Changes in Version 4.7.0 ------------------------ 1. New feature: WSJT can now be used simultaneously with Spectran on the same computer. You can start Spectran by selecting "Use Spectran for input" on the Setup menu. This feature also allows the user to select the sound card to be used for input. 2. Bug fix: in some circumstances, switching from a JT65 long-format message to a shorthand message did not work properly. Fixed. 3. Bug fix: for operators in the southern hemisphere and longitudes more than 90 degrees east or west, the wrong "Hot spot" was sometimes identified as the best direction for sporadic meteors. Fixed. Changes in Version 4.6.1 ------------------------ 1. Bug fix: I mistakenly shifted the RX data by 2 seconds, rather than the intended 1 s, when changing the DT range. Therefore in v4.6.0 the actual DT range is not -1 to +5 s as stated, but rather 0 to +6 s; moreover, the displayed values of DT are too small by 1.0 s. This has been fixed. 2. New feature: thanks to Akira, JM1SZY, I learned that occasionally a file (or an average of several files) will decode better with the AFC feature turned off. Consequently I have added a checkbox that must be ticked to activate AFC in the JT65 modes. If you can trust the frequency stability of the signal you are receiving, and especially if you are trying to receive a signal at -27 dB or weaker, leaving this box unchecked may yield a slight improvement in decoding. In most cases, especially at 144 MHz and above, I recommend leaving the AFC turned ON. 3. Thanks to Chris, GW4DGU, for pointing out that the Gx series of prefixes is no longer legal for reciprocal license operating in the countries of the UK. The valid prefix series for such operation is now the M-series, i.e., M MD MI MJ MM MU MW. I have changed the prefix table accordingly. Version 4.6 ----------- This is the first full release of WSJT since version 3.0. New users can install Version 4.6 directly, without upgrading from a previous installation. Of course, you can also upgrade from an earlier version in the usual way. All download files can be found on the WSJT home page, http://pulsar.princeton.edu/~joe/K1JT. The new release includes an entirely new "WSJT 4.6 User's Guide." This document is about one third the length of the former "User's Guide and Reference Manual," but contains nearly everything you need to know to use the program. A copy of the new Guide is included in the version 4.6 distribution files. You can also download it directly from http://pulsar.princeton.edu/~joe/K1JT/WSJT_User_460.pdf. Even if you are an experienced WSJT user, you should definitely print and read this document. New features in WSJT Version 4.6 include the following: 1. Improved automatic frequency control in JT65 modes. If you have lost JT65 QSOs because of unstable oscillators, this is for you. 2. The acceptable range for DT in JT65 mode is now -1 to +5 s. This range is a better fit for EME communication than the former -2 to +4 s. It will allow for somewhat greater clock errors before inter-station synchronization fails on an EME path. Note to experienced users: this means that the plotting scale for the "blue curve" now runs from -1 to +5 s. EME signals should normally produce a blue peak near the center of the plot area. 3. When the blue window displaying moon coordinates has been toggled to display coordinates for the DX station as well as the home station, it now displays MaxNR in place of SD. MaxNR is the maximum path non-reciprocity in dB. This effect arises from the combination of spatial polarization shift plus Faraday rotation; it is what causes "one way propagation" between stations that use fixed linear polarization. 4. A facility for generating the file ID.WAV for station identification is now built into WSJT. 5. The "Save Decoded" menu item now saves files with decoded shorthand messages as well as normal messages. 6. JT65 has a new shorthand message "ATT" (for "Attention!"). It is intended as an aid to help two stations find each other by determining the correct DF. 7. Visual aids for evaluating JT65 shorthand messages "by eye" are provided if you click on the sync-tone frequency in the Big Spectrum display. 8. For DXpeditions: a country prefix preceded by "/" may be substituted for the grid locator in a type 1 JT65 message. 9. Alternatively, a signal report of the form "-NN" or "R-NN" may be substituted for the grid locator in a type 1 JT65 message. For example, -24 might indicate that signals were being received at -24 dB. The minus sign is required, and NN must lie between 01 and 30. 10. The receiver noise level reported by Measure mode (the level of the "green line") has been increased by 2 dB to be consistent with levels reported by the other operating modes. Changes in Version 4.5.1 ------------------------ Bug fixes: 1. JT6M did not transmit properly in version 4.5.0 because the program failed to switch its wavefile generator into JT6M mode. Fixed. 2. WSJT consumed a large fraction of CPU time in JT65 mode, even when the program was supposedly doing nothing. Fixed. 3. Monitoring for long periods in JT65 mode would occasionally produce a Fortran "output conversion error". Fixed. 4. Clicking the "Add" button with nothing in the Grid box would cause a program crash. Fixed. Enhancements: 1. In the FSK441 modes, messages longer than 3 nonblank characters and starting with R26, R27, RRR, or 73 are no longer transmitted as shorthand messages. 2. Onscreen labels now indicate active status of the "Save Decoded", "Save All", and "Save text in File DECODED.CUM" features. 3. Small improvements have been made in the decoding of shorthand messages in modes FSK441B and C. 4. Alphabetic characters in message templates (on the Setup | Options screen) are now case-insensitive. Principal New Features in WSJT Version 4.5 ------------------------------------------ 1. The JT65 modes employ an entirely new decoding algorithm that uses "soft decisions" to recover the transmitted message. The message format and Reed-Solomon encoding are unchanged, so the JT65A, B, and C modes are fully compatible with earlier versions. However, the new decoder is more sensitive by slightly more than 1 dB. It is based on software licensed from CodeVector Technologies, LLC, and protected under United States Patent 6,634,007. As usual, however, I am making WSJT freely available for amateur radio use. Changes to the JT65 modes are "under the hood," and except for the improved performance you will find them mostly invisible. The new decoder can be somewhat slower than the one in v4.3.4, depending on details of the received data. 2. Like JT65, FSK441 now provides three submodes. FSK441A is identical to the "classical" FSK441. The two new modes, FSK441B and FSK441C, use the same 4-tone frequency shift keying at 441 baud. However, they use forward error correction (FEC) on a character-by-character basis, to improve message reliability. They also provide shorthand messages which are more robust than the single-tone messages of traditional FSK441. Cross-mode communication will not work: a transmission in FSK441B must be received in FSK441B, etc. FEC information in FSK441B and FSK441C is conveyed by sending additional channel symbols (tones) for each character, using special codes designed to optimize the error rate and sensitivity with very short pings. The additional symbols are redundant when the S/N is high, but they allow recovery from transmission errors when the S/N is low. FSK441A, which provides no redundancy, transmits 3 symbols per character. Modes B and C use 4 and 7 symbols per character, respectively. The raw throughput of user information is summarized in the following table: FSK441A FSK441B FSK441C ---------------------------------------------------------------- Sequential tones per character 3 4 7 User data rate, characters/second 147 110 63 Time for a 12-character message, ms 82 109 190 The