Hello everyone! 👋

I want to share my latest project for the M5Stack Cardputer ADV. It’s called Wikipedia Offline ADV.

I wanted to see if it was possible to fit all of human knowledge into this tiny device and make it searchable without any Wi-Fi connection. The result is a hyper-optimized, standalone Wikipedia search engine wrapped in a retro-futuristic "ELIZA OS" terminal interface.

🚀 Key Features

• 6.4 Million Articles Offline: The entire database is packed into an 18.3 GB archive stored on your SD card.
• Lightning Fast: I built an O(log N) binary pointer search engine. It bypasses standard FAT32 limits by packing file IDs and local offsets into 64-bit pointers. Searches take mere milliseconds.
• Pseudo-AI Assistant: It’s not just a dumb search bar. The onboard NLP engine parses your intent (e.g., "Who is the president of..." or "What is the capital of..."). It uses a custom TF-IDF Lite algorithm to score sentences and extract the exact answer before printing the full 32KB article buffer.
• Fuzzy Levenshtein Radar: Typo on the tiny keyboard? No problem. The engine scans a 250-article sliding window on the SD card, calculates Levenshtein distance, and offers a paginated list of the closest matches.
• Zero-Flicker CRT Rendering: Custom Unified Canvas with hardware clipping for a smooth 60FPS CRT scanline effect.

🛠 How to Install

1. Download Wikipedia Offline ADV from M5Burner.
2. The Database: The firmware requires the database to work! Download the 6.54 GB .zip archive from the official GitHub release (hosted on Google Drive).
3. Extract the files and place them directly into the root directory of a FAT32-formatted MicroSD card (32GB or larger required). Do not put them in folders!
4. Insert the SD card and boot up.

🔗 GitHub Repository & Full Guide: Zeloksa/Wikipedia-Offline-ADV

If you like my work, you can support my late-night coding sessions here: Boosty ☕

Hello everyone! I've just released my new firmware for the Cardputer: IMU Acoustic Analyzer.

This project transforms your device into a professional-grade acoustic scleroscope, allowing you to measure the structural hardness and elasticity of real-world surfaces.

⚠️ Important: How it works (and how to drop it)
This is a precision physical instrument. It does not come with a pre-filled encyclopedia of materials-you train it yourself.

By standing the Cardputer on its edge (SD card slot facing UP) and letting it naturally tip over and fall flat onto a table or floor, the internal IMU records the micro-impact. It calculates the material's properties and saves the acoustic profile. Later, it can instantly identify which surface you dropped it on.

The Drop Technique: Let it fall naturally. Do not apply downward force. Even a 1-degree tilt upon impact or hidden supports under a table will alter the acoustic stiffness!

Durability: Do not be afraid to use it. During development, I dropped my Cardputer over 1000 times using this exact edge-drop method (including onto hard tiles). Because it simply tips over from a standing position, the device easily survives without any damage.

✨ Technical Features
4-Dimensional DSP Engine: Polling the BMI270 at 1600Hz, the custom DSP doesn't just look at peak impact. It calculates MaxG (hardness), Ratio (elastic rebound), flight time, and AUC (Area Under Curve) to measure the true physical impulse and viscosity of the impact.

"Zero-Shot" Physical Insight: Even if a surface is not saved in your database, the analyzer evaluates the raw physics and gives you an instant diagnostic (e.g., Ultra-Hard | Elastic or Firm | Dampening).

🕹️ Controls
[ 1 ]: Start 3-Drop Measurement.

[ 2 ]: Add New Material (5-Drop Calibration).

[ 3 ]: Open Database List.

[ < / , ] & [ > / / ]: Navigate Database Pages (Left / Right).

[ 1 - 4 ]: Delete a specific material from the current Database page.

[ ESC / ` ]: Global Cancel / Return to Main Menu.

[ - / _ ] & [ = / + ]: Decrease / Increase system volume (emits an acoustic test click).

📥 Download
You can flash it right now via M5Burner - just search for IMU Acoustic Analyzer or Zeloksa.

You can also find the .bin release on GitHub:
🔗 https://github.com/Zeloksa/IMU-Acoustic-Analyzer

I would love to hear your feedback. Let me know what materials you scan and how your calibration profiles look!

