6.2 Beta Release (March 2026)
The changes in this beta will also be available in the next Source version 6.10 production release.
This release includes:
SWP (Source Work Program)
Fixed a bug where the Supply Point receives demand during the Order Phase, where Water User functions are evaluated using flow values from the previous time step. Since these functions are not re-evaluated during the Flow Phase even when they depend on Flow Phase variables, the resulting water extraction is based on outdated values and can be incorrect. This was resolved by re-evaluating Water User functions during the Flow Phase at the first connected Supply Point, ensuring that demand is updated using current time step conditions when it is requested.
Applied a change from MDBA relating to the ownership of storage routing links when a division in the link transitions from live to dead storage. Updated regression tests results that were affected by this change.
Added the ability to centrally enable or disable individual constituents in models. When a constituent is disabled, it is ignored during simulation. This feature is also accessible via Scenario Input Sets.
Optimized Function Manager tree performance with lazy loading and resolved memory issues due to referencing functions to support projects with large amount of functions.
Function Manager toolbar operations (Expand All, Collapse All, Show Referencing Functions) are now dynamically enabled based on function and variable count, with related settings moved to Scenario Options and referencing functions also accessible via the context menu.
For more details on what’s changed please see this page : 6.2.99.14122 - Alpha Release (February)
Aggregation of partial periods occurring at the start and end of a timeseries due to run boundary dates can now occur. This occurs in-line with reporting of periods with missing data in the middle of a timeseries.
Fixed a bug relating to priority ordering at confluences when upstream branches have different travel times.
Resolved a slow project load time in the .NET 8 version of Source when opening the Barwon-Darling model. Improvements were made to project loading that bring load times in line with the pre-.NET 8 version for that project.
Added a new Dead Storage Volume recorder for Storage Routing Links, which can be used by Context Variables. Enabled the Storage Volume on Storage Routing Links to be used by Context Variables.
Ensured that the "Area Volume Table" option persisted correctly for On-Farm Storages under the Storages > {Storage Name} > Area > {Date} UI.
Fixed a bug where rainfall runoff model parameters did not display "inf" (infinity) when they are large.
Fixed a bug where previously user trying to filter "Subcatchment" columns by Geo Regions could cause an error.
Hidden the "AllowCarryover" checkbox when the user selects anything other than "Water Year" in the Supply Point > Demand Constraints > Usage Limit UI, and persisted the Textbox entry.
Removed the Demand Model > Crops > {crop name} > Crop recorder from the recorder tree for the Irrigator model.
Fixed a bug where the recorders would not update if the recorders had been opened following a run error in the previous run.
Fixed an issue where the Scenario Mass Balance Recorder was not refreshing correctly after each run.
Fixed a bug where Scenario Mass Balance changes overwrote previous run results as well as Latest Run in Results Manager
Fixed a bug that crashes Source while using the Project option “Force Comments On Save”.
Added a configurable base folder path in Results Manager -> Options menu item, enabling Result Manager Configurations to store relative file paths alongside absolute paths for improved portability and easier sharing across different environments.
Fixed a bug where the run progress bar would appear stuck after a scenario run is completed for Replicate Analysis runs.
Roadmap Initiatives
Marker Routing : Added a new ‘Experimental Feature’ (available under Application Settings) to alter the way Marker Routing information is managed. This work is still in development, and the feature is only intended for specific clients to test with.
Source Performance Improvements : Micro-optimizations were made to the following 4 areas of Source to improve Source run-times: Water User, Constraints, Ordering and Ownership, Recording, and Functions. A ~4% reduction in mean runtime was achieved across the three analyzed models (GBCCL, Namoi, and River Murray 50) through improvements made to several functions identified as hotspots during profiling. These gains were achieved without any major restructuring of the codebase.
Conversion to GUID recorders : Converted the Pipe Junction node's Connections From and Connections To list recorders to the GUID-based system.
River Operations
Performance Improvements : Improved run time performance for River Operations by optimizing iteration logic and reducing initialization time. This seems to have a 20% reduction in run time for 'RiverMurrayOps' project.
Fixed a bug in Results Manager where repeatedly clicking on a chart would incorrectly increase the left Y-axis range after manually setting it.
Fixed project summary export errors related to forecast model properties (Trend, Monthly Average, Expression, and Timeseries) that depend on Reflected Items, which occurred when exporting before a scenario was fully initialized.
Fixed an exception that occurred when right-clicking on the Time Column in the Tabular Editor.
Regression Testing
There is no tolerance in our regression test suite to allow for earlier identification of results being different on different machines or versions of operating systems.
See: 6.2 Regression Test Changes
Full Release Documentation