Nexus 1.15.0
Released 26 September 2026. The current release is 1.24.0. Download it · Every version
Added
-
The Logbook now works from the keyboard.
Tab into the list, then the arrow keys, Page Up and Page Down, and Home and End move through your contacts; Enter edits the one you are on, and Delete asks before deleting it. A Remote browser without the station's edit rights can move through the list but not edit or delete.
-
The Sub receiver of an IC-7610 or IC-9700 now has its own row on the Phone and CW screens.
In Phone's Receiver pane and in CW's rig controls, a SUB row shows the second receiver: its frequency where Nexus knows it (the uplink during a satellite pass), and sliders for its RF gain and, on an IC-7610, its AF gain and squelch. The main receiver's controls are labelled MAIN while the SUB row is there. It appears when the radio runs through Nexus's own CI-V control, and the sliders change only the Sub — the main receiver's controls are untouched. The Remote page shows the same row, and its sliders work while you hold control of the station; a station still on an older version of Nexus shows no SUB row there. Nexus does not read the Sub's levels back from the radio yet, so each slider shows the last value the radio accepted from Nexus, and a knob turned on the radio itself is not reflected. An IC-9700 gets RF gain only: Icom's documentation does not say its Sub has an audio stage of its own. In CW the SUB row hides with the RX DSP controls. Other dual-receiver radios (FTDX101, TS-990S, IC-9100, IC-910H, FTDX5000), an Icom run through Hamlib, and every radio with one receiver look exactly as before. Needs checking on a real IC-7610 and IC-9700.
-
A station on your watch list now stands out on the Call Roster, the Stations list and Spots.
The watch list (Settings ▸ Spots & Alerts) sounded one loud alert when a station you asked for was decoded, and after that nothing on the screen said which row it was. Every station the list names, by call or prefix, by DXCC entity or by grid square, now carries a WATCH tile. It is the same lime tile on all three lists, and hovering it names the entry that matched. A watched station counts as a need: Needed only keeps it, Hide worked keeps it even after you have worked it (on the roster and on Spots), and the Stations list's Needed counts it too. There is nothing to set up. It reads the list you already keep and follows it as you edit it. In a Remote browser it reads that browser's own watch list, the one its alerts already use.
-
The Satellites section can now work out when you and another station can both see the same bird.
Nexus could always tell you when a satellite rises over your own QTH, which is half of what a satellite QSO needs: the other operator has to be able to see it at the same moment, and working that out meant two pass lists, two windows on the screen and a lot of arithmetic. Open "Sked with a station" under the schedule, type their grid square or their callsign, and you get the windows over the next fortnight when one of your ★ birds is up for both of you at once — the day, the times, how long the window lasts, which bird, and where to point. Give it a callsign and it finds their square from the last time you worked them. Nothing else is asked for: your own square and your favourites are already known. Two things it is careful about, because both are easy to get wrong by hand. A window needs the satellite at least 5° above the horizon at BOTH ends at the same instant — two passes merely overlapping on the clock is not a contact, and 5° is about where the trees and the rooftops stop. And each window shows how high the bird gets for you and for them separately, which on a real pass are nothing like each other: one of you can have it overhead while the other is scraping the treeline. Windows far enough out that the orbital elements will be over a fortnight old by then are marked "re-check" rather than quietly presented as certain. If there is no window at all, it says so and tells you how far apart you are — for a low-orbiting bird, two stations much more than about 3300 km apart simply cannot see it together, and that is an answer rather than a failure.
-
The Satellites transponder list now shows a downlink's data rate beside its mode.
A packet or telemetry downlink reads
AFSK · 1200 bdorGMSK · 9600 bdinstead of the mode alone, so you can tell a 1200-baud bird your TNC copies from a 9600-baud one before the pass starts. The rate is the one the SatNOGS database lists for that transmitter; where it lists none — most voice, CW and SSTV entries — nothing is added, rather than a guessed rate or a zero. -
PSK now has editable F-key macros, the way RTTY does.
The PSK dock's four macro keys were fixed in place: the messages could not be changed, there was no way to add one of your own, and the keys drawn on them — F1, F2, F3, F4 — did nothing when you actually pressed them, so the one thing the buttons were labelled for was the one thing they would not do. There are now eight keys, in two sets you switch between: Everyday, which starts as the four messages PSK has always sent plus four blanks to fill, and Contest, a run and search-and-pounce set laid out on the same keys as RTTY's so working both modes does not mean learning two keyboards. Click the pencil on a key to write its title and its message, with buttons that drop in your callsign, theirs, the report and your contest exchange. Press F1 to F8 and the message goes out — from anywhere in the cockpit, including while you are typing in the compose bar. Your keys are saved as you make them and are there at the next start, and any key, or a whole set, can be put back to what it shipped with. (#316)
-
Alerts can now be kept to the parts of the world you actually chase.
Settings ▸ Spots & Alerts has continent boxes and a country search, and anything you pick there narrows the new-one, new-grid and CQ alerts to those places — Europe only, or Europe plus Japan, or just France. It is the same way of naming a place the Spots panel's "Spotted from" chips and Band Activity's hide-by-continent already use, so there is one idea to learn rather than three. Two things always get through regardless of what you have picked: a station calling YOU, and anything on your watch list. Leave every box clear and nothing changes — that is how it ships, and an operator who never opens the setting hears exactly what they heard before. (#174)
-
The POTA/SOTA board marks the activation you are hunting.
After you press HUNT, that activator's row at that park is highlighted and scrolled into view, and it stays marked while you scroll. An activator moving a few kHz within the same park does not scroll the board under you. The mark goes when you clear the hunt or log the contact.
-
The Phone cockpit is laid out as a receiver and a transmitter, and has AF gain, RF gain, squelch and an S-meter.
The two panes named after Hamlib's terms, which put the speech processor and VOX among the receive controls, are now Receiver (an S-meter with a printed scale, then filter width, the attenuator and preamp, RF gain, NB, NR, the notches, AGC, AF gain and squelch) and Transmitter (mic gain, speech processor, VOX and the monitor level). AF gain, RF gain and squelch are new, through Hamlib and through Nexus's own CI-V. With AF nearly off, or the squelch up outside FM, the control shows DECODE?: if your soundcard is fed from the radio's speaker or headphone jack, AF silences FT8, RTTY and PSK as well, and on most radios a closed squelch mutes the USB audio too. A control your radio has not reported is named once at the foot of its pane, Not on this radio: …, instead of taking a row, and without rig control every row stays on screen, greyed, under one line that says they need CAT. The automatic notch is now labelled Auto notch in Phone and CW, and the manual one Manual notch, on every radio: on a Yaesu the old Notch and MN read the wrong way round.
-
Phone shows what goes out when you key, and the transmit meters keep their last reading.
A line above the meters gives the frequency your next over goes out on, with any split and XIT already in it, then the mode, the split offset, XIT, and your power beside the watts the radio measured on the last over, marked where the radio confirmed a figure and where Nexus only commanded it. On FM through a repeater it gives no transmit frequency rather than the one you are listening on, and with VOX on it warns that Stop TX cannot unkey a transmitter your voice is keying. In Phone and CW the SWR, ALC, power and compression meters now keep the last over's reading, dimmed, until you key again, instead of vanishing when you unkey. The power meter's full scale is the radio's Rated power (W), a new setting for each radio in Settings ▸ Radio ▸ Rig & CAT (100 W unless you change it), so a QRP set reads across the whole bar instead of sitting on the first mark.
-
The attenuator, the preamp and the transmit monitor are now controls you can reach.
Nexus had no way to set any of the three, through Hamlib or its own CI-V, so an operator fighting a strong neighbour on a crowded band had to reach past the app to the radio's own front panel for the one control that fixes it. The pad and the preamp now sit at the head of the Phone cockpit's Receiver pane, where the signal meets them, and the monitor level sits at the end of the Transmitter pane with the rest of what shapes your audio. The pad and preamp are offered as the steps your radio actually has — one 20 dB pad on some rigs, 6/12/18 dB on others — rather than a slider, because a radio will refuse or quietly round a pad it does not hold. A radio that tells Nexus it has no attenuator says so in one line at the foot of the pane; a radio that never published its list keeps the control on screen and explains why it cannot be offered, instead of leaving you guessing whether the feature is missing from your radio or from Nexus.
-
VFO A and B can now be swapped from the tuning strip.
The A and B buttons were there and the swap was not, so the one gesture split operating is built on — listen on B, work on A, put them back — meant two presses and a lost frequency, or a trip to the radio. The new A⇄B button sits with the pair it acts on; both VFOs keep their frequencies and simply change places.
-
Beam at the station you are working, the short way or the long way, from Phone too.
The rotator strip's → CALL button, which turns the antenna toward the station you are working, was on the CW and Operate cockpits only; Phone has it now. Beside it, LP turns the antenna to the long-path heading. They are two buttons rather than a switch, so the path you asked for is never in doubt, and the message after each names the path it took. The Remote page's rotator strip is unchanged. (#338)
-
The log shows a contact's date and its time as two columns instead of one.
One "Time (UTC)" column printed the whole instant —
2026-09-14 00:58Z— in a single cell, so the dates could not be read down the column and the time of day sat wherever the date's width left it. They are now Date and Time, side by side; sorting by either still orders the log by when the contact happened, because that is the one thing a contact records. Nothing else on the row moved: the pair shares the width the single column had. One thing you will see change: a contact imported from a log that carried a date and no time of day now shows a dash in the Time column rather than a midnight it never had. That is what "More columns" already showed you in the record itself, and it is why those contacts go unmatched at LoTW and eQSL, which match on time. (#239) -
A contact's full record is now one double-click away.
The logbook was a list of contacts with no way to open one, so everything the table has no column for was being logged and then hidden: the other station's name and QTH, their grid, the rig and antenna they gave you, your own power and mode, the comment you typed at the time, which services the contact has been uploaded to and which have confirmed it, and any ADIF fields that arrived with an imported log and had nowhere to go. Double-click a row and it is all there, laid out and read-only. Fields the contact does not carry are left out rather than shown blank, so a sparse record reads as a short card instead of a page of dashes. (#313)
-
The RTTY decoded-text window now shows what you sent, alongside what you received.
