CPU: per-mode decode benchmarks + live/max demod effort (Pi-deployable)#101
Conversation
bench/cpu.sh measures x-realtime per mode. First measurement caught the demod-campaign trade: AIS 3.0x and ADS-B 2.2x realtime on an M-series Mac - unusable on Pi-class hardware. Fixes: - AIS: cached GMSK waveform tables (were re-integrated per burst), trellis traceback matrix instead of O(n^2) path clones, stride-2 template hunt with low-metric span skipping. 3.0x -> 8.6x, and the hunt change GAINED a frame (51 -> 52). - ADS-B: --demod-effort live|max. Max (file-decode default) scans all eight 1/8-sample grids: 5.3x, 161 unique. Live (SDR default) scans one half-sample grid: 16.6x, 156 unique (97%). Per-command defaults, flag overrides. All four bench gates green (bench uses max). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
|
Warning Review limit reached
More reviews will be available in 41 minutes and 26 seconds. Learn how PR review limits work. Your organization has run out of usage credits. Purchase more credits in the billing tab to continue. ⌛ How to resolve this issue?After more reviews become available, a review can be triggered using the We recommend that you space out your commits to avoid hitting the rate limit. 🚦 How do rate limits work?CodeRabbit enforces hourly rate limits for each developer per organization. Our paid plans include higher PR review limits than trial, open-source, and free plans. In all cases, reviews become available again over time. During sustained high-volume PR review activity, CodeRabbit may temporarily slow when the next review becomes available. Please see our Fair Usage Limits Policy for further information. ℹ️ Review info⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (10)
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: b334f24a1a
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| include: tune.filter_labels.clone(), | ||
| exclude: tune.exclude_labels.clone(), | ||
| }, | ||
| demod_effort: tune.demod_effort.unwrap_or(runtime::DemodEffort::Live), |
There was a problem hiding this comment.
When xng tui --file ... is used without --demod-effort, this unconditionally selects Live, so ADS-B recordings opened in the TUI use the reduced timing-grid scan and can miss frames that xng decode finds by default. This contradicts the new option help/default policy for file decoding and only affects TUI sessions backed by a file; choose the default based on whether file is Some before falling back to Live for actual SDR use.
Useful? React with 👍 / 👎.
1 participant
What
Task #38:
bench/cpu.shmeasures ×-realtime decode speed per mode — and the first measurement caught the demod campaign's hidden trade: AIS at 3.0× and ADS-B at 2.2× realtime on an M-series Mac, unusable on Pi-class hardware.*first measurement startup-dominated; methodology corrected (looped capture).
AIS: cached GMSK waveform tables, traceback matrix replacing O(n²) path clones, stride-2 template hunt with low-metric span skipping.
ADS-B:
--demod-effort live|max— per-command defaults (file decode = max, SDR = live), flag overrides.Pi math (5–8× slower): live ADS-B ≈ 2–3×, AIS ≈ 1.4× — both deployable.
Testing
Workspace green; all four bench gates green; coh_unit 202/202 bit-true after the traceback rewrite.
🤖 Generated with Claude Code