shorthand messages in FSK441B and FSK441C use alternating tones at two specific frequencies, as follows: Shorthand Low tone High tone message (Hz) (Hz) ------------------------------ R26 861 1206 R27 861 1550 RRR 861 1895 73 861 2239 Tests with my digital simulator show that shorthand messages in FSK441B and FSK441C are several dB more sensitive than the ST messages of FSK441A. At the same time they give a far lower rate of false positives. With multi-tone messages the sensitivities of the three submodes are nearly the same, but they have different trade-offs. Mode A is 25 percent faster than mode B, but mode B is more accurate and will produce much less on-screen "gibberish". Mode C is about half the speed of mode A but has still stronger FEC code. Parameters of the B and C modes were selected in the expectation that FSK441B might become the mode of choice for meteor scatter work on 144 MHz (and possibly also 222 MHz), while FSK441C will likely prove best at 50 MHz where the pings are longer. These suppositions need to be tested, of course. Experienced WSJT users should have no difficulty making the new modes work. Just select the desired mode from the Mode menu (or use the appropriate hot-key combination), and away you go. Other Changes in Version 4.5.0 ------------------------------ 1. Bug fix: In JT65 mode, if you sent 73s and then started a QSO with a new station, the program would sometimes continue sending the 73 message even though TX message #1 had been checked. This has been fixed. 2. Performance enhancement: Shorthand messages in JT65 were sometimes suppressed because of an apparent low-value Sync detection that produced no decoded message. This has been fixed; there is no longer any need to set your Sync threshold to a higher value when you are expecting to receive a shorthand message. 3. The JT65 "Filter" function has been removed. It was confusing to some, and anyway was generally deemed of little value. The new decoder provides a much better solution. 4. The "suggested report" has been deleted from FSK441 decoded text lines. It is replaced by S/N, the measured signal-to-noise ratio in dB. Note that the familiar "dB" measurement of (S+N)/N is still available, as well. 5. Finer adjustment intervals are provided for "S", the FSK441 ping detection limit. As in earlier version, these numbers refer to (S+N)/N. In contrast, the "Single Tone" or "Shorthand" detection limits refer to S/N, a more useful parameter at very low signal levels. (See below for more details on signal to noise ratios, if interested.) 6. A button labeled "Add" just below the grid locator box will cause the displayed callsign and grid to be entered into the CALLSIGN.TXT database. 7. A date and time stamp is now added to the DECODED.CUM file at program startup or when you first enable writing to this file. Request to Users ---------------- As usual, I will appreciate hearing from users about the new features in WSJT. Let me know, of course, if you find bugs or other problems in the program. In particular, let me know of your experiences with FSK441B and FSK441C. Remember, my guess is that FSK441B should work very well on 2 meters, while FSK441C may work best on 6 meters. After you have gained some experience with FSK441C on 6 meters, I would be interested to know whether you think JT6M should be retired. In North America, at least, I do not think it is being used very much. Why use both S/N and (S+N)/N ? ------------------------------ WSJT has traditionally measured the level of FSK441 signals as the ratio (signal plus noise)/(noise) = (S+N/N), in dB. This quantity is approximately what S-meters try to measure; it has the advantage that it goes to zero when there is no signal, while at high signal levels it increases as you would expect, in proportion to signal strength. At low signal levels, however, the numbers for (S+N)/N in dB behave in a way that may be counter-intuitive. When WSJT reports that a meteor ping had strength 3 dB, it means that signal plus noise was 3 dB higher than noise alone. That means that signal and noise were equal in power, so the corresponding value of S/N must be 0 dB. If the same signal had been transmitted with half the power, it would have had S/N = -3 dB, and (S+N)/N would have been 10*log(0.5+1.0) = 1.76 dB. Yes, cutting the TX power in half would only reduce the ping level from 3.0 dB to 1.76 dB! That's why, for many purposes, S/N is a more useful number -- and why I am now listing both numbers in FSK441 decoded text lines. The table below will allow you to convert easily between S/N and (S+N)/N, both as numerical ratios and as dB. S/N S/N (S+N)/N (S+N)/N (dB) (dB) ---------------------------------------- 10.0 10.000 11.000 10.41 9.0 7.943 8.943 9.51 8.0 6.310 7.310 8.64 7.0 5.012 6.012 7.79 6.0 3.981 4.981 6.97 5.0 3.162 4.162 6.19 4.0 2.512 3.512 5.46 3.0 1.995 2.995 4.76 2.0 1.585 2.585 4.12 1.0 1.259 2.259 3.54 0.0 1.000 2.000 3.01 -1.0 0.794 1.794 2.54 -2.0 0.631 1.631 2.12 -3.0 0.501 1.501 1.76 -4.0 0.398 1.398 1.46 -5.0 0.316 1.316 1.19 -6.0 0.251 1.251 0.97 -7.0 0.200 1.200 0.79 -8.0 0.158 1.158 0.64 -9.0 0.126 1.126 0.51 -10.0 0.100 1.100 0.41 Beta Release 4.3.4 ------------------ Beta Release 4.3.4 of WSJT is now available for free download at the WSJT home page, http://pulsar.princeton.edu/~joe/K1JT. The principal change from version 4.2.1 is to offer three JT65 submodes. The submodes differ in tone spacing and total bandwidth as follows: Mode Spacing Total BW ------------------------- JT65A 2.7 Hz 177.6 Hz JT65B 5.4 355.3 JT65C 10.8 710.6 Note that JT65A is identical to the original JT65. If you want to work people who have not yet upgraded to v4.3.4, be sure to select mode JT65A. Otherwise, be sure to use the same mode that your QSO partner is using. Cross-mode contacts will not work. JT65B should be nearly as sensitive as JT65A, and it will be twice as forgiving of frequency instabilities. On balance, with existing "stock" radios, JT65B will probably be better than JT65A. JT65C is less sensitive by a small amount, perhaps 1 dB, but will be even more lenient on stability issues. By all means experiment with the different submodes, and be sure to let me know your conclusions about them! I am presently inclined to recommend that JT65B should become the "standard" JT65 mode. If this tentative conclusion holds up, future versions of the program may no longer support the A and C modes. Other changes from version 4.2.1 include the following: 1. Further improvements have been made to the JT65 decoding algorithm. These improvements apply to all three submodes. Some wave files that would not decode with v4.2.1 now decode properly, especially in averages over several minutes. 2. The frequency width W of the sync tone (the "red spike") is now measured and displayed in Hz after DF in the main text box. In any of the three JT65 modes, W should be no more than 2-4 Hz under good conditions. Uncorrected frequency drifts, excessive oscillator phase noise, and certain propagation effects can make the width larger. Anything over about about 4 Hz will impair copy in JT65A. Similarly, widths greater than about 7 and 15 Hz will begin to impair copy in JT65B and C, respectively. 3. The utility program CWID.EXE now accepts lower case letters on the command line. It also permits you to specify the audio frequency of the tone in the wave file. You may wish to place the tone at 600 Hz or lower so that it lies well below the tones generated by any of the WSJT operating modes. 4. The "Clip" function has been improved in several ways. The yellow and magenta curves in the Big Spectrum display no longer disappear when Clip > 0. Setting Clip = 3 does hard clipping, as before, but it also blanks out any data regions with average power well above the "baseline" of the green curve. Experimenting with different values of Clip may help you to recover good copy from noisy data. 5. I believe that the text window displays in Monitor mode, and when you are using the Include/Exclude buttons, now function correctly. 