Skip to content

Commit 356a03b

Browse files
author
FastJava Team
committed
docs: Add comprehensive hyperlinks to technical terms across all sections
1 parent 9f509d7 commit 356a03b

1 file changed

Lines changed: 33 additions & 33 deletions

File tree

‎MAP.md‎

Lines changed: 33 additions & 33 deletions
Original file line numberDiff line numberDiff line change
@@ -10,7 +10,7 @@ FastJava is **minimal, deterministic, zero-bullshit** — built for bots, automa
1010

1111
## 🚀 Features
1212

13-
- **Native Speed** — DirectX, Win32, DWM, IOCP, SIMD, CUDA
13+
- **Native Speed** — [DirectX](https://learn.microsoft.com/en-us/windows/win32/directx), [Win32](https://learn.microsoft.com/en-us/windows/win32/apiindex/windows-api-list), [DWM](https://learn.microsoft.com/en-us/windows/win32/dwm/dwm-overview), [IOCP](https://learn.microsoft.com/en-us/windows/win32/fileio/i-o-completion-ports), [SIMD](https://en.wikipedia.org/wiki/SIMD), [CUDA](https://developer.nvidia.com/cuda-zone)
1414
- **Zero-Copy Everywhere** — Screen, Images, IPC, GPU
1515
- **Unified JNI Core** — FastCore as foundation
1616
- **Agent-Friendly** — FastAI, FastToolBridge, FastContext
@@ -69,52 +69,52 @@ FastJava, FastCore, FastPlugin
6969

7070
| Module | Explanation |
7171
|--------|-------------|
72-
| **FastWindow** | Java cannot control foreign windows. FastWindow finds, focuses, moves windows via Win32 - important for multi-window bots. |
73-
| **FastProcess** | ProcessHandle is too limited. FastProcess reads thread IDs, handle count, real CPU affinity for process isolation. |
74-
| **FastTheme** | Java doesn't know Dark Mode. FastTheme reads Windows 11 theming (Dark/Light, Accent, Mica) and adapts Java apps. |
75-
| **FastOverlay** | HUDs over games need transparent overlay windows. FastOverlay uses DirectX for ESPs, debug visuals. |
72+
| **FastWindow** | Java cannot control foreign windows. FastWindow finds, focuses, moves windows via [Win32](https://learn.microsoft.com/en-us/windows/win32/apiindex/windows-api-list) - important for multi-window bots. |
73+
| **FastProcess** | ProcessHandle is too limited. FastProcess reads thread IDs, handle count, real [CPU affinity](https://learn.microsoft.com/en-us/windows/win32/procthread/setting-thread-affinity) for process isolation. |
74+
| **FastTheme** | Java doesn't know Dark Mode. FastTheme reads [Windows 11 theming](https://learn.microsoft.com/en-us/windows/apps/desktop/modernize/apply-windows-themes) (Dark/Light, Accent, Mica) and adapts Java apps. |
75+
| **FastOverlay** | HUDs over games need transparent overlay windows. FastOverlay uses [DirectX](https://learn.microsoft.com/en-us/windows/win32/directx) for ESPs, debug visuals. |
7676
| **FastWindowEvents** | Java gets no events when windows are moved. FastWindowEvents notifies when the target window changes. |
77-
| **FastSystemMetrics** | Mouse speed and drag threshold are only accessible via Win32. Important for human-like bot input. |
78-
| **FastDWM** | VSync and frame latency are important for frame-locked rendering. FastDWM reads Desktop Window Manager timing. |
79-
| **FastFileWatch** | Java's WatchService is slow and loses events. FastFileWatch uses ReadDirectoryChangesW for immediate notifications. |
77+
| **FastSystemMetrics** | Mouse speed and drag threshold are only accessible via [Win32](https://learn.microsoft.com/en-us/windows/win32/apiindex/windows-api-list). Important for human-like bot input. |
78+
| **FastDWM** | VSync and frame latency are important for frame-locked rendering. FastDWM reads [Desktop Window Manager](https://learn.microsoft.com/en-us/windows/win32/dwm/dwm-overview) timing. |
79+
| **FastFileWatch** | Java's WatchService is slow and loses events. FastFileWatch uses [ReadDirectoryChangesW](https://learn.microsoft.com/en-us/windows/win32/api/winbase/nf-winbase-readdirectorychangesw) for immediate notifications. |
8080
| **FastProcessWatch** | Watch process start/stop for trigger bots (e.g., "when game starts, activate bot"). |
8181

