System Requirements

MixModeler is a browser-based application that runs entirely in your web browser with no installation required. All data processing happens locally on your device, ensuring complete privacy and security.


Supported Browsers

Google Chrome (Version 90+)

  • ✅ Full feature support including GPU acceleration

  • ✅ Optimal performance with large datasets

  • ✅ Best WebAssembly (WASM) performance

  • Recommended for: Power users, large datasets, Bayesian modeling

Microsoft Edge (Version 90+)

  • ✅ Full feature support including GPU acceleration

  • ✅ Chromium-based, similar performance to Chrome

  • ✅ Excellent compatibility

  • Recommended for: Windows users, enterprise environments

Mozilla Firefox (Version 88+)

  • ✅ Full feature support

  • ✅ Good WASM performance

  • ✅ Strong privacy features

  • Recommended for: Privacy-focused users

Safari (Version 14+)

  • ✅ Core features fully supported

  • ⚠️ GPU acceleration limited on some Macs

  • ✅ Good performance on Apple Silicon (M1/M2/M3)

  • Recommended for: Mac users

Minimum Browser Requirements

  • JavaScript: Must be enabled

  • Cookies: Required for authentication only

  • Local Storage: Required for application state

  • WebAssembly (WASM): Required for computation

  • Modern ES6+ Support: Required

Not Supported:

  • ❌ Internet Explorer (all versions)

  • ❌ Opera Mini

  • ❌ Very old browser versions (pre-2020)


Hardware Requirements

Minimum Specifications

For Basic MMM Models (up to 50 variables, 100 observations):

  • CPU: Dual-core processor, 2.0 GHz

  • RAM: 4 GB

  • Storage: 500 MB free space (browser cache)

  • Display: 1280x720 resolution

  • Internet: Required for initial load only

Performance: Adequate for small models with OLS estimation

For Standard MMM Models (50-200 variables, 100-300 observations):

  • CPU: Quad-core processor, 2.5 GHz or better

  • RAM: 8 GB or more

  • Storage: 1 GB free space

  • Display: 1920x1080 resolution

  • Internet: Broadband for smooth experience

Performance: Good performance for typical MMM workflows

High-Performance Specifications

For Large MMM Models (200+ variables, 300+ observations, Bayesian models):

  • CPU: 6-core or better, 3.0 GHz+

  • RAM: 16 GB or more

  • GPU: Dedicated graphics card (NVIDIA/AMD) for GPU acceleration

  • Storage: 2 GB free space

  • Display: 2560x1440 or higher

  • Internet: Broadband connection

Performance: Excellent performance, full GPU acceleration benefits


Operating System Compatibility

Fully Supported

Windows

  • Windows 10 (64-bit)

  • Windows 11 (64-bit)

  • Note: Windows 7/8 not recommended due to outdated browser support

macOS

  • macOS 10.15 (Catalina) or newer

  • macOS 11+ (Big Sur and later) recommended

  • Full Apple Silicon (M1/M2/M3) support

  • Note: GPU acceleration available on Apple Silicon

Linux

  • Ubuntu 20.04 LTS or newer

  • Fedora 34 or newer

  • Debian 11 or newer

  • Other distributions with modern browsers

  • Note: Best performance with recent kernel versions

ChromeOS

  • ChromeOS 90 or newer

  • Works well on modern Chromebooks

  • Limitation: GPU acceleration may be limited


Network Requirements

Internet Connection

During Initial Load:

  • Required to load application

  • Download size: ~5-10 MB (first visit)

  • Subsequent visits use browser cache

During Use:

  • NOT required for data processing

  • All computation happens locally

  • Your data never leaves your device

For Updates:

  • Automatic updates when you reload the page

  • No manual installation needed

Bandwidth Recommendations

  • Minimum: 1 Mbps (initial load)

  • Recommended: 5+ Mbps (smooth experience)

  • Note: After loading, internet is not needed

Firewall & Security

  • No special ports required

  • Works through standard HTTPS (port 443)

  • Compatible with corporate firewalls

  • No VPN issues (data stays local)


Data & File Limitations

Subscription Plan Limits

Free Plan

  • Maximum 20 variables per dataset

  • Suitable for small businesses and testing

  • All features available

Professional Plan

  • Maximum 500 variables per dataset

  • Suitable for mid-size businesses

  • Full feature access

Business Plan

  • Unlimited variables

  • Suitable for enterprises

  • Premium support

File Size Limits

Excel/CSV Upload

  • Maximum file size: 50 MB

  • Typical datasets: 1-5 MB

  • Larger files process more slowly

Observations (Time Periods)

  • Minimum: 26 weeks (6 months)

  • Recommended: 52+ weeks (1+ years)

  • Maximum: 500+ weeks supported

Processing Time

  • Small models (50 vars): Seconds

  • Medium models (200 vars): 10-30 seconds

  • Large models (500 vars): 1-3 minutes

  • Bayesian MCMC: 2-10 minutes depending on settings


Performance & Acceleration

WebAssembly (WASM) Acceleration

Automatic Acceleration:

  • All modern browsers support WASM

  • 5-10x faster than JavaScript

  • No configuration needed

  • Always available

GPU Acceleration (Optional)

