diff --git a/DESIGN.md b/DESIGN.md
index 4ad964d..4f34064 100644
--- a/DESIGN.md
+++ b/DESIGN.md
@@ -186,8 +186,11 @@ triplet.
 - Pressure fixture: `etaf-research-shelf-fixture-size` defaults to 256 and
   `etaf-research-shelf-page-size` defaults to 12; existing local rows are
   preserved and missing fixture rows are topped up with fresh IDs.
-- Performance: one Data mutation -> one Runtime generation/publication; normal
-  warm actions remain under the accepted 105ms p50 target.
+- Performance: one Data mutation -> one Runtime generation/publication. The
+  current active latency lane uses 5 unmeasured warmups and 30 measured samples
+  at fixed 1413×62 geometry; every scenario must satisfy p95 and max at or below
+  50ms. The accepted 105ms p50 wording is historical design context, not the
+  current evaluator gate; `DOC-PERF-001` keeps its formal disposition in M0b1.
 - Compatibility: any same-basename `.etaf`/`.el` example is discoverable;
   `.ecss` is optional, and the bundled Research Shelf remains only one
   consumer.
@@ -195,6 +198,18 @@ triplet.
   refresh tests, SQLite temp-file integration tests, mount/remount tests,
   repeated selection/mutation tests, and clean GUI screenshots.
 
+## Performance evidence lanes
+
+The latency lane, trace lane, and GUI lane share one
+repository/environment/scenario/fixture/build identity but remain separate
+measurements. The latency lane owns the uninstrumented 5/30 p95/max 50ms gate.
+The trace lane owns a representative instrumented invocation with cost class,
+work counters, turns, allocation, and GC. The GUI lane owns the real Emacs
+action sequence, screenshots, recording, and temporal-review verdict. Batch
+verifier duration is not GUI first paint; action-start through forced redisplay
+completion remains an explicitly separate future first-paint gate and an M0a
+observed gap. No one lane may be used to claim another lane passed.
+
 ## Open questions
 
 - [ ] User may rename the product after reviewing the first HTML/ETAF render;
diff --git a/DESIGN.zh-CN.md b/DESIGN.zh-CN.md
index 3db17cd..d7095a0 100644
--- a/DESIGN.zh-CN.md
+++ b/DESIGN.zh-CN.md
@@ -108,12 +108,24 @@ ETAF 私有 registry 变量；Component setup 中创建的 Controller 自动归
 - 用 inert reader、source tab/session、未保存 source refresh 和 SQLite 临时文件
   测试 mount/remount、筛选、分页、重复选行、mutation、错误状态和 cleanup；GUI
   用干净 fullscreen 截图验证真实布局。
-- 目标：一次 Data mutation 对应一次 Runtime generation/publication；warm 交互
-  维持已接受的 105ms p50 预算。
+- 目标：一次 Data mutation 对应一次 Runtime generation/publication。当前有效的
+  延迟 lane 在固定 1413×62 几何下执行 5 次未计时 warmup 与 30 个计时 sample，
+  每个场景的 p95 和 max 均须不超过 50ms。已接受的 105ms p50 是历史设计上下文，
+  不是当前 evaluator gate；其正式处置仍由 M0b1 的 `DOC-PERF-001` 负责。
 - 压测入口：`etaf-research-shelf-fixture-size` 默认 256，
   `etaf-research-shelf-page-size` 默认 12；已有本地记录保留，不足部分使用新 ID
   补齐 fixture。
 
+## 性能 evidence lanes
+
+延迟 lane、追踪 lane 与 GUI lane 共享同一组仓库、环境、场景、fixture 和 build identity，
+但保持为三种独立测量。延迟 lane 负责无 instrumentation 的 5/30、p95/max 50ms
+门禁；追踪 lane 负责一次代表性 instrumented invocation 的 cost class、work counters、
+turns、allocation 与 GC；GUI lane 负责真实 Emacs action sequence、截图、录像与
+temporal-review verdict。Batch verifier duration 不是 GUI first paint；从 action-start
+到 forced redisplay completion 的 first-paint 计时仍是独立 future gate，也是 M0a
+明确记录的 observed gap。任何一条 lane 都不能替另一条 lane 宣称通过。
+
 ## 待 review
 
 - [ ] 用户可调整产品命名或 palette；pair 边界、组合/复用原则和公开交互合同不变。
diff --git a/README.md b/README.md
index 25ecd37..cd4fb23 100644
--- a/README.md
+++ b/README.md
@@ -110,6 +110,26 @@ For repeated absolute-latency runs, use `make perf-prepare` once after source
 changes, let compilation activity settle, then run `make perf-evaluator` without
 rebuilding the dependency graph.
 