Your own keyed text went to the air and nowhere else, so the one window that is supposed to be the record of the contact held only half of it — you could read their side of an exchange but not your own, and there was no way to check what actually went out. Every over you send now appears in that window a character at a time as it goes out — an F-key macro, a line sent with Enter, an Auto call, or text typed with continuous TX — in the accent colour with a rule down its left edge, so your side and theirs are never confused. Stop TX ends it where the transmission was cut: nothing that stayed off the air appears.
-
The RTTY cockpit shows what you are sending, at the bottom beside the typing field.
Working a contest on the F-keys, nothing on screen said what an over said or whether it had finished — which is exactly the doubt a call that goes unanswered leaves. The line to the left of the typing field shows the over going out, or the last one: the part already sent is underlined, the rest follows it, and an over you stopped shows what never went out struck through. It covers F-key macros, Enter, Auto calls and continuous TX, needs no setting, and takes no room from the log strip. (#379)
-
A satellite tag can be corrected or taken off a contact without leaving Nexus.
A contact carrying the wrong satellite — or carrying one when it was never on a bird at all, which is what the 0.24–0.27 tagging bug left behind — could only be repaired by closing Nexus and hand-editing
log.adiin a text editor. The Logbook row now has a SAT▸ menu beside the QSL one: pick a satellite to set or correct the name, or pick Not via satellite to remove the tag.PROP_MODEandSAT_NAMEare written as a pair and removed as a pair, because TQSL rejects a record carrying one without the other and one rejected record can take a whole signed batch with it. The menu lists only the satellites LoTW accepts rather than letting you type a name, for the same reason the automatic tagging refuses to guess one: TQSL matches the name against its own designators, and a logged contact is a permanent record of a typo. Nexus still rewrites nothing on its own — only you know which pass you were actually on. The row's ✎ edit form deliberately still carries neither field: it treats a blank box as "leave this alone", which is what stops a busted-callsign fix quietly stripping a satellite tag off a contact that earned it, so removing a tag is its own decision and not a side effect of saving the form. -
RTTY Auto call now has a timer you can set.
Auto call sends CQ, listens, and sends CQ again if nobody comes back — and how long it listened was fixed at 30 seconds with no way to change it, so an operator who wanted a quicker cycle could only click the button by hand. Settings ▸ Digital ▸ RTTY now has Listen before repeating, in seconds, and Repeats before giving up, which is how many times Auto call asks a station again inside a contact before letting them go. Both ship set to what Auto call was already doing, so nothing changes unless you change it, and the listen clock starts when your over has finished going out rather than when it was queued. The repeat count does not limit CQ: an unanswered CQ still repeats until you stop it. (#304)
Changed
-
The Needed board now puts every station on your watch list at the top whenever it is heard.
By call, prefix, country or grid, worked or not, as the WATCH tile already does on the Call Roster, the Stations list and Spots, and it carries that same lime WATCH tile on the board and in Band Activity. The Watch list chip shows exactly those stations. Any calls left in the old hidden "Wanted watch list" setting are added to your watch list (Settings ▸ Spots & Alerts) once, with a note saying so.
-
Big logbooks no longer weigh on every window.
Each window now asks Nexus for just the contacts it shows instead of holding its own copy of the whole log. With 150,000 contacts that is about 5 MB of a window's memory instead of 67 MB (6 MB instead of 208 MB at 500,000), the Logbook opens in a quarter of a second instead of over two, and recording an upload no longer sends the whole log (112 MB at 150,000 contacts) to every open window again.
-
Sorting, searching or filtering the Logbook now takes you back to the first contact.
The list used to stay at the same scroll depth, showing whatever contacts had landed there in the new order. A contact logged while you are scrolled down still leaves the rows you are looking at exactly where they are.
-
When the logbook is too busy to take a change, nothing changes and Nexus says so.
A change that keeps meeting other changes on its way in — from a big import, another window or a sync — is tried again a few times. If it still cannot be made, nothing is changed and Nexus says the logbook was busy, so doing it again retries. That covers the Logbook's edits, QSL marks, satellite tags and deletes (the row and your typing stay as they were) and an import, a LoTW, eQSL or QRZ sync, the Field Day merge and the POTA stamps. Mark on LoTW is the one exception. It marks a few thousand QSOs at a time, so a big log does not hold the radio up for the whole of it at once. If it stops partway — the logbook stays busy, or Nexus closes — the QSOs it had marked stay marked, and when the logbook was busy Nexus says how many of how many it reached. Marking again finishes the rest and marks none of them twice.
-
A change made in another window on the same logbook no longer makes the radio wait.
Two Nexus windows can share one logbook, and each used to read the whole log again, while the radio waited, whenever the other one changed it: about a second on a log of 150,000 contacts, nearly four at 500,000. A contact marked uploaded or QSL sent in the other window now costs this one nothing, and any other change is read in while the radio carries on.
-
Turning Field Day on no longer stops the radio to read a busy logbook.
Starting a contest session reads the contacts it checks for duplicates. When the logbook is too busy to answer — a big import, another window, a sync — Nexus tries again for a few seconds at most, then leaves Field Day as it was and says the logbook was busy, so trying again in a moment works. At start-up, a Field Day session the logbook is too busy to resume within a few seconds is left off, and Nexus says to turn it on again when you are ready.
-
Your logbook now lives in a database, and
log.adiis kept as an up-to-date copy of it.Every change to the log used to rewrite the whole of
log.adi— every stamp, every confirmation, every edit — and on a big log that rewrite was long enough to make the radio and the screen hesitate each time a contact was saved or an upload was marked. The log is now kept inlog.sqlite3besidelog.adi, where a change writes only the contacts it touched. The first time this version starts it converts your log — a lifetime log can take several seconds, once, and the start-up screen says so while it works, in your language — and keeps the file exactly as it was, aslog.adi.pre-sqlite. Starting Nexus again while it converts does not open a second copy; on Windows it says Nexus is already moving the logbook and will open by itself.log.adistays where it always was and keeps up with every change, a moment later, so other loggers, backup scripts and sync tools that read it still see every contact. Starting Nexus no longer rewrites the log. The country and US state Nexus shows for a contact that arrived without them are now saved into the log, in the background once the window is up — on the first start of each new version and the first start after a new country or callsign file arrives — so the database andlog.adihold exactly what you see. Quitting waits for the last change to reach the disk. A data folder on a network drive keeps the log inlog.adialone, as before, and so does a start where the database cannot be opened. Either way Nexus says so once on screen, with the reason and what to do. Two radio windows on one data folder now see each other's corrections and deletions as they are: a contact corrected in one window is corrected in the other, not logged a second time beside the old one, and a contact deleted in one stays deleted instead of coming back when the other window next saves. Two things behave differently, and both are the safe side of a change. If something other than Nexus writes tolog.adi— an older Nexus on the same folder, a restored backup — Nexus takes that file's contacts in the way the Logbook's Import does: new contacts are added and confirmations come across, but a contact changed or removed only in that file is not changed or removed in your log. And if Nexus stops in the middle of importing a large file, the contacts it had already taken in are kept; importing the file again adds the rest without duplicating any. -
If Nexus cannot open a logbook at all, it says why and does not start.
Your logbook always lives in a database now: on disk, or in memory when
log.adiis the log's home, as on a network drive. In the rare case that not even the one in memory can be opened, Nexus stops before anything can be logged and says why — in a message box on Windows, and in its diagnostic log on every system. -
The dated backups of your logbook keep several days of copies again, however big the log.
Nexus keeps dated copies of
log.adiin thebackups/folder beside it — one a day while the log changes, and one before anything makes it smaller. That folder was capped at 64 MB, which was room for nine copies of the biggest logs of its day and for only ONE copy of a 150,000-contact log, so the copy taken before a delete replaced yesterday's. It now holds up to four copies of your log (never less than 64 MB): about 190 MB more beside a 150,000-contact log. How to go back to one of them is in the install guide, under "Restoring the logbook from a backup copy" — now that the logbook is a database, dropping a copy overlog.adiis not enough on its own. -
Closing Nexus while the logbook is still saving now says so, and asks before it drops anything.
If you close the window while changes are still on their way to the disk — right after a big import, or on a slow drive — Nexus takes the radio off the air first, as it always has, then keeps the window open with Saving your logbook… and the number of changes still to write, and closes the moment they are saved. A quick save shows nothing at all. If the disk has not taken them after a minute, or refuses one, Nexus asks: Keep trying, or Quit without the last N changes. Restarting after an update, and installing one on Windows, save the same way first.
-
A contact the logbook could not save is kept and sent again, and the screen says so.
If the disk is full, fails a write, or another program holds the logbook for too long, Nexus keeps the change in memory and sends it again (after a few seconds, then about once a minute) until the logbook takes it. A message stays on screen while that is happening, with the reason, and another says so when every change is saved again. A change the logbook refuses outright, which sending again cannot fix, is kept for the session instead, and the quit asks about it. At the quit, Keep trying sends those changes again, so freeing some disk space and pressing it saves them.
-
IC-7610 on native CI-V: the S-meter, the receive controls (AF, RF gain, squelch, NB, NR, notch, AGC, attenuator, preamp) and the CTCSS tone now always act on the Main receiver
, even with the Sub band selected on the radio. The frequency and mode Nexus shows are now Main's too, so selecting the Sub band on the radio, or tuning it, no longer moves them, and a frequency or mode change made from Nexus goes to Main as well. Split works exactly as before. Needs checking on a real IC-7610.
-
IC-7610 on native CI-V: the attenuator offers all fifteen of the radio's steps, 3 to 45 dB in 3 dB steps
, rather than the 6, 12 and 18 that Hamlib lists for it. The list is the one Icom documents for the radio. Where the Receiver pane is too narrow for every step in one row, the steps wrap onto a second row rather than running off the edge. Needs checking on a real IC-7610.
-
On a cross-band satellite pass, the licence check and the band strip follow the band you transmit on.
