Sunday, July 16, 2017


Have you tried FT8 yet? Man, that is one fast digital mode! Self-completing QSO's make it easier that one might initially think after enabling the mode and just watching a few exchanges take place among the many users of this new mode.

Bas PE4BAS wrote about it a few days ago and I decided to give it a try this morning. Uninstalling the old version of WSJT-X and re-installing 1.8.0 was a two-minute operation with my config files being automatically saved and applied to the new version. I was decoding FT8 signals and making my first QSO within 5 minutes of opening the program.

Whereas JT65 gives you about 12 seconds to respond to a station calling CQ, you only have about two seconds with FT8. Transmission cycles are 15 seconds each and, once a QSO is initiated, the remaining exchanges take place automatically.

LoTW doesn't currently support the mode and if you upload a file containing FT8 QSO's you'll get an error message. You can map FT8 QSO's to generic a "Data" mode and then upload your logs but your QSO partner must do the same thing in order to get a match (QSL). Instructions for doing that are here.

I personally plan to wait until the next version of TQSL is released - it will reportedly support FT8. Those stations mapping their FT8 QSO's to "Data" will not QSL match with future QSO's uploaded as FT8 mode.

I'll be interested to see if this new mode is deemed fast enough to be considered by DXpeditioners. While it is certainly fast for a digital mode, it is no match for the speed of CW or phone. But its SNR capability may make it a worthy consideration for tough paths or for QRP DXpeditioners.

Coupling the efficiency and speed of this mode with the portability of QRP might yield some low-cost DXpeditions of a unique nature where the KX2 and a small laptop or tablet become the tools of the trade. In fact, the FT8 FAQ mentions that a feature to be made available in a future version of FT8 is for FT8 to Auto-respond to the weakest responder to your CQ. Lots of implications in that statement for DXers and QRPers...


PE4BAS, Bas said...

To comment on your last sentence. I would absolutely like to see the auto respond function in all JT modes, or at least a choice to switch it on or off. The feature of responding to the weakest signal would be great again for all modes. However, we now are close to a completely automated QSO. Just like in SIM31, which is absolutely possible in WSJT-X as well. This would be (too) controversial I think. 73, Bas

John AE5X said...

Hi Bas - I think it is inevitable that we will eventually see a completely automated QSO digital mode. This auto respond function is a big step in that direction and in operating FT8 I notice that the software recognizes whether or not I get a response when I answer a CQ. If not, the software calls the station again; if I am answered, the next exchange is automatically made.

The upcoming "respond to the weakest caller" will be very significant I believe. Paths that are normally ignored or little understood will become known in a way that WSPR often makes them known, only with 2-way QSO credit.

Bob M. Read said...

Great mode, fun to operate - I hope LoTW gets it sorted!

Unknown said...

I REALLY like the idea of auto respond to the *weakest* signal. If that limitation were not present, a DXpedition running FT8 would cause everyone to slowly crank up their power to get the response similar to the "big guns" on SSB. This is very similar to what we have now. I have been in JT65 and as soon as a hot call sign is heard, many operators automatically crank their power up to 40-60 watts to be sure they are the ones getting through.

This would also add another challenge to the digital modes which have really made radio operation boring and trivial. The challenge will be to use just enough power to get through.

To make JT modes more exciting for myself, I regularly call CQ on 0.5 to 2 watts. People cannot believe my last 73 of "TUHALFWATT73". With a modest vertical I regularly cross the Atlantic on 2 watts and the Pacific on 7. When conditions are really bad and I want to do DX I will go to 15 or so but I regularly look at PSK Reporter to guide my power levels.

BTW, here's a little secret I've noticed. I often get the QSO when I respond to CQ with a reasonable power level. I have also started responding to the weaker signals when I call CQ. I like to think of it as a reward.

John AE5X said...

I think LoTW will be updated fairly quickly. I have read that Joe K1JT asked for a bit of time first while any possible issues are worked out with the mode. We are all actually beta-testers at the moment.

John AE5X said...

I like this feature too and can't wait to see how its implementation will manifest itself. I also think it might draw other people to digital modes who might not otherwise be attracted to them.

Photon said...

Very interesting post and responses. I don't find any value in simply collecting DX points for the sake of it using automated means. That's a bit like WSPR, which has its place and is of immense value technically, but has no meaning in terms of human contact.

I, too, like the idea of responding to the weakest stations, which many of us will already do. But, again, fully automated QSOs are not enough in themselves - it's little moer than a permanent 'contest' situation. Well, what are your points worth when you're dead? Better enjoy some ragchewing whilst you can, I think...

Mike KJ4Z said...

I've been playing around with JT65 (and now FT8) on a Raspberry Pi, with the intention of making totally automated QSOs. I have been planning to use its API to reply to CQs and make complete QSOs while I'm sleeping, or at work, etc. But it sounds like some of the latest changes may do almost all the work for me. Ultimately, the goal would be to send it along to a DX locale and let it make auto-Qs that get uploaded periodically via satellite. The technology to do this is here already, and while I understand it may be controversial, it's also essentially inevitable unless the rules are changed to prevent it. And there are a lot of fascinating problems to be solved on the periphery before a robust solution could be introduced. Fun times!

John AE5X said...

Hi Mike,

I really like the idea of automated QSO's. I know it's controversial among those who think all activity has to be QSO-based rather than propagation study - the mode you're describing would serve both and there's nothing wrong with that. We have so much spectrum available for narrow-band modes - those who find it controversial need not participate.

I agree - it is inevitable, just a matter of time. Glad you're working on it. Please consider me a willing beta-tester!

Anonymous said...

Have I notice that the FT8 logging file refreshes each time you start up a new session of WSJT-X Ver 8.1? I worked 7 FT8 QSO's yesterday and then quit the program and closed my laptop. I started my laptop and then WSJT-X and the log was gone in the directory that I saw before I closed everything. I agree with you and ill wait for LOTW to catch up.

John - N7AME

Anonymous said...

... next step will be old fax machine dialing godzillions of numbers to find another old fax machine adopted for the same purpose by another "piece of roasted pork". I do understand convenience, I do understand, that ARRL need to keep all the "dead licensees" interested in purchasing their wallpapers, but for the life of me I can not understand what in it for the operator who gets home at the end of the day, prints/looks/saves log and goes to bed? And all over next day, and so on... Haven't used any other mode but CW in the last 25 years or so, hence my misunderstanding, PSE excuse me if someone will find it offensive.

John AE5X said...

Maybe such a mode would have more widespread approval if, instead of being thought of as a QSO mode, it were seen as a mode whose main purpose is to study propagation?

Imagine stations around the world, working each other on all bands with the details of time, location, freq and signal strength being uploaded to the internet in a similar way as WSPR is now. But this would be 2-way contacts (as opposed to 1-way with WSPR). The potential for detailed indications of propagation would exceed anything that has ever been implemented, by hams or anyone.

Why the reluctance to having a narrow sliver of each band dedicated to such a mode? Because DXers who use other modes would that their accomplishments would be devalued. This need not be the case if current awards program looks at automated modes separately. This is a similar argument (with a similar solution) that contesters presented when CW skimmers were brought online - separate categories for use and non-use of skimmers.

This is amateur radio, not CB. We need not be engaged in on-the-air conversation in order to advance the state of the art.