 - 20 of 42
Locked Missing decodes #AudioIssues
|
Hi all,
Over the last few days, I have noticed that WSJTx isn't decoding some time slots. This appears to be random. I am using an IC7800 via a microHam MKII to an AMD A8-9600 Radeon R7 with 16GB RAM running 64bit Win10Pro. I looked in the WSJTx folder, and checked the wsjtx_syslog.txt I see the following lines: [SYSLOG][2022-10-07 13:04:29.431052][00:36:11.110627][warning] Detected dropped audio source samples: 1920 (0.04 S) [SYSLOG][2022-10-07 13:05:29.378859][00:37:11.055690][warning] Detected dropped audio source samples: 1104 (0.023 S) [SYSLOG][2022-10-07 13:05:59.410405][00:37:41.085883][warning] Detected dropped audio source samples: 1008 (0.021 S) [SYSLOG][2022-10-07 13:06:29.389255][00:38:11.039565][warning] Detected dropped audio source samples: 1104 (0.023 S) Can anyone suggest where to look, or what to change to reduce this issue? The Windows Resource Manager isn't showing any issues, so I am not sure what is happening. 73 de Phil GU0SUP |
|
|
|
Karza <kari.sillanmaki@...>
Hi Phil,
On 7.10.2022 18.03, Phil Cooper via groups.io wrote: Hi all,First check that your computer time is accurate. Go to https://time.is/ to check your clock. It shoud report "Your Time is Exact". If not, install some good time scynronization program. Those dropped audio samples happen all the time, no need to worry about those too much. 73's de Kari, oh2gqc |
|
|
|
Hi Kari,
toggle quoted message
Show quoted text
My PC time is good, as I use Dimension4 to keep it good, and this is updated every minute. Maybe that is too often? Maybe the sync is confusing things? 73 de Phil GU0SUP -----Original Message-----
From: "Karza" <kari.sillanmaki@...> Sent: Friday, 7 October, 2022 16:44 To: main@WSJTX.groups.io Subject: Re: [WSJTX] Missing decodes #AudioIssues Hi Phil, On 7.10.2022 18.03, Phil Cooper via groups.io wrote: Hi all,First check that your computer time is accurate. Go to https://time.is/ to check your clock. It shoud report "Your Time is Exact". If not, install some good time scynronization program. Those dropped audio samples happen all the time, no need to worry about those too much. 73's de Kari, oh2gqc |
|
|
|
Alan G4ZFQ
On 07/10/2022 15:59, Phil Cooper via groups.io wrote:
is updated every minute. Maybe that is too often?Phil, Definitely. Use far less than that, maybe several hours or more unless your PC clock is broken. In fact the proper NTP of Meinberg is recommended. 73 Alan G4ZFQ |
|
|
|
Brian Morrison
On Fri, 7 Oct 2022 16:59:27 +0100 (BST)
"Phil Cooper via groups.io" <pcooper@...> wrote: My PC time is good, as I use Dimension4 to keep it good, and this isIt depends exactly how the clock time is altered, if it makes significant jumps this can cause all sorts of trouble. If it is gently tweaked in without upsetting things much then it should be fine. Examples of the latter approach are Meinberg NTP on Windows, ordinary ntpd on *nix and also the more modern chronyd approach on *nix. I have no personal experience with Dimension4. -- Brian G8SEZ |
|
|
|
Karza <kari.sillanmaki@...>
My PC time is good, as I use Dimension4 to keep it good, and this is updated every minute. Maybe that is too often? Maybe the sync is confusing things?Phil, from what I've heard, Dimension4 is not very good time sync application. And once a minute could well be way too often too. Have a look at this email by late Bill Somerville, G4WJS: https://wsjtx.groups.io/g/main/message/28195 Perhaps switching to Meinberg or similar for time keeping could help? I don't use Windows myself so I can not recommend any particular time sync application. 73's de Kari, oh2gqc |
|
|
|
Tim Dawson
To clarify a point, Meinbers is nothing more than a compile of "ordinary ntpd on *nix" (which already supported a windows buildP with a windows installer and a few other things added for ease of use. It is the exact same core software . . .
toggle quoted message
Show quoted text
FWIW . . . On October 7, 2022 11:20:23 AM CDT, Brian Morrison <bdm@...> wrote:
On Fri, 7 Oct 2022 16:59:27 +0100 (BST) --
Sent from my Android device with K-9 Mail. Please excuse my brevity. |
|
|
|
Many have found that too frequent time sync can cause dropouts as the time is a higher priority that FT8 when it comes to CPU resources.
