diff --git a/doc/site/performance.markdown b/doc/site/performance.markdown
index 9c2173779..09eb8ddca 100644
--- a/doc/site/performance.markdown
+++ b/doc/site/performance.markdown
@@ -6,90 +6,80 @@ seem to like them, so here's a few:
Method Call
- wren | 0.12s |
+ wren0.4 | 0.079s |
- luajit (-joff) | 0.16s |
+ luajit2.1 -joff | 0.091s |
- ruby | 0.20s |
+ ruby3.0 | 0.104s |
- lua | 0.35s |
+ lua5.4 | 0.121s |
- python3 | 0.78s |
-
-
- python | 0.85s |
+ python3.11 | 0.171s |
-DeltaBlue
+
+Delta Blue
- wren | 0.13s |
+ wren0.4 | 0.086s |
- python3 | 0.48s |
-
-
- python | 0.57s |
+ python3.11 | 0.107s |
+
Binary Trees
- luajit (-joff) | 0.11s |
+ luajit2.1 -joff | 0.074s |
- wren | 0.22s |
+ ruby3.0 | 0.116s |
- ruby | 0.24s |
+ python3.11 | 0.137s |
- python | 0.37s |
+ lua5.4 | 0.139s |
- python3 | 0.38s |
-
-
- lua | 0.52s |
+ wren0.4 | 0.141s |
+
Recursive Fibonacci
- luajit (-joff) | 0.10s |
+ luajit2.1 -joff | 0.055s |
- wren | 0.20s |
+ lua5.4 | 0.087s |
- ruby | 0.22s |
+ ruby3.0 | 0.119s |
- lua | 0.28s |
+ wren0.4 | 0.148s |
- python | 0.51s |
-
-
- python3 | 0.57s |
+ python3.11 | 0.156s |
-**Shorter bars are better.** Each benchmark is run ten times and the best time
-is kept. It only measures the time taken to execute the benchmarked code
-itself, not interpreter startup.
-
-These were run on my MacBook Pro 2.3 GHz Intel Core i7 with 16 GB of 1,600 MHz
-DDR3 RAM. Tested against Lua 5.2.3, LuaJIT 2.0.2, Python 2.7.5, Python 3.3.4,
-ruby 2.0.0p247. LuaJIT is run with the JIT *disabled* (i.e. in bytecode
-interpreter mode) since I want to support platforms where JIT-compilation is
-disallowed. LuaJIT with the JIT enabled is *much* faster than all of the other
-languages benchmarked, including Wren, because Mike Pall is a robot from the
-future.
+**Shorter bars are better.** Each benchmark is run ten times and the median
+time is calculated. It only measures the time taken to execute the benchmarked
+code itself, not interpreter startup.
+
+These were run on a machine with an AMD Ryzen 9 5950X processor. Tested
+against Lua 5.4.4, LuaJIT 2.1.0-beta3, Python 3.11.3, ruby 3.0.5. LuaJIT is
+run with the JIT *disabled* (i.e. in bytecode interpreter mode) since I want
+to support platforms where JIT-compilation is disallowed. LuaJIT with the JIT
+enabled is *much* faster than all of the other languages benchmarked,
+including Wren, because Mike Pall is a robot from the future.
The benchmark harness and programs are
[here](https://github.com/wren-lang/wren/tree/main/test/benchmark).