On an IC-9700 or IC-905 using native CI-V the uplink goes out on the Sub band, so the band strip now shades the phone segment of the uplink's band rather than the downlink's. The transmit lock judges the same frequency it always did. The new refusals are an uplink whose mode Nexus no longer sets because you changed it mid-pass, in FT and the other data modes, PSK31, RTTY sent as audio tones, or CW sent from the sound card, near a segment edge: it now stays locked unless both sides of the carrier are inside your privileges.
-
Nexus now refuses to put your data and log folder on a network drive, and tells you why.
Settings ▸ Config ▸ Data & log folder used to accept a NAS or a mapped network drive, and the manual suggested exactly that for a multi-PC shack. That was survivable while the logbook was a plain ADIF file — the worst a share could do was cost you the most recent contacts, and the rest of the file still read back. The logbook now lives in a database, and a database is not so forgiving: file locking across a network is the classic way to corrupt one, and the worst case is a log that will not open at all. So a Windows share (
\\nas\ham\nexus) or a folder on an NFS or SMB mount is refused now, with the reason and what to do instead. A folder already in use is never moved for you. IfNEXUS_DATA_DIRor adata-dir.jsoncopied in from another machine already points at a network drive, your contacts are in there, and quietly opening a different folder would read as losing them. Settings says so in Folder in use instead, every time you open it. -
A data folder that looks like Dropbox, OneDrive, iCloud or Google Drive is flagged rather than refused.
Those are ordinary local drives, and nothing about the one-Nexus-at-a-time rule has changed — two machines writing one logbook through a sync client still make a conflicted copy, not a merged log. The check is a guess from the folder's name, so it will miss a sync folder you have renamed, which is why it only ever warns.
Fixed
-
Switching radios no longer reaches for the port of the radio you just left.
Nexus keeps the radios you are not using connected in the background, and when you switched radios (by band routing, the radio buttons or a Remote browser) that background connection could try to open the radio you had just left before Nexus had let go of it. On Windows this logged "Access is denied." (seen with an IC-9700 on native CI-V), and the radio could fall back to Hamlib's rigctld instead of its native CI-V connection. The radio you leave is now picked up only after Nexus releases it.
-
A radio switch can no longer leave the radio you switched away from transmitting.
If the switch landed at one particular moment in Nexus's radio cycle, the connection to the radio you were leaving was closed even when its key-up had failed, and a radio connected through Hamlib then had nothing left that could key it up. The radio you leave is still keyed up at once, as before, but its connection is now closed only by the switch itself, which waits while the radio is still transmitting.
-
A radio with its own PTT port now switches instantly, and a radio keeps its keying line after a switch.
A radio keyed by RTS or DTR on a port of its own was closed and reopened at every switch to it. It now switches as quickly as any other radio and keys on the port you configured. The radio you leave lets go of its PTT port at the switch, so two radios keyed through one controller port (SO2R) each key after a switch. A radio keyed by RTS or DTR on its CAT port could, after a switch, key by another method than the one you chose (usually the radio's own CAT command). It is now reopened at every switch to it, so it keeps keying on its line. The connection that keeps your other radios live in the background never opens a PTT port.
-
Choosing a radio from a Remote browser keeps its keying line, and two radios sharing one PTT port can both be chosen.
A radio keyed by RTS or DTR on its CAT port, chosen from a Remote browser, could key by another method than the one you chose (usually the radio's own CAT command). It is now reconnected for the selection, keying on its line. Choosing between two radios keyed through one controller port (SO2R) was always refused. The radio you leave now lets go of the port once its PTT is released, so the radio you choose can use it, and a selection that fails hands the port back.
-
A radio switch no longer sends the new radio's requests to the radio you leave.
If the switch landed at one particular moment in Nexus's radio cycle, requests meant for the radio you chose could reach the radio you were leaving instead. That covered a noise-blanker or other DSP switch, a filter width, RIT/XIT, a scope span or scope mode, and a voice-memory playback asked for by a logger over Nexus's CAT link, which that radio would then transmit. Those requests now wait for the radio you chose. The voice-memory playback is the exception: it is dropped, not played later on the other radio, so press it again. A playback asked for while a switch waits for a still-transmitting radio to stop, or while Test CAT holds the radio's port, is dropped the same way. Earlier releases could send that one to the radio being left too.
-
A tune-up or voice-memory playback pressed just before a radio switch no longer goes to the radio you switch to.
An ATU tune-up pressed for one radio could start on the other if you switched radios, or a band change switched them for you, within a few seconds of the press. A voice-memory playback a logger asked for at that moment could play on the new radio. Both are now dropped at the switch; press again for the radio you chose.
-
Switching radios now stops a voice memory the radio you leave may be playing.
After a switch, Stop TX reaches only the radio you switched to, so a voice memory still playing on the radio you left had nothing in Nexus that could stop it. The switch now asks the radio you leave to stop its voice memory, right after unkeying it. That includes a radio chosen from a Remote browser. A radio whose CAT link offers no such stop is unaffected.
-
Icom radios on Nexus's native CI-V connection now take a voice-memory stop.
On the IC-7300, IC-7610, IC-9700, IC-705 and IC-905 with native CI-V, the stop that a radio switch sends, and one a logger sends through Nexus's CAT link, used to be refused. It is now sent as the radio's own Voice TX memory stop.
-
Stop TX now also stops a voice memory the radio is playing.
Nexus never keyed a voice memory started by a logger through its CAT link, or from the radio's own panel, so Stop TX sent that transmission nothing, and a PTT release may not end one. Stop TX, a logger's Halt TX and the high-SWR cutoff now also ask the radio to stop its voice memory, after they unkey it. A radio whose CAT link offers no such stop is unaffected.
-
After a radio switch, an Icom on Nexus's own CI-V connection uses its own Data mode setting.
The D1, D2 or D3 chosen for the radio you switched to could be replaced by the previous radio's, and on an Icom that setting picks which audio input it transmits from. The setting now follows the radio.
-
Awards' upload and push buttons always act on exactly the contact they list.
If a contact was deleted in another window while Awards was open, the Upload to LoTW and Push to QRZ, ClubLog or eQSL buttons could upload or push the neighbouring contact instead.
-
The Logbook no longer overwrites a change made elsewhere.
If a contact was edited or deleted in another window (or by a sync) before your edit, QSL mark, satellite tag or delete reached it, nothing is changed, Nexus tells you so, and the Logbook shows the contact as it is now.
-
The Logbook no longer slides contacts out from under the pointer when the log changes.
With the list scrolled down, a contact the FT sequencer logged went in above everything on screen and pushed every row below it down by a row's height, so a click aimed at one contact's ✎ or ✕ could land on its neighbour. A contact deleted from another window or a Remote browser slid them the other way. The rows you are looking at now stay exactly where they are. At the top of the list nothing changes: a new contact still appears there, as it always did.
-
The Logbook's edit mark and an open comment stay on their own contact.
When a contact above them was deleted from another window or a Remote browser, both moved onto the neighbouring contact. They now stay with the contact you opened them on.
-
Awards can push a contact you logged after opening it.
Under "Confirmations — why isn't this credited?", the Push to QRZ, ClubLog or eQSL button looked the contact up in the copy of the log Awards took when it opened, so a contact logged since then answered "Could not find that QSO in the log". It now sends the contact the diagnosis names.
-
An open comment in the Logbook no longer leaves a gap, or gets covered, when the log changes.
With a contact's comment or note open, a contact logged or deleted above it left a blank gap where the open row had been, or drew the rows below over its note, until those rows were drawn again. Sorting or searching the list did the same. The open row now keeps its height in its new place. In a Remote browser's Logbook, an open row also no longer flickers out of place for a moment when the page refreshes with a new contact.
-
The callsign card no longer shows "New DXCC!" for a country you have already worked while the log is still loading.
On the log strip and the Operate card, the need badge now appears once the log has answered for that call.
-
After a question or a dialog closes, the keyboard goes back to where you were.
Answering a confirmation, or closing the Getting started guide, the setup wizard, a contact's details, the CHIRP how-to, the satellite question or the logbook-saving notice, left the keyboard on the page itself, so the next Tab started again from the top. It now returns to the button you opened it from, and after a delete to the same button on the next item.
-
A paper QSL card you mark as received now counts toward your awards straight away.
Ticking the card in the Logbook recorded the card but not the confirmation that Awards, the Needed board and the Journey count, so the new confirmation only appeared after Nexus was restarted. It now counts the moment you tick it, and unticking a card ticked by mistake takes the credit back just as quickly.
-
A contact you import now carries its US state from the moment it is imported.
Nexus works out the state from the callsign for a US contact that arrives without one. For an import, a QRZ download or a Field Day merge that used to happen only after the next restart. It now happens as the contact arrives, as it always did for a contact you log yourself.
-
Exporting a big logbook, or starting a LoTW upload, no longer makes the radio wait.
The ADIF and CSV exports (the whole log, a date range, one operator or one activation) and the lists of operators and activations to choose from went through every contact while the radio waited for them, and so did the LoTW upload while it gathered the contacts to sign — a pause the decoder and the waterfall felt on a log of a hundred thousand contacts. They now read the logbook database while the radio carries on, and the files are the same, byte for byte. If a change you made is not in the database ten seconds after you ask for an export — still on its way, or refused by the disk — the file is still written with what the database holds, so an export still rescues your log from a failing disk, and Nexus tells you how many recent changes the file is missing and whether they are still being saved.
-
Keeping
log.adiup to date no longer needs memory the size of your log.Every change used to build the whole file in memory before writing it — about 64 MB on a log of 150,000 contacts, about 210 MB on 500,000, each time.
log.adiis now written straight from the logbook database a few thousand contacts at a time, byte for byte the same file. -
Nexus now checks that a shared
log.adihas not changed before it saves over it.When two computers keep one
log.adion a network drive, a contact one of them added while the other was saving could be lost; Nexus now looks again the moment before it replaces the file, and if anything else has written to it since it last looked, it reads that in first. -
With a big logbook, an upload being marked no longer makes Awards, the Needed board, the Journey and the statistics start over.
