TL;DR by me: Biggest BeyondATC update yet, major refinements, VFR stage 1 for supporters.  

 

'BeyondATC Toolbar V2.0.0

Changelog on Discord:

NEW: Added D-ATIS toolbar tab to view the ATIS of your arrival and departure airports
NEW: Added CPDLC logon code under frequencies tab 
NEW: Added ability to queue messages in the toolbar when traffic is speaking
NEW: Added version footer in the toolbar```
The toolbar is a way to have the core BeyondATC UI features directly within MSFS. Useful for VR or any user who wants direct access to BeyondATC from within MSFS.

Download from the Toolbar link at https://www.beyondatc.net/download

 

Major Version 1.8.6 EA

Changelog on Discord:

⚙️ Core Refactors and Stability

* New feature: Complete rewrite of the arrival sequencer. It is a new system that uses a "compression inversion" algorithm to model non-deterministic conversion points along flight tracks. In simple terms this will sequence aircraft from over 100+ miles out to ensure they arrive nicely spaced out. The player and traffic will now receive speed instructions if the system detects a collision based on airframe characteristics and stage of flight. This is still a work in progress, but represents a massive step up from the previous system.
* New feature: Complete rebuild of the vectoring system (v3). This is the most advanced and complex vector system yet - taking wind, terrain, drift and reaction time into consideration to produce minimal commanded accurate vector instructions. Paths can rebuild themselves in real time to match changing situations.
* Complete rework of the Comms platform with an emphasis on stability.
* Complete rework of speech recognition; added semantic cognate token matching for much more accurate readback handling.
* Added layered fallback handling for bad microphone transcriptions - automatically grant readbacks when it's detected the transcription was corrupted.
* Complete rework for runway queue handling; improved queuing from planes coming from multiple directions onto the same runway.
* Full rework of the taxi pathing algorithm, should be much faster, more stable and produce far better results.
* Rework for taxi path phrasing, long repetitive readbacks should be much shorter now.
* Full rework for traffic data functions with a focus on stability and performance.
* Runway data completely rewritten to use mid points and non-sim data for length and extended thresholds, giving much more reliable runway positioning and allowing aircraft to take off from any entry point in an extended threshold.
* Takeoff queue handling significantly reworked - you'll be handed to tower as soon as you're in the queue, and can end up deep in the queue on busy days without issues.

🗣️ Local Voice Model

* New feature: Local voice accents added! The model is still being trained but you have the same breadth of accents now that premium voices have all locally, with no cloud dependency or downtime and lightning fast generation time with minimal performance overhead.
* The local voice model no longer relies on an external exe which should lead to more stability.
* UK voice mapping updated to balance male/female ratio and better segment English, Irish, and Scottish accents in line with premium voices.
* US voice mapping updated to balance male/female ratio and better.
* Standardized speech speeds across US, UK, and accent models with additional pauses, particularly for ATIS.
* VC++ redistributable added to the installer process for local voice support (silent install if needed).

🛫 Departure, Gate and Runway Handling

* Improved SID altitude handling. [#11394807]
* Tower/Ground will no longer constantly tell you to start and stop taxi when in a queue or heading in the same direction as the aircraft in front — only triggers when converging from different taxiway directions.
* When crossing a completely inactive runway (no departures or arrivals either end), hold short instructions are removed and the crossing is included in the taxi instruction.
* Fixed backtrack runway departures being triggered when landing aircraft are on short final, with longer clearance windows for backtrack departures.
* Backtracking aircraft are now always cleared for takeoff once the backtrack is complete rather than being told to hold short.
* Departure runway selection improved based on taxi route and busyness.
* Gate designators (empty, C, L, R) now treated as a group — prevents parking on empty/C while also using L or R, but allows L and R together.

🛬 Arrival and Sequencing

* Reduced approach aircraft distances that block takeoffs for better flow with faster runway clearing.
* Resume speed releases at/before IAF/IF.
* Braking force after landing now scales based on proximity of following approaches so a longer runout when nothing is behind.

