Drop HRTF claim from audio description (it's stereo, not HRTF)

Audio is per-player stereo mixed from each agent's position and
orientation, not true HRTF — adjust README + home page accordingly.

README.md

@@ -35,7 +35,7 @@ This dataset turns those demos into rendered, frame-accurate, fully-annotated tr
 - **Behaviour cloning / VLA policies.** `(frame, audio) → action` for vision-conditioned action models.
 - **Inverse Dynamics Models (IDM).** `(frame_t, frame_{t+k}) → action` to recover actions from unlabelled video — the workhorse for VPT-style pre-training.
 - **Forward dynamics / world models.** `(frame, action) → frame_{t+1}` with all 10 player POVs of the same world state available as supervision.
-- **Spatial-audio conditioning.** Audio is recorded per-player with HRTF positional cues, so models can learn to localize footsteps, gunfire, and callouts.
+- **Spatial-audio conditioning.** Per-player stereo recorded relative to each agent's position and orientation, so models can learn to localize footsteps, gunfire, and callouts.
 - **Multi-agent training.** All 10 perspectives of the same round are kept aligned tick-for-tick — useful for collaborative policies, opponent modelling, and multi-player world models.
 
 ## What's recorded
@@ -43,7 +43,7 @@ This dataset turns those demos into rendered, frame-accurate, fully-annotated tr
 For every chunk (≤ 1 minute, one player POV):
 
 - **Video** — 1280×720 @ 32 fps, near-lossless H.264. One stream per player; ten POVs per round all tick-aligned.
-- **Audio** — per-player stereo with HRTF positional cues (footsteps, gunfire, callouts).
+- **Audio** — per-player stereo, mixed from each agent's position and orientation (footsteps, gunfire, callouts).
 - **Inputs** — every tick: keyboard state, mouse delta, fire/jump/use, weapon switches.
 - **World state** — every tick, for all 10 players: position, velocity, view yaw/pitch, camera intrinsics, health, armor, ammo, primary/secondary weapon, alive flag.
 - **G-buffers (coming soon)** — per-pixel luminance, depth map, and vertex-ID map for self-supervised pretraining and dense prediction heads.

src/routes/+page.svelte

@@ -125,7 +125,7 @@
 		{
 			icon: SpeakerHighIcon,
 			title: 'Spatial-audio conditioning',
-			body: 'Per-player HRTF audio with footsteps, gunfire, and callouts.'
+			body: 'Per-player stereo, mixed from each agent’s position and orientation in the world.'
 		},
 		{
 			icon: UsersThreeIcon,

src/routes/+page.ts

@@ -2,8 +2,11 @@ import type { PageLoad } from './$types';
 import { listMatches, listAllRounds } from '$lib/api/hf';
 
 export const load: PageLoad = async ({ fetch }) => {
+	// The dataset's index/ shards may be missing during a re-upload window.
+	// Tolerate that with empty arrays so the page renders an empty browser
+	// instead of crashing with a 500.
 	const [matches, rounds] = await Promise.all([
-		listMatches({ fetch }),
+		listMatches({ fetch }).catch(() => []),
 		listAllRounds({ fetch }).catch(() => [])
 	]);
 	return { matches, rounds };