Each of them went through every contact again after every upload to QRZ, ClubLog, eQSL or LoTW was marked on a contact, although an upload changes none of them. They now read the logbook database, and only when a contact, a confirmation or an edit could change what they show.
-
Logging a Field Day contact no longer makes the radio wait for the disk.
Every Field Day contact rewrites the contest journal, and that write used to finish on the disk before the radio could go on — including when the FT sequencer logged a contact itself, in the middle of its timing. On a slow disk (an SD card, a busy USB drive) that was a pause the decoder and the waterfall felt. The journal is now written on a thread of its own a moment later, in the same order and with the same contents; a contact you log by hand is still on the disk before the logging window confirms it, and quitting waits for the last write. The queue of held Tempo messages and the JS8 inbox are written the same way.
-
An upload to QRZ, ClubLog or eQSL is now marked on the contact that was actually sent.
With two contacts with the same station on the same band, mode and day, sending the earlier one — from the Logbook, or when an upload was retried, or when Nexus caught up on contacts a service had missed — marked the later one as sent instead. The earlier contact was then sent again at every catch-up, and the later one, marked as sent, was never sent at all. The mark now goes on the contact that went, and a contact corrected or deleted while its upload was on its way is not marked in its place.
-
A big logbook now takes far less memory.
Loading the log kept a spare block of memory beside it as large as all of its contacts put together, for as long as Nexus ran: about 220 MB extra for a 150,000-contact log, and twice that at 500,000. It is now released as the log loads.
-
With a big logbook, the radio no longer waits while Nexus reads the whole log.
The Needed board, a satellite pass's needs, a sked looked up by callsign, the log statistics, the confirmation diagnostics on Awards, the pounce alert's refresh, and a Remote browser's Log and Confirmations each went through every contact while holding the lock the radio needs every 20 ms: up to a sixth of a second at 150,000 contacts, and most of a second for the diagnostics at 500,000. Each now takes a list of the contacts, which costs about a millisecond, lets go, and does its work after. The Needed board, a satellite pass and the pounce alert also share one picture of what is worked and needed instead of each rebuilding it on every refresh.
-
With a big logbook, logging a contact no longer holds up the radio afterwards.
After every contact, the next screen refresh rebuilt the worked-before (B4) marks from the whole log; after an upload to QRZ, ClubLog or eQSL, the next contact rebuilt the NEW DXCC, NEW GRID and NEW PARK marks, looking up the country of every contact again; after an edit or a delete, the next contact rebuilt the duplicate check; and during a contact with a station whose grid was not yet known, every refresh searched the whole log for it. Each happened while holding the lock the radio needs every 20 ms — up to a third of a second at 150,000 contacts. All of these are now kept up to date one contact at a time. A side effect: after a LoTW download or a POTA park import, the "confirmed on this band" filter and the NEW PARK marks now change straight away, where they used to wait for the next logged contact.
-
With a big logbook, nothing the logbook does holds up the radio for more than 5 ms.
Each of the logbook's actions that take the lock the radio needs every 20 ms now lets go of it within 0.8 ms on a log of 150,000 contacts and within 1.6 ms on one of 500,000: logging a contact, a QSL card or QSL-sent mark, a satellite tag, an edit or a delete, an upload stamp, an import, a LoTW, QRZ or POTA sync, "already uploaded", the first launch's fill of countries and states, and clearing the whole logbook (0.2 ms and 0.1 ms). A call's history and the band map's worked calls hold it for under 0.1 ms. The log itself takes 42.8 MiB of memory at 150,000 contacts and 89.7 MiB at 500,000, plus 3.9 MiB of the database's own. Measured on a quiet 32-core machine, at a load of 0.61 when the run began (2.4 for a call's history and the worked calls). Two limits: the 5 ms covers the logbook's own work under that lock, and settings written while holding it are an app-wide pattern it does not cover; and for the snapshot the radio takes after each contact it measures the logbook's side only, a median of 0.001 ms.
-
Filling in the logbook's countries and states no longer reports a failure that did not happen.
After an update Nexus saves the country and the US state or Canadian province of contacts that lack one, and the diagnostic log said how many it saved. It counted every contact outside the US and Canada as lacking a state, which none of them can have, so a log of mostly DX contacts read "country and state saved into the logbook for 0 of the 4847 contacts that lacked one". It now counts only the contacts that could take one, and says the logbook is up to date when none lacks one. What it saves is unchanged.
-
A POTA spot with an emoji after the callsign no longer stops the Needed board.
On 1.13 and 1.14 a real spot on the POTA feed, an activator's call followed by a coffee-cup emoji, crashed the lookup that turns a callsign into a country. The Needed board and the need chips beside the decodes stopped updating for as long as that spot was up, and a browser on Nexus Remote lost its connection each time it read the Needed list. Nexus now reads a callsign only up to its first character outside the plain keyboard set (an emoji, an accented or full-width letter), so that activator counts as the station it is. Nothing that arrives from outside (a spot, a cluster line, an imported log) can crash the lookup any more. Four other places had the same weak spot and are fixed with it: a DXpedition date read from NG3K's list, a contest exchange in an imported log, the APRS iGate's check of a packet's sender, and the line-tidying the diagnostic log does before it writes.
-
A slow space-weather, POTA/SOTA or callbook server can no longer hold up the rest of Nexus.
Eight requests made their wait on one of the few threads that answer the screen: the Kp forecast, the NOAA scales and alerts, the aurora oval, the MUF map, the POTA and SOTA spot lists, a callbook lookup and the QRZ Logbook connection test. Each held its thread for as long as the server took, up to its time limit, and with several going at once on a slow connection everything else could queue behind them. They now wait off to the side, the way logbook uploads always have.
-
A park reference next to a WWFF area or a SOTA summit no longer disappears when you import a log.
Several loggers write a combined activation as a WWFF or SOTA reference plus a
POTA_REF. Importing that record into Nexus kept the WWFF or SOTA reference and silently dropped the POTA one, and a SOTA record lost any POTA tags it carried. They now stay with the contact and are written back out with it. -
A two-fer activation, a Field Day contact after a restart and a contact imported without a mode are now written as proper ADIF.
A two-fer you activate is written with one park in
MY_SIG_INFOand both inMY_POTA_REF: the pair inMY_SIG_INFOnamed a park that does not exist, inlog.adiand in every upload built from it. A Field Day contact in FT4, Q65, FST4 or another MFSK submode keeps its submode when Nexus restarts mid-event, instead of coming back as plainMFSKin the file you submit and in your logbook when you merge. A contact imported with no mode or no grid no longer writes an emptyMODEorGRIDSQUAREfield. -
Getting started now tells a WSJT-X operator which switch keeps JTAlert and GridTracker working.
Its "Coming from WSJT-X?" note said both keep working, and they do, but only once WSJT-X UDP API in Settings ▸ Logging & Connectors ▸ Integrations & Feeds is on, and it ships off. A GridTracker2 user went looking for an ADIF file instead. The note now names the switch, says it is off by default and that its address already matches WSJT-X's, and links straight to that section. The manual's pages that made the same promise say so too. (#353)
-
Spanish and French no longer show
\u00a0in the middle of a label.Twelve strings in each language printed the code for a non-breaking space instead of the space itself: the Flex pan and rig scope buttons on the CW and Phone cockpits, five lines of Getting started, two Settings hints and a line of the setup wizard. They show the space now, and a check stops a translation from bringing the code back.
-
An IC-9700 is no longer offered XIT.
The radio has no XIT: Icom's CI-V reference for it lists RIT and no transmit offset. Nexus drew the XIT buttons anyway. Over Nexus's own CI-V connection an XIT change went out as Icom's transmit-offset commands, whose offset register on the 9700 is the RIT offset, and over Hamlib the radio refused it. In both cases Nexus went on showing an XIT offset the radio never applied, on the transmit line and in the frequency it checks against your licence privileges. On an IC-9700 the XIT buttons and the transmit line's XIT figure are gone, and an XIT change aimed at one is refused, including from a browser. RIT is unchanged, and so is XIT on every other radio.
-
An Icom on Nexus's own CI-V connection has its manual notch and speech-processor depth back.
Turning native CI-V on, the connection that brings the Icom panadapter, took the manual notch button and the processor depth slider off the Phone cockpit without a word, because the native connection could not answer for either. It now can. The notch frequency cannot be set over native CI-V, the same as through Hamlib.
-
The transmit lock now judges the sideband you picked in Phone.
Picking USB, LSB or AM in the Phone cockpit, or recalling a saved memory in its own sideband, changed what the radio sent, but the licence check went on judging the sideband the band normally uses. So a General could pick USB at 7.299 and transmit across the top of 40 m, or LSB at 14.226 and go below the bottom of the 20 m General phone segment, and nothing locked. Both now read 🔒 TX locked, like any other frequency outside your privileges, and a Remote split or XIT that would land there is refused the same way. Nothing that was locked before is unlocked, and with the mode on AUTO nothing changes.
-
On FM through a repeater, the transmit lock now judges the repeater's input, where you transmit, not the output you listen on.
The shift stays set when you tune away from a repeater with the knob, so near a band edge the input could fall outside the band (on 70 cm the shift is 5 MHz) with nothing locked. APRS and FM satellites are still judged simplex. A shift set only on the radio's own front panel is still invisible to Nexus.
-
The transmit lock now checks RTTY sent as audio tones on the sideband the radio is on.
It took the tones to sit below the dial, which is right on LSB, the usual way. With the radio on the upper sideband (RTTY-R, or a data mode on USB) the signal sits above the dial, so near the top of a segment it could go out past the edge with nothing locked: on 20 m a dial of 14.1495 sends at 14.1516, past the top of the data segment at 14.150. The radio's own report of its mode decides, and when Nexus cannot read it, both sides must be inside your privileges.
-
A station in the Stations list no longer runs under its own SNR badge.
At the 1024×768 floor the list is about 250 px wide, which left the callsign room for three or four letters, so every call ran under the SNR number and the chips carried on over the Work button. A long country, grid and distance line could do the same even on a wide screen. Chips that do not fit beside the call now start a second row, the SNR badge moves under the call when the list is that narrow, and the second line trims itself with "…" as it was always meant to. On a full band the cards are the same height as before; on a quiet band they are now that same compact height.
