diff --git a/CLAUDE.md b/CLAUDE.md
index 77675a3..613226e 100644
--- a/CLAUDE.md
+++ b/CLAUDE.md
@@ -97,10 +97,17 @@ exi_bba.synth --seeds 8`:
 
 | domain | target | result |
 |---|---|---|
-| `clk` (sync, HFOSC) | 24 MHz | 28.1–30.8 MHz — **PASS on all 8 seeds** |
-| EXI pin domain | 27 MHz | 57.7–66.8 MHz — **PASS on all 8 seeds** |
+| `clk` (sync, HFOSC) | 24 MHz | 27.3–29.1 MHz — **PASS on all 8 seeds** |
+| EXI pin domain | 27 MHz | 50.0–63.9 MHz — **PASS on all 8 seeds** |
 
-Best seed **6**: `clk` 30.83 MHz, EXI 57.74 MHz. Flash `build/seed<N>/top.bin`.
+Best seed **3**: `clk` 29.08 MHz (1.21×), EXI 63.89 MHz. Flash
+`build/seed<N>/top.bin`.
+
+Area is **4305 ICESTORM_LC** (81% of the UP5K), identical on every seed. The
+documentation/dead-code pass took it 4350 → 4305 by deleting a dead `rx_rptr`
+SyncFIFO and the unused W5100 debug-peek path; `clk` moved 29.74 → 29.08 on the
+best seed, which is placement noise, since none of the removed logic was on a
+critical path.
 
 Measured after making the SPRAM response combinational (the `+1` descriptor
 fix), which put SPRAM's output → byte mux → `tx_fifo` write data on `sync`'s