FT8 in VK

Thanks Ron,
For clarity, the write-up v 1.71 includes info on FT8 which, as this write-up says, only comes in the v 1.8 (presently RC1) of the software, so this write-up should really be called v 1.8. A very complete write-up in any case.

Ed.

Hi Paul,

I’ve used an empty beer carton. Folds flat, light and can be opened out to enclose a netbook. An empty wine carton would be fine. Either way you may need to empty a few before finding one you like.:yum:

Check your screen brightness by going into Control Panel and Select Display and wind up the brightness. That might just be enough.

73
Ron
VK3AFW

You’ve just given me a great idea! Here in Germany, the beer crates are similar to the old school milk crates you may remember from your youth. They’re fairly strong and size wise could make a great housing for a “Go kit”! Doubling as a seat as well…

Ed.

Of course I’d have to empty it of those bottles first but that shouldn’t be a problem with some good weissbier …

1 Like

Time Synching

As most would understand having an accurately set clock on the PC is important to success with FT8 and other digital modes because of the sequencing between Tx and Rx.

On Sat midday I had Thinking Man’s Dimension 4 running on my SOTA notebook so my time as measured by time.us was “accurate”, within about .01 something or other. I then switched the notebook off and took it with me to a summit. Didn’t get to use it but that’s another story. About 50 hours later I then resunk (is that a word?) it and it was .4 seconds out. I’d say that the internal clock in Win 10 managed to keep up so that if I was on a summit I would have been OK.

This is also backed up by me qualifying a summit on FT8 a couple of weeks earlier when I must have been in sink (sic) doing the same synch before I go.

Compton
Waiting for my first 23cm FT8 S2S.

Hi Compton,

Some comments.

Your computer will keep better time if it’s not doing anything. If It’s grinding away there will be lots of interrupts and the clock will run slow. Some computers no doubt keep better time than others. Time between syncs and temperature will be factors.

In the good old days we had hardware clocks. They could be trimmed to keep very good time. Today the designers assume you live on-line and so the software is designed to keep you sort of in sync. Is 3 seconds the legal maximum discrepancy for bank transactions?

When using D4 use a local time source. Connecting overseas may involve, coax, optical fibre, satellite, microwave link, 600 ohm copper line, all together or in varying combinations. The connection might go from a long path to a very long path. The synchronizing protocol measures the one way time delay and assumes it is fixed for the transactions. It often isn’t. Networks have their own rules about when to switch and which path to switch in.

Yes the local tome source may involve different transmission media too but it is likely to be much more stable. When in VK3 I use Deakin Uni and consistently sync to 0.02 seconds.

If you don’t sync you are sunk.

Disable the silly “advanced feature” which D4 uses to select a different time source each time it syncs. Therein lies error. And madness.

My backpack packable 1296 MHz rig is FM only. Possibly could run FT8.

My SSB 1296 MHz station can only be carried a short distance - it’s not light as it includes 432 MHz and 144 MHz . A significant amount of preparation required, maybe later in the year.

73
Ron
VK3AFW

We still do. It’s the BIOS clock that has a 32kHz crystal to maintain time across power downs.

Because having a net connection is the norm. Using NTP et al. costs nothing in hardware and allows time to be maintain to high degrees of accuracy without adding to the hardware cost of every PC sold. This is important because most people do not care about accuracy to more than a minute. NTP syncing allows cheap hardware to have high accuracy timing. It doesn’t fit our needs but our needs are a special case (accurate time when not synced).

It’s just as well they thought about that when designing the (S)NTP spec so that latencies and jitters in the path between timing source and timing user is accounted for. Clever these programmers.

A bigger issue is that typically NTP software (such as the built in Windows stuff) does not set the BIOS clock. It’s designed to run all the time and so the BIOS clock only needs to be sort-of accurate and is used to set the time as the PC boots. As soon as the networking stack is up and connection solicited, then the hardware BIOS clock is superfluous and the NTP syncing can get you a muhc more accurate time.

I haven’t played with the Dimension 4 software, so it may actually set the BIOS clock as well as the Windows clock.

Time… such a simple concept :slight_smile:

Just about every system out there that has a synchronised clock service (Linux, Windows, Mac, Solaris, AIX, etc) all use NTP by name, or a variant of NTP called something else. On Windows, it’s called w32time, and unless you’re joining AD (if I recall correctly), it should be able to sync to a proper NTP server.

NTP is designed to work over the public internet and can synchronise to millisecond accuracy easily across a switched network. If it doesn’t, there’s something broken in your configuration, or you’re attached to a server reporting bogus time and you don’t have others in your pool to compare with (which is a problem with your configuration). 100ms should be expected across public internet, 10ms or better if you have a nearby clock source.

NTP uses round-trip delay, not one way time delay, as per RFC 958. It does not assume it’s fixed for the transactions involved. Depending on your configuration, it may take some time for NTP to synchronise your computer, but in general it’s fairly quick. A clock that keeps time to sub-1s over 2 days as Compton’s did would synchronise so quickly as to be considered immediate for our purposes.

If by some chance you’re not using something other than NTP as a time protocol, I suggest you change - it’s about the only thing that regularly works across all modern operating systems.

