From 11e25f6a5fedc8765f380a6bc48f4ab7907d49d3 Mon Sep 17 00:00:00 2001 From: Alan Silva Date: Sat, 7 Nov 2026 22:03:52 +0000 Subject: [PATCH] Profile mobile chat text rendering --- app/notes/mobile-chat-text-profile.md | 81 +++++++++++++++++++++++++++ app/tool/chat_perf_fixture.dart | 4 ++ app/tool/chat_perf_probe.mjs | 14 +++-- 3 files changed, 95 insertions(+), 4 deletions(-) create mode 100644 app/notes/mobile-chat-text-profile.md diff --git a/app/notes/mobile-chat-text-profile.md b/app/notes/mobile-chat-text-profile.md new file mode 100644 index 00000000..c1899e8d --- /dev/null +++ b/app/notes/mobile-chat-text-profile.md @@ -0,0 +1,81 @@ +# Mobile chat text profile + +## Scope + +Diagnostic-only profile of the real `ChatThreadView` with its isolated fake +`ApiGateway` (`tool/chat_composer_probe_entry.dart`). The synthetic transcript +contains prose, headings, inline emphasis/code, fenced Dart, links, lists and a +3-column Markdown table; the stream adds 80 incremental text updates. This is a +focused widget fixture rather than a full shell/session-list capture. Runs were +made in a quiet worktree on release web, headless Chrome at 560x1000 with +software rasterisation. These are Chrome wheel/software-raster comparison +numbers, not Android FPS; no authorized Android device was connected. + +## Reproduction + +From `app/`, after checking that no competing Flutter build, test or capture was +running: + +```sh +/Users/alan/development/flutter/bin/flutter build web --release \ + --target tool/chat_composer_probe_entry.dart \ + --output build/mobile-chat-text-profile +node --check tool/chat_perf_probe.mjs +node tool/chat_perf_probe.mjs build/mobile-chat-text-profile 20 80 3 \ + > build/mobile-chat-text-profile-short.json +node tool/chat_perf_probe.mjs build/mobile-chat-text-profile 300 80 3 \ + > build/mobile-chat-text-profile-long.json +``` + +Each repetition uses fresh pages for open, 120 wheel gestures (60 up/60 down), +and streaming. The JSON reports are in the ignored build directory. The probe +was changed only to make the phone-width/repeat count explicit; the fixture was +extended with the table above. + +## Measurements + +Values below are medians across three repetitions. Frame timings are Flutter +`FrameTiming` values. `over16` is the count of measured frames over 16 ms in the +software Chrome harness, not a device refresh-rate assertion. + +| workload | frames | build p50 / p90 ms | raster p50 ms | total p50 / p90 ms | over16 | +| --- | ---: | ---: | ---: | ---: | ---: | +| open, 20 messages | 13 | 2.6 / 4.3 | 26.2 | 29.6 / 31.8 | 13 | +| open, 300 messages | 15 | 2.5 / 4.3 | 25.7 | 29.3 / 32.5 | 15 | +| scroll, 20 messages | 120 | 1.4 / 6.7 | 25.0 | 28.0 / 36.0 | 120 | +| scroll, 300 messages | 245 | 1.5 / 6.8 | 24.7 | 27.6 / 35.9 | 245 | +| stream, 20 messages | 165 | 1.3 / 3.7 | 24.7 | 27.6 / 30.1 | 165 | +| stream, 300 messages | 160 | 4.8 / 5.8 | 24.8 | 29.4 / 32.0 | 160 | + +## Ranked findings + +1. **Measured: raster dominates this scrolling harness.** Build p50 is 1.4–1.5 + ms while raster p50 is about 25 ms. These numbers cannot identify whether + text, tables, selection, or CanvasKit compositing owns the raster cost, so + they do not justify a Markdown rewrite. +2. **Measured: the long live answer costs more build time.** Stream build p50 is + 4.8 ms for 300 messages versus 1.3 ms for 20. This is consistent with the + growing live Markdown turn and text layout, but it is not isolated proof that + Markdown parsing alone is responsible. +3. **Measured: opening history is not strongly message-count-sensitive here.