Availability:

  • Requires dedicated GPU (NVIDIA, AMD, or Apple Silicon)

  • Chrome/Edge provide best GPU support

  • Automatically enabled if available

  • Provides 10-100x speedup for large models

GPU Requirements:

  • NVIDIA: GTX 10-series or newer

  • AMD: RX 5000-series or newer

  • Apple: M1/M2/M3 chips

  • Intel: Integrated graphics supported (limited acceleration)

When GPU Helps Most:

  • Large datasets (200+ variables)

  • Bayesian modeling with MCMC

  • Correlation matrix calculations

  • Batch variable testing

Performance Indicators

Acceleration Badge:

  • Green indicator shows active acceleration

  • Click to see which method is active (WASM/GPU)

  • Performance metrics displayed after operations


Browser Settings

Required Settings

JavaScript

  • Must be enabled

  • Required for application to function

Cookies

  • Used for authentication only

  • First-party cookies required

  • No third-party tracking cookies

Local Storage

  • Required for application state

  • Stores preferences and settings

  • No sensitive data stored

Pop-ups

  • May need to allow for downloads

  • Excel export uses browser downloads

Optional Settings

GPU Acceleration

  • Chrome: chrome://flags/#enable-webgl

  • Should be enabled by default

  • Restart browser if changed

Memory Limits

  • Modern browsers handle automatically

  • Chrome allows ~2GB per tab

  • Upgrade RAM for very large models


Data Privacy & Security

On-Device Processing

Your Data Never Leaves Your Computer:

  • All computation in browser

  • No server uploads

  • No cloud storage

  • Complete data isolation

What Gets Sent to Servers:

  • ONLY authentication tokens (login)

  • NO marketing data

  • NO model results

  • NO personal information

Security Features

HTTPS Encryption

  • All connections encrypted

  • Secure authentication

  • No data interception risk

No Data Storage

  • We don't store your data

  • We can't access your data

  • Complete privacy compliance

GDPR & CCPA Compliant

  • Privacy by design

  • No personal data collection

  • Full user control


Troubleshooting Performance Issues

Slow Performance

If models are running slowly:

  1. Close unnecessary browser tabs

    • Each tab uses memory

    • Close tabs to free resources

  2. Clear browser cache

    • Chrome: Settings > Privacy > Clear browsing data

    • Reload MixModeler after clearing

  3. Check system resources

    • Task Manager (Windows) or Activity Monitor (Mac)

    • Close memory-intensive applications

  4. Reduce model complexity

    • Start with fewer variables

    • Use OLS before Bayesian

    • Test on subset of data

  5. Use recommended browser

    • Chrome/Edge provide best performance

    • Update to latest browser version

Memory Issues

"Out of Memory" errors:

  1. Reduce dataset size

    • Fewer variables or observations

    • Process in batches

  2. Increase available RAM

    • Close other applications

    • Restart browser

    • Upgrade computer memory

  3. Use 64-bit browser

    • Allows more memory usage

    • Check browser version


Mobile & Tablet Support

Current Status

Limited Support:

  • ⚠️ Viewing models possible

  • ⚠️ Basic interactions supported

  • ❌ Model building not recommended

  • ❌ File uploads may have issues

Why Desktop is Recommended:

  • Complex interface designed for desktop

  • Large screens needed for charts and tables

  • Mobile browsers have memory limitations

  • Touch interactions less efficient

Future Plans

  • Improved tablet support coming

  • Mobile viewing optimization planned

  • Simplified mobile workflows under development


Enterprise Considerations

IT Requirements

No IT Department Burden:

  • No software installation

  • No server setup

  • No firewall changes

  • No VPN requirements

Works With:

  • Corporate firewalls

  • Proxy servers

  • VPNs

  • SSO systems (coming soon)

Compliance

Data Residency:

  • Data never leaves user's device

  • No data center concerns

  • Complies with local regulations

Audit & Security:

  • No data exposure risk

  • Audit-friendly architecture

  • Enterprise-grade privacy


Support & Updates

Automatic Updates

  • Updates deployed automatically

  • Refresh browser to get latest version

  • No manual installation

  • Backward compatible

Browser Compatibility Testing

We continuously test on:

  • Latest 3 versions of Chrome

  • Latest 3 versions of Edge

  • Latest 3 versions of Firefox

  • Latest 2 versions of Safari

Getting Help

Performance Issues:

Feature Requests:

  • Submit via feedback form in app

  • Roadmap available on website


Optimal Configuration for MMM

Hardware:

  • 16 GB RAM or more

  • Quad-core CPU (3.0 GHz+)

  • Dedicated GPU (optional but beneficial)

  • SSD for faster browser cache access

Software:

  • Windows 10/11 or macOS 11+

  • Google Chrome (latest version)

  • Browser cache enabled

  • JavaScript and WASM enabled

Network:

  • Broadband internet (5+ Mbps)

  • Stable connection for initial load

  • No connection needed during analysis

Display:

  • 1920x1080 or higher resolution

  • Dual monitors recommended (optional)

  • Color-calibrated for accurate charts

This setup ensures optimal performance for all MixModeler features, including large models and Bayesian inference.

Last updated