Hello everyone! I've just released my new firmware for the Cardputer: AI Offline Chat ADV.

This project turns your device into a fully autonomous, offline virtual therapist.

⚠️ Important: What this is (and isn't)

This is NOT ChatGPT or an LLM. It does not know facts, cannot write code, and does not connect to the internet.

Instead, it is a highly optimized C++ port of the legendary 1966 ELIZA program, running entirely on the ESP32-S3's SRAM. She acts as a Rogerian psychotherapist. You type out your feelings, thoughts, or problems in English (e.g., "I am feeling anxious today" or "My mother never listens to me"), and she uses a custom NLP engine with JIT-compiled RegEx to analyze your grammar, invert pronouns, and ask deep follow-up questions.

The magic happens when you engage with her as if she were a real listener.

✨ Technical Features

• 100% Offline NLP: Heavy RegEx parsing done locally. Includes a custom sanitizer to prevent RTOS Stack Overflows during deep recursive sentence analysis.
• Persistent SD Memory: She remembers. With an SD card, it saves your name, mentioned entities (family/objects), and the emotional context of your session. If you talk about grief, the UI adapts, and she will check on your feelings after a reboot.

🕹️ Controls

• [ Full Keyboard ]: Type your thoughts.
• [ ENTER ]: Send message.
• [ DEL ]: Backspace.
• [ ESC / ` ]: Open System Menu (Start New Session / Wipe SD Memory).
• [ ; / . ]: Scroll chat history.

📥 Download

You can flash it right now via M5Burner - just search for AI Offline Chat ADV or Zeloksa.

You can also find the .bin release on GitHub:
🔗 https://github.com/Zeloksa/AI-Offline-Chat-ADV

I would love to hear your feedback and see how your conversations with her go!

Hi everyone!

I want to share my new open-source project for the Cardputer: WiFi Remote Display ADV (V1.0).
It turns your Cardputer into an ultra-low latency wireless mini-monitor for your Windows PC.

How it works:
It uses the Cardputer as a USB HID device to seamlessly inject a Python-based UDP streaming script directly into your PC's PowerShell. It then receives the high-speed adaptive desktop video over your local Wi-Fi network.

Key Features:

• 🔒 100% Open Source & Safe: The Cardputer types the script visibly in PowerShell and pauses at the end. It will NOT execute until you manually press ENTER on your PC, so you have full control.
• ⚡ True Adaptive Quality: Dynamically downgrades PC-side JPEG compression if lag is detected to maintain smooth FPS.
• 🎮 Interactive Controls: Live Pan & Zoom using the Cardputer keyboard.
• 💾 NVS Memory: Saves your Wi-Fi credentials autonomously (no SD card needed).

[WARNING]
CRITICAL REQUIREMENTS (Must Read):

1. Python MUST be installed on the target Windows PC and added to the system PATH. The streaming engine relies on it.
2. PC and Cardputer must be on the SAME Wi-Fi network.
3. The target PC keyboard layout MUST be set to English during the USB injection.

Where to get it:

• Download via M5Burner: Search for WiFi Remote Display ADV or Zeloksa.
• Source code and .bin files on GitHub: https://github.com/Zeloksa/WiFi-Remote-Display-ADV

I'd love to hear your feedback or answer any questions. Enjoy! 🚀

Hello M5Stack Community! 📡

I am excited to announce a massive V2.0 update to my tracking tool for the Cardputer. What started as a high-speed Wi-Fi scanner is now the Wi-Fi & BLE Radar ADV (V2.0).

This update transforms the Cardputer into a full-fledged dual-band tactical direction-finding tool. I've added a completely independent Bluetooth Low Energy (BLE) engine, asynchronous background scanning, and cyber-security analytics to bypass modern MAC randomization.