6. Minor bug fixes: the program no longer crashes in EME Echo mode if you select "EME Calc | Load | Cancel". The correct "S" value is listed on the status bar in JT6M mode. 7. The program's "Fit and finish" is improved in several not very important ways. WSJT Version 4.2.1 ------------------ Version 4.2.1 contains a number of enhancements and bug fixes, mostly related to the new JT65 mode. Changes from version 4.1.1 include the following: 1.1 Message averaging now works correctly 1.2 Many small improvements to the decoding algorithm 1.3 Decoding speed improved by 50% 1.4 JT65 monitor mode is properly implemented 1.5 TX message can be changed up to t=59 s of preceding RX period 1.6 Switch to a shorthand TX message at any time 1.7 Freeze works properly for shorthand messages 1.8 Decodes with failed FEC (forward error correction) are optionally displayable 1.9 "Garbage filter" provided so that questionable decodes appear only if they contain some recognizable text 1.10 Automatic station ID, as in FSK441 and JT6M modes 1.11 Companion program to generate a CW ID.WAV file is included 1.12 The birdie zapper now works in JT65 mode 1.13 "Clip" function has been reactivated 1.14 F5 help screen updated to reflect JT65 practices 1.15 "OOO" message handled more transparently 1.16 Optional display of Moon Az/El at DX station, replacing Sun Az/El 1.17 Right/Left audio out now works properly 1.18 DT displayed as blank rather than 0.0 for shorthand messages 1.19 No program crash if ToRadio or Grid left empty 1.20 No program crash if attempting to decode 60 s file in JT6M 1.21 All other reported problems causing crash have been fixed +-------------------------------------------------------------+ | Quick Start Guide to Using WSJT version 4 and the JT65 Mode | +-------------------------------------------------------------+ Version 4 of WSJT marks a significant departure in the program's evolution. The list of features is no longer a full superset of those available in all previous versions. In particular, the JT44 mode is gone; it is replaced by a new mode called JT65 that I hope will be a significant improvement for making QSOs via EME and extremely weak tropospheric paths. The procedure for upgrading to version 4 is identical to previous upgrades, with one exception: the setup procedure will install a new program in your existing WSJT directory, but will keep the old version alive as well. The two versions of the program can coexist peacefully. During a testing period you will probably want the older version and JT44 to be available as well. What is different about JT65, compared to JT44? Here's a short list: 1. T/R period 60 s; actual TX audio duration 46.8 s. (Yes, this will put somewhat more stress on your PA. If its cooling is marginal, take appropriate action.) 2. Modulation uses 64 data tones plus a sync tone. 3. Tones are spaced by 2.7 Hz in frequency and 0.372 s in time. The total occupied bandwidth of a JT65 signal is about 180 Hz. 4. Transmissions consist of 63 data tones or "symbols," each carrying 6 bits of information. In addition there are 63 sync symbols for establishing time and frequency synchronization. 5. Software AFC (automatic frequency control) can follow drift rates up to about +/-10 Hz/minute. 6. User messages are tightly "source coded" into 72 bits. 7. Strong FEC (forward error correction) coding is used to mitigate transmission errors. 8. User-level mesage formats are designed to permit maximum possible efficiency in EME or similar QSOs. A valid message must be be one of three possible types: 1. "Call1 Call2 Grid" or "Call1 Call2 Grid OOO" 2. "RO", "RRR", or "73" (so-called shorthand messages) 3. "any text you want" (up to 13 characters selected from a 42-character alphabet) Instead of a callsign, the first field of a message type 1 may contain "CQ" or "QRZ". Other such "special tokens" may be added later. The available alphabet of characters for message type 3 is: 0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ +-./? 9. Instead of JT44's very effective intra-message averaging of the even, odd, and "last N" characters of a message, JT65 offers even more powerful special shorthand messages for RO, RRR, and 73. These have much higher sensitivity than the other message types. They work reliably down to about -30 dB. If you succeed in exchanging callsigns and "OOO" by means of type 1 messages, you should certainly be able to complete the QSO. 10. Message averaging over subsequent transmissions works as it did in JT44, with one exception. The "OOO" signal report typically used for EME can be appended to message #1 and the message will continue averaging without any need to start over. The presence or absence of "OOO" will be detected if sync is achieved, whether or not full decoding has been successful. 11. Receiving FEC-decoded transmissions takes some getting used to. You will find *much* less gibberish on your screen in JT65 than in JT44. The FEC algorithm has a good idea whether it has succeeded or not, and the program will suppress output unless it is "pretty sure" that it has got the message right. You will discover that when the FEC procedure fails and the program has not realized it, text is occasionally produced looks like plausible (but quite wrong) callsigns or grid locators. You must mentally reject such garbage messages, when they occur. In the small amount of on-the-air testing that JT65 has enjoyed so far, my experience is that that only a few percent of decoded messages are displayed incorrectly. Most of the time, received text is either "letter perfect" or is left blank. 12. You may optionally have the program display text for instances when the decoder in uncertain of its results. Doing so makes use of an "expected message", and you can specify a minimum number of characters thst must match before uncertain text is displayed. By default the expected message is "MyCall HisCall HisGrid" where MyCall is your own callsign, HisCall is the one entered in the onscreen "To Radio" box, and HisGrid is the first four characters of the grid locator in the "Grid" box. 13. At least one design choice used to definine the JT65 algorithm is still subject to change. If the 2.7 Hz tone spacing turns out to be problematic because of propagation anomalies or inadequate oscillator stabilities, the spacing could be increased. There are significant advantages to the smaller spacing, however, so I am sticking with it for now. Beta Release 4.1.1 ------------------ +-------------------------------------------------------------+ | Quick Start Guide to Using WSJT version 4 and the JT65 Mode | +-------------------------------------------------------------+ Version 4 of WSJT marks a significant departure in the program's evolution. The list of features is no longer a full superset of those available in all previous versions. In particular, the JT44 mode is gone; it is replaced by a new mode called JT65 that I hope will be a significant improvement for making QSOs via EME and extremely weak tropospheric paths. The procedure for upgrading to version 4 is identical to previous upgrades, with one exception: the setup procedure will install a new program in your existing WSJT directory, but will keep the old version alive as well. The two versions of the program can coexist peacefully. During a testing period you will probably want the older version and JT44 to be available as well. What is different about JT65, compared to JT44? Here's a short list: 1. T/R period 60 s; actual TX audio duration 46.8 s. (Yes, this will put somewhat more stress on your PA. If its cooling is marginal, take appropriate action.) 