⚙️ General Improvements

* Reworked how ground speed of traffic is calculated, using wind values at the player aircraft rather than unreliable sim data, with improved TAS calculation at higher altitudes.
* Allow scenery that only exposes one start on a runway to work (Flightbeam KDEN 25 for example).
* On departure a runway change request will be granted to any active departure runway even if our logic says it is too short for the player aircraft.
* Ensure a runway is always available that suits the player's aircraft size.
* Map voices to tail number prefixes by country for GA / Bizjets (or any aircraft without an airline based callsign).
* Line up and wait clearance given to player when traffic on short final fixed.
* Prevent hold short nodes that are near the runway centre line (i.e. incorrectly on the runway) being considered hold short (fixes data issue in EGCC Inibuilds).
* Correct handling for CF waypoints at LEBL, EDDL and other airports with long approaches.
* Latest AIRAC and airline data.

📡 ATIS Improvements

* Improved ATIS handling for METAR codes such as FG, DZ, GR, SQ, SS, DS, PL, PE, IC, SG, FC, NSC, VV and more to allow for more dynamic ATIS generation.
* Removed 5G notice in effect from ATIS.
* Fixed issue where there was no space between WIND CALM and cloud information. [#06711097]
* Wind is now rounded to 10 degree increments.
* Wind variable now reports wind direction and variable direction.
* Added various differences between FAA and ICAO ATIS reporting (example, saying "knots" and "degrees" in ICAO, but omitting in FAA).
* In ICAO land - ATIS is reported in meters below 5km, and kilometers between 5 and 10km.
* Added 'sm' to visibility in FAA land.
* Added support for MPS (Meters Per Second) parsing for wind information in China and Russia.
* FAA reports ATIS visibility in statutory miles, ICAO reports it in kilometers.