** + Open build and total medians are effectively unchanged from 20 to 300 rows. + The transcript uses the existing row construction and `ChatMarkdownCache`; + this does not prove all native-device layout costs are equivalent. +4. **Evidence-backed next candidate, not a defect finding:** + `ChatMarkdownView` renders tables with `IntrinsicColumnWidth`, so a focused + visible-table A/B profile is warranted. It was included in this workload but + not isolated from the rest of rich Markdown. + +## Next small experiment + +Use the existing full-app `tool/perf/perf_fixture.dart` rich-chat fixture to +isolate table-free versus table-rich visible answers, repeating touch and wheel +separately at 560px. Then compare the same workload on an authorized native +Android device before making a mobile performance claim. A diagnostic selection +on/off comparison may follow only if the table experiment does not explain the +raster cost; selection behavior must remain unchanged. + +No production code or behavior was changed. Selection/copy, rich Markdown, +scroll anchors and drafts remain unchanged. No app tests were run because this +visit changed only profiling fixture/probe code and a note; `node --check` and +the release build succeeded. diff --git a/app/tool/chat_perf_fixture.dart b/app/tool/chat_perf_fixture.dart index 949704e2..0f187ba9 100644 --- a/app/tool/chat_perf_fixture.dart +++ b/app/tool/chat_perf_fixture.dart @@ -23,6 +23,10 @@ String _answer(int index) => '```dart\n' '${List.generate(30, (line) => 'final v$line = $index * $line;').join('\n')}' '\n```\n\n' + '| check | result | owner |\n' + '| --- | --- | --- |\n' + '| unit tests | passed | worker $index |\n' + '| static analysis | passed | worker $index |\n\n' '- one\n- two\n- three\n'; /// The fixture history; the newest message is the operator's, so a streamed diff --git a/app/tool/chat_perf_probe.mjs b/app/tool/chat_perf_probe.mjs index e280b223..e2f86e1b 100644 --- a/app/tool/chat_perf_probe.mjs +++ b/app/tool/chat_perf_probe.mjs @@ -4,7 +4,7 @@ // flutter build web --release --target tool/chat_composer_probe_entry.dart \ // --output build/chat-perf // then: -// node tool/chat_perf_probe.mjs build/chat-perf [messages=300] [deltas=200] +// node tool/chat_perf_probe.mjs build/chat-perf [messages=300] [deltas=200] [repeats=3] // // Three phases, each on a fresh page: open the transcript (the frames that do // work after the history reply), wheel-scroll it up and back down, @@ -17,13 +17,14 @@ import { openBrowser } from './capture/browser.mjs'; import { flutterPage } from './capture/page.mjs'; import { serveBuild } from './capture/serve.mjs'; -const [dir, messagesArg = '300', deltasArg = '200'] = process.argv.slice(2); +const [dir, messagesArg = '300', deltasArg = '200', repeatsArg = '3'] = process.argv.slice(2); if (!dir) { console.error('usage: node tool/chat_perf_probe.mjs [messages] [deltas]'); process.exit(2); } const messages = Number(messagesArg); const deltas = Number(deltasArg); +const repeats = Number(repeatsArg); function stats(frames) { const pick = (i) => frames.map((f) => f[i]).sort((a, b) => a - b); @@ -67,7 +68,7 @@ async function phase(origin, query, drive) { }); const opened = Date.now(); try { - await page.open(`${origin}/?${query}`, { width: 880, height: 1000 }); + await page.open(`${origin}/?${query}`, { width: 560, height: 1000 }); } catch (error) { throw new Error(`${error.message}; page errors: ${JSON.stringify(page.errors.slice(0, 5))}`); } @@ -89,6 +90,8 @@ async function phase(origin, query, drive) { const server = await serveBuild(dir); try { + const results = []; + for (let repeat = 0; repeat < repeats; repeat++) { const open = await phase(server.origin, `perf=${messages}`, async ({ opened }) => { // The fixture flushes the timings 2.5 s after the history reply. const end = Date.now() + 20000; @@ -124,7 +127,10 @@ try { return { streamed: marks.includes('PERF_STREAM_END'), stream: stats(stop()) }; }, ); - console.log(JSON.stringify({ build: dir, messages, deltas, open, scroll, stream }, null, 2)); + results.push({ repeat: repeat + 1, open, scroll, stream }); + console.log(JSON.stringify({ repeat: repeat + 1, messages, deltas, open, scroll, stream })); + } + console.log(JSON.stringify({ build: dir, messages, deltas, repeats, results }, null, 2)); } finally { await server.close(); }