2. Modulation uses 64 data tones plus a sync tone. 3. Tones are spaced by 2.7 Hz in frequency and 0.372 s in time. The total occupied bandwidth of a JT65 signal is about 180 Hz. 4. Transmissions consist of 63 data tones or "symbols," each carrying 6 bits of information. In addition there are 63 sync symbols for establishing time and frequency synchronization. 5. Software AFC (automatic frequency control) can follow drift rates up to about +/-10 Hz/minute. 6. User messages are tightly "source coded" into 72 bits. 7. Strong FEC (forward error correction) coding is used to mitigate transmission errors. 8. User-level mesage formats are designed to permit maximum possible efficiency in EME or similar QSOs. A valid message must be be one of three possible types: 1. "Call1 Call2 Grid" or "Call1 Call2 Grid OOO" 2. "RO", "RRR", or "73" (so-called shorthand messages) 3. "any text you want" (up to 13 characters selected from a 42-character alphabet) Instead of a callsign, the first field of a message type 1 may contain "CQ" or "QRZ". Other such "special tokens" may be added later. The available alphabet of characters for message type 3 is: 0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ +-./? 9. Instead of JT44's very effective intra-message averaging of the even, odd, and "last N" characters of a message, JT65 offers even more powerful special shorthand messages for RO, RRR, and 73. These have much higher sensitivity than the other message types. They work reliably down to -30 dB and below. If you succeed in exchanging callsigns and "OOO" by means of type 1 messages, you should certainly be able to complete the QSO. 10. Message averaging over subsequent transmissions is intended to work as it did in JT44, with one exception. The "OOO" signal report typically used for EME can be appended to message #1 and the message will continue averaging without any need to start over. The presence or absence of "OOO" will be detected if sync is achieved, whether or not full decoding has been successful. [Note: message averaging is only partially functional in the first beta release of the JT65 mode. I will need some experience with the mode in order to optimize the code here.] 11. Receiving FEC-decoded transmissions takes some getting used to. You will find *much* less gibberish on your screen in JT65 than in JT44. The FEC algorithm has a good idea whether it has succeeded or not, and the program will suppress output unless it is "pretty sure" that it has got the message right. You will discover that when the FEC procedure fails and the program has not realized it, text is occasionally produced looks like plausible (but quite wrong) callsigns or grid locators. You must mentally reject such garbage messages, when they occur. In the small amount of on-the-air testing that JT65 has enjoyed so far, my experience is that that only a few percent of decoded messages are displayed incorrectly. Most of the time, received text is either "letter perfect" or is left blank. 12. The program makes use of what it calls the "expected message". By default this message is "MyCall HisCall HisGrid" where MyCall is your own callsign, HisCall is the one entered in the onscreen "To Radio" box, and HisGrid is the first four characters of the grid locator in the "Grid" box. The AFC algorithm, in particular, can work more effectively when the actual message received matches the expected message. 13. Several design choices used in defining the JT65 algorithm are still subject to change. If the 2.7 Hz tone spacing turns out to be problematic because of propagation anomalies or inadequate oscillator stabilities, the spacing could be increased. About 1 dB of additional S/N could be obtained by omitting the "Grid" field of the standard message format. This would also reduce the maximum "plain text" message length from 13 characters to 10. 14. Please note that the "Clip" and "ST" are not functional in WSJT v4.1.1. Their settings do not affect program operation. Beta Release 3.8.1 ------------------ This release contains the following new features: 1. An optional large spectral display. Its axes are reversed with respect to those of the FSK441 and JT6M waterfall displays; in the new plot, frequency runs from left to right and time from top to bottom. 2. A new birdie zapper that is especially effective in the FSK441 and JT6M modes. You can watch what it is doing by observing the waterfall displays before and after the "Zap" box is checked. Click "Decode" to refresh the displays after checking or uncheking "Zap". 3. Message decoding in FSK441 mode has been improved in several ways. DF is more accurately determined. Better synchronization is achieved with the precise timing of 25-sample tone bursts comprising each message symbol. Folding of messages is no longer attempted. (It seems that whenever a ping is long enough to make folding effective, it's also strong enough to make folding unnecessary.) As in previous versions, the left mouse button allows decoding of single-tone messages while the right button suppresses them. 4. It is now permissible to click "Gen Std Msgs" while transmitting in FSK441 or JT6M mode. When this button is clicked, the Tx message number will reset to #1 if and only if the callsign in "To Radio" has changed. 5. Message formatting templates for FSK441 and JT6M may now include the special codes %G and %L to insert your 4-digit grid or full 6-digit locator, respectively, in the message. 6. "Reset Defaults" now resets the QRN level to 5, as it should. 7. The decoding parameters in use are no longer listed in the Status Bar at lower right. This listing had become redundant. 8. In Monitor mode, the name assigned to the Rx wave file begins with "Mon_" instead of the "To Radio" callsign. 9. In FSK441 mode, all messages with more than 3 non-blank characters are transmitted in multi-tone mode. Beta Version 3.6.4 ------------------ The previous version had the threshold set very low for all mouse-picked decoding attempts in JT6M mode. This was useful to me for testing, but produced excessive gibberish. Version 3.6.4 has a reasonable threshold and produces much cleaner output. In addition, the message averaging algorithm has been improved. I find that I seldom need to use either "Freeze" or "FixAve" now. These buttons are still present in version 3.6.4, but unless I hear that people are finding them useful I may remove them soon. An option has been added to the Setup menu so that you can determine whether you want the "Tx Stop" button to kick you out of Auto Mode, or not. The default is not to do so. ***IMPORTANT*** If you can make WSJT v3.6.4 crash in normal operation, please let me know (and tell me how). Also let me know if you find anything that does not work as intended. I believe the present code behaves well and is nearly ready to be upgraded from "Beta" status to a full release. WSJT Beta Version 3.6.3 ----------------------- Version 3.6.3 has a number of small improvements, many of them not immediately visible. Several bugs have been exterminated. The JT6M decoder works better and is more sensitive. JT6M now has a "FixAve" checkbox; when this is checked, average messages will be sought only at the lengths of the automatically generated messages. WSJT Beta Version 3.6.2 ----------------------- Version 3.6.2 corrects a problem that on a few machines caused a fatal error that could only be corrected by deleting the initialization file, WSJT361.INI. WSJT Beta Version 3.6.1 ----------------------- Version 3.6.1 corrects a problem that caused an immediate termination on program startup on some users' machines (typically older ones). If you need this upgrade, you need it bad! It also fixes a minor bug that could cause a divide-by-zero error if you clicked on the plot area when no data was available to analyze. WSJT Beta Version 3.6.0 ----------------------- The JT6M mode is only a week old, but many hundreds of QSOs have been made with it all over the world. Thie mode is expressly designed for meteor scatter on 6 meters, but it may be useful with some other propagation modes as well. It has characteristics that place it about midway between FSK441 and JT44 in both speed and sensitivity. A second beta release of WSJT with JT6M, Version 3.6.0, is now available for downloading at http://pulsar.princeton.edu/~joe/K1JT/UPD360.EXE The download file is about 0.75 MB in length. Comparison of Versions 3.6.0 and 3.5.1 -------------------------------------- 1. Version 3.6.0 has a better and faster decoder for JT6M. Further improvements in this area are still to come. 2. Several bugs that could cause crashes in V3.5.1 have been fixed. Version 3.6.0 appears to be quite stable (at least on my own computers). 