⚡ V2.0 Technical Highlights:

• Dual-Band Radar (Wi-Fi & BLE): Independent tracking engines with strict antenna sharing logic to prevent radio coexistence interference.
• Anti-Tracking Bypass & OUI DB: Modern phones use randomized MAC addresses to hide. The V2.0 engine parses IEEE address types on the fly to expose randomized targets and uses an embedded OUI database to instantly resolve vendors (Apple, Xiaomi, Garmin, Huami, etc.) directly on the screen.
• Asynchronous TDM Scanning: The new Split-Screen Device List continuously multiplexes between Wi-Fi and BLE in the background (3-second intervals). It can track up to 1000 devices per protocol in real-time without freezing the UI.
• Visual Stability Engine: Added UI Throttling. While data is captured at maximum speed, visual sorting updates every 1.2s to prevent "text shaking" in highly crowded RF environments.
• Active Probing & Log-Normal Math: Still features the aggressive 40 FPS Wi-Fi probing from V1.5, combined with real-time signal attenuation math to estimate physical distance (in meters).
• Security Tags: Open Wi-Fi networks are now explicitly flagged with a yellow [OPEN] tag (turns red when targeted) so you can spot unsecured networks instantly.

⚠️ Important Hardware Limitation:
The ESP32-S3 chip inside the Cardputer supports BLE only, not Bluetooth Classic. It will easily track modern smartphones, smartwatches, and IoT beacons, but it will not see legacy devices like old headphones, car stereos, or PS3/PS4 controllers.

📥 How to Install:
Available right now in M5Burner:

1. Open M5Burner.
2. Search for "Zeloksa" or "WiFi & BLE Radar ADV".
3. Download version V2.0 and burn to your Cardputer.

📖 Manual & Feedback:
To get the most accurate direction-finding results (using the "Body Shield" technique), please read the full user manual on my updated GitHub repository. If you find bugs or have suggestions, feel free to open an issue:
👉 https://github.com/Zeloksa/Cardputer-ADV-WiFi-BLE-Radar

☕ Support Development:
If this tool is useful for your security audits, research, or hobby projects, consider supporting my work. It helps me create more advanced tools for the Cardputer!
👉 https://boosty.to/zeloksa

Let me know what you think of the new dual-band engine!

Hello M5Stack Community! ⚗️

I've just released a new project for the Cardputer: Alchemy ADV.

Remember the classic element-combining puzzle games? I brought that cozy, addictive gameplay to the Cardputer! You start with just 4 basic elements (Water, Fire, Earth, Air) and combine them to discover a whole world of items, ranging from nature and animals to technology and myths.

⚡ Features:

186 Elements: A massive library of unique combinations to unlock.

Dynamic Environment: Real-time Day/Night cycle with a fully animated background. Keep an eye out for random events like rockets, airships, and flying witches! 🚀🌙

Smart Hint System: Stuck? Press 'H' to highlight a valid, undiscovered combo using your existing items.

Auto-Save: Automatically saves your progress directly to your MicroSD card so you never lose your library.

📥 How to install:
Available now in M5Burner:

Open M5Burner.

Go to User Custom.

Search for "Zeloksa" or "Alchemy ADV".

🛠 Source & Support:
Check out the repository for updates, bug reports, and feedback:
👉 https://github.com/Zeloksa/Cardputer-Alchemy-ADV-Game

☕ Support my work:
If you enjoy this game and my other projects, feel free to support further development:
👉 https://boosty.to/zeloksa

Happy crafting! Let me know if you manage to unlock all 186 elements!

Hi everyone!

I just published a new utility for the Cardputer ADV. It's a highly optimized, state-machine-based flashlight that repurposes the TFT display as a multi-mode light source.

⚡ Key Features

• Smart Controls (Press ANY Key):
  • Single Click: Power ON/OFF. Features a Smart Memory system that remembers your last used color (White or Red).
  • Double Click: High-speed Strobe mode (50ms interval).
  • Hold (0.5s): Tactical Red Light mode to preserve night vision.
• Deep Battery Optimization: The ESP32-S3 CPU is automatically underclocked down to 80MHz. This guarantees stable I2C/SPI operation while drastically reducing battery consumption and preventing any heat generation during long-term use.
• Hardware-Level Display Control: When turned off, the firmware kills the TFT backlight brightness to 0 and fills the VRAM with TFT_BLACK to completely halt power draw from the screen matrix.