📡 CPDLC Improvements

* Many overall tweaks.
* Fixed chance for CPDLC messages with MRN's assigned to be silently consumed when they are actually meant to be requests.
* Fixed CPDLC soft errors showing on the UI as being from the player not ATC.
* Improved user facing error messages.
* Added support for pushback request via TELEX.
* Added support for taxi request via TELEX.
* Fixed issue where CPDLC departure runway would not match ATIS if recently updated and would not cause runway to be put into a locked state.
* Added LFEE, LJLA, SBRE, TTZP, UGGG, UBBA, UTAK, UTAA, UTAV, OAKX, OPLR, OPKR station logon codes.
* Fixed issue with auto tune not working when responding to 'wilco' to a contact new station message.
* Hide "handover" message from UI as this is a system message.
* Fixed CPDLC log messages not being outputted to production version.
* Added GOOO station code to missing frequencies. [#94909585]
* Always add the initial climb to the SID. [#13020537]
* Changed center name of Brindisi Military to Brindisi. [#19210877]
* When getting a CPDLC handoff, fixed a chance where auto tune would tune to the incorrect frequency. [#14286630]
* Fixed issue where after being handed off with an END SERVICE - you would be kept logged on within the server, leading to BATC acting like you were still logged in. [#36823081]
* Add explicit rejection of step climbs. [#10478508]
* Fixed issue where leading zeros in callsigns can lead to issues with CPDLC message delivery. [#33934962]
* Fixed issue where initial climb of 1000 feet was reported in CPDLC PDC. [#63298660]
* Issue where DPFREQ would show up when no departure frequency is specified. [#25701504]

```🐞 Bug Fixes

* Large number of fixes for stuck traffic in taxi, on runways, and during pushback.
* Finally fully fix the vectors after altitude change bug.
* Fixed clearance being given when an aircraft is on very short final.
* Fix AI ATC repeating calls [#22278488].
* Fix historic METAR not loading over weekends [#44165420].
* Fix traffic receiving the same instruction repeatedly [#82442706].
* Fix departure runway changes not being possible at airports without a clearance frequency.
* Fix "Request departure" display hanging or not responding.
* Fix aircraft sometimes being deleted because they can't find a runway exit even though one exists.
* Fix DECTalk ATIS saying "period" after runways when arrival/departure runway set are the same.
* Fix LLM query for next waypoint sometimes returning the previous waypoint.
* Fix fatal crash in enroute traffic data generation.
* Fix tug spline fatal crash.
* Fixed altitude readbacks being rejected even when spoken correctly (FL100, FL200, FL300, FL400).
* Fixed taxi readbacks being rejected when route has separate number and letter taxiways (e.g., "2, C").
* Fixed frequency readback failing when speech recognition fuses digits with callsign.
* Fatal exception fixes [#44914924], [#40514399], [#83565855] + others.
* Fix traffic getting wrong approach initially that can lead to incorrect instructions or early landing state [#06429848].
* Prevent aircraft being given lineup and wait when aircraft on short final if there is another departing aircraft on the takeoff roll.
* Fix traffic aircraft potentially overlapping each other in a traffic queue at some airports.
* Add a fallback method for determining runway turnoffs to help prevent aircraft de-spawning after landing at some airports.
* Fix the issue with the center controller having a potentially wrong accent - Dutch on London Control for example.
* Pick STARs and STAR transitions for traffic more consistently to prevent vectors being given incorrectly.
* Fix TOD calc if there is no STAR and some route waypoints received low cruise alts from Simbrief.
* Fix the feature that allowed you to use the current transponder value set in your aircraft when restarting a flight mid-flight (helps with Vatsim use).

🛬 Airport Ops Updates

* Improvements for: EBCI, EDDK, EFHK, ESMT, GMME, GVAC, KATL, KBZN, KDBL, KDFW, KLRD, KMDW, KOAK, KOSH, KPBI, KSAT, KSDF, LFKJ, LHBP, LICJ, LIMC, LKPR, LOWI, MPTO, OERK, OMDB, PHLI, RCTP, RJTT, RPLL, VTSG, ZGSZ```
Representing many ongoing months of stability fixes and improvements, we finally have the system in a fantastic space to work on our next features. Work does not stop though and we expect regular fixes into the EA build in the coming days and weeks.'

 

Major Version 1.9.0 Experimental

Changelog on Discord:

🛩️ VFR Stage One — Core Systems

* New feature: Three dedicated phraseology regions for VFR — UK, US (Canada also uses the US system) and AUS — with custom scripts and procedure handling per region. Action button text adapts to the region you are flying in. Where you are not flying in one of these regions, the default is UK phraseology to best align with ICAO.
* New feature: Airport Circuit System with support for Touch and Go, Low Approach, Full Stop Landing, and Stop and Go (US and AUS).
* New feature: Dynamic circuit direction system that avoids circuit patterns over runways and near terrain. If no conflict is detected the direction defaults to the left, and you may request a circuit direction change on the ground or while flying in the circuit.
    - Note: You may also request a different runway before or during your circuit.
* New feature: BATC remembers your last circuit intention and presents it as a quick option when reporting downwind/midfield, with the option to change it before or after reporting.
* New feature: VFR Taxi Back — after a full stop landing, choose between taxiing back to the runway or to parking. Parking can be auto-assigned to a GA-appropriate gate, or you can select a specific spot, so you can practice full stop landings and continue the flight.
* New feature: Intersection departures — request an intersection departure before requesting taxi. BeyondATC lists the available intersections if you do not suggest one in the initial call. VFR only in this release; IFR and traffic support planned.
* New feature: Submenus for action button requests. Buttons that require follow-up information (for example pattern intention) now expand into sub-options instead of ATC asking you. Voice users keep the original flow provided the original call contains all the needed information. VFR only for now, IFR expansion planned.
* When starting a VFR flight in the air you will be set into an uncontrolled state and can then tune to your arrival or departure airport to request entry for circuits/pattern or landing.
* Stage one does not expand support for non-towered airfields. The airport you fly to must have a tower controller to allow for VFR circuit management with ATC. Improvements are planned for stage two.

🛫 VFR Departure, Arrival and Join Handling

* New feature: Enter and exit controlled airspace at your departure or arrival airport at any time, including multiple times in a flight.
* New feature: Leave controlled airspace via a frequency change (US/AUS) or "VFR departure" (UK). In the US you are automatically moved into an uncontrolled state at 6nm from the airport to simulate airspace where calling ATC for a frequency change is not required.
* New feature: Request to re-join controlled airspace with the option for landing or pattern/circuit work. Supported join types: Downwind, Straight-In, and Overhead (UK).
* New feature: VFR traffic deconfliction (work in progress). When in the circuit and reporting downwind/midfield, ATC will extend your downwind or ask you to orbit if required. If you cannot be slotted in after extending, you will be asked to navigate back to downwind. ATC also provides traffic information in certain calls based on traffic conditions. This system currently targets traffic within the circuit/pattern only, improvements planned between stage one and two.
* ATC will provide traffic in sight calls to VFR aircraft in the circuits
* Stage one does not fully simulate control zone entry/exit, realistic zone transits, VRPs or full traffic avoidance — improvements are planned through stage one and into stage two.