3. Some buttons have been moved from their familiar positions. 4. Gadgets that Microsoft calls "Up-Down Controls" have replaced the +/- buttons used for various decoding parameters. 5. You can now drag the mouse pointer across any portion of the large plot area to cause decoding of that portion of a recorded file. This works in both FSK441 and JT6M mode. The area you selected is marked when the plot is refreshed. 6. A yellow curve now appears above the green curve in JT6M mode. The yellow line represents measured power received at the sync-tone frequency and in the sync-tone intervals. I haven't yet written up any detailed instructions for the JT6M mode. However, if you are an experienced WSJT user you probably won't need any. Here are a few things it will be helpful to know about JT6M. 1. Like JT44, JT6M uses 44-tone FSK with a "sync tone" and 43 possible data tones -- one for each character in the supported alphanumeric set. The sync tone is at 1076.66 Hz, and the 43 other possible tones are spaced at 21.53 HZ intervals up to 2002.59 Hz. The tones are keyed at a rate of 21.53 baud, so each one lasts for 1/21.53 = 0.04644 seconds. The sync tone is ON during every 3rd transmission interval; tones representing two data characters follow each sync tone. The transmission rate of user data is therefore (2/3)*21.53 = 14.4 characters per second. The transmitted signal sounds a bit like piccolo music. 2. Basic operation is very similar to FSK441. Just select JT6M from the WSJT "Mode" menu (or use "Shift-F7") and then proceed as usual. 3. The green line and waterfall display have their usual meanings. 4. The program attempts to decode both single pings and an "average message". The average is the last line displayed during each decoding attempt, and is flagged with an asterisk as in FSK441. The number to the right of the asterisk is the inferred message length (for example, the number should be 10 for the message "W8WN K1JT "). If the program finds the wrong length, the average will be garbled or meaningless. All message lengths should be even numbers, because odd-length messages are padded with an extra space at transmission time in order to make them even. 5. Clicking with the left mouse button decodes a 4-second block of data near the mouse pointer. The right button uses a longer segment of 10 seconds. Drag the mouse with the button down to select any desired region. Experiment for best decoding as necessary. 6. As soon as you see some properly decoded text, set the DF box to the measured DF and check the "Freeze" box. Alternatively, you can set the Tolerance to a low value (say 25 Hz) after the desired signal has been identified. Use RIT, if necessary, to bring subsequently measured DF's down to a small value. 7. Default settings for the decoder are S > -12 dB, Tol=400 Hz, DF=0 Hz. 8. JT6M can work with signals that are up to 13 dB weaker than those required for FSK441. In the brief on-the-air tests I've made with W8WN, we have both found that mouse-clicking on the smooth green line, even where nothing was heard and nothing can be seen, sometimes causes both callsigns to pop up out of the noise! Version 3.0 ----------- A new major release of WSJT, Version 3.0, is now available for free download. Instructions for upgrading and for complete installations can be found at http://pulsar.princeton.edu/~joe/K1JT. What's new in Version 3.0? -------------------------- 1. The "User's Guide and Reference Manual" has grown to 51 well illustrated pages. The new manual includes four major new sections which together cover: a) the EME Echo mode, including the "Measure" sub-mode and the "EME Calc" utility for estimating the strength of your echoes from the moon; b) the nature of the various astronomical calculations done within WSJT, and a summary of their accuracies; c) several pages describing availability of Fortran source code for the DSP algorithms in WSJT, as well as a suite of test programs for generating simulated data and testing the FSK441 and JT44 encoding and decoding algorithms; d) a short essay on possible future developments for WSJT. 2. The astronomical calculations in WSJT have been overhauled and thoroughly checked. In practice the differences from V2.9 will seem minor, but as described in the new manual's Appendix B, the accuracies of computed positions for the sun and moon, and for the EME Doppler shift, are now well documented. Computed positions are accurate to within about 0.04 degrees, and Doppler is better than 1 Hz at 144 MHz. Exactly what is meant by the displayed data is now described in the manual, as well. 3. A new feature has been added on the Help menu. It pops up a screen summarizing the standard message exchanges used for minimal QSOs using FSK441 and JT44. No more excuses for not being sure about which message you should send next! 4. The Measure mode can now be left running indefinitely, with its output written to a file. A few individuals have been wanting to use the program for radio astronomy purposes, and the new version provides a minimal facility for doing this. 5. The display screen for the EME Calc utility has been cleaned and tightened up. 6. A few other small niceties, all minor in scope. Version 2.9.0 ------------- I am pleased to announce the availability of an upgrade to WSJT Version 2.9. The upgrade provides most of the features of the soon-to-be-released Version 3.0; I am releasing it now as Version 2.9 because many users have asked for early access to its new capabilities. A full release of Version 3.0 must await some updating of the Users Guide and Reference Manual. Probably a few additional features will be added by then, as well. Please be patient! Version 2.9 is available only as an upgrade. As usual, it can be downloaded from the WSJT web site, http://pulsar.princeton.edu/~joe/K1JT, or from the European mirror site http://www.dk5ya.de. New features of the program include the following: 1. EME Echo mode now works on certain computers (generally older, slower ones) that previously refused to run Echo mode properly. The new version runs fine under Windows 95 on my ancient 100 MHz Pentium with 32 MB of RAM. 2. A new feature known as "Measure" can be selected from the main screen in EME Echo mode. Click the Measure button and your system will record the received audio for one second, compute the level of the noise and display the result in units of dB relative to the nominal WSJT "0 dB" level. The program will repeat this measurement cycle every 2 seconds and plot the results as a green line in the graphical screen area. You can use this mode to measure Sun noise, antenna temperature, ground noise, preamp gain, and a host of other useful quantities, relative to a chosen reference level. 