8282
### Display & Graphics
8383

8484
| Module | Explanation |
8585
|--------|-------------|
86-
| **FastGraphics** | Java2D is unsuitable for 60+ FPS. FastGraphics uses DirectX/Vulkan for GPU rendering without JVM heap. |
86+
| **FastGraphics** | Java2D is unsuitable for 60+ FPS. FastGraphics uses [DirectX](https://learn.microsoft.com/en-us/windows/win32/directx)/[Vulkan](https://www.vulkan.org/) for GPU rendering without JVM heap. |
8787
| **FastImage** | BufferedImage allocates 200-300MB heap. FastImage uses ByteBuffer off-heap for fast pixel operations. |
8888
| **FastImageView** | JFrame with image is slow. FastImageView renders 1:1 pixels in 200ms startup time for debugging. |
89-
| **FastDisplay** | Displays framebuffer directly without copy. Foundation for all GPU rendering modules. |
90-
| **FastColorSearch** | Pixel loops in Java are 100x too slow. FastColorSearch uses SIMD (SSE/AVX) for 10GB/s pattern matching. |
89+
| **FastDisplay** | Displays [framebuffer](https://en.wikipedia.org/wiki/Framebuffer) directly without copy. Foundation for all GPU rendering modules. |
90+
| **FastColorSearch** | Pixel loops in Java are 100x too slow. FastColorSearch uses [SIMD](https://en.wikipedia.org/wiki/SIMD) ([SSE](https://en.wikipedia.org/wiki/Streaming_SIMD_Extensions)/[AVX](https://en.wikipedia.org/wiki/Advanced_Vector_Extensions)) for 10GB/s pattern matching. |
9191

9292
### Data & I/O
9393

9494
| Module | Explanation |
9595
|--------|-------------|
96-
| **FastIO** | Java NIO has too much overhead for real-time. FastIO uses unbuffered I/O and IOCP for constant latency. |
97-
| **FastMemoryScan** | For modding and reverse engineering: reads foreign process memory for pattern scans and pointer chains. |
98-
| **FastGPUCopy** | GPU↔CPU transfers are the bottleneck in ML pipelines. FastGPUCopy uses DMA for zero-copy. |
99-
| **FastIPC** | Shared memory and named pipes for separation of bot engine and AI model - both run isolated. |
96+
| **FastIO** | Java NIO has too much overhead for real-time. FastIO uses unbuffered I/O and [IOCP](https://learn.microsoft.com/en-us/windows/win32/fileio/i-o-completion-ports) for constant latency. |
97+
| **FastMemoryScan** | For modding and reverse engineering: reads foreign process memory for [pattern scans](https://guidedhacking.com/threads/signature-scanning-tutorial.12132/) and [pointer chains](https://guidedhacking.com/threads/pointer-scanning-tutorial.12170/). |
98+
| **FastGPUCopy** | GPU↔CPU transfers are the bottleneck in ML pipelines. FastGPUCopy uses [DMA](https://en.wikipedia.org/wiki/Direct_memory_access) for [zero-copy](https://en.wikipedia.org/wiki/Zero-copy). |
99+
| **FastIPC** | [Shared memory](https://en.wikipedia.org/wiki/Shared_memory) and [named pipes](https://learn.microsoft.com/en-us/windows/win32/ipc/named-pipes) for separation of bot engine and AI model - both run isolated. |
100100
| **FastClipboard** | Java's Clipboard is buggy. FastClipboard uses native APIs for stable copy/paste. |
101101

102102
### Audio
103103

104104
| Module | Explanation |
105105
|--------|-------------|
106-
| **FastAudioCapture** | WASAPI instead of Java Sound for <10ms latency. Loopback capture for audio trigger bots. |
106+
| **FastAudioCapture** | [WASAPI](https://learn.microsoft.com/en-us/windows/win32/coreaudio/wasapi) instead of Java Sound for <10ms latency. Loopback capture for audio trigger bots. |
107107
| **FastAudio** | Audio output for feedback/voice output. Counterpart to FastAudioCapture. |
108-
| **FastTTS** | Text-to-Speech for voice output. Piper (local), Coqui, ElevenLabs API, System TTS. For voice bots and accessibility. |
108+
| **FastTTS** | Text-to-Speech for voice output. [Piper](https://github.com/rhasspy/piper) (local), [Coqui](https://github.com/coqui-ai/TTS), [ElevenLabs API](https://elevenlabs.io/docs/api-reference), System TTS. For voice bots and accessibility. |
109109
| **FastSTT** | Speech-to-Text for speech recognition. Whisper (local), Vosk, System STT. For voice commands and transcription. |
110-
| **FastOCR** | Read text in screenshots. Tesseract (local), ONNX Runtime (neural OCR), System OCR. For quest logs, chat messages, UI elements, forms. |
110+
| **FastOCR** | Read text in screenshots. [Tesseract](https://github.com/tesseract-ocr/tesseract) (local), [ONNX Runtime](https://onnxruntime.ai/) (neural OCR), System OCR. For quest logs, chat messages, UI elements, forms. |
111111

