---
title: "DX-Ball 1.07: From 63 Bytes of x86 to Verified C"
description: "Following a brick-hit sound call into 0x00406400, reconciling decompiler blind spots with raw x86 instructions, and verifying 3,205 oracle cases."
canonical: "https://askgina-blog.pages.dev/posts/dx-ball-sound-pan"
---

# DX-Ball 1.07: From 63 Bytes of x86 to Verified C

[HTML version](https://askgina-blog.pages.dev/posts/dx-ball-sound-pan)

Published October 7, 2026 · REA Native Team

Following a brick-hit sound call into 0x00406400, reconciling decompiler blind spots with raw x86 instructions, and verifying 3,205 oracle cases.

**Abstract.** Decompilers frequently hide x87 stack-passing conventions and inline float constants. In DX-Ball 1.07, the sound-panning routine at `0x00406400` spans just 63 bytes of x86 (The initial pseudocode misses the stack input at \[EBP+0x8\]. The brick-hit caller at 0x411f40 passes 20 + 30 \* tile\_x. Source: [rea.tools/showcase/dx-ball](https://rea.tools/showcase/dx-ball/).), yet recovering its exact C implementation requires reconciling caller disassembly, raw memory reads, and 3,205 oracle comparisons. The published 7 October 2026 checkpoint reports 55 maintained C functions, 45,380 comparisons with original code, and 33 functions with matching compiled bytes.

## 63 Bytes at 0x00406400

The steps below follow the published REA 4.1.0 showcase (Byte reads established 1.5625 (0x420068), 500.0 (0x420070), and the initial pan\_scale of 1.0 (0x4210a0). Source: [rea.tools/showcase/dx-ball](https://rea.tools/showcase/dx-ball/).). Walk through the five steps from the stack input to the final `__ftol` truncation. The slider simulates the reconstructed C formula in JavaScript at a pan scale of 1.0. It does not run the x86 oracle.

Interactive figure: [DX-Ball Sound-Panning: x86 Assembly to Verified C](https://askgina-blog.pages.dev/posts/dx-ball-sound-pan#dx-disassembly).

## Oracle Verification & Investigation Trace

REA 4.1.0 supplied the instructions, caller information, and byte reads. The `N0zoM1z0/dx-ball` project compared the C and original x86 return values for all 3,205 cases: integer x from 0 to 640 at pan scales 0, 0.5, 1, 20, and −1. A separate VC4.0 compiler replay matched all 63 bytes (The replay matches the function at 0x406400–0x40643e after applying reviewed relocations and checking the referenced constants. Source: [rea.tools/showcase/dx-ball](https://rea.tools/showcase/dx-ball/).). The waterfall below uses a synthetic 5-span timeline to show the sequence of steps. Its 940 ms duration is not measured latency telemetry.

Interactive figure: [REA Target Surface & Investigation Span Waterfall](https://askgina-blog.pages.dev/posts/dx-ball-sound-pan#dx-verification).

## Notes & References

1.  Vercel Labs (2026). UI SDK agent docs, project settings screen benchmark (vercel-labs/ui, benchmarks/agent). [ui-sdk.dev/docs/agents](https://ui-sdk.dev/docs/agents)
2.  REA Project (2026). “Reverse Engineer Anything: One MCP for reverse engineering across binaries, applications, and runtime behavior (v6.8.0, 229 tools).” [github.com/morluto/rea](https://github.com/morluto/rea)
3.  REA Showcase. DX-Ball 1.07 project checkpoint, 7 October 2026 (REA 4.1.0). [rea.tools/showcase/dx-ball](https://rea.tools/showcase/dx-ball/)
4.  REA Showcase (2026). Notion Desktop 7.6.1 clipboard IPC trace (REA 4.1.0) and separate cached web-asset probes. [rea.tools/showcase/notion](https://rea.tools/showcase/notion/)
5.  Editorial design reference: SID-2 Technical Report (three-column reading grid, inline figure toggles, margin sidenotes). [sid.ai/research/sid-2-technical-report](https://www.sid.ai/research/sid-2-technical-report)