As an ex-admin there, I can inform you that Deakin sync off Melbourne Uni’s Stratum 0 servers. The publicly available NTP server was at Stratum 2 and routes out to the public internet via a tortured route that’s often impacted by sudden changes in latency, congestion and the like. As you can see 20ms is easily possible despite that. Last I’d heard, they’d disabled public access to the time server though, so I might have to report you :wink:

Any bank that works to that standard now would be out of business, particularly since NTP can be implemented for peanuts and give millisecond accuracy. Banks are now implementing PTP, an even more extreme version of NTP, giving accuracies in the nanoseconds.

They use it for things like High Frequency Trading, and in that situation, they control every single parameter. For each customer allowed to access their HFT network, they’ll put in a rack in their datacenter with the same length network cabling as every other customer to ensure there’s an even playing field on latency. Then they synchronise using PTP to ensure they’re all on the same time. A ex-colleague of mine spent time measuring RTT latency down to nanoseconds on cables for one of the major US banks, trimming copper off as appropriate until it met spec.

Pretty sure FT8 would work under those circumstances :slight_smile:

Hi Andy,

Thanks, the one way sync was entrenched when I was last involved so good to know a smarter system has been accepted as a standard. Now whether everyone meets the standard might be another argument.

I have found “internet” time inadequate for JT type programs so don’t bother with it for that activity.

The 32 kHz crystals used are I’m sure fit for purpose but not for keeping accurate time.

Time is as simple as the three laws of relativity. Once it was only needed to a second by astronomers and mariners. It was much simpler for average folk until the railways arrived. Later on Einstein got into the act and then some blokes at NPL built a MASER as a time and frequency standard and now frankly if you can’t keep time to a few nanoseconds per day you aren’t in the game.

73
Ron
VK3AFW

You’re right Ron, the crystals are near enough that the time is about right when you restart a PC. But most will not be trimmed or have triming facilities. The other thing with NTP is that you should get your clock sync’d to about 1sec accuracy very quickly once the PC is booted and online etc. But it can take some hours to get a very accurate clock sync.

The issue we see is the window is very short on FT8 so accuracy of the clocks is every so important, more so if we are using small lower performance tablets etc.

Hi Andrew,
A bit OT.

I have just deleted my initial “banks and clocks” response in order to avoid having the men in black suits call. Suffice to say I expect to see a profit driven gap between what is achievable and what is done.

For many other legal time stamps my understanding is that an uncertainty of 1 second wrt UTC is accepted.

I have had computers that had batteries replaced and were “off line” for 2 months and required manual intervention to get the year, date, day and time right. Still using them. One uses XP and the other W7.

Before you report me for using the Deakin time server you should chat to my adviser, Michael. He has coffee at 10 am at a well known cafe in Coburg.:sunglasses:

BTW I have found that the Melb Uni site gets overloaded frequently and I time out waiting but that rarely happens on Deakin. (Oops I’ve said too much). Repeated syncs on either yield similar corrections, varying randomly from 0.001 seconds to 0.5 seconds. So in spite of two-way corrections the delay time seems to be only partly compensated.

FT8 and JT65HF can accommodate a 1 second error but if I am NAD on UTC then the other person has a little more leeway. I did have a GPS time sync on one computer but it has died and I need to find the software disc before running it on my current machine.

In the meantime I’m a D4 user.

73
Ron
VK3AFW

Every single bank in Australia and the vast majority worldwide already have the capability ingrained in software they’ve already purchased; it’s zero cost to them. I know of one very large company in Australia that’s syncing their entire server fleet off an ancient, 1999-era white box server. The moral? Good time keeping is so cheap it gets forgotten about :slight_smile:

The amount of people syncing off stratum 0/1 servers when their local stratum 2/3 is sufficient is a major issue. It doesn’t generally yield a much better accuracy, just a faster sync. Use your ISP’s stratum 3 time server or the AU NTP pool servers and be done with it.

Note that the correction is not the same as how far out your clock is, as NTP will (in general) slew the clock to match, so a sudden jump in latency to 500ms from a 10ms won’t yield 500ms of inaccuracy, unless that latency continues for a reasonable length of time. The aim is to ensure your clock monotonically increases, because most software assumes time flies like an arrow

Hi Andrew,

Easily said, but you assume I know more than I do. I know enough to know I should stay out of the operating system and not change system files.

D4 does not care who my ISP is - that isn’t constant- so while things work I’m not going to spend time with the hood up poking at stuff I’m liable to mangle.

73
Ron
VK3AFW

So. Is anyone up for a S2S then?
We seem to have sorted the clock thing out!
Compton

Hi Compton,

Won’t be able to activate this weekend but maybe the following weekend. Could be some dx ops about this weekend.

73
Ron
VK3AFW

For those using an ICOM IC-7300 (and possibly other ICOM models) - beware the latest versions of ICOMs RS-BA1 remote control software, which installs a multi-port serial driver which impacts the ability of Hamlib (used by wsjtx) to communicate with the rig. I found this after installing wsjtx v1.8RC2 and it not working. I went back to RC1 and it also did not work. I had recently installed the latest version of ICOM’s RS-BA1 software and when I de-installed ICOMs multi-port serial driver Hamlib (and wsjtx) was happy again!.

73 Ed.

Hi Ed,

What a nuisance. Sometimes the promises of plug and play get broken. At least you know what works and what doesn’t. Maybe there will be some updates that fix this.

73
Ron
VK3AFW

1 Like