-
The spots on the 3-D map no longer flicker.
Every dot on the Connect globe was rebuilt about three times a second, so the spots blinked constantly, and the ring that marks your QTH restarted before it could spread. Each is now redrawn only when it changes.
-
The Linux AppImage no longer prints
Failed to load module "canberra-gtk-module"when it starts.The 1.14.0 AppImage stopped looking in your system's own GTK folders, so the add-ons your desktop asks every GTK program to load — the event-sound module, and on some desktops the helpers for its theme and window decorations — could not be found, and GTK said so in the terminal once or twice at every start. It looks there again, after its own bundled copies, the way 1.13 did. Nothing else about the AppImage changes, and the .deb was never affected.
-
A transponder card in the Satellites section now shows the downlink's mode, not just the uplink's.
On a card whose uplink lists a mode of its own — RS-44's linear transponder, many FM repeaters, the packet digipeaters — the ↑ side showed its mode and the ↓ side showed none, so RS-44 read
↑ 145.935–145.995 LSBand never said the USB you listen on. It now reads↓ 435.610–435.670 USBas well. The line under the list that says which sideband the transmit (split) VFO takes had the same fault and never appeared at all; it now shows once you pick RS-44 or another inverting transponder. -
Working KOSEN-1 no longer puts its uplink in FM.
Its transponder takes CW on 21.125–21.150 MHz and sends it down as AFSK on 435.525 MHz, and Nexus set the transmit (split) VFO from the downlink — so the uplink went to FM, on a part of 15 m where FM is not allowed, from every section. The transmit VFO now takes CW when you work it from Phone or from CW on the radio's own keyer (reverse CW if you have that set), and the data mode a keyed tone needs if you use the soundcard CW keyer, the same as on every other band. Digital is unchanged for now: what an FT mode should do on a CW-only uplink is its own decision. The note under the transponder list says which mode the transmit VFO takes, and the mode shown beside the uplink in the Doppler readout no longer claims the uplink always runs a different mode from the downlink. KOSEN-1 is the only satellite listed this way today.
-
FT8 and the other data modes now go up an inverting transponder on the right sideband.
On a bird like RS-44 or AO-7's mode B, the transponder turns the passband over, so the uplink has to go up on the opposite sideband. Nexus already did that for voice — USB down, LSB up — but left a data mode's uplink on the downlink's side, and an FT8 over sent up that way comes back down with its tones reversed, where nothing can decode it. The uplink now takes the mirrored data mode (DATA-L up for DATA-U down, and the other way round). Only the transmit VFO's sideband changes: FT timing, sequencing and audio are exactly as before, and a non-inverting bird is untouched.
-
The transmit lock now checks a data uplink on the sideband it actually goes out on.
In Digital the lock works out where your signal lands, the dial plus your audio offset, and it took the side from the dial. Up an inverting transponder the data uplink goes out on the opposite sideband, so the signal sits below the uplink frequency, not above it. That is new in this release for most radios, and was already so in 1.14.0 on a radio set to Data modes use plain SSB. Within a couple of kHz of a band or segment edge, the uplink could cross the edge with nothing locked. The lock now checks both sides, the one it always checked and the one the transmit VFO is set to, whether the uplink rides the Sub band or VFO B, and reads 🔒 TX locked if either is outside your privileges. Nothing that was locked before is unlocked, and away from a satellite pass nothing changes.
-
The transmit lock now checks PSK31 and RTTY on the sideband the transmit VFO is set to.
Up an inverting transponder the uplink goes out on the opposite sideband from the one you listen on. So PSK31, which the lock always took to be above the dial, goes out below the uplink frequency, and RTTY sent as audio tones goes out above it rather than below. As for FT8, that is new in this release for most radios and was already so in 1.14.0 on a radio set to Data modes use plain SSB, and within a couple of kHz of a band or segment edge the signal could cross the edge with nothing locked. The lock now checks both sides and reads 🔒 TX locked if either is outside your privileges. In RTTY it also checks the side Nexus set when your radio reports the other one, because Nexus can put the radio back on its own side before the over. True FSK RTTY is unchanged, and nothing that was locked before is unlocked.
-
Changing your licence class now cancels an FT over that is about to go out.
Nexus prepares each over a moment before it keys it, and a class changed in the first-run wizard or in Settings during that moment did not stop it, so the over went out under the old class. It is now dropped, and the next over is checked under the new class. Picking the class you already have cancels nothing.
-
A same-band satellite pass through an inverting transponder now logs the frequency you sent on.
When the uplink and downlink share a band, the log records both, and for FT8 and the other data modes it put your uplink signal one audio offset above the uplink frequency. The uplink goes out on the opposite sideband, below it, so the logged frequency was 3 kHz off at the usual 1500 Hz. The received frequency and every other contact are logged as before.
-
RTTY contacts, and contest contacts in phone, CW and PSK31, are logged on the frequency your signal was on.
A contact the RTTY auto-sequencer logged was given the dial plus the audio offset FT8 uses instead of its mark, so its frequency was off by up to a few kHz, in every version since 0.12.0. And in builds made after 1.14.0, which never reached a release, a contest or Field Day contact was logged by that FT8 offset in every mode: phone at 14.250 went in as 14.2515, and CW at 14.030 as 14.0315. Each is now logged where its signal was: phone and CW on the dial, CW on the soundcard keyer on its tone, RTTY on its mark and PSK31 on its centre. FT8 and the other data modes keep the dial plus their offset, as before. The Cabrillo file you submit always had the right frequency and is unchanged. Contacts already in your log keep the frequency they were given.
-
The transmit lock now checks CW sent through the soundcard keyer where its tone goes.
The soundcard keyer sends a tone a pitch away from the dial, above it on 20 m and below it on 40 m, but the lock only checked the dial. So near a band or segment edge the tone could go out past the edge with nothing locked: at 14.3496 a 600 Hz tone goes out at 14.3502, past the top of 20 m. Up an inverting transponder the tone goes out on the opposite side of the uplink too. The lock now also checks the tone, on the side the transmit VFO is set to, and reads 🔒 TX locked if it is outside your privileges. CW keyed by the radio itself goes out on the dial and is checked exactly as before. Nothing that was locked before is unlocked.
-
XIT is no longer offered on the other radios that have none.
Besides the IC-9700 (above), 56 radios Nexus lists have no XIT that Hamlib can set: 22 more Icoms (the IC-703, the IC-706 family, the IC-718, IC-725, IC-726, IC-728, IC-729, IC-735, IC-746 and IC-746PRO, the IC-756 series, and the IC-910, IC-7000, IC-7100, IC-7200, IC-7410 and IC-9100), the Yaesu FT-817, FT-818, FT-857, FT-897, FT-100, FT-450, FT-890, FT-847 and FT-736R, the Kenwood TS-790, TS-140S, TM-D710 and TM-V71, the Xiegu X108G, X5105, X6100, X6200 and G90, four Ten-Tecs, two Alincos, and control through FLRig, TRX-Manager, PowerSDR, Thetis, SDR Console, Malachite, the QRP Labs QMX and Hamlib's SmartSDR backend. Nexus showed the XIT buttons on them anyway and kept the offset you set. The transmit line showed it and the transmit lock checked it, but the radio transmitted without it. So near a band edge the lock could refuse a transmission your radio would have made legally, or allow one it should not have. On these radios the XIT buttons are gone, a Remote XIT request is refused, and the lock checks the frequency the radio really transmits on. That can lift a lock that was only there because of an offset the radio never used, and it can lock where that offset hid a frequency outside your privileges. Every other radio keeps XIT as it was.
-
XIT is no longer offered through OmniRig.
OmniRig passes frequency, mode, PTT and split to the radio but not XIT, so an XIT offset set on a radio controlled through OmniRig never reached it, while Nexus showed it on the transmit line and the transmit lock checked it. Through OmniRig the XIT buttons are now gone, a Remote XIT request is refused, and the lock checks the frequency the radio really transmits on. As on the radios above, that can lift a lock that was only there because of that offset, and it can lock where the offset hid a frequency outside your privileges.
-
The transmit lock now checks a satellite uplink in its own mode right after each Doppler correction too.
During a pass Nexus retunes the uplink every few seconds. For the moment between each correction and the radio taking it, the lock judged the uplink in the downlink's mode. Up an inverting transponder that put your signal on the wrong side of the uplink frequency, so within a couple of kHz of a band or segment edge it could cross the edge for that moment with nothing locked. The lock now checks the uplink's own mode then as well. Nothing that was locked before is unlocked.
-
The CW ID after an FT 73 is now checked where it is sent.
With CW ID after 73 on, Nexus sends your call in CW after the final 73, and the lock checked it as if it were the FT8 signal, one audio offset above the dial. The CW goes out elsewhere: from the soundcard keyer one CW pitch from the dial, and from the radio's own keyer on the dial. So near the bottom of a band or of your segment it could go out below the edge. An ID that would go outside your CW privileges is now not sent. When the ID is sent and what it sends are otherwise unchanged, and nothing that was locked before is unlocked.
-
QO-100's narrowband transponder is now worked in SSB, not FM.
The satellite database labels every narrowband segment on QO-100 as FM up and FM down, the "SSB only" segments included, and Nexus believed it: picking one put both legs in FM, routed the pick by your FM rules and set up the FM repeater plumbing — on a transponder whose band plan allows no FM at all. Anything between 10489.500 and 10490.000 MHz down is now treated as the linear transponder it is, so Phone gets USB on both legs, CW gets CW, and Digital gets DATA-U. Nothing else about the pick changes, and every other satellite is worked exactly as before.
-
The filter no longer stays at 6 kHz when you switch from phone or CW into a digital mode.