3. A pop-up utility labeled "EME Calc" can be selected from the EME Echo screen. It provides an easy way to predict whether you should be able to detect your own echoes from the moon, as well as your ability to work another station by EME. Boxes are provided to enter your TX power, TX feedline loss, RX noise figure, RX feedline loss, antenna gain, ground gain, ground noise, and sky temperature. Similar quantities can be entered for a second station, and you must also specify the operating frequency. When you click "Compute," the program will calculate the maximum expected echo strengths for the "Home Station" and the "DX Station" individually, as well as the maximum expected signal strength of each station at the other location. The program also estimates the averaging time that would be required to detect echoes at the predicted signal level. Signal strengths are quoted relative to the WSJT standard, the noise power in a 2500 Hz bandwidth. If the computed result for your echo exceeds about -38 dB, you have a chance of being able to detect your echoes using WSJT. In comparison, echoes are detectable by the human ear only if they exceed about -14 dB on the same scale, or equivalently +3 dB in a 50 Hz bandwidth. Note that the estimated signal strengths are supposed to be the maximum values expected for the specified conditions. There are many reasons (Faraday rotation, ionospheric scintillation, libration fading, ...) why the actual signal strength may be different, and deviations are much more likely to be downward than upward. The predicted echo strength for my present 144 MHz station at a reasonably good time of the month is around -25 dB. My experience has been that the predictions are fairly good if enough time is spent to be sure of catching a Faraday rotation peak. A number of smaller enhancements have been made in the EME Echo mode of WSJT. These include the following: 4. The program is much better behaved when operating at higher frequencies, in particular 1296 MHz and above. If you fail to enter an RIT setting or specify one that would make the return echo fall outside the audio frequency range 900 - 2100 Hz, the program will suggest a better RIT value for you to use. 5. The amount of programmed frequency spread of your transmitted signal (the "Dither" magnitude) can be set to any value in the range 0 to 500 Hz. It defaults to 50 Hz. 6. Instead of accumulating average echo parameters indefinitely, you can set a parameter "Tavg" that specifies a time constant for averaging. The default value is 5 minutes; at this setting the average echo spectrum will build up as before for the first 5 minutes, but thereafter it will track the signal characteristics over the most recent 5 minutes. In other words, the average gradually "forgets" the signals received more than Tavg minutes ago. Setting Tavg to a large number, say 999 minutes, will closely approximate the program's previous behavior. If you can detect your EME echoes easily and want to see how they vary with time, you might set Tavg to 1 minute, start a "Measure" sequence, and take down the signal level readings at one minute intervals. 7. Information sent to the main text window every 6 seconds now accumulates, with the text window scrolling as necessary. The output is also (optionally) written to file DECODED.CUM so that you can study the data later. Other miscellaneous improvements and bug fixes include the following: 8. In JT44 mode, if the "Grid" box is left empty then no EME Doppler shift will be displayed. 9. The minimum "Dsec" increment has been reduced from 1 second to 0.5 second. This will permit more precise on-the-fly correction of the Windows clock for use by WSJT, should that be necessary. 10. JT44 messages are always exactly 22 characters in length, and any additional characters are ignored. To make this behavior more obvious, any excess characters are now visibly removed from the screen when transmission of a message begins. 11. In Version 2.3.0, hitting the F4 key while in echo mode would cause the program to crash. Fixed. 12. Switching between modes could cause the Auto Period button to be stuck in the disabled or "grayed out" state. Fixed. 13. The last character in a manually edited FSK441 message was not sent if the default trailing blank and "<" character were erased. Fixed. Version 2.3.0 ------------- This release of WSJT is the first to include the EME Echo mode. This mode allows you to detect and measure your own lunar echoes, even if they are far too weak to hear. The mode can be highly useful for evaluating your station performance, even if you prefer to use CW rather than JT44 for your EME QSOs. If you are a present user of WSJT with no interest in detecting and measuring your EME echoes, you will find no significant advantages to upgrading to WSJT Version 2.3.0. With the exception of a minor bug fix, the FSK441 and JT44 modes are essentially unchanged. You can download the upgrade from the WSJT home page, http://pulsar.princeton.edu/~joe/K1JT, and soon also from the European mirror site http://www.vhfdx.de/wsjt. To upgrade an existing WSJT installation of Version 1.9.4 or later you should download and execute the file UPD230.EXE, which will replace your existing files WSJT.EXE and WSJT1.DLL with new files of the same name. I have not yet produced a new full distribution of the latest version. Doing this will require extensions to the User's Guide and Reference Manual, and will probably be accompanied by further improvements to the program. If you wish to do a full installation of Version 2.3.0 from scratch you should download the installation file WSJT222.EXE, run it to install Version 2.2.2, and then upgrade to Version 2.3.0 as described above. Minor Bug Fix: In WSJT Version 2.2.2 and earlier, if you dismissed the "Setup | Options" page by clicking on the "X" in the upper right corner of the form, instead of by clicking the "Done" button, the home station callsign ("My Call") would revert to its default value "K1JT". (Contrary to popular opinion, this was not an subtle ploy designed to increase my own VUCC totals; it was simply a coding mistake.) The bug has now been fixed. EME Echo mode is presently a plain, no-frills implementation. You can activate it from the Mode menu or by striking function key F9. Most of the familiar WSJT buttons will then disappear from the screen, leaving just a few that are essential for controlling the Echo mode. If your station is already operational in the FSK441 and JT44 modes and you have provided the proper signal levels, all you need to do for an echo test is to start WSJT Version 2.3.0, hit F9 to switch to EME Echo mode, aim your antenna at the moon, pick a clear frequency, and toggle Auto Period On. The program will then start cycling through the following loop: 1. Transmit a fixed tone for 2.0 s 2. Wait about 0.5 s for the start of your echo 3. Record the received signal for 2.0 s 4. Analyze and plot the results 5. Repeat from step 1 The loop cycle time is 6 seconds, so the transmitter duty cycle is only 2/6 or 33%. Your transmitter will think it is loafing. At the start of each transmission the frequency of the transmitted tone is randomly dithered by an offset up to +/- 100 Hz around a nominal value of 1500 Hz. The programmed offset is removed from the computed spectrum of each recording before it is added into the accumulating average. This procedure helps to minimize the effect of birdies in the receiver passband: in the average spectrum a fixed-frequency birdie will be smeared out over a 200 Hz range, while the desired signal remains sharply defined. Two curves are plotted in WSJT's main plot area during each pass through the Tx/Rx loop. Each represents the spectrum of received power over a 400 Hz range centered on the expected echo. The curve in gray is a reference spectrum that you can use to be sure you have chosen a reasonably birdie-free passband. It is aligned so as to remove the EME doppler shift computed at the start of your run. The alignment will not be subsequently adjusted for changes in doppler shift or for the random dithering of the transmitted frequency. Stable birdies will therefore stay fixed in the blue spectrum, making them easy to recognize and evade if necessary. The red curve displays the desired EME echo signal. Spectra computed for each 2-second receive period are shifted to correct for changing doppler shift and for the programmed frequency dithering, and are