112112
---
113113

114114
### FastTTS (Details)
115115

116116
**APIs:** `speak(text)`, `speak(text, voice)`, `setVoice(id)`, `setRate(speed)`, `setPitch(value)`, `saveToFile(text, path)`, `getVoices()`
117-
**Native:** Windows SAPI, macOS NSSpeechSynthesizer, Linux Speech Dispatcher, ONNX Runtime, ggml-TTS (Piper, Coqui)
117+
**Native:** Windows SAPI, macOS NSSpeechSynthesizer, Linux Speech Dispatcher, [ONNX Runtime](https://onnxruntime.ai/), ggml-TTS ([Piper](https://github.com/rhasspy/piper), [Coqui](https://github.com/coqui-ai/TTS))
118118
**Features:** Low-latency streaming, Voice selection, Local/offline TTS, File output (WAV/MP3)
119119
**Use-Cases:** Voice-Bots, Screenreader, Agent-Feedback, Audio-UI
120120

@@ -123,7 +123,7 @@ FastJava, FastCore, FastPlugin
123123
### FastSTT (Details)
124124

125125
**APIs:** `transcribe(audioBuffer)`, `transcribeFile(path)`, `startStreaming(callback)`, `stopStreaming()`, `setModel(model)`, `setLanguage(lang)`
126-
**Native:** Whisper.cpp (ggml), ONNX Runtime, Windows Speech API, macOS Speech Recognition, Vosk
126+
**Native:** [Whisper.cpp](https://github.com/ggerganov/whisper.cpp) (ggml), [ONNX Runtime](https://onnxruntime.ai/), Windows Speech API, macOS Speech Recognition, [Vosk](https://alphacephei.com/vosk/)
127127
**Features:** Streaming STT, Real-time transcription, Multi-language, Noise-reduction
128128
**Use-Cases:** Voice-Commands, Dictation, Voice-Bots, Meeting-Transcription, Audio-Automation
129129

@@ -137,23 +137,23 @@ FastJava, FastCore, FastPlugin
137137
| **FastSIMD** | [SSE](https://en.wikipedia.org/wiki/Streaming_SIMD_Extensions)/[AVX](https://en.wikipedia.org/wiki/Advanced_Vector_Extensions)/[NEON](https://developer.arm.com/architectures/instruction-sets/simd-isas/neon) wrapper for Java. 10x faster for vector operations, pixel processing, physics. |
138138
| **FastString** | Java Strings are immutable and UTF-16. FastString is mutable, UTF-8, zero-copy - for parsing. |
139139
| **FastBytes** | ByteBuffer with String API and SIMD operations. For binary parsing without GC. |
140-
| **FastHash** | xxHash3/BLAKE3 are 100x faster than SHA-256. For checksums in real-time pipelines. |
141-
| **FastJSON** / **FastParse** | Jackson/Gson are too slow for high-frequency. FastJSON is zero-copy, 50x faster. |
142-
| **FastRegex** | Java Regex has backtracking problems. FastRegex uses Hyperscan engine for SIMD pattern matching. |
143-
| **FastSort** | Radix Sort is O(n) for integers. 10x faster than Java's Dual-Pivot Quicksort for large arrays. |
144-
| **FastCompress** | LZ4 is 10x faster than gzip. For netcode and log compression where speed matters. |
145-
| **FastPathfinder** | A* in Java is too slow for large grids. Native implementation for real-time navigation. |
140+
| **FastHash** | [xxHash3](https://github.com/Cyan4973/xxHash)/[BLAKE3](https://github.com/BLAKE3-team/BLAKE3) are 100x faster than SHA-256. For checksums in real-time pipelines. |
141+
| **FastJSON** / **FastParse** | [Jackson](https://github.com/FasterXML/jackson)/[Gson](https://github.com/google/gson) are too slow for high-frequency. FastJSON is zero-copy, 50x faster. |
142+
| **FastRegex** | Java Regex has backtracking problems. FastRegex uses [Hyperscan](https://github.com/intel/hyperscan) engine for SIMD pattern matching. |
143+
| **FastSort** | [Radix Sort](https://en.wikipedia.org/wiki/Radix_sort) is O(n) for integers. 10x faster than Java's Dual-Pivot Quicksort for large arrays. |
144+
| **FastCompress** | [LZ4](https://github.com/lz4/lz4) is 10x faster than gzip. For netcode and log compression where speed matters. |
145+
| **FastPathfinder** | [A*](https://en.wikipedia.org/wiki/A*_search_algorithm) in Java is too slow for large grids. Native implementation for real-time navigation. |
146146
| **FastEventBus** | JNI callbacks are expensive. FastEventBus batches events and uses lock-free queues. |
147147
| **FastHookChain** | Order hooks by priority: Input → PreProcess → AI → Action → PostProcess. |
148-
| **FastDetour** | Function hooking for modding and instrumentation. Trampoline for original call. |
148+
| **FastDetour** | [Function hooking](https://en.wikipedia.org/wiki/Hooking) for modding and instrumentation. [Trampoline](https://en.wikipedia.org/wiki/Hooking#Trampoline) for original call. |
149149
| **FastThreading** | Java thread pools have too much overhead. Lock-free queues and thread affinity for <1ms latency. |
150-
| **FastBenchmark** | Micro-benchmarks with QueryPerformanceCounter. For performance marketing and optimization. |
150+
| **FastBenchmark** | Micro-benchmarks with [QueryPerformanceCounter](https://learn.microsoft.com/en-us/windows/win32/api/profileapi/nf-profileapi-queryperformancecounter). For performance marketing and optimization. |
151151

