Professional floating-point conversion and analysis tool
- Download APK: Click the button above or download from Releases
- Enable Unknown Sources: Settings → Security → Unknown Sources (if needed)
- Install: Open the downloaded APK file and follow the prompts
- Launch: Find "Cool FP Converter" in your app drawer
💡 Tip: The app works completely offline once installed - perfect for engineering work anywhere!
A precision-engineered tool for converting between binary, hexadecimal, and decimal representations across multiple floating-point formats. Built for engineers, researchers, and students who need accurate floating-point analysis and IEEE 754 visualization.
- Real-time conversion between Binary ↔ Decimal ↔ Hexadecimal
- 6 floating-point formats supported:
- FP32 (IEEE 754 Single) - 32-bit standard precision
- FP64 (IEEE 754 Double) - 64-bit double precision
- FP16 (IEEE 754 Half) - 16-bit half precision
- BF16 (Brain Float) - 16-bit Google Brain format
- FP8 (E4M3) - 8-bit emerging ML format
- NVFP4 (E2M1) - 4-bit NVIDIA experimental format
- IEEE 754 compliant conversions with full standard support
- Special value handling: NaN, ±Infinity, ±Zero, subnormal numbers
- Bit-level accuracy with proper rounding and overflow handling
- Input validation with clear error messaging
- Interactive bit breakdown showing sign, exponent, and mantissa
- Format specifications with bias, range, and precision details
- Real-time updates as you type in any input field
- Copy-to-clipboard functionality for easy sharing
- Progressive Web App with offline capabilities
- Android APK available for direct installation
- Responsive design optimized for mobile and desktop
- Touch-friendly interface with accessibility support
Clean, professional interface optimized for desktop workflows
Native Android app with intuitive touch interface and bit visualization
Live demonstration of real-time floating-point conversion
Educational visualization showing sign, exponent, and mantissa components
Get started in 30 seconds!
# Method 1: Direct download
wget https://github.com/phoenixjyb/floating-point-conve/raw/main/android/app/build/outputs/apk/release/app-release.apk
# Method 2: Install via ADB (for developers)
adb install app-release.apk✅ Features included:
- ✨ Offline capable - works without internet
- 🔋 Battery optimized - minimal resource usage
- 📱 Touch-friendly - designed for mobile use
- 🚀 Fast startup - launches in under 2 seconds
- 💾 Small footprint - only 4.4MB installed
Instant access - works on any device with a modern browser
- ✅ No Installation - runs instantly in your browser
- ✅ Cross-Platform - works on Windows, Mac, Linux, iOS, Android
- ✅ Always Updated - automatically gets latest features
- ✅ PWA Support - can be installed as desktop/mobile app
- ✅ Shareable - send links to colleagues instantly
- ✅ Responsive Design - optimized for phones, tablets, and desktops
- 🌍 Online Use: Open Web App - No download required
- 💻 Install as App: Click "Install" when prompted (PWA support)
- 🔗 Bookmark: Add to favorites for instant access
- 📱 Add to Home Screen: iOS/Android - use "Add to Home Screen" option
Deploy your own instance in minutes:
# Quick deployment
git clone https://github.com/phoenixjyb/floating-point-conve.git
cd floating-point-conve
npm install && npm run build
# Upload 'dist/' folder to any web server
# Works with: Netlify, Vercel, GitHub Pages, Apache, Nginx, etc.🚀 Easy Deploy Services:
- GitHub Pages: Automatic deployment on push to main
- Netlify:
netlify deploy --prod --dir=dist - Vercel:
vercel --prod - Firebase:
firebase deploy
# Clone the repository
git clone https://github.com/phoenixjyb/floating-point-conve.git
cd floating-point-conve
# Install dependencies
npm install
# Start development server
npm run dev- Android 7.0+ (API level 24+)
- 4.4MB storage space (installed size)
- ARM64/ARM32/x86_64 architecture support
- No additional permissions required
- Offline functionality - works without internet
- Modern browsers with ES2020 support
- JavaScript enabled
- Local storage for PWA functionality
- Node.js 18+
- npm 8+ or yarn 1.22+
- Android Studio (for Android builds)
- Java 17+ (for Android compilation)
├── React 18 + TypeScript # Core framework
├── Vite # Build tool and dev server
├── Tailwind CSS # Utility-first styling
├── Radix UI # Accessible component primitives
├── Capacitor # Cross-platform native runtime
└── PWA # Progressive web app features
src/
├── components/ # UI Components
│ ├── ui/ # Reusable UI primitives
│ ├── BitVisualization.tsx # Floating-point bit breakdown
│ ├── ConversionCard.tsx # Main conversion interface
│ ├── FormatSelector.tsx # Format switching component
│ ├── ValueRange.tsx # Format range display
│ └── PWAInstall.tsx # PWA installation prompt
├── lib/
│ ├── floatingPoint.ts # Core conversion algorithms
│ └── utils.ts # Utility functions
├── styles/ # Global styling
└── assets/ # Static resources
android/ # Android platform files
├── app/ # Android app configuration
├── gradle/ # Build system
└── capacitor.config.json # Capacitor configuration
public/ # PWA assets
├── manifest.json # Web app manifest
├── sw.js # Service worker
└── icon.svg # App icon
// Convert decimal to IEEE 754 binary representation
function decimalToBinary(decimal: number, format: FloatFormat): string
// Convert binary representation to decimal value
