Video summary
UEFN Verse Insights: How to Find and Fix Verse Performance Issues
Main summary
Key takeaways
Summary
The video demonstrates how to use Verse Insights in UEFN to find and fix code that causes periodic lag on a Fortnite island.
Capture the slowdown
Timing Insights connects to the running island and records activity over time. Because Verse runs on the server, slow Verse code can cause in-game latency even when the client’s frame rate does not appear low. In the demo, the server is captured while pumpkins are harvested.
Trace the costly work
In the Frames panel, tall bars identify slow frames. The Timing View shows nested timing events. Following a wide event down to the lowest relevant level helps locate the work consuming time.
Understand the timing categories
Verse code appears as bytecode, executed by the Verse virtual machine. Built-in functions appear as native code. Slow bytecode suggests revising the code itself, while slow native time may indicate that built-in functions are being called inefficiently.
Identify the root cause
Each harvest recalculated the total soup by looping through more than 150,000 previously harvested pumpkins. The small soup from pumpkin function was not inherently slow; the problem was calling it so many times.
Fix and verify
The code was changed to maintain a running total, adding only the soup from each newly harvested pumpkin. A follow-up capture showed shorter frame bars and no harvest lag.
AI-assisted option
The video also shows Claude, through UEFN MCP, finding and fixing the same performance issue.
Guides and resources
- A previous video about Timing Insights is recommended for viewers who want to learn how to use the tool.
- A UEFN MCP getting-started video is linked in the description.
- Verse Insights and profiling-tool resources are also linked in the description.
Main speaker/source: A Fortnite Developers narrator demonstrating profiling and code changes in UEFN.
Rate this summary
Your feedback will help improve summaries.
Improve this summary
Reprocess with a stronger model when the summary feels incomplete or inaccurate.
Translate summary in another language
Ask questions to this video
Chat for follow-up questions, clarifications, and source-backed answers.