```🔤 Callsigns, Squawks and Phraseology

* New feature: Callsign shortening for general aviation. Depending on region, ATC and auto-responses shorten your callsign where appropriate — "G-ABCD" becomes "G-CD" in the UK, "N42PC" becomes "Piper 2PC" in the US. The full and shortened forms both display at the bottom right of the flight interface (e.g. "G-ABCD (G-CD)") and you can transmit either to ATC.
* New feature: VFR squawk code generation (UK only, work in progress). Common UK VFR squawk code blocks are assigned on arrival or clearance. After exiting controlled airspace, aircraft are told to squawk conspicuity and enter an uncontrolled state. US/AUS use default squawks. 
* New feature: QFE information at UK aerodromes — calculated on demand and available via voice or keyboard request to ATC.
* New feature: Airframe Name Library — over 880 airframes mapped to ICAO codes so traffic deconfliction calls and scripts can refer to airframes by name instead of ICAO.
* Added the word "runway" before hold short instructions — "hold short of 22" is now "hold short of runway 22".
* Removed "ready to go" from the departure request script so IFR aircraft consistently use "ready for departure".
* ATC and traffic now explicitly use "wilco" for compliance with instructions, and "roger" / "affirm" etc. for acknowledging information.
* In ICAO regions, taxi and hold short instructions now better align to ICAO phraseology, including ensuring 'runway' and 'hold short' instructions are explicitly stated.

⚙️ General Improvements

* Improved wake turbulence — landing aircraft now set wake on the runway, enabling wake separation for subsequent departures.
* LLM split into two contexts, one for IFR and one for VFR, giving more context-aware prompts depending on your flight type.
* D-ATIS now refreshes each time you open the frequencies menu.

🐞 Bug Fixes

* Fixed an issue where a blank push-to-talk could lock up the action system.
* Fixed an issue where, after a crossing instruction, ATC could say the name of the cross instruction twice.
* Fixed typo in CPDLC "PRE DEPARTURE" PDC request.
* Fixed the speech engine saying "ATC" incorrectly.
* Added a guard against autotune triggering for the same frequency.
* Fixed an issue where the "Waiting to Cross" button could appear for your own departure runway at airports with parallel runways.
* Improved proximity detection for the automatic crossing prompt from ATC.
* Fixed an issue with the readback scoring system where the lowest value script could win.
* Fixed a fatal where an aircraft transferred to tower with no assigned runway.```
This release lays the foundation for VFR operations in BeyondATC. As with any experimental build we expect issues to surface, and we will be posting known issues and follow-up fixes over the coming days and weeks. Thank you for helping us test stage one — we hope you enjoy it.'

Edited April 6Apr 6 by MrLamb

"US voice mapping updated to balance male/female ratio and better." 😲

I hope it's now three-quarters in favor of women?

 

4 hours ago, Vitold69 said:

"US voice mapping updated to balance male/female ratio and better." 😲

I hope it's now three-quarters in favor of women?

