Amateur Radio – W6CQZ

IF you have an INTEL based Macintosh you may be able to run JT65-HF using Darwine. I’m still testing this, but, at this time I can run the program and use all its functions on a Macbook Pro running OS X 10.5

If you are a Mac user with an Intel Mac running OS X 10.4 or 10.5, can download Darwine (about 30MB) and would like to try JT65-HF then please join the alpha testing group. See www.w6cqz.org/?p=172 for details on joining.

The specfic version of Darwine I’m using is DarWine-X86-1.1.21

You will need to follow some specific instructions to get this working and those will be posted in the alpha testing group. So, if you want to try this join the group and leave a message indicating you’ve joined and would like to get started with this on Mac OS X.

Again. This will ONLY WORK WITH INTEL Macs.

73 – Joe – W6CQZ

Spots Per Month

Spots/Month updated daily. Click image to see a full sized, easier to read version.

JT65-HF is continuing to progress toward a general release. Slowly, yes, but it is progressing. The alpha testing group is currently running the 19th alpha test release build and things continue to go well. Some builds are better than others, but, over all, the process seems to be moving forward.

I am now opening the alpha test program to any interested parties, but with three requirements.

1 – You must be registered at jt65.w6cqz.org (The RB network and chat system).

2- You must join the alpha test group at http://groups.google.com/group/jt65-hf

3- You must be willing to put up with bugs, program crashes and all the joys of testing alpha test level software.

JT65-HF is currently still a windows only program, but, it is known to work well under Linux using Wine. A native Linux build will certainly happen some day, but I have to focus my limited resources on one thing at a time. Mac OS X might also work with its variation of Wine if you have an Intel processor, but this has not been tested.

If you would like to join the test group head to the URL listed above and request membership. If your email address doesn’t contain your call sign please place it in the area allowed for comments when you sign up. Please be patient after requesting membership as I have to process each request manually.

73 – Joe – W6CQZ

It’s been a while since I gave a progress report on JT65-HF, so, here it is.

Other commitments have slowed my progress with the new program. I’m still working on the code, but it’s progressing very very slowly due to my lack of time.

The alpha testing group continues to provide a great deal of information on how well the program does/does not work and at this point I’m working to fix the bugs found so far and run one more round of alpha testing before a general beta release.

As to timing of a more wide-spread release I’m not able to give any firm date. I think I have about 30..40 hours of programming time to get the program to where it needs to be, but that’s only a guess. If that estimate turns out to be correct I’d be looking at about 4…5 more weeks with the time I’m able to spend on programming. So… hopefully I’ll be able to get something out for all interested by mid to late May.

I apologize for the slow pace of development, but, it will become more widely available eventually, just not as fast as I had hoped.

73 – Joe – W6CQZ

On, of all days, Friday – March 13th, the server hosting the RB network, JT65 chat logger and other business related sites crashed. This was due to a concerted effort to hack into the system via an input point unrelated to the ham system. The hack ultimately failed, but it did manage to crash the system in such a way that a full system reload and data restore was required. I have updated and enhanced my security measures so, hopefully, this will not be an issue again.

The status of the system as of 20:30 – March 17th UTC is as follows;

JT65 chat logger – Functional

RB Network – Functional

RB Database – Functional, but it has not been re-populated with previous data. All stations having submitted reports to the RB network have had their total counts and unique calls heard counts reset to zero. I will, when time permits, correct that as no data was lost. It was simply faster to restart the RB database from empty and worry about the old data later.

Statistical Pages/Graphs – Offline. That functionality is little used and will be restored when I have a bit more spare time.

While not strictly related to the server issue JT65-HF work has been impacted by my loss of time in dealing with the server. I hope to get back to coding again soon and my goal is to have a version out for this weekend to the Alpha testing group.

Thanks to all who offered their support, well wishes and thanks during the server recovery process and since.

73 – Joe – W6CQZ

I had intended to expand the test group for JT65-HF this week and push out the first transmit enabled version as well, but due to a couple of problems discovered that will need to wait.

I’m currently working out a problem identified with audio input code and another with a DLL dependency which will delay both the expansion of the test program and roll out of the TX enabled version. I anticipate being able to resolve both issues soon, but until resolved, I need to back down from the informal release schedule I had intended to follow.

I will post an update here when I have a better idea of a new time line for expansion of the testing group and a more general release to all.

73 – Joe – W6CQZ

I’m making reasonable progress with getting JT65-HF compiled on Linux. Still have a bit of work ahead to get it up to parity with the Windows build, but it’s pretty close already!

If things continue to progress well I hope to have my pre-arranged Linux alpha tester try this out in the next day or so. My goal is to get both Windows and Linux versions into beta at same point so there will be no delay for either set of users for the final release. That being said, there is Mac OS X to consider. I’m optimistic that the changes I’m making for the Linux version will mostly handle the work needed for Mac OS X, but my G5 Mac died some time back so it’s not so easy for me to get on a Mac system and do some coding. If I can get my system running again I’ll try to get a Mac version up as well, but no promises on that one just yet.