A Flex over SmartSDR CAT answers "done" to the 3 kHz data filter Nexus asks for and then keeps its own 6 kHz SSB filter. Nexus already read the filter back and put the right one in — but only on the retunes it drives itself, not on the one gesture that shows the fault: clicking a digital section yourself from phone or CW. That click now checks the filter the same way, sets it again if the radio ignored it, and, when the radio still will not take it, says on the CAT line which filter you are on and which one the mode needs — instead of confirming the mode and saying nothing about a passband twice as wide as FT8 wants. Moving the dial inside a band still leaves your filter exactly where you put it. (#349)
-
A park activator you work from the Call Roster or Band Activity is now logged with their park.
Nexus tags a contact with the activator's park — the reference your hunter credit comes from — when you start it from the HUNT button, the map or the Needed board. A double-click in the Call Roster, Band Activity or the station list, which is how most FT8 contacts start, never did: those contacts went into the log with no park at all, no hunter credit, and the station stayed on the Needed board as a new park after you had worked it. Starting the contact from any of those now tags the park, whenever the POTA/SOTA spots show the station at exactly one activation right now. When they show more than one — a summit that is also a park, spotted to both programmes — nothing is tagged, the same rule the Needed board follows, and a park you picked yourself on the POTA/SOTA board is left as you picked it. The call itself goes out exactly as before. (#351)
-
"Needed only" and "Hide worked" no longer keep the station you just worked.
A contact starts out unconfirmed, so the moment you log one the station can pick up a LoTW chip — a confirmation still to come — and that chip counted as a need, so both Call Roster filters, and the Needed chip on the Classic station list, went on showing the station you had just worked. A confirmation and nothing else no longer keeps a station on those lists once you have worked it on the band you are on: another contact with them cannot confirm anything the first one will not. The LoTW chip still shows with the filters off, and a station you worked on another band, or one that still has something real to offer — a new state, a park you have not worked in this activation — stays on the list as before. The needs also refresh the moment Nexus logs a contact on its own after RR73 or 73; until now that waited for the next 30-second update. (#350)
-
Moving your data folder now takes the logbook's safety copies with it.
"Copy my log and data there" in Settings carried
log.adiand nothing else that belongs to it: the untouched copy of your log as Nexus first opened it (log.adi.bak), the dated copies in thebackupsfolder, and the copy kept from before the logbook database, all stayed behind in the folder Nexus stops reading. They now move with the log, checked byte for byte like everything else. The logbook database moves too, and it is copied through the database itself rather than file by file: a contact you have just logged can still be in the database's working file, and a plain file copy of a database that is open can miss it. The copy is checked row for row against the original before the new folder is used, and if anything does not match, nothing is changed. -
Deleting a contact while an upload to LoTW is running no longer marks the wrong contacts as uploaded.
TQSL takes a while to sign and send a batch, and the log stays open while it works. Nexus then recorded the result on contacts by where they sat in the log, so deleting one in that time shifted every contact after it: a contact that was never sent could be marked uploaded, and a contact marked uploaded is never offered to LoTW again, while one that was sent stayed unmarked. The result now goes on exactly the contacts that were sent, found by their identity rather than their place in the log. A contact you correct while TQSL is working is left unmarked too, because LoTW has the version from before your correction; it is offered again with your next upload. The connection log says when either happens, and an upload you start also shows the counts in a toast.
-
The SSTV screen now says where the switch is that keeps your radio in the data mode.
On an IC-7300 or IC-7100 the rig drops back to plain USB between pictures, taking its data-mode filter settings with it, and two operators independently ended up setting USB-D by hand every single time. The switch that stops it — "Hold the data mode while SSTV is receiving", per radio — has shipped since 1.13.0, inside a collapsed Advanced section that neither of them found. With the receiver running and the hold switched off, the SSTV screen now says the radio goes back to plain USB between pictures and offers a link straight to that switch, which opens the section it is in rather than leaving you to spot the arrow. Searching Settings for "LSB-D", "back to USB" or "stays in USB" now lands there too. Nothing about what the radio is commanded to do has changed. (#191)
-
When another program already has the sharing port, Nexus now says so where you would look.
If you run your own rigctld on its default port 4532 before starting Nexus, Nexus cannot open its own sharing port — and Settings ▸ Radio went on printing
127.0.0.1:4532beside a Copy button, an address to paste into WSJT-X that nothing was listening on. The refusal only ever reached the connection log. That block now says what happened instead of showing the address: which port is taken, that another program is already listening on it, and the two ways out by name — switch off "Share this radio with other programs" if the other program is the one you want, or give "Sharing port" a number nothing else uses. It clears itself the moment the port comes free. Sharing behaves exactly as it did; only what you are told about it has changed. (#165) -
A repeater search that could not reach one state now says so, instead of quietly handing you a short list.
Nexus pulls RepeaterBook a whole state at a time, and a search near a state line asks for more than one. If one of those requests failed — the directory was briefly unreachable, or the search came round again too soon after the last one — that state's machines were simply absent from the results, and nothing on the panel suggested anything had gone wrong. A search that reached only half of what it asked for looked exactly like a search that found everything there was, so the honest reading of a short list, "there are no repeaters near you", was sometimes wrong. The panel now names the state it could not read and tells you to fetch again. Nothing about the fetching itself changed: same requests, same timing, same cached copies — only what the results are allowed to leave unsaid.
-
A satellite contact made during Field Day now scores for the club, on the bird's band.
The log strip in the Satellites section was the one strip that never switched to the Field Day entry, so a pass contact typed there during FD went into your general log and earned the club nothing — with Field Day visibly running everywhere else in the app. It switches now, sharing the same exchange and the same dupe check as every other position. The part that is not just wiring is the band: everywhere else the FD log stamps each contact with the band the radio is on at the moment you type it, so wiring this strip up the ordinary way would have filed a 70 cm pass written up after LOS on whatever HF band the club had gone back to — in the Cabrillo, and on the N1MM / N3FJP wire. Here the band, the frequency and the satellite tag come from the transponder you were holding, so you can work the pass with both hands on the rotator and log it once the bird has set. The
PROP_MODE/SAT_NAMEpair survives into the contest export and into your lifetime log when you merge, so a Field Day pass contact still earns LoTW satellite credit and still reaches the Sat VUCC and satellite-needs boards. -
Field Day now scores satellite contacts the way ARRL's rules actually read.
Three things that were simply not modelled before. A bird counts as its own band, so a station worked through RS-44 and again on 70 cm from the ground is two contacts and both count — the dupe check used to refuse the second one, and the club lost a QSO that was always legal. A single-channel FM satellite is limited to one contact per station whatever the mode, which is ARRL's own limit; a linear transponder is not, so RS-44 on SSB and again on CW is still two contacts, and Nexus tells the two kinds apart from the satellite catalogue's record of each transponder rather than guessing from its passband. And the 100-point satellite bonus is on the Field Day checklist as a sixteenth box — you tick it yourself, like every other Field Day bonus, and it is one bonus however many birds you work.
-
A previous contact's comment can be read in the callsign card without hovering it.
The "Previous contacts" list in the lookup card clipped each comment to one line and put the rest in a tooltip, which is the same dead end the Logbook's Comment column had before it opened in place. Click the small ▾ beside a comment and it opens to its full length right there in the card; click it again and it folds back. Clicking anywhere else on the line, the comment's own text included, still takes you to that contact in the Logbook, and so does Enter — reading a comment no longer throws you out of the card in the middle of working someone. (#162)
-
The setup wizard can now reach every rig in the catalog, not just the common ones.
The wizard's radio list showed about fifty curated rigs and had no way to see past them, so if yours was not among them there was nothing you could do in the wizard about it — Settings has always had a "Show all models" box, and the wizard simply did not. That is what was behind "my radio is missing": radios reported as absent were in the catalog the whole time, just out of reach on the one screen a new operator actually meets. The same box is now on both of the wizard's radio pickers, unticked by default so the short list stays the default, and the full Hamlib catalog loads the first time you tick it. The IC-7600, which a tester had taken to be unsupported, is now in the short list itself.
-
The FT-890 now gets its one true CAT rate instead of being left on the default.
That radio talks at 4,800 baud and nothing else, so picking it and being left on the 38,400 default meant it simply never answered — the same failure the fixed-rate table was built to end, on a rig the table had no row for. It has one now. The reason it was missed is the part worth saying: the rate table is checked against a generated dump of what each backend actually declares, the FT-890 was absent from that dump, and every check ran over the dump — so the one radio that needed catching was the one radio nothing could see. The dump is rebuilt, and it is now checked against the radio list itself, so a rig added later cannot go unnoticed the same way.
-
A radio found by Detect my radio is named for the radio, not for its USB chip.
With a radio whose USB port is a bridge chip (reported on an FTDX10 with a CP2105), the chip's name, "CP2105 Dual USB to UART Bridge Controller", was stored as the radio's, and every contact logged after that recorded it as your rig. Nexus now uses the catalog's name, and at start-up corrects a stored name the catalog shows to be wrong. Contacts already logged keep the rig they were given, and a label the detection gave the radio in Settings is left for you to rename.
-
"NO RF POWER" no longer shows while the radio is transmitting normally.
An FT-710 reported a power setting below the lowest its own CAT says it can be set to, and Nexus took that as no power at all. A reading under the radio's declared minimum is now treated as a bad reading; radios that really can be set to zero still get the warning.
-
Nexus Remote: re-sending a logging request no longer reports the station as busy when the contact had already been logged.
If a browser's reply went missing and it asked again, the station answered the repeat with "the station was busy" instead of the receipt for the entry it had already written — so the browser could not find out whether the contact was in the log. The repeat was being made to wait for the radio, which it never needed: the answer was already on file. It is now served from that record straight away, and only genuinely new work waits for the radio.
-
"Hide confirmed" no longer hides a station calling you either.
Same fix as the −B4 one below, on the other decluttering chip: it hides every station from an entity you have already confirmed on the band, which took your partner's RR73 away the moment the contact was logged.
-
Garbled text on the band can no longer be read as a message to you — or answered.