152152
### Debugging & Development
153153

154154
| Module | Explanation |
155155
|--------|-------------|
156-
| **FastCallTrace** | Who is calling me? Stack introspection 10x faster than Throwable.getStackTrace(). |
156+
| **FastCallTrace** | Who is calling me? [Stack introspection](https://en.wikipedia.org/wiki/Stack_trace) 10x faster than Throwable.getStackTrace(). |
157157
| **FastWatch** | Watch variables in real-time (60-240 Hz) without Swing overhead. For game dev debugging. |
158158
| **FastReplay** | Record and playback input. For UI testing and bot validation. |
159159
| **FastTest** | Combines FastScreen + FastRobot for visual assertions. UI testing engine. |
@@ -164,13 +164,13 @@ FastJava, FastCore, FastPlugin
164164
|--------|-------------|
165165
| **FastAI** | One interface for all AI providers (Ollama, OpenAI, Claude). No JSON hassle, just prompts. |
166166
| **FastAIClient** | Provider-specific implementations behind FastAI interface. |
167-
| **FastEmbedding** | Local embeddings with ggml. For RAG without cloud dependency. |
168-
| **FastVectorDB** | In-process vector database for KNN search. For semantic search and RAG retrieval. |
167+
| **FastEmbedding** | Local embeddings with [ggml](https://github.com/ggerganov/ggml). For RAG without cloud dependency. |
168+
| **FastVectorDB** | In-process vector database for [KNN search](https://en.wikipedia.org/wiki/K-nearest_neighbors_algorithm). For semantic search and RAG retrieval. |
169169
| **FastToolBridge** | AI models can call Fast* modules directly. MCP-compatible for agents. |
170170
| **FastContext** | Agent memory with profiles and long-term knowledge. For personalized AI agents. |
171171
| **FastMemory** | Persistent embedding-based agent memory. Stores facts with automatic summarization and retrieval. Uses FastEmbedding + FastVectorDB. |
172-
| **FastRAG** | Complete RAG pipeline for local and cloud models. Chunking, embedding, vector search, context injection in one module. |
173-
| **FastMCP** | MCP (Model Context Protocol) for FastAI. Connects tools, memory, RAG, files via MCP servers with FastAI as model backend. |
172+
| **FastRAG** | Complete [RAG](https://en.wikipedia.org/wiki/Retrieval-augmented_generation) pipeline for local and cloud models. Chunking, embedding, vector search, context injection in one module. |
173+
| **FastMCP** | [MCP](https://modelcontextprotocol.io/) (Model Context Protocol) for FastAI. Connects tools, memory, RAG, files via MCP servers with FastAI as model backend. |
174174

175175
### Platform
176176

0 commit comments

Comments
 (0)