4 Commits

Include renames

Author	SHA1	Message	Date
kferjaoui	59d8ed8a0c	docs(deck): 15 pt body, wider margins, and a layout audit that reports clean ... Content box 0.70-12.70 -> 0.50-12.90, applied as an affine map over every module-level layout call. Interstitial thesis and list set at body size, with per-entry measured pitch. ~30 slide cuts: numbers and remarks stay, the sentence explaining them moves to notes(). Restores three code panels that had been reduced to bare list literals -- slide 33's frame-147 walkthrough, 34's recommended pattern, 15's barrier comparison. Figures re-laid wherever text met text (pinning, tile, resultpath, streams, arc, pagefault, pedtiming and others). New deckgate.py holds the projection floor and placement table so make_figs_kernel.py is gated too; its three figures had been set at 7-8 pt, unchecked. audit_layout.py splits lines at a column gutter and drops the overlap threshold to 10%: it was reporting every two-column slide and missing a bullet running under a code panel title. Deck now audits 0 collisions, 0 overruns, all figure text >= 10 pt.	2026-08-27 19:28:11 +02:00
kferjaoui	bc6121ba5f	docs: update deck	2026-08-27 16:05:14 +02:00
kferjaoui	a9ee1fc074	deck: identify Test3 as the CPU/CPU divergence channel ... Slide 30 reported an 8/11 cluster mismatch between ClusterFinder and ClusterFinderFrozen without naming its cause. Two experiments now do. Instrumented: 974 divergent pixels partition onto the three decisions with no remainder -- Test1 619 flips / 0 clusters, Test3 347 / 11, local-max gate 8 / 8. Ablated: the 11 go to exactly zero. Corrects a claim the data refuted -- Test1 initiates independently, ~7x slower, and never creates a cluster. AARE_BRANCH_TRACE defaults to 0 and folds away at compile time, so the CPU baseline the deck quotes is unaffected. AARE_TEST3_ENABLED defaults to 1. New annex A7 (2 slides): part 1 restores the third test slide 4 omits and derives c3 from variance addition; part 2 carries the measurement. fig_test3 painted a completed branch map with a "scan is here" cursor over it -- now the finished frame, arrows meaning raster order, shadow given its own fill so a photon stops looking 3x4 wide, frozen in red so amber keeps one meaning. fig_overlap_9x9: row spacing now clears the tag, not the lane. New docs/deck/QA.md: the code and algorithm questions this raised, with where each is settled. Also retired 3 figures placed on no slide; annex count in report and README was stale (6/53, now 7/55).	2026-08-26 17:14:31 +02:00
kferjaoui	9665df975c	docs: detail opt2, op3 and opt5->opt6 explanaitons	2026-08-26 14:38:23 +02:00