🧲 Practical Use Cases

• Hands-Free Work: Thanks to the built-in magnets inside the Cardputer case, you can attach the device to server racks, car hoods, or any metal surface to illuminate your workspace.
• Night Vision Preservation: The dedicated Red Light mode is perfect for astronomy, reading in the dark, or tactical environments where standard white light is too harsh.
• Emergency Beacon: The high-frequency Strobe mode can be used as a distress signal or a highly visible marker on the road.

🛠 How to install

• M5Burner: The firmware is proprietary. You can install the binary directly by opening M5Burner and searching for Flashlight ADV or Zeloksa.
• Project Page / Issue Tracker: https://github.com/Zeloksa/Cardputer-Flashlight-ADV

☕ Support the Project

If you find this utility useful for your EDC (Everyday Carry), consider supporting further development here: https://boosty.to/zeloksa

Hello M5Stack Community! 📡

I am excited to introduce my new tool for the Cardputer: Wi-Fi Radar ADV (V1.5).

Unlike standard passive scanners, this firmware is an advanced, high-speed Wi-Fi tracking and spatial direction-finding tool. It uses a custom Active Probing Engine and real-time log-normal attenuation math to not just list networks, but to physically track their location and estimate distance in meters.

⚡ Technical Highlights:

• Active Probing Engine (40 FPS): Instead of passively waiting for beacons, the radar aggressively sends Probe Requests directly to the target. This forces an instant response (even from Hidden networks) and drops channel scan time down to 20ms for lag-free tracking.
• Log-Normal Attenuation Math: Converts raw RSSI signal strength into an approximate physical distance (in meters).
• Mesh-Network Support: Intelligently separates physical hardware. If a building has 5 routers sharing the same SSID, the radar plots 5 distinct dots and locks onto the closest physical node.
• Target Lock & Sonar: Enter "Single Track" mode to focus 100% of the ESP32-S3's processing power on one specific target. Features a Geiger-counter style audio sonar for eyes-free tracking.
• CRT Retro-Futurism UI: Custom-built rendering engine with 3D perspective grids and hardware-accelerated scanlines for an authentic 80s tactical terminal look.

📥 How to Install:

Available right now in M5Burner:

1. Open M5Burner.
2. Go to the User Custom section.
3. Search for "Zeloksa" or "Wi-Fi Radar ADV" (Make sure to pick the Wi-Fi Radar, not the CSI one!).

📖 Manual & Feedback:

To get the most accurate direction-finding results (using the "Body Shield" technique), please read the full user manual on my GitHub. If you find bugs or have suggestions, feel free to open an issue:
👉 https://github.com/Zeloksa/Cardputer-ADV-WiFi-Radar

☕ Support Development:

If this tool is useful for your security audits, research, or hobby projects, consider supporting my work. It helps me create more advanced tools for the Cardputer!
👉 https://boosty.to/zeloksa

Let me know what you think of the new Active Probing engine!
Cardputer, WiFi, Radar, ESP32-S3, Tracking

Hello M5Stack Community! 👋

I’m back with a major update (V1.43) for the WiFi CSI Radar ADV. Since the last release, I’ve focused on stabilizing the 2D spatial math and pushing the packet capture rate to its absolute theoretical limit on the ESP32-S3.

This firmware turns your Cardputer into a professional-grade Channel State Information (CSI) sensing tool, capable of "seeing" motion through solid obstacles by analyzing radio wave distortions.

🚀 What's New in V1.43:
Enhanced 2D Spatial Profiles: Train the radar to recognize up to 21 unique spatial zones with improved accuracy.

UI Overhaul: New high-contrast HUD elements and real-time PPS (Packets Per Second) counter.

Stability Fixes: Improved ADC filtering for the battery sensor to prevent interference during heavy Wi-Fi injection.

Optimized Dual-Core Architecture: Core 0 now handles 100% of the radio stack, while Core 1 is dedicated to the math engine and UI rendering.

⚙️ Technical Specs:
Capture Rate: 200+ PPS (Physical limit of the 2.4GHz stack).

Detection Engine: Log-normal shadow fading model for distance estimation.