73 – Joe – W6CQZ

I am far enough along with my latest project for the RB system that I can begin to talk about it a bit in public. Please understand that this is not ready for use by anyone other than myself and a set of ALPHA testers who are helping me validate the program’s functionality. But, at least for windows users, a beta version should not be more than a few weeks away (barring any unexpected bugs in the alpha). I will start with a screen capture of the new software in operation.

And, PLEASE, understand this is not ready for use by a general audience yet and I have all the testers I can handle for now.

What you are seeing above is a screen capture of my new program, JT65-HF. It’s coded in Pascal using the Free Pascal Compiler + Lazarus RAD environment. Basically, those two comprise an open source clone of Delphi. Unless you’re a programmer who knows what that means I’m sure that’s of little interest. (If you are a programmer who knows what that means and do find it of interest, please let me know, I could use some coding help eventually!). Back to the main story…

With JT65-HF you can currently receive JT65 signals and spot them to the RB network without need for WSJT and the RBC program pair as is currently required. And you can decode more than one signal per sequence without manually having to do so. As testing progresses I will be enabling JT65-HF transmit functions and hope that it will eventually become not only an RB tool, but, also an entry level program for running JT65A on HF.

Note that I am being very specific in the wording of this. The new program will only receive and (eventually) transmit JT-65A. I may expand it to support JT-4A, but I do not anticipate ever adding any WSJT modes not referenced here. Remember, I said, it’s intended to be an RB tool and an entry level program for JT65 on HF. An experienced WSJT operator will probably not find it of use (beyond those who run RB stations) as it will be constrained not only in the WSJT mode it supports, but, also in its operation as it will be VERY rigidly coded to not allow some of the poor operating practices so often seen with new (and some not so new) JT65 HF operators. I suppose the question must be answered as to why we need yet another implementation of JT65 when we have WSJT and MultiPSK…

Both MultiPSK and WSJT provide their own set of positive and negative features, much as, I hope, my program will. But. Neither provide a direct hook for the RB system and it is becoming increasingly difficult for me to support and expand the RB network under the current method of having a program that attempts to glean the necessary data from another program that has no ‘knowledge’ of why another program would want said data. So, the primary focus of JT65-HF will always be the RB system, but it certainly can’t hurt to have (I hope) an easily used program that will foster some new interest in JT65 on HF and offer a less steep learning curve for those who might like to give it a try. More details will follow as I get closer to an expanded then general release.

I also need to write a bit about how I’ve implemented the program. I have spent the last year or so teaching myself how WSJT works from the inside, studying the source code file by file while teaching myself Fortran (which is what most of the important bits is written in). At first it seemed I would never grasp the overall picture as WSJT is by no means a simple or small bit of work, but I finally managed to reach a point where I understood enough to pick out the parts I needed to begin coding my own software. Where possible I have translated routines written in Fortran or C or Python to Pascal for a ‘native’ implementation and, where I couldn’t do (or haven’t yet done) so, I am calling routines compiled into a shared library (or dll in windows speak) from C and/or Fortran routines I have modified directly from WSJT’s source code base. That being said, WSJT is GPL software as is JT65-HF and the full source is available (I will provide a link to the sources in the near future or immediatly upon individual request). I will be making a post to the WSJT developer’s list as well with information about this project.

Direction of development.

My primary focus is to get a reasonably bug free version ready for general distribution on Windows XP/Vista. I will not be testing it on any Windows version older then Win2K and will not deal with Win2K as a supported platform, but it should work under 2K… no idea about 98. Once the Windows release is stablized I will shift to Linux and continue work with the codebase there until a stable Linux version is produced, or (hopefully not!) it’s seen that such a thing can’t be done. For Linux I will develop on a modern distribution, probably Ubuntu 8 (or whatever its current stable distribution version is) and, I hope to avoid much of the problems seen with WSJT on Linux by my abstraction of DSP code from GUI code via the shared library. With Linux out of the way I will begin work on Mac OS X, again continuing until a general release or an admission of failure. In theory it should be relatively ’simple’ to produce Linux and Mac OS X versions, but theory may not translate to practice. More info on those items as I begin to actually attempt the ports.

Enough for now.

73 – Joe – W6CQZ

I have re-enabled reporting of total report count and unique call signs heard on the RB information page. The unique call sign count eats some serious CPU time so it may have to go away again, but the totals should remain. My apologies to those who have missed seeing those numbers. I plan to rebuild the master database sometime in the next few months and do so in a way that will improve the response time and not use so much CPU time on my server… I really never anticipated having anywhere near the number of reports we have collected, so, I didn’t code the database to work well with searching/indexing nearly a million records! I’m not complaining though, just amazed!

Thanks to all who have submitted reports and put out the word about this system! I hope in the near future to release some software to increase our ability to collect spots and, perhaps, even generate some new interest in HF JT65, but it’s premature to speak in detail about that. Soon though.

73 – Joe – W6CQZ

Each point represents a station having made a report to the RB system since its beggining (June 29, 2007). You may click the image for a full sized version.