then averaged. The EME echo should apprear as a narrow spike near the middle of the red curve, close to DF = 0. In addition to the graphical display, a line is presented in the WSJT text window in the following form: N: 16 Sig: -26.3 dB DF: -1.3 Hz Width: 0.7 Hz Q: 9 This information gives the number N of Tx/Rx cycles that have been averaged, the mean signal strength in dB, the measured frequency offset of the detected echo from the expected frequency, the spectral width of the echo, and a relative quality indicator for the detection on a 0 - 10 scale. Signal strength is measured in the same units as used in the FSK441 and JT44 modes, i.e., in dB relative to the received noise power in a 2500 Hz bandwidth. Low values of Q represent dubious detections, in which case the values of signal strength, DF, and width may be meaningless. A note about doppler calculations will be helpful here. Most computer programs in amateur EME stations use approximate formulae to compute the position and distance of the moon and the expected doppler shifts of echos. WSJT is no exception, and I cannot presently quote a firm figure on the accuracy of its doppler calculations or describe how their errors depend on lunar coordinates or the accuracy of your station location. EME Echo mode uses a calculated doppler shift to align received spectra so that the echo should appear at DF = 0. I have found that on 2 meters the return signals generally fall within 10 Hz of DF = 0. Further improvements in accuracy of the doppler routine will be forthcoming. If you can hear your own EME echoes you should see a spike in the red curve within a few seconds after toggling Auto Perion On. If your echoes are 10-15 dB below the audible threshold you should see a significant spike on the red curve within a few minutes. To give you a better example of what to expect, consider the parameters of my station. On 2 meters I run up to 400 Watts to a 17.6 dBd antenna (4 x 9 el yagis) aimed at the horizon. I have never heard my own CW echoes with this system. However, when the moon is in my elevation window at 0 - 10 degrees I can easily detect my echoes at any time of the month using WSJT in its EME Echo mode. Indeed, I can turn off the PA and use my FT-847 barefoot, delivering 35 W to the antenna, and still detect my echoes rather easily, even with the moon in a "bad" part of the sky and the path degradation as large as -8 dB. I seem to be able to detect my echoes reliably down to relative signal levels around -36 dB. The spectral analysis done in EME Echo mode provides a frequency resolution of 0.67 Hz. If your transmitter or receiver has short-term frequency stability much worse than this value, so that something drifts or wobbles by more than about 1 Hz in 2 seconds, your echo sensitivity will be degraded. Most modern radios have no difficulty in meeting this standard on the 6 and 2 meter bands, but the higher UHF and microwave bands will be more problematic. I don't have enough experience to know what the spectral width of an echo on the 432 MHz or 1296 MHz band should be after averaging for a few minutes. I have been measuring widths of 2 Hz or less on 2 meters. By default WSJT will assume that your receiver and transmitter are tuned to the same frequency. An on-screen box labeled "RIT (Hz)" is provided so that you can inform the program of any offset receiver tuning, for example to accommodate a large doppler shift. Suppose you are running a test on 70 cm and the predicted doppler shift at the start of the run is -1087 Hz. That would cause echoes from the 1400 - 1600 Hz transmitted audio tone to come back as low as 313 Hz, probably well below the low-frequency cutoff in your receiver's passband. Use your transceiver's RIT control to offset the receiver tuning by some round number within a few hundred Hz of the predicted value -- say -1000 Hz in this example -- and enter this offset in the RIT box before starting the echo measurement. The program will accomodate subsequent changes in the doppler shift up to 800 Hz or so, if necessary, without any further adjustments. Your echo should appear at the center of the red curve, as usual. You won't need to use the RIT feature on 6 or 2 meters, where doppler shifts are much smaller and echoes always fall well within the receiver's SSB passband. Please note that I have so far tested the EME Echo mode only on 2 meters. You will discover that the software implementation is not yet highly polished; a number of improvements are already in the works, but I want to gain the advantage of feedback from other users before I go too much further. If you use the EME Echo mode -- especially on bands other than 2 meters -- please send me your comments, experiences, and suggestions! I am anxious to know how well it works for you. Version 2.2.2 ------------- This is a minor maintenance release. As usual, you can download it from the WSJT web page, http://pulsar.princeton.edu/~joe/K1JT. In addition to a short update file, UPD222.EXE, a full distribution of WSJT Version 2.2.2 is available as WSJT222.EXE. I had thought that Version 2.2.1 had already fixed a sometimes baffling bug appearing in earlier versions: if WSJT was terminated when in its "minimized" state, it could get "stuck" on your Windows taskbar and refuse to return to a full size display. It's not always easy for me to test program revisions on all available versions of Windows, and it seems that the V2.2.1 fix did not solve the problem on at least some versions of Windows 98. This time, with V2.2.2, I believe it's *really* fixed! If WSJT is stuck in the minimized state you should fix it as follows, and then upgrade to Version 2.2.2: A) Start WSJT. It should appear in minimized form on the taskbar at the bottom of your screen. B) Right-click on the WSJT taskbar label and select "Move". C) Press the "left arrow" and/or "up arrow" keys a few times and then move the mouse. You should start to see a "dotted frame" indicating the location of the WSJT screen. D) Click the left mouse button, and you should be back in business. Then you should download and install UPD222.EXE and upgrade your system to Version 2.2.2. Version 2.2.1 ------------- This is a minor maintenance release. You can download it from the WSJT web page, http://pulsar.princeton.edu/~joe/K1JT. In addition to the short update file, a new full distribution of WSJT Version 2.2.1 is available, as well as a new version of the manual addressing the new program features. As always, I will be pleased to receive comments and suggestions at email address k1jt@arrl.net. Please note, however, that I will be on vacation and not reading email from July 7 through 21. Version 2.2.1 fixes the following minor bugs in Version 2.2.0: 1. When first started without a valid INI file, the v2.2.0 would fail to "Generate Std Messages" when asked to do so. Once you have switched modes, say from FSK441 to JT44, the program worked correctly. 2. Local hour angles greater than 180 degrees are now displayed as negative angles. 3. If you did not check the menu item "File | Save text in File DECODED.CUM", the v2.2.0 would create an unwanted file named "fort.21" and write all decoded JT44 text there. 