ROCm Optiq (Beta) release history

ROCm Optiq (Beta) release history#

2026-02-11

2 min read time

Applies to Linux and Windows

Version

Release date

Beta 0.2.0

February 11, 2026

Beta 0.1.0

December 10, 2025

ROCm Optiq (Beta) 0.2.0#

Added#

  • Summary View: Displays the top ten kernels by execution time using pie charts, bar charts, or tables.

  • Minimap: Provides a compact overview of event density and counter values across the entire trace, enabling rapid navigation of large datasets.

Changed#

  • Timeline View: Improved navigation and selection. Added context menu option to create a time range filter from a selected event or events.

  • Advanced Details Panel: Aggregate by Column drop-down groups the results by the selected column. Options to size columns to fit in Event Table and Sample Table. Event Details now shows the function call’s arguments, if available.

  • Time Range Filtering: Improved time range selection.

  • Histogram: Shows event density in two display modes: “Normalization: All Tracks” and “Normalization: Visible Tracks”.

  • Multi-node: Multi-node data and a new multi-database yaml file format are supported.

ROCm Optiq (Beta) 0.1.0#

Initial release of ROCm Optiq (Beta).

Added#

  • System Topology View: Displays a hierarchical representation of the hardware or system components, such as nodes, processes, as well as the GPU queues, memory operations, threads, and more that belong to them.

  • Timeline View: Shows CPU and GPU activities, events, and performance metrics in chronological order for a detailed temporal analysis. ROCm Optiq allows you to zoom, filter, and bookmark data for fine-grained inspection. You can correlate GPU workloads with in-application CPU events and performance with hardware resource usage, enabling easy identification and remediation of performance blockers.

  • Advanced Details Panel: Provides an in-depth view of profiling data, enabling you to analyze performance metrics and event-specific information. It offers SQL-like filters and group-by operations.

  • Histogram: Shows the event density across all visible tracks and highlights the zoomed-in region to quickly identify hotspots.

  • Time Range Filtering: Select a specific time interval to filter events and counter samples for focused analysis.

  • Event Search: Quickly locate target events.

Contents