I would think a time sync of NO LESS than 30 minutes would be great for all but the driftyist clock chips. Typically, 1-3 hours would probably work. Just keep an eye on the Band Activity window and watch the DT column. If it starts approaching 0.5s overall, then adjust your time to a bit more often, but don't overdo it. It will NEVER be 0.0s as everybody's clock is slightly different. And you have a 2.5s window for FT8 and 1.0s for FT4. I echo the others in saying that Dimension4 has long outlived its usability as there are better ways to sync your clock. I use BktTimeSync (its FREE), but that's just my choice since it can also use a GPS dongle to do all the heavy lifting, which I do. If you choose this one, make sure to click the lower left option "Reduce in system tray" which continues to run in the background and not lower right option "Close" which will exit and shut down. -- 73's George - WB5JJJ Hams over IP #100105 |
|
|
|
Hi to Alan, Brian and Kari,
toggle quoted message
Show quoted text
That is very interesting! I changed the time update period earlier, and that did seem to make a difference. I think I went from 1 minute to 5 minutes, and noticed it didn't drop as much. I will change that to 30 minutes over the weekend, and see whether that improves things. As for Meinburg vs Dimension4, well, that seems to be one of those things. I know many folk that run D4 happily, and hate Meinburg, and also vice versa, so I will stick with D4 for now. Thanks for the tips, all much appreciated. I will keep an eye on things, and see what happens. Very 73 Phil GU0SUP -----Original Message-----
From: main@WSJTX.groups.io [mailto:main@WSJTX.groups.io] On Behalf Of Alan G4ZFQ Sent: 07 October 2022 17:17 To: main@WSJTX.groups.io Subject: Re: [WSJTX] Missing decodes #AudioIssues On 07/10/2022 15:59, Phil Cooper via groups.io wrote: is updated every minute. Maybe that is too often?Phil, Definitely. Use far less than that, maybe several hours or more unless your PC clock is broken. In fact the proper NTP of Meinberg is recommended. 73 Alan G4ZFQ |
|
|
|
Robert Rose <robert.rose@...>
Just for completeness, I note that I was using using Meinberg on my Mac, and the time differential was 2.1 to 2.3 seconds. I switched to an Apple server in Cupertino (I am in San Diego) and the difference dropped to 0.1 seconds.
toggle quoted message
Show quoted text
I believe the WSJT-X User Guide recommends that you use the closest time server. It takes 0.13 seconds for light to circumnavigate the earth. Add in time for hops and switches, and this sounds like good advice. 73 Bob KN6UXD On Oct 7, 2022, at 2:46 PM, Phil Cooper via groups.io <pcooper@...> wrote: |
|
|
|
Tim Dawson
ntpd (which Meinberg is also a compile of) is designed to compensate for propagation delay. It queries the time server, listens for the reply, and can deduce the transit time pretty well, so that should not be a problem *IF* the Mac is using a proper ntp distribution, and not some cut down sntp kludge.
toggle quoted message
Show quoted text
On October 7, 2022 9:40:16 PM CDT, "Robert Rose via groups.io" <robert.rose@...> wrote:
Just for completeness, I note that I was using using Meinberg on my Mac, and the time differential was 2.1 to 2.3 seconds. I switched to an Apple server in Cupertino (I am in San Diego) and the difference dropped to 0.1 seconds. --
Sent from my Android device with K-9 Mail. Please excuse my brevity. |
|
|
|
Some interesting thoughts in this thread. so I might as well throw in my 2 cents worth!
I use D4 with the update time set to 15 minutes (it could probably be longer). Bob (KN6UXD) mentioned switching to a nearby server, which is a great idea, except... If you select a single server for any time keeper you run a risk of that server being down or unreachable for whatever reason in which case you won't be syncing. When first installed, D4 has a list of servers all over the world which it cycles through. It does measure response time and ignores responses longer than some specified interval. I deleted any in the list that were more than about 500 miles away and ever since, 'Time.is' has always reported my time as 'exact'. One other thing to note, as many have mentioned here, for FT8 using WSJT-X, the time need not be precise to the millisecond, but there are things we do in ham radio which may or may not require more accuracy. I've been playing with FT4 on the satellites. Doing that not only requires adequate time accuracy for the FT4 decoding process, but the time also gets ground into the orbital calculations (including Doppler correction) for the satellite. -- John P. WA2FZW |
|
|
|
Dave Garber <ve3wej@...>
I had done the latest update to w11, and my audio went erratic. did your's
toggle quoted message
Show quoted text
update? I returned to older version of w11, and all is working fine. Dave Garber VE3WEJ / VE3IE On Sat, Oct 8, 2022 at 12:11 AM John P <j.m.price@...> wrote:
Some interesting thoughts in this thread. so I might as well throw in my 2 |
|
|
|
Chuck Gelm
On 10/7/22 22:40, Robert Rose via groups.io wrote:
Just for completeness, I note that I was using using Meinberg on my Mac, and the time differential was 2.1 to 2.3 seconds. I switched to an Apple server in Cupertino (I am in San Diego) and the difference dropped to 0.1 seconds.IIRC, my NTP client monitors RTT latency with multiple time servers and adjusts the time accordingly. Time adjustments are done s l o w l y in adjustments of thousandths of seconds, not as an 'at once' synced jump. $0.02 It seems to me that MS-Windows users should either: * use a NTP client that maintains their computer clock or * use a time sync client immediately before starting WSJT-X. 73, Chuck |
|
|
|
If you were 2+ seconds off your Meinberg was obviously not correctly installed or set up. It should be with no more than 50 or so Milliseconds with that as checked by timeis.