4. If WSJT was terminated when in the "minimized" state, it could get "stuck" on your Windows taskbar. If you are stuck in this mode you should fix it as follows, and then upgrade to Version 2.2.1: A) Start WSJT. It should appear in minimized form on the taskbar at the bottom of your screen. B) Right-click on the WSJT taskbar label and select "Move". C) Press the "left arrow" and/or "up arrow" keys a few times and then move the mouse. You should start to see a "dotted frame" indicating the location of the WSJT screen. D) Click the left mouse button, and you should be back in business. Version 2.2.0 ------------- WSJT Version 2.2.0 provides several significant enhancements, a number of smaller improvements, and four minor bug fixes. To upgrade to v2.2.0 you should download the self-extracting zip file UPD220.EXE and execute it to extract its contents, directing the resulting files to your WSJT installation folder. The new version includes the following changes: 1. The JT44 mode now has an adjustable parameter called "Clip." It can be controlled with +/- buttons just below analogous ones for the "Sync" parameter. The value of Clip defaults to 0, where it has no effect. By increasing Clip to 1, 2, or 3 you can introduce "soft," "moderate," or "hard" clipping of any sudden increases in signal strength that might ruin the decoding of an otherwise usable signal. I have found that setting Clip to 2 or 3 permits me to use JT44 in the presence of summertime QRN that renders v2.0.1 useless. Clipping also helps to accommodate occasional meteor pings in a JT44 QSO, recovering the program's ability to synchronize on a weak residual signal. You can leave the clipping turned on; note, however, that using hard clipping on a signal that does not require it can cost you about 1 dB in message S/N. I recommend generally leaving Clip set to 0 and increasing it only when necessary. 2. JT44 mode has a new checkbox labeled "Zap Birdies." It does just what you would hope such a command would do -- and it can turn a totally spoiled signal into good copy! (In the upgrade file I've included an example wave file recorded via EME from W7FG, in the presence of a strong birdie at my station. To become a believer, try decoding this file both with and without "Zap Birdies" checked. He was sending me the message "K1JT W7FG EM26 ".) The Zap algorithm works best with birdies that are steady in both amplitude and frequency. A sure indicator that you have a birdie problem is a persistent extra spike (or spikes) in the red-line plot, in addition to the one corresponding to the JT44 sync tone. If the birdie is higher in frequency than the sync tone by 20 to 465 Hz, you will probably see a number of identical garbage characters in the line of decoded text. If this happens, check the "Zap Birdies" box and hit "Decode Again", and your copy should improve. It will work best if you have already identified the correct value of DF and checked the "Freeze" box to lock onto the Sync tone. Do not expect miracles! Keeping birdies out of your receiver or QSYing to avoid them will always work better than trying to deal with them in software. Nevertheless, this birdie-killer can make the difference between a successful QSO and one that fails miserably. 3. The JT44 mode has a second new checkbox labeled "Fold Msg." For messages having identical content in the first and second half, this feature can yield a signal-to-noise improvement of 1.5 dB. The JT44 default message formats have been modified slightly to maximize the opportunities for useful message folding. For example, if K1AA is working G2ZZ, the first EME-style message will now be generated as "G2ZZ K1AA G2ZZ K1AA ". (Notice the two spaces in the middle of the message and at the end.) If the "Fold Msg" box is checked, the message will be decoded simply as "G2ZZ K1AA ". Try decoding a marginal signal both with and without the "Fold Msg" box checked. QSB conditions might make one or the other preferable at a particular time. 4. In WSJT versions 2.0.0 and 2.0.1 the JT44 decode algorithm produces a single-character average of the last four character positions in a message. In Version 2.2.0 the averaging limit has been changed to equal the number of "O" characters (for EME messages) or "R" characters (for non-EME messages) at the trailing end of default TX message #2. For example, if clicking "Generate Std Texts" produces "G2ZZ K1AA OOOOOOOOOOOO" for message #2, the program will produce a single-character average based on the last 12 received character positions. This choice gives you the best possible chance of properly decoding an "O" or "R" report in message #2, and it also gives you a good chance at snagging the "RRRRRRRRRRRRRRRRRRRRRR" message under very marginal conditions. For steady signals the procedure can yield a 5.4 dB S/N advantage over single characters and a 2.4 dB advantage over the four-character average. 5. WSJT now remembers whether you were using FSK441 or JT44 mode when you last exited the program. On startup it restores the most recently used mode, including values of W, S, and Sync, as appropriate. 6. All decoded text in JT44 mode is now written to the cumulative file DECODED.CUM. In previous versions of WSJT, text was saved only in FSK441 mode. 7. The default Sync setting in JT44 mode is now 1 rather than 2. I believe nearly everybody runs with Sync = 1. 8. WSJT v2.2.0 is more complete and more consistent about saving information about the starting directory and the form size of the "File | Open" dialog box. 9. The displayed azimuths for "Hot A" and "Hot B" (direction headings to use for optimal sporadic meteor reflections) now wrap correctly at 0 and 360 degrees. You will no longer see, for example, values like 368 or -8 degrees if you are working someone to your north. 10. In JT44 mode the program now displays the local hour angle of the Moon, in degrees. You will appreciate this if you have a polar mounted EME array. 11. The UTC Offset may now be specified as a floating-point number -- that is, with significant digits after a decimal point. 12. The count of available records displayed in the average message window behaved illogically when "Decode Again" and "Include" were used. This has been fixed. 13. There was a bug in the JT44 display routine that caused error messages to appear when the moon's right ascension was very close to 00:00. The error could appear at most once a month, and it would persist for an hour or so. The bug has been fixed. 14. There was an apparent logical inconsistency in program behavior if "Exclude" was clicked after "Clear Avg" had been executed. This has been fixed. 15. In V2.0.1 if you hit F8 more than once and then hit F7, the "Width" parameter in FSK441 mode would be set to 200 ms. This has been fixed. Version 2.0.1 ------------- This is a minor maintenance release. Principal changes from Version 2.0 include the following: 1. The installation scripts for both the full installation and the upgrade to v2.0.1 are more robust and much easier to use. 2. The decoding parameters for FSK441 mode are saved correctly after you have used JT44 mode. 3. The controls for "Decode Again", "Include", and "Exclude" are now fully available during the Tx period. 4. Certain colors (e.g., the backgrounds of a few labels) now display better on older machines. 5. The background colors behind certain Tx messages in JT44 mode have been corrected. 6. The manual has been updated to conform with the new installation scripts, and a number of other small changes (mostly correcting typos) have been made. 7. The Tab key now moves the focus between various on-screen controls and text boxes in a rational way. Version 2.0 ----------- A major new release of WSJT, Version 2.0 is now available for free download. Instructions for upgrading and for new installations can be found at http://pulsar.princeton.edu/~joe/K1JT. What's new in Version 2.0 ? -------------------------- The JT44 mode for extreme weak signal work was first introduced in beta release 1.8.0. This mode has now matured, the program is stable, and tons of EME QSOs (among others) are being made with it. New features and fixes in Version 2.0 include the following: 1. Full monitor mode with separate averaging of 1st and 2nd sequences. 2. Mouse-selected value of DF for decoding when "Freeze" is checked. 3. Program is much faster at certain critical points, and now runs reliably on a 75 MHz Pentium with 24 MB of RAM. 4. The dreaded "always starts minimized" bug has been fixed. 5. Certain dates (such as "2002 VIII 21" now display properly in machines configured for European format. 7. A number of other small niceties.