 

Can you please explain your issue in more detail? 

Edited April 6Apr 6 by pemigris

39 minutes ago, pemigris said:

Can you please explain your issue in more detail? 

I guess it's about a discussion taking place a few months ago about the real life amount of male or female pilots/controllers. Some people thought, there were too many women on BATC voices. Others accused them of being sexist. So, the usual... 😉

Edited April 6Apr 6 by Watsi

I mean if VATSIM is in any way a reflection of reality, women should be the white elephants of aviation. I only ever heard TWO female controllers (one controlls Polaris Control the other the Frankfurt Sector) and one pilot. 

The rest is all men.

 

Then again this is only a reflection of what men do in their free time as opposed to what women do in their free time. 

Nontheless I did see the point back then when the discussion started. There ARE too many female voices on BeyondATC, or rather the ratio of their appearance is too high. If one really wants a realistic expeirence it should definately not be 50-50 like it used to be.

 

 

All that aside, I'm surprised the new version was released already. I followed the discord on the experimental build and the issues that came with the new features and they seemed aplenty until recently so that I didn't expect a public release for another month. Glad to see they think they got it in shape so soon.

Edited April 6Apr 6 by Farlis

I'm pretty sure the amount of women in the cockpit or at a radar screen is pretty high nowadays (IRL). (at least more than VATSIM, since we are all aviation geeks - and mostly male). But you can say that the amount of women was still too high in BATC.

I never cared, actually... 🙂

5 minutes ago, Watsi said:

I'm pretty sure the amount of women in the cockpit or at a radar screen is pretty high nowadays (IRL). (at least more than VATSIM, since we are all aviation geeks - and mostly male). But you can say that the amount of women was still too high in BATC.

I never cared, actually... 🙂

Neither do I, but I understand where the people who do are coming from. The ratio is unrealistic. Then again there are not enough distinct voices available to simply use more male voices. If you reduce the ratio it will get more repetive because there are not enough different voices to make up for the gap.

Edited April 6Apr 6 by Farlis

Yeah, it may also depend on the country you're flying in/over. Some countries have more women in the aviation business than others. But I'm more happy about other essential 'bug fixes' and improvements than this one. A 50/50 in general would be totally fine with me.

I honestly haven't given that ratio any thought 😂 I actually like the variety.

I'm so interested in VFR ATC that even at Stage 1 and Experimental... let's finally buy it.

Edited April 6Apr 6 by vbazillio

8 minutes ago, vbazillio said:

I'm so interested in VFR ATC that even at Stage 1 and Experimental... let's finally buy it.

I'm curious as well I just have to download the update.  Default for vfr has been semi ok for me so far.

For our VFR testing users, we have a wiki page going through the implemented features in this stage one and to help you get through your first flight: https://wiki.beyondatc.net/knowledge-base/vfr/

For those interested, one of the real world controllers that helped us with VFR procedures made a video to showcase VFR in the US:

A note, the video shows an early, in development build of BeyondATC, you will find some readback errors in the video and the video is not indicative of the final product. VFR is release into the experimental build of BeyondATC while it is matured over time, it is not feature complete, nor bug free but the team will work hard to get all those issues sorted!

good now to try vfr flights now since its been released at last

5 hours ago, Watsi said:

I never cared, actually... 🙂

The 50/50 ratio was actually great. I felt rejuvenated as Benny Hill in a girls' dorm.

After having read the feedback in the experimental branch subjection of discord I thought it would take longer until this is released publcly and after my first flight I think they should have waited and further refined.

Getting a 280knot speed restriction at FL350 makes no sense. First of all it should be a Mach restriction at this height and then it should make sense. There was no traffic ahead of me to slow me down for so early. 

And then I saw an AI aircraft that came from the south be vectored way to the north of the destination, completely outside of any arrival routing. Also made zero sense.

 

So many of the issues reported in the experimental branch remain in this first version. I know they will keep polishing but I would have waited before releasing it publicliy in this state.