README: sync table from an idle box

Two sweeps taken while no other session had a build or test running: 1 thread
4.5 to 4.6 M at 217 ns, 2 threads 7.6 to 9.5 M, 4 threads 16.4 to 17.4 M,
8 threads 25.1 to 26.8 M, 16 threads up to 35.8 M (the WSL-down run). The
earlier 2-thread reading of 6 to 7 M was other sessions' load, not the
library: pinned to distinct physical cores the two threads run at 210 ns each.
The intro now says the WSL VM was shut down for the Kestrel and comparison
rows specifically.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

Author: Astn

README.md

@@ -238,7 +238,7 @@ dotnet run -c Release --project TestServer_Console                  # menu: Ente
 dotnet run --project samples/WasmHost                               # browser: "Run benchmark" on the page
 ```
 
-All numbers below are from an AMD Ryzen 7 7800X3D (8 cores / 16 threads, 4.2 GHz), 64 GB, Windows 11, .NET 10, Release, Server GC, with the built-in serializer, measured 2026-09-23 with the WSL virtual machine shut down (idle, it was taking 15 to 25 % of the box). Where a row gives two figures they are the spread over that day's runs. A single benchmark thread on this machine varies with whatever else lands on its core's SMT sibling, so the 1-thread rows are from runs on an idle core.
+All numbers below are from an AMD Ryzen 7 7800X3D (8 cores / 16 threads, 4.2 GHz), 64 GB, Windows 11, .NET 10, Release, Server GC, with the built-in serializer, measured 2026-09-23. Where a row gives two figures they are the spread over that day's runs on an otherwise idle machine; the WSL virtual machine, which takes 15 to 25 % of the box when idle, was shut down for the Kestrel and comparison runs. A single benchmark thread on this machine varies with whatever else lands on its core's SMT sibling, so the 1-thread rows are from runs on an idle core.
 
 ### Sync: the library alone
 
@@ -248,11 +248,11 @@ All numbers below are from an AMD Ryzen 7 7800X3D (8 cores / 16 threads, 4.2 GHz
 
 | Threads | RPC/s | ns per request per thread | Allocations per request |
 | ---: | ---: | ---: | --- |
-| 1 | 4.4 M to 4.6 M | 224 | 0 bytes for numeric shapes, one string for `StringMe` |
-| 2 | 6.0 M to 7.3 M | 276 | |
-| 4 | 12.6 M to 15.1 M | 265 | |
-| 8 | 23.6 M to 23.8 M | 336 | |
-| 16 | 31.2 M to 35.8 M | 446 | |
+| 1 | 4.5 M to 4.6 M | 217 | 0 bytes for numeric shapes, one string for `StringMe` |
+| 2 | 7.6 M to 9.5 M | 222 | |
+| 4 | 16.4 M to 17.4 M | 230 | |
+| 8 | 25.1 M to 26.8 M | 298 | |
+| 16 | 30.6 M to 35.8 M | 446 | |
 
 Per-thread cost rises with thread count because the 16 threads share 8 physical cores.
 

benchmarks/charts/render.py

@@ -108,11 +108,11 @@ def chart(title, subtitle, rows, filename, width=920):
     "JSON-RPC.Net alone, by thread count",
     "Byte entry point in a loop, no scheduler, no transport. 8 cores / 16 threads.",
     [
-        ("1 thread", 4.4e6, 4.6e6, "ours", "224 ns per request, 0 bytes allocated"),
-        ("2 threads", 6.0e6, 7.3e6, "ours", ""),
-        ("4 threads", 12.6e6, 15.1e6, "ours", ""),
-        ("8 threads", 23.6e6, 23.8e6, "ours", ""),
-        ("16 threads", 31.2e6, 35.8e6, "ours", "SMT: two threads per core"),
+        ("1 thread", 4.5e6, 4.6e6, "ours", "217 ns per request, 0 bytes allocated"),
+        ("2 threads", 7.6e6, 9.5e6, "ours", ""),
+        ("4 threads", 16.4e6, 17.4e6, "ours", ""),
+        ("8 threads", 25.1e6, 26.8e6, "ours", ""),
+        ("16 threads", 30.6e6, 35.8e6, "ours", "SMT: two threads per core"),
     ],
     "sync-threads.svg",
 )