diff --git a/src/default/render.rs b/src/default/render.rs
index e56c7b7..0004373 100644
--- a/src/default/render.rs
+++ b/src/default/render.rs
@@ -24,8 +24,6 @@ impl UiRenderer {
     pub fn draw(&mut self) {
         let output = match self.surface.get_current_texture() {
             CurrentSurfaceTexture::Success(texture) => texture,
-            // Still drawable; the surface has just changed under us, and
-            // reconfiguring is what the next frame wants rather than this one.
             CurrentSurfaceTexture::Suboptimal(texture) => {
                 self.surface.configure(&self.device, &self.config);
                 texture
@@ -34,7 +32,6 @@ impl UiRenderer {
                 self.surface.configure(&self.device, &self.config);
                 return;
             }
-            // Nothing to draw into this frame.
             CurrentSurfaceTexture::Timeout
             | CurrentSurfaceTexture::Occluded
             | CurrentSurfaceTexture::Validation => return,
@@ -81,8 +78,6 @@ impl UiRenderer {
     pub fn new(window: Arc<Window>) -> Self {
         let size = window.inner_size();
 
-        // The display handle is what GLES needs to present on Wayland, and it
-        // has to be the one the surface is made from.
         let instance = Instance::new(InstanceDescriptor {
             backends: Backends::PRIMARY,
             display: Some(Box::new(window.clone())),
diff --git a/tests/draw_cost.rs b/tests/draw_cost.rs
index d7ea2a3..36a1005 100644
--- a/tests/draw_cost.rs
+++ b/tests/draw_cost.rs
@@ -4,20 +4,19 @@
 //!
 //!     cargo test --release --test draw_cost -- --ignored --nocapture
 //!
-//! **Read the instruction count, not the clock.** Wall time here swings by 2x
-//! between runs of one binary on this machine -- more under `cargo test` than
-//! run directly -- while instructions retired are stable to 0.1%:
+//! Wall time is the wrong number to read for anything under a few percent --
+//! it varied by 2x between runs of one unchanged binary where instructions
+//! retired varied by 0.1%. Count those instead:
 //!
 //!     perf stat -e instructions:u target/release/.../draw_cost-* --ignored
 //!
-//! Measured that way on 2026-09-13, drawing each primitive through its own
-//! `PrimitiveRender` rather than a match in the renderer costs **6
-//! instructions per list drawn**, which is 0.1% of a frame at both 256 and
-//! 1024 layers. Recording one list into the pass costs wgpu ~5,400.
+//! That is how `PrimitiveRender` was measured against a match in the renderer:
+//! 6 instructions per list drawn, against the ~5,400 wgpu spends recording
+//! one.
 //!
-//! The instance is leaked on purpose. Dropping the last one makes the Vulkan
-//! loader unload Mesa's ICD, which faults when a thread that touched Vulkan
-//! exits -- and libtest runs every test on a spawned thread.
+//! The instance is leaked deliberately. A Vulkan loader may unload the driver
+//! when the last one drops, which can fault as a thread that used it exits --
+//! and every test runs on a spawned thread.
 
 use std::time::Instant;
 
@@ -30,13 +29,13 @@ use wgpu::{Color as GpuColor, *};
 
 const SIZE: u32 = 1024;
 const FRAMES: u32 = 200;
-/// Reported as the best of this many batches. The mean moves by 15% between
-/// runs on this machine, which is more than the thing being measured.
+/// Reported as the best of this many batches, since the mean moves by more
+/// than the thing being measured.
 const BATCHES: u32 = 8;
 
 fn gpu() -> Option<(Device, Queue)> {
-    // Probed rather than assumed: this machine's Vulkan device comes and goes,
-    // and GL is what is left when it is gone.
+    // Probed rather than assumed: there may be no Vulkan adapter, and GL is
+    // what is left when there is not.
     let all = Instance::new(InstanceDescriptor::new_without_display_handle());
     let instance = match pollster::block_on(all.request_adapter(&RequestAdapterOptions::default()))
     {