A decoded line only counts as a real exchange now if BOTH callsigns in it are callsigns. It used to be enough for the last word to look like a report, a grid or a 73, so a line such as
YOURCALL AGN 73was treated as a signoff sent to you by a station named "AGN". Two things followed. JS8 traffic rides the same waveform as FT8 and a JS8 message starts with the callsign it is aimed at, so those lines landed in the FT8 Rx Frequency pane even when they were nowhere near your receive frequency (#303). Worse, while you were calling CQ the sequencer could take one of them as an answer, stop calling, and put an over on the air addressed to a station that does not exist. This is the rule WSJT-X applies to the same two fields before it will send a standard message. One deliberate cost: a free-text goodbye between two things that are not callsigns —HPE CUAGN 73— no longer counts toward the CQ+73 chip. A 73 between two real callsigns, which is what that chip is for, is unchanged. -
The waterfall now says "TRANSMITTING — display held" during a tune, too.
On RTTY, PSK, JS8 and SSTV the receive picture freezes while the radio is keyed, and since 1.13.0 a label has said so — but only for a message you sent. Pressing Tune froze it exactly the same way with nothing on screen to explain it, which is the "the waterfall has died" look the label exists to prevent. (#230)
-
The RR73 that finishes your QSO no longer disappears from the Rx Frequency pane.
With −B4 on, the other operator's
RR73was on screen for a moment and then gone: logging the contact marks them worked, the sequencer lets go of them as your partner, and the decode list is rebuilt from the logbook on every update — so −B4 took the row away while you were reading it. A decode addressed to you is now never hidden by −B4, in either pane, whatever the logbook says: a station putting your callsign on the air is traffic for you, not clutter. −B4 still hides a worked station calling CQ, which is what it is for. (#268) -
A repeater tone the radio refused is now sent again instead of being written off as done.
Nexus pushes the shift, the offset and the CTCSS tone once, on change, and then stops — but it recorded that as done the instant the commands were sent, not when the radio took them. So a single refusal, from a rig busy on a band change or a momentary CAT hiccup, left you working the machine with no tone for the rest of the session, and restarting Nexus was the only thing that cleared it. It now waits for the radio's own answer, retries a refusal on the next pass, and if the radio keeps saying no it says so on the status line rather than leaving you to find out by being unheard. (#319)
-
One missed reading no longer costs a radio its S-meter for the rest of the session.
A rig that failed to answer a single poll — busy on a band change, or a momentary CAT hiccup — could be written off as having no S-meter at all, and the reading stayed blank until Nexus was restarted. A transient miss is now treated as what it is, and the meter comes back on the next successful read.
-
Nexus stops asking a transmitting radio for meters it has never answered.
While you are keyed, SWR, ALC, power and compression are read in turn several times a second. Every other reading Nexus takes gives up on a radio that will not answer it — these four did not, so on a rig whose CAT has no such meter Nexus kept firing unanswered commands at it for the whole of every over, for the whole session. Three tries each and it leaves that meter alone, then quietly checks again later while you are receiving, so a meter lost to one busy over comes back. Nothing changes on a radio that answers, and there is nothing new to configure. (#331)
-
When the radio refuses CAT CW keying, the warning now quotes what the radio said.
It used to be one fixed sentence, which read the same whether the rig's Hamlib backend has no keying command at all or the reply simply never arrived — two different faults needing two different answers. (#332)
-
LoTW confirmation downloads no longer give up on a report that is still arriving.
Download Confirmations gave the whole download sixty seconds and then failed, however well it was going. LoTW answers straight away and then sends the report while it builds it, so that limit was a budget on how fast LoTW works, not on whether the connection is alive — and a first sync asks for your entire confirmation history, which for an established log does not finish in a minute. Because the sync marker only moves on a download that completed, every retry was the same request, so the confirmations never arrived at all. Nexus now keeps reading for as long as the report keeps coming, and stops only if it goes quiet for two minutes or the whole download passes ten. The message when it does stop says which of those happened and how much had arrived, instead of offering two possible causes and picking neither.
-
A private note in the log opens in its row, the way a long comment already does.
1.13.0 gave the Comment column a click-to-open; the private Note beside it stayed a 📝 marker with its text only in a hover tooltip — the same "there is no way to read this" the comment fix answered. Clicking 📝 now opens the note to its full length in that row, line breaks and all, because a note is the one field in the log you type paragraphs into; clicking again folds it. A row carrying both shows both: the comment keeps the line it had and the note takes one underneath it.
-
The Logbook shows when a contact ended, and lets you fix it.
1.14.0 stopped Ham Radio Deluxe seeing 00:00 as the end time of a contact confirmed before logging — but Nexus itself still had nowhere to show that time, so a wrong one could not be noticed, let alone repaired. More columns now carries an End (UTC) column, and the edit form an End (UTC) box beside the time the contact started: 24-hour UTC, the same plain text box as the start, on the contact's own date. An end time earlier in the day than the start is read as the contact running past midnight, which is what it is. Leaving the box empty keeps whatever end time the record already had — correcting one is offered, clearing one is not. On a browser away from the shack the column shows, but the box does not: the remote edit cannot carry an end time yet, and a box that quietly dropped what you typed would be worse than none.
-
A contact QRZ or eQSL never took now goes back when you fix the credential.
1.14.0 made every upload service say so in the Connections log when it gives up on a contact after its retries — but only Club Log ever went back for them, so on QRZ and eQSL the contact was named and then left. Saving your QRZ Logbook API key, or your eQSL password, now sweeps the log the way saving the Club Log app-password already did: every contact that service never accepted is re-queued, one every 15 seconds so the catch-up doesn't look like a flood, and the Connections log says how many and roughly how long it will take. A contact one service already holds is only sent to the other, so nothing is uploaded twice. HRDLog, N3FJP, Cloudlog and World Radio League keep no per-contact record of what they already hold, so they have no sweep — for those it is still the button on the row, or an ADIF export. The catch-up lines in the Connections log now also name the service actually catching up; every one of them said Club Log before.
-
Your private note on a contact is no longer sent to logbook services.
The Notes field, which the Logbook shows as a private note, went out with every contact sent to QRZ, Club Log, eQSL, HRDLog and Cloudlog/Wavelog, and in the WSJT-X UDP feed, although the Comment is the field marked as shared on the QSL. It now stays in your own records:
log.adi, your ADIF exports, and DXKeeper or HRD Logbook if Nexus forwards contacts to them. A note an earlier version sent is not taken back. -
The A/B indicator now follows the radio, not just what Nexus last asked for.
Nexus told the rig which VFO to use and never asked it back, so pressing A/B on the radio's own front panel left the indicator showing the other one — and it stayed wrong until you changed VFO from Nexus. It is now read back from the radio every few seconds on rigs that can answer the question, which Nexus checks for once when it connects. A rig that cannot answer behaves exactly as before, showing the VFO Nexus selected, and your own A/B press in Nexus still takes effect immediately rather than waiting for the radio to confirm it.
-
FT4 contacts now export as ADIF the way WSJT-X writes them, so LoTW and the rest accept them.
FT4 is not an ADIF mode in its own right — it is a submode of MFSK, as Q65, FST4 and FST4W are — and Nexus was writing a bare
MODE=FT4. That spelling is not in the mode list every service checks against, so an FT4 contact could be rejected outright by TQSL/LoTW and read wrongly by anything else that takes an ADIF file: QRZ, eQSL, Club Log, HRD, and the live feed JTAlert and Log4OM listen to. They now go out asMODE=MFSKwithSUBMODE=FT4, which is exactly what WSJT-X writes for the same contact. Nexus already read that spelling correctly, so importing an FT4 log from WSJT-X always worked; it was only our own export that was wrong. Q65, FST4 and FST4W are fixed with it. Contacts already in your log are unchanged — re-export them and they will come out right. -
Contest serials: the station you called first is no longer forgotten when you move on.
1.14.0 said that calling one station, getting no answer, working somebody else and coming back would give the first station the number they already copied. That only held if you pressed Clear in between. Typing the next callsign straight over the box was read as "you mis-copied the call", so the first station's number was handed to the second and the first was forgotten — and if the station you typed already had a number of their own from earlier, theirs was overwritten instead. Both stations now keep what they copied, and no new number is burned either way. Two stations can end up holding the same serial; both of them copied it from you, so both match at check-in.
-
Clicking a spot during a contest now ties the serial to that station.
The callsign followed the click and the serial did not, so an F-key macro went out with the new station's call and the previous station's number. On phone it was worse: the cockpit showed that number as the exchange to read aloud, with no macro on screen to contradict it.
-
The contest dupe warnings name only what the contest's rule checks.
In Sweepstakes a station counts once on any band, so one you worked on 40 m is a dupe on 20 m, but the log strip's warning named 20 m and your mode, and the Contest dupe badge named 20 m too, so checking 20 m found nothing. In CQ WW and CQ WPX, which count by band alone, the strip also named the mode. The band and the mode are now named only where the rule uses them. Field Day's warning is unchanged.
-
The transmit SWR meter no longer gives you a figure to act on for a rig it cannot calibrate.
Nexus only arms the high-SWR cutoff for rigs whose SWR scale it can vouch for — but on every other rig the meter was still turning red past 2:1 and advising you to keep it under 2:1. On a Xiegu, 1.2:1 on the radio's own panel arrives here as 6:1. The reading stays, because it still moves the right way while you tune, but the colour coding and the threshold advice are gone and the label reads
SWR?. -
On a FlexRadio, Settings no longer calls the high-SWR cutoff verified when nothing is measuring SWR.
The cutoff reads the Flex's own SWR meter, which arrives only while Flex native panadapter (early access) is on, and that is off by default, so on a standard setup the cutoff could not stop anything while Settings said it was verified. Settings now says so and names the switch. The cutoff itself is unchanged.
-
Turning Field Day on or off now re-arms the RTTY auto-sequencer after a finished contact.
The sequencer rests in Done once a contact completes, and the switch only took effect from Idle. So Field Day starting mid-session left every following contact going out with your casual exchange — and because that exchange is not a Field Day one, those contacts were written to the general logbook and never reached the contest log.
-
A contest phone contact now records which sideband, and an FM one is no longer logged as SSB.