+## Performance evidence lanes
+
+Performance evidence has three separate lanes correlated by one
+repository/environment/scenario/fixture/build identity:
+
+- The latency lane runs the fixed 1413×62 batch evaluator with 5 unmeasured
+  warmups and 30 measured samples. The current active gate requires both p95
+  and max to be at or below 50ms for every scenario.
+- The trace lane runs a separate instrumented representative invocation. Its
+  cost classes, work counters, turns, allocation, and GC data are not inserted
+  into the timed latency distribution.
+- The GUI lane captures the real Emacs interaction sequence, screenshots, and
+  temporal-review verdict. Batch verifier duration is not GUI first paint;
+  action-start to forced-redisplay-complete first-paint timing remains a
+  separately identified future gate and an observed M0a gap.
+
+Evidence from one lane supports only that lane's conclusion. A
+performance-complete or user-visible-non-regression claim requires the
+applicable gates from all three lanes.
+
 The bundled Research Shelf example installs a deterministic 256-record SQLite
 fixture with 12 records per page. Bind `etaf-research-shelf-fixture-size` and
 `etaf-research-shelf-page-size` for smaller tests or larger pressure runs. Its
diff --git a/tests/etaf-playground-docs-tests.el b/tests/etaf-playground-docs-tests.el
new file mode 100644
index 0000000..5a0c963
--- /dev/null
+++ b/tests/etaf-playground-docs-tests.el
@@ -0,0 +1,50 @@
+;;; etaf-playground-docs-tests.el --- Documentation contracts -*- lexical-binding: t; -*-
+
+;;; Code:
+
+(require 'ert)
+
+(defconst etaf-playground-docs-tests--root
+  (file-name-directory
+   (directory-file-name
+    (file-name-directory (or load-file-name buffer-file-name))))
+  "Absolute etaf-playground repository root.")
+
+(defun etaf-playground-docs-tests--contents (name)
+  "Return the literal contents of repository file NAME."
+  (with-temp-buffer
+    (insert-file-contents
+     (expand-file-name name etaf-playground-docs-tests--root))
+    (buffer-string)))
+
+(defun etaf-playground-docs-tests--normalized-contents (name)
+  "Return NAME with runs of whitespace collapsed for prose assertions."
+  (replace-regexp-in-string
+   "[[:space:]]+" " " (etaf-playground-docs-tests--contents name)))
+
+(ert-deftest etaf-performance-docs-distinguish-evidence-lanes ()
+  "Latency, trace, and GUI evidence remain explicitly separate lanes."
+  (dolist (file '("README.md" "DESIGN.md" "DESIGN.zh-CN.md"))
+    (let ((text (etaf-playground-docs-tests--normalized-contents file)))
+      (should (string-match-p "latency lane\\|延迟 lane" text))
+      (should (string-match-p "trace lane\\|追踪 lane" text))
+      (should (string-match-p "GUI lane" text))
+      (should (string-match-p
+               "repository/environment/scenario/fixture/build identity\\|仓库、环境、场景、fixture 和 build identity"
+               text))
+      (should (string-match-p
+               "not GUI first paint\\|不是 GUI first paint"
+               text)))))
+
+(ert-deftest etaf-performance-docs-separate-current-gate-from-historical-budget ()
+  "The active 50ms gate must not be confused with the historical 105ms text."
+  (dolist (file '("DESIGN.md" "DESIGN.zh-CN.md"))
+    (let ((text (etaf-playground-docs-tests--normalized-contents file)))
+      (should (string-match-p "5.*warmup.*30.*sample\\|5.*warmup.*30.*sample" text))
+      (should (string-match-p "p95.*max.*50ms\\|p95.*max.*50ms" text))
+      (should (string-match-p "105ms p50" text))
+      (should (string-match-p "historical\\|历史" text))
+      (should (string-match-p "M0b1" text)))))
+
+(provide 'etaf-playground-docs-tests)
+;;; etaf-playground-docs-tests.el ends here