Tweaks to sfox_remove_tone.
More work on tone removal.
Minor tweaks to sfrx.f90.
Reconfigure sfrx to call sfrx_sub.
Fix argument handling in sfoxsim.f90.
Bring sfoxsim over from superfoxtx repository.
Updates to remove_tone() routine. Currently disabled pending more tests.
fortran_charlen_t should be size_t, not int, with Clang on MacOS.
This fixes the all-zeroes SNR issue on MacOS.
An attempt to fix the msk144 crash on macos.
Update mac executables to commit 9a3db75.
Update mac executables to sf commit 6e3f191.
Update mac x86 executables.
Update MacOS executables.
Update MacOS executables.
Change lower case nsps to upper case NSPS in sfox_wave_gfsk.f90.
Use GFSK for SuperFox fox transmissions.
Only change Tx duration when mode==Fox and when SuperFox is checked.
Set the correct transmit duration for SuperFox fox transmissions.
Update macos executables.
Add updated macos executables.
Update mac binaries.
Update sfrx binary.
Add updated binaries for macos.
Revert "Update lib/superfox/mac/sfrx"
Update lib/superfox/mac/sfrx
Add updated mac sf binaries.
Change permissions for installed sfrx, sftx, foxchk to 755.
Add macos executables for sfrx, sftx, foxchk.
Tweak erasure probability in ftrsd3.f90
Add noise to the output of watterson.f90 instead of the input.
Add Eb/No to sfoxtest output.
Merge remote-tracking branch 'refs/remotes/origin/superfox' into superfox
Make watterson.f90 spreading function the same as that used in ITU report ITU-R F.1487.
Make ftrsd3 work better for a (127,48) Reed Solomon code.
Make ftrsd3 work with (127,48) 128-ary code.
Reconcile the SNR of simulated complex and real noisy signals.
Only hash the callsign of stations that we are hearing.
fastosd240_74.f90: Fix a couple of typos. No impact on WSJT-X decoding performance.
Changes to enable simulating (240,74) FST4W code with noncoherent 4FSK on AWGN or Rayleigh fading channels
ldpcsim174_91: Add the option to simulate the (174,91) code with noncoherent 8FSK on AWGN and block Rayleigh channels.
Make spot counter work for WSPR.
Protect against out-of-bounds array index. Thanks N2BP.
Minor change needed for compilation on MacOS - comment out an unused variable.
Add hash22calc to installation list.
A small tweak to fst4sim output formatting for 50-bit messages.
When unpacking FST4W messages, discard those with idbm.lt.0 or idbm.gt.60.
fst4sim: Tweak output format for FST4W messages.
Minor to hash22calc.f90
Build and install hash22calc.
New command-line program hash22calc will print the 22-bit hash value of a valid callsign.
FST4W: Add -Y option to jt9, which is like -W except that 22-bit hash values are printed for type 3 decodes.
Replace sprintf with snprintf so that clang doesn't complain.
Updates to NEWS and Release notes.
FST4/W: Improve calibration for SNR estimates.
Upload FST4W-900 spots with tr=15 because WSPRnet does not recognize tr=16.
Merge branch 'develop' of bitbucket.org:k1jt/wsjtx into develop
Edits to Release Notes.
Merge branch 'feat-ft8_sync_improvement' into develop
Merge branch 'develop' of bitbucket.org:k1jt/wsjtx into develop
Minor tweaks to FT8 decoder, remove some debug print statements.
FT8: disable a7 decoding for Fox/Hound and when Enable AP is not checked.
FT8: Avoid a crash when a wav file is processed multiple times.
Improve sync for early decodes.
Improve ft8 decoding on crowded bands.
Remove some debug statements.
Revert commit f5291f2 for WSPR mode only.
Revert a change that was introduced in commit f84ba8 because it breaks JT9 mode.
Fix a crash caused by calling Reset Cabrillo Log when Active Stations window is not open. Thanks to Mike W9MDB./wsjtx &
Allow FST4W transmissions on 30m. Ensure that correct frequency is reported to WSPRnet for FST4W decodes when Rx spin box frequency is not 1500 Hz.
Merge branch 'echo-mode' into develop
Merge branch 'echo-mode' into develop
Fix scaling of waveform in wspr_wav when SNR > 90dB.
Merge branch 'arrl-digi-dx' into develop
Merge branch 'arrl-digi-dx' of bitbucket.org:k1jt/wsjtx into arrl-digi-dx
Make wspr mode work properly with 3 character callsigns like N1D.
Make minimum SNR -24 dB for a7 decodes.
Fix a bounds error.
Merge branch 'feat-a7-UR' into develop
Make SNR calculation for a7 decodes consistent with that of regular decodes.
Fix the failure to auto-sequence when a decode contains a low-confidence a-priori designator, such as "? a3".
Revert "Repair a defect where Auto Seq failed for messages with apriori code."
Make wspr mode work properly with 3 character callsigns like N1D.
Fix the failure to auto-sequence when a decode contains a low-confidence a-priori designator, such as "? a3".
W3SZ patch: Initialize AP params and set npasses=2 before first call to q65_dec0().
Change version number to 2.6.0-rc1.
Ensure that tx1 and txb1 are properly enabled/disabled when changing special operating activity.
Another change to make Q65 autosequencing work the same way as FT8.
First cut at User Guide changes to section 6 to replace JT9 with FT8.
Make incomplete messages like 'K9AN G4WJS R-' encode as free text.
More work on User Guide, section 6.
Add new FT8 sample file.
Remove old FT8 sample file.
Edits to User Guide section 6.
Merge branch 'release-2.5.0' of bitbucket.org:k1jt/wsjtx into release-2.5.0
minor edit to FT8 tutorial.
another edit to FT8 tutorial.
Another attempt to make nagain work correctly.
FT8: ensure that nagain only decodes near nfqso.
Revert "Another attempt to make nagain work correctly."