Unreal Engine Research & Evidence-Based Practice 5 — Questions and Answers
Question 1: Which Unreal Engine feature lets a designer run data-driven experiments on AI behavior trees without modifying code?
- Gameplay Ability System attributes driving Blackboard keys (Correct answer)
- Chaos solver iterations console variable
- Physics Asset collision channels
- Sequencer master sequence blending
Correct answer: Gameplay Ability System attributes driving Blackboard keys
Exposing Gameplay Ability System attributes to Blackboard keys allows designers to swap numeric parameters at runtime, enabling data-driven behavioral experiments without code changes.
Question 2: A team researching texture streaming efficiency should analyze which Unreal Engine stat category?
- stat engine
- stat streaming (Correct answer)
- stat net
- stat audio
Correct answer: stat streaming
`stat streaming` displays real-time texture streaming throughput, pool usage, and pending requests, giving direct evidence of streaming efficiency.
Question 3: When citing Unreal Engine source code behavior in technical documentation, which practice best supports reproducibility?
- Reference the engine version and specific file path and line range in the GitHub mirror (Correct answer)
- Quote the behavior from memory without a source
- Link to a community wiki that may be edited later
- Describe the behavior verbally without code citation
Correct answer: Reference the engine version and specific file path and line range in the GitHub mirror
Pinning a citation to a specific engine version, file path, and line range ensures readers can verify the exact code you describe, even as the engine evolves.
Question 4: Which profiling approach provides evidence for optimizing tick frequency of non-critical Actors in Unreal Engine?
- Enable 'Start with Tick Enabled = false' on all actors without measuring first
- Use Unreal Insights to identify Actors with high cumulative tick time, then increase their TickInterval (Correct answer)
- Reduce polygon count of the Actor mesh
- Switch the Actor to a Static Mesh Component
Correct answer: Use Unreal Insights to identify Actors with high cumulative tick time, then increase their TickInterval
Identifying the highest-cost ticking Actors via Unreal Insights and increasing their TickInterval based on measured cost is an evidence-driven optimization.
Question 5: A developer proposes that Virtual Shadow Maps (VSMs) reduce shadow rendering cost. What constitutes sufficient empirical evidence for this claim?
- Epic's marketing page stating VSMs are better
- Profiling the same scene with traditional shadow maps vs. VSMs using GPU Visualizer and recording GPU shadow pass time (Correct answer)
- Noticing the scene looks nicer with VSMs enabled
- The VSM setting being marked 'Recommended' in Project Settings
Correct answer: Profiling the same scene with traditional shadow maps vs. VSMs using GPU Visualizer and recording GPU shadow pass time
Recording GPU shadow pass time under both shadow systems on the same scene using GPU Visualizer provides quantitative, reproducible evidence for the performance comparison.
Question 6: Which practice best ensures that performance research findings in Unreal Engine are valid across target hardware tiers?
- Test only on the developer's workstation
- Run benchmarks on a minimum-spec, mid-tier, and high-end target device and report results separately (Correct answer)
- Average results from one device type
- Use only synthetic benchmarks unrelated to Unreal
Correct answer: Run benchmarks on a minimum-spec, mid-tier, and high-end target device and report results separately
Testing across minimum-spec, mid-tier, and high-end hardware and reporting results separately ensures findings reflect the real performance envelope for all target players.
Question 7: How does enabling 'r.ProfileGPU' (the GPU profile dump command) differ from using Unreal Insights for ongoing research?
- r.ProfileGPU captures a single-frame GPU hierarchy snapshot; Unreal Insights records multi-frame traces for trend analysis (Correct answer)
- r.ProfileGPU only measures CPU cost; Unreal Insights only measures GPU cost
- They are identical tools with different names
- r.ProfileGPU requires an Nvidia GPU; Unreal Insights works on AMD only
Correct answer: r.ProfileGPU captures a single-frame GPU hierarchy snapshot; Unreal Insights records multi-frame traces for trend analysis
`r.ProfileGPU` (or the 'profilegpu' command) dumps one frame's GPU pass hierarchy, while Unreal Insights records continuous multi-frame GPU data suitable for trend and regression analysis.
Unreal Engine Research & Evidence-Based Practice 5
Which Unreal Engine feature lets a designer run data-driven experiments on AI behavior trees without modifying code?