toggle quoted message
Show quoted text
Outlook LT Gil W0MN Hierro Candente Batir de Repente 44.08226 N 92.51265 W EN34rb -----Original Message-----
From: main@WSJTX.groups.io <main@WSJTX.groups.io> On Behalf Of Robert Rose via groups.io Sent: Friday, October 7, 2022 9:40 PM To: main@wsjtx.groups.io Subject: Re: [WSJTX] Missing decodes #AudioIssues Just for completeness, I note that I was using using Meinberg on my Mac, and the time differential was 2.1 to 2.3 seconds. I switched to an Apple server in Cupertino (I am in San Diego) and the difference dropped to 0.1 seconds. I believe the WSJT-X User Guide recommends that you use the closest time server. It takes 0.13 seconds for light to circumnavigate the earth. Add in time for hops and switches, and this sounds like good advice. 73 Bob KN6UXD On Oct 7, 2022, at 2:46 PM, Phil Cooper via groups.io <pcooper@...> wrote: -- W0MN EN34rb 44.08226 N 92.51265 W Hierro candente, batir de repente HP Laptop |
|
|
|
Robert Rose <robert.rose@...>
Chuck’s 2 cents is certainly good advice, but all this begs the question:
toggle quoted message
Show quoted text
What happens in the Zombie Apocalypse? If that seems overly dramatic, then I ask: What do we do if the the World Wide Web is down world wide? In other words, should we prioritize GPS non-Internet solutions for time sync, regardless of operating system? 73 Bob KN6UXD On Oct 8, 2022, at 3:22 AM, Chuck Gelm <nc8q-aredn@...> wrote: |
|
|
|
William Smith
Yes, but if the Zombie Apocalypse arrives, FT8 is the least of my worries.
toggle quoted message
Show quoted text
JS8Call has a way to sync to the constellation of other users, and there are other options as well. If ZA _and_ GPS still up, then a local NTP time server or a GPS puck isn’t a bad idea. 73, Willie N1JBJ On Oct 8, 2022, at 8:00 AM, Robert Rose via groups.io <robert.rose@...> wrote: |
|
|
|
Chuck Gelm
On 10/8/22 00:11, John P wrote:
I deleted any in the list that were more than about 500 miles away and ever since, 'Time.is' has always reported my time as 'exact'.~ 500/186000=0.002688172043011 I suggest that there may be no NTP servers within ( apparent ) 500 miles. Please do a test to see if any NTP server has ~<= 3 ms latency. +1 A first test everyone can do is to see if /time.is/ reports your time as 'exact'. If not, then your clock is errant and WSJT-X decodes may fail. IIRC, regarding your computer's clock using MS-Windows: Dimension4 syncs time...a leap at a time. Meinberg manages time...adjusting your clock speed in very small increments. 73, Chuck |
|
|
|
Chuck Gelm
On 10/8/22 08:00, Robert Rose via groups.io wrote:
Chuck’s 2 cents is certainly good advice, but all this begs the question:Hi, Bob: Good point in spite of the "overly dramatic"s. ;-) A NTP client with both internet and GPS time servers will favor the servers with less 'jitter'. If the internet servers are unavailable then the GPS servers will still be available and 'automagically' chosen. There is no need for operator intervention. https://en.wikipedia.org/wiki/Jitter IMHO, use a time manager, not a time sync'er. 73, Chuck |
|
|
|
Reino Talarmo
I suggest that there may be no NTP servers within ( apparent ) 500 miles.Please do a test to see if any NTP server has ~<= 3 ms latency. Hi Chuck, You seems to have a really high speed internet connection! My typical Delays are tens of milliseconds while Meinberg keeps time below +/- 10 ms. 73, Reino OH3mA |
|
|