Monitoring: Built-in Web Server for remote data visualization and spatial mapping.

📥 Installation:
The updated firmware is available now in M5Burner:

Open M5Burner.

Search for "Zeloksa" or "WiFi CSI Radar" in the User Custom section.

Burn and enjoy!

🛠 Manual & Feedback:
For detailed calibration instructions and the full user manual, visit the official repository:
👉 https://github.com/Zeloksa/Cardputer-ADV-CSI-Radar

☕ Support the Research:
Development of these ADV tools takes a lot of time and RF testing. If you find this project useful for your security audits or research, consider supporting me here:
👉 https://boosty.to/zeloksa

Looking forward to your feedback and new PPS records! 📡

Tags: Cardputer, CSI, Radar, ESP32-S3, Wi-Fi Sensing

Hello M5Stack Community! 🎹

I've just released a new project for the Cardputer: Piano Tiles ADV.

It’s a high-performance rhythm game optimized specifically for the Cardputer's hardware. I focused on achieving zero audio-visual lag to make the gameplay smooth and challenging.

⚡ Features:

• Optimized Audio: Low-latency sound engine for real-time feedback.
• ADV Performance: Smooth 60 FPS rendering on the Cardputer screen.
• Increasing Difficulty: The game speeds up as you progress.
• Score Tracking: Keeps track of your best performance.

📥 How to install:

Available now in M5Burner:

1. Open M5Burner.
2. Go to User Custom.
3. Search for "Zeloksa" or "Piano Tiles".

🛠 Source & Support:

Check out the repository for updates and feedback:
👉 https://github.com/Zeloksa/Piano-Tiles-ADV

☕ Support my work:

If you enjoy this game and my other projects, feel free to support further development:
👉 https://boosty.to/zeloksa

Have fun and let me know your high scores!!

Hello M5Stack Community! 👋

What if your Cardputer could see through walls?

I’ve been working on pushing the ESP32-S3 to its absolute physical limits and I'm excited to share WiFi CSI Radar ADV (V1.42). This firmware turns your device into a high-performance Channel State Information (CSI) radar that analyzes invisible radio wave reflections to detect motion and spatial presence.

🔥 WHAT IT ACTUALLY DOES:

🛡️ 1D Invisible Tripwire: Ultra-precise motion detection. It senses subtle movements (like walking or even breathing) through solid doors and walls. Perfect for perimeter monitoring.

📍 2D Spatial Tracking (Active Beta): Tracks your physical position inside a room in real-time via a built-in Web UI by profiling Wi-Fi reflections in the corners of the room.

🗺️ Hybrid SLAM Room Mapping: Draw a digital floor plan of your environment simply by holding the device flat against your chest and walking around.

⚙️ UNDER THE HOOD (For the Geeks):
To make this work without lag, I had to bypass standard limitations:

200+ PPS Capture Rate: Achieved the physical limit of the 2.4GHz stack.

Dual-Core FreeRTOS Architecture: Core 0 handles aggressive radio/injection (NAT Bypass), while Core 1 crunches the heavy math and UI rendering.

Smart Power Management: Advanced ADC noise filtering to prevent false charging detections under heavy Wi-Fi load.

📥 HOW TO INSTALL:
The firmware is available directly in M5Burner.

Open M5Burner.

Go to the Cardputer -> User Custom (or Community) section.

Search for WiFi CSI Radar ADV or Zeloksa.

Burn and enjoy!

📖 MANUAL & FEEDBACK:
Wi-Fi sensing requires exact calibration (e.g., leaving the room to set a baseline). Please read the full User Manual before your first launch!
If you find bugs or have ideas for improving the 2D spatial math, drop an issue here:
👉 https://github.com/Zeloksa/Cardputer-ADV-CSI-Radar

☕ SUPPORT THE PROJECT:
If this tool helps with your research, pentesting, or hobbyist projects, consider supporting further development. It keeps the updates coming!
👉 https://boosty.to/zeloksa

Looking forward to seeing your test results and PPS records! Let me know how it performs on your local routers. 📡

Tags: Cardputer, CSI, Radar, ESP32-S3, Sensing, FreeRTOS