Timebase 1.3 Release Notes
Overview of the release and full release notes for all versions of Timebase 1.3
Upgrade: Download Version 1.3.0
Smarter ingestion. Faster writes. More control.
Release date: January 30, 2026
Timebase 1.3 is a foundational release focused on a problem every industrial system eventually faces: reality is messy.
Networks drop. Devices buffer. Data arrives late. Volumes spike. Systems restart. Engineers need the historian to behave predictably when the world does not.
This release strengthens Timebase across three core areas:
-
reliable ingestion under real plant conditions
-
higher-throughput and lower-memory data movement
-
better operational tooling for users and integrators
The result is a historian that handles disorder without sacrificing performance or trust.
What’s new at a glance
Timebase 1.3 introduces:
-
native late data handling in the historian
-
major performance improvements in ingestion and indexing
-
improved state boundary management for large datasets
-
wildcard subscriptions for scalable real-time access
-
authentication support in Timebase Explorer
-
parallel write support in the Ignition module
Each of these changes is covered in detail below.
Historian enhancements
Late data handling
Timebase Historian now supports ingesting data that arrives out of chronological order.
Late data is common in industrial systems, especially when:
-
edge devices buffer during network outages
-
store-and-forward queues flush after reconnect
-
multiple collectors publish asynchronously
Previously, out-of-sequence data was rejected to protect performance and ordering guarantees. In 1.3, Timebase can safely accept late data within defined constraints.
Key characteristics:
-
configurable Late Data Tolerance, default 100 ms
-
accepted only within active blocks (last 30 days)
-
merged automatically during the next indexing cycle
-
typically queryable within approximately one minute
Late data points are marked using a vendor-specific quality bit so analytics can distinguish them without altering original quality semantics.
A new system metric, Dataset.Writes.Late, allows monitoring of late data ingestion rates.
This change allows Timebase to tolerate real network behavior without compromising chronological integrity.
Improved state management
Historian tag state tracking has been refactored to improve scalability.
Boundary and transition metadata is now stored in block-level partitioned files rather than centralized structures. This significantly reduces memory pressure and improves performance when working with:
-
high-frequency state tags
-
long-running datasets
-
large numbers of transitions
The behavior of state queries remains unchanged, but performance under load improves substantially.
Dataset existence API
A new endpoint is now available:
GET /dataset/{datasetId}/exists
This lightweight call allows client applications and services to verify dataset availability without loading full metadata.
Common uses include:
-
startup validation
-
client-side caching
-
automated provisioning workflows
Additional historian improvements
-
faster TVQ ingestion path
-
optimized dataset size calculations
-
improved query logging with full query visibility
-
enhanced tag metadata envelopes
-
fixed point-in-time query accuracy
-
deprecated single-tag writes in favor of batch operations
These changes improve both performance and diagnosability in production environments.
Wildcard subscriptions for real-time access
Timebase WebSocket subscriptions now support MQTT-style wildcard patterns.
This allows clients to subscribe to large tag groups using naming structure instead of explicit tag lists.
Supported features include:
-
single-level wildcard
+ -
multi-level wildcard
# -
support for
/and.delimiters -
automatic inclusion of newly created tags
Rules:
-
matches one and only one hierarchy level
-
cannot span multiple levels
-
may appear anywhere in the pattern
Examples:
enterprise/site/+/temperature
Matches:
enterprise/site/line1/temperature
enterprise/site/line2/temperature
Does not match:
enterprise/site/temperature
enterprise/line1/zone1/temperature
When a wildcard subscription is created:
-
the last known value for all matching tags is delivered
-
future updates are streamed automatically
-
new tags matching the pattern are included without resubscription
This enables scalable real-time consumers such as:
-
UNS subscribers
-
dashboards
-
alerting services
-
AI pipelines
Direct tag subscriptions continue to work unchanged.
Explorer improvements
User authentication
Timebase Explorer now supports authenticated users.
The UI displays the currently logged-in user and provides login and logout workflows tied to configured authentication providers.
This prepares Explorer for multi-user deployments and controlled access environments.
Save As behavior
When modifying the path of a visual, users are now prompted to Save As instead of overwriting the original file.
This prevents accidental loss of existing visuals and aligns with expected engineering workflows.
Band management improvements
Creating new bands now includes confirmation dialogs when items remain in the current band, reducing accidental layout issues when organizing trends.
Collector updates
Sparkplug B reliability improvements
The Sparkplug B collector plugin now includes:
-
configurable retry counts for birth messages
-
adjustable retry timers
-
improved cancellation token handling
These changes significantly reduce connection edge cases and prevent endless retry loops when brokers are unavailable.
Ignition module release 8.3.2
The Timebase Historian Ignition module was released alongside 1.3 and requires Historian 1.3 or newer.
This update focuses heavily on performance and memory efficiency.
Key improvements include:
-
parallel writes to Historian
-
configurable maximum concurrent writes
-
improved chunking and sorting logic
-
reduced memory pressure during large transfers
These changes are most visible during Ignition store-and-forward recovery, where throughput improvements can be substantial without requiring configuration changes.
Additional fixes include:
-
corrected missing query results
-
full string tag support
-
improved URL encoding
-
trimmed configuration values to prevent whitespace errors
There are no breaking changes.
Service stack improvements
-
automatic endpoint configuration when adding components
-
improved Service Starter UI property editing
These changes simplify deployment and reduce manual configuration steps.
Upgrade considerations
Before upgrading, review the following:
Late Data Tolerance
The default tolerance is 100 ms. Systems using store-and-forward or buffered edge devices may want to increase this value.
Explorer Authentication
Authentication providers must be configured if login control is required.
Ignition Module
Historian 1.3 is required for Ignition module version 8.3.2.
Why this release matters
Timebase 1.3 is not about features for features’ sake.
It is about closing the gap between how industrial systems behave in theory and how they behave in practice.
Late data happens.
Backfill happens.
Burst writes happen.
Networks fail.
This release ensures that when those things occur, the historian does not become the bottleneck or the source of doubt.
Timebase 1.3 makes the system more tolerant, more scalable, and more predictable under real operating conditions.