The contest log filed every phone contact under its scoring class and nothing else, so the exported
.adisaidSSBfor all of them: no contact said upper or lower, and an FM contact — scored as phone, exactly like SSB — went out claiming an emission that was never on the air. A phone contact now carries the mode the radio was actually on, and it is written the way your everyday logbook has written it since 1.13.0: the mode staysSSBand the sideband rides as theSUBMODE, with FM and AM as modes in their own right. The sideband comes from your radio or not at all — without CAT there is no evidence of one and the contact still logs as plain SSB rather than Nexus guessing it from the band. It survives a restart of the contest log, it reaches N1MM and N3FJP as a mode they already understand, and where a contest's Cabrillo asks forCATEGORY-MODEan FM entry can now say so instead of declaring SSB. This applies to contacts logged from here on. A contest you have already merged into your main logbook keeps what it was given — see the note below. -
A contest contact keeps its frequency on the way into your main logbook.
Merging a contest into the lifetime log — the file that goes to LoTW, QRZ and Club Log — wrote the band and threw the dial away, so every row from a contest weekend arrived with no frequency at all. On VHF that lost the segment with it: 144.200 SSB and 146.520 FM both landed as plain
2m. The frequency each contact went out on now travels with it, with the one you listened on beside it when you worked split, and a contact that never knew its dial still writes no frequency rather than a zero. Field Day is included — its log is written by the same code that restores it after a restart, so leaving the frequency out of the file was also leaving it out of the recovery, and a restart mid-event quietly blanked the dial on every contact already in the log. -
Correcting a busted callsign now reaches the Cabrillo you submit, not just the log you can read.
Fixing a wrong callsign or a wrong band on a contest contact updated your log but stopped there. The Cabrillo is written from the contest log, so the file you sent the sponsor still carried the busted call while the log on screen showed the corrected one — and nothing told you the two disagreed. A correction now reaches both. The contact's country, continent and prefix are worked out again from the corrected callsign, so it no longer counts for the wrong country or the wrong prefix multiplier, and if the callsign or band you corrected changes which contacts are duplicates, every affected contact is re-marked and your QSO count and score follow. Correcting the exchange itself is not included yet.
-
A contest you merged before this release is not repaired by it, and re-merging will not fix it.
Contacts are matched into the lifetime log by a stable id, so a second merge reports everything as already there and changes nothing — by design, because that is what stops a re-merge duplicating your log. If you merged a contest on 1.14.0 or earlier, those rows keep the band without the dial, and their phone contacts keep plain
SSB. Repairing them means editing the rows; nothing in this release does it for you. -
The Phone and CW scope's frequency numbers grow with the UI scale too (#215).
1.13.0 fixed this on the FT8 waterfall and left the scope beside it alone, so on Phone and CW the DIAL plate and the kHz figures along the bottom — the ones that tell you where a click will land — still painted at the same size whatever you set Settings ▸ Appearance ▸ Workspace ▸ UI scale to, and looked smaller the further you scaled up. They now follow the scale like the waterfall's do.
-
Field mode is in Settings, beside Theme and UI scale (#215).
Maximum contrast and larger type have been one click away since 1.1.0 — but only as a chip marked Field in the top bar, named for operating outdoors, and Settings said nothing about it. An operator who went looking for a high-contrast or large-text setting found the size half under UI scale and no contrast at all, and searching Settings for "contrast" returned nothing. There is now a Field mode row in Appearance ▸ Workspace, and the same words find it. It is the same switch as the chip, not a second one: flipping either moves both.
-
High contrast is now its own switch, separate from Field mode (#215).
Field mode raises contrast and makes everything bigger, and it was the only way to get the stronger contrast — so an operator who just wanted a more readable screen at the size they had already chosen could not have one. Settings ▸ Appearance ▸ Workspace ▸ High contrast is that setting: the same stronger text-against-background Field mode applies, in either theme, and it changes nothing else — your UI scale stays exactly where you put it. Field mode is unchanged and still turns contrast on as part of the outdoor treatment; the two do not fight. Your High contrast choice is remembered separately, so turning Field mode off puts the screen back to the contrast you chose rather than clearing it, and while Field mode is on the High contrast row says so.
-
A contest contact merged into your logbook keeps the state or province the other station sent, and both signal reports.
In CQ WW RTTY, the QTH a US or Canadian station sends (599 05 MA) stayed in the contest log and never reached the contact's State field, and the reports never reached its RST fields, so World Radio League, QRZ and the other logbooks a merge uploads to got neither. The state now goes in as the standard ADIF code (NF and LB become NL, NWT becomes NT, PEI becomes PE), and the reports go in for any contest whose exchange has them. World Radio League has no field for a CQ zone, so the zone stays in your own log and in your Cabrillo file.
Corrections to 1.14.0
A released section is history and is not rewritten, so two credits it got wrong are corrected here.
-
#290
was credited as done. Only the give-up notice shipped in 1.14.0. The catch-up sweep that ClubLog has reaches QRZ and eQSL in this release (see Fixed); HRDLog, N3FJP, Cloudlog and World Radio League still have none.
-
#334
was credited as done. Only the band-display half shipped — a rig answering
PKTUSBon its FT8 channel now matches its band-plan channel instead of reading "custom". The spontaneous band jumps reported alongside it are not addressed.
Every version
- 1.24.0 11 Nov 2026
- 1.23.0 5 Nov 2026
- 1.22.0 1 Nov 2026
- 1.21.0 26 Oct 2026
- 1.20.0 22 Oct 2026
- 1.19.0 16 Oct 2026
- 1.18.0 12 Oct 2026
- 1.17.0 7 Oct 2026
- 1.16.0 3 Oct 2026
- 1.15.0 26 Sept 2026
- 1.14.0 20 Sept 2026
- 1.13.0 16 Sept 2026
- 1.12.0 14 Sept 2026
- 1.11.1 8 Sept 2026
- 1.10.3 4 Sept 2026
- 1.10.2 3 Sept 2026
- 1.10.1 1 Sept 2026
- 1.10.0 31 Aug 2026
- 1.9.2 29 Aug 2026
- 1.9.1 26 Aug 2026
- 1.9.0 25 Aug 2026
- 1.8.1 23 Aug 2026
- 1.8.0 23 Aug 2026
- 1.7.6 21 Aug 2026
- 1.7.5 20 Aug 2026
- 1.7.0 18 Aug 2026
- 1.6.1 17 Aug 2026
- 1.6.0 16 Aug 2026
- 1.5.0 16 Aug 2026
- 1.4.0 15 Aug 2026
- 1.3.0 13 Aug 2026
- 1.2.0 10 Aug 2026
- 1.1.0 10 Aug 2026
- 1.0.5 8 Aug 2026
- 1.0.2 6 Aug 2026
- 1.0.1 5 Aug 2026
- 1.0.0 4 Aug 2026
- 0.28.2 3 Aug 2026
- 0.28.0 3 Aug 2026
- 0.27.0 2 Aug 2026
- 0.25.0 1 Aug 2026
- 0.24.0 31 Jul 2026
- 0.21.5 29 Jul 2026
- 0.21.0 29 Jul 2026
- 0.20.0 28 Jul 2026
- 0.19.7 27 Jul 2026
- 0.19.6 26 Jul 2026
- 0.19.4 26 Jul 2026
- 0.18.0 25 Jul 2026
- 0.17.22 25 Jul 2026
- 0.17.21 25 Jul 2026
- 0.17.20 25 Jul 2026
- 0.17.19 25 Jul 2026
- 0.17.18 25 Jul 2026
- 0.17.17 25 Jul 2026
- 0.17.16 25 Jul 2026
- 0.17.15 25 Jul 2026
- 0.17.14 25 Jul 2026
- 0.17.13 25 Jul 2026
- 0.17.12 25 Jul 2026
- 0.17.11 25 Jul 2026
- 0.17.6 25 Jul 2026
- 0.17.5 25 Jul 2026
- 0.17.4 25 Jul 2026
- 0.17.3 25 Jul 2026
- 0.17.2 24 Jul 2026
- 0.17.1 24 Jul 2026
- 0.16.4 24 Jul 2026
- 0.16.3 24 Jul 2026
- 0.16.2 24 Jul 2026
- 0.16.1 24 Jul 2026
- 0.16.0 24 Jul 2026
- 0.15.24 24 Jul 2026
- 0.15.23 24 Jul 2026
- 0.15.22 24 Jul 2026
- 0.15.21 24 Jul 2026
- 0.15.20 24 Jul 2026
- 0.15.18 24 Jul 2026
- 0.15.17 24 Jul 2026
- 0.15.16 23 Jul 2026
- 0.15.15 23 Jul 2026
- 0.15.14 23 Jul 2026
- 0.15.1 22 Jul 2026
- 0.15.0 21 Jul 2026
- 0.14.0 21 Jul 2026
- 0.13.0 19 Jul 2026
- 0.12.0 18 Jul 2026
- 0.11.1 18 Jul 2026
- 0.11.0 18 Jul 2026
- 0.10.0 17 Jul 2026
- 0.9.7 17 Jul 2026
- 0.9.6 16 Jul 2026
- 0.9.5 16 Jul 2026
- 0.9.4 16 Jul 2026
- 0.9.3 16 Jul 2026
- 0.9.2 15 Jul 2026
- 0.9.1 15 Jul 2026
- 0.9.0 15 Jul 2026
- 0.8.9 15 Jul 2026
- 0.8.8 14 Jul 2026
- 0.8.7 14 Jul 2026
- 0.8.6 14 Jul 2026
- 0.8.5 14 Jul 2026
- 0.8.4 13 Jul 2026
- 0.8.3 13 Jul 2026
- 0.8.2 13 Jul 2026
- 0.8.1 12 Jul 2026
- 0.8.0 12 Jul 2026
- 0.7.1 12 Jul 2026
- 0.7.0 12 Jul 2026
- 0.6.0 11 Jul 2026
- 0.5.2 11 Jul 2026
- 0.5.1 10 Jul 2026
- 0.5.0 10 Jul 2026
- 0.4.1
- 0.4.0
- 0.3.0
- 0.2.0 3 Jun 2026
- 0.1.0