function binaryToDecimal(binary: string, format: FloatFormat): number
// Extract sign, exponent, and mantissa components
function getBitBreakdown(binary: string, format: FloatFormat): BitBreakdowninterface FloatFormat {
name: string; // Human-readable format name
totalBits: number; // Total bit width
signBits: number; // Sign bit count (always 1)
exponentBits: number; // Exponent field width
mantissaBits: number; // Mantissa/fraction field width
bias: number; // Exponent bias value
description: string; // Format description
range: { // Numerical range limits
min: number; // Most negative finite value
max: number; // Most positive finite value
minNormal: number; // Smallest normal number
maxNormal: number; // Largest normal number
smallestPositive: number; // Smallest positive value
precision: number; // Decimal digits of precision
};
}| Format | Bits | Sign | Exponent | Mantissa | Bias | Precision | Range |
|---|---|---|---|---|---|---|---|
| FP32 | 32 | 1 | 8 | 23 | 127 | ~7 digits | ±3.40×10³⁸ |
| FP64 | 64 | 1 | 11 | 52 | 1023 | ~15 digits | ±1.80×10³⁰⁸ |
| FP16 | 16 | 1 | 5 | 10 | 15 | ~3 digits | ±65,504 |
| BF16 | 16 | 1 | 8 | 7 | 127 | ~2 digits | ±3.39×10³⁸ |
| FP8 | 8 | 1 | 4 | 3 | 7 | ~1 digit | ±448 |
| NVFP4 | 4 | 1 | 2 | 1 | 1 | ~1 digit | ±6 |
- ±Zero: Positive and negative zero representations
- ±Infinity: Overflow handling for each format
- NaN (Not a Number): Invalid operation results
- Subnormal numbers: Gradual underflow near zero
- Denormalized values: Extended precision for tiny numbers
// Convert decimal 3.14159 to FP32 binary
decimalToBinary(3.14159, FLOAT_FORMATS.fp32)
// Returns: "01000000010010010000111111010000"
// Convert binary back to decimal
binaryToDecimal("01000000010010010000111111010000", FLOAT_FORMATS.fp32)
// Returns: 3.141590118408203// Analyze bit structure
getBitBreakdown("01000000010010010000111111010000", FLOAT_FORMATS.fp32)
// Returns: {
// sign: "0", // Positive number
// exponent: "10000000", // Biased exponent = 128
// mantissa: "10010010000111111010000", // Fractional part
// signValue: 0,
// exponentValue: 128, // Actual exponent = 128 - 127 = 1
// mantissaValue: 4788186
// }// Same value in different formats
const value = 1.5;
decimalToBinary(value, FLOAT_FORMATS.fp32) // 32-bit representation
decimalToBinary(value, FLOAT_FORMATS.fp16) // 16-bit representation
decimalToBinary(value, FLOAT_FORMATS.bf16) // Brain float representation# Install Node.js 18+
curl -fsSL https://deb.nodesource.com/setup_18.x | sudo -E bash -
sudo apt-get install -y nodejs
# Install Android Studio (for mobile builds)
# Download from: https://developer.android.com/studio# Clone and install
git clone https://github.com/phoenixjyb/floating-point-conve.git
cd floating-point-conve
npm install
# Configure environment
cp .env.example .env.local# Development
npm run dev # Start dev server (http://localhost:5173)
npm run build # Production build
npm run preview # Preview production build
npm run lint # Run ESLint
# Mobile Development
npm run build:android # Build and open Android Studio
npx cap add android # Add Android platform
npx cap sync android # Sync web assets to Android
npx cap run android # Run on Android device/emulator
# Deployment
npm run build # Build web app
cd android && ./gradlew assembleRelease # Build APK
cd android && ./gradlew bundleRelease # Build AAB for Play Store# Unit tests
npm run test
# E2E tests
npm run test:e2e
# Conversion accuracy tests
npm run test:precision# Build for production
npm run build
# Deploy to static hosting (Vercel, Netlify, etc.)
npm run deploy# Generate release APK
cd android
./gradlew assembleRelease
# APK location: android/app/build/outputs/apk/release/app-release.apk
# Generate AAB for Play Store
./gradlew bundleRelease
# AAB location: android/app/build/outputs/bundle/release/app-release.aabThe app automatically prompts for installation when accessed via HTTPS on supported devices.
"Install blocked" or "Unknown sources"
- Go to Settings → Security → Unknown Sources
- Enable "Allow installation of apps from unknown sources"
- Try installing the APK again
"App not installed" error
- Make sure you have 4.4MB+ free space
- Uninstall any previous versions
- Download the APK again (file might be corrupted)
Can't find the downloaded APK
- Check your Downloads folder
- Use a file manager app to locate
app-release.apk - Tap the file to install
App crashes on startup
- Make sure your device runs Android 7.0+
- Restart your device and try again
- Clear the app data: Settings → Apps → Cool FP Converter → Storage → Clear Data
Need help? Open an issue or email [email protected]
We welcome contributions! Please see our Contributing Guide for details.
- Fork the repository
- Create a feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
- TypeScript strict mode enabled
- ESLint configuration with React hooks rules
- Prettier for code formatting
- Conventional commits for commit messages
This project is licensed under the MIT License - see the LICENSE file for details.
- IEEE 754 Standard for Binary Floating-Point Arithmetic
- Google Brain team for BF16 format specification
- NVIDIA for FP8 and experimental format research
- React and TypeScript communities for excellent tooling
- Radix UI for accessible component primitives
- Issues: GitHub Issues
- Discussions: GitHub Discussions
- Email: [email protected]
