Laminar Engine

Transform your idle Integrated GPU (iGPU) into a dedicated Asynchronous Co-Processor!

Laminar is a cutting-edge, low-overhead optimization suite designed to bridge your CPU and its built-in integrated graphics card (Intel HD Graphics, AMD Radeon) using explicit OpenCL 1.2 pipelines.

By offloading the most crippling CPU-bound tasks in real-time, Laminar ensures your server TPS stays locked at 20.00 while freeing up your main processor to focus purely on delivering maximum FPS.

---

✨ The Four Pillars of Laminar

🧟 Project "Hive" (GPU Flow Field Pathfinding)

Traditional A* pathfinding destroys server performance when handling hundreds of mobs.

• Macro-Micro Handoff: The iGPU calculates a global 3D vector field within a 24-block radius (110,592 voxels) and traces a highly optimized path asynchronously. Vanilla physics still handle the "Micro" movements (precise jumping, swimming, and climbing).
• Acoustic Swarm Alert: Mobs communicate dynamically. If a zombie spots you, it alerts only the mobs within a realistic 12-block auditory radius that share an unoccluded physical path to you.

🔊 Project "Echo" (GPU-Accelerated Acoustics)

Real-time audio occlusion and reverberation without the severe micro-stutters caused by CPU-based raycasting mods.

• Fibonacci Sphere Raycast: Dispatches 16 parallel rays per active sound directly on the iGPU, simulating wave propagation and room-size reflections.
• Multiplayer-Safe: Echo performs its own independent, thread-safe voxel scanning on the client-side, completely isolated from server-side calculations.
• Real-Time Volume Dampening: Automatically overrides the active OpenAL channel gain in real-time as you move behind walls.

👁️ Project "Aura" (GPU Line-of-Sight)

Offloads the extremely heavy, CPU-bound entity hasLineOfSight() checks.

• Mass Raymarching: Casts parallel rays from mob eye level to player eye level on the iGPU.
• Shadow Cache Lookups: Stores visibility states in a multi-player safe, nested cache, providing O(1) zero-latency lookups on the main server thread.

🧠 Adaptive Entity AI Throttling (LOD AI)

• Dynamic 3-Zone LOD: Throttles idle mob brains based on spatial proximity (Zone 1: 20 TPS | Zone 2: 5 TPS | Zone 3: 1.6 TPS).
• UUID-Staggered Ticking: Tick execution is staggered across frames using each mob's unique UUID, completely flattening CPU usage.

---

📈 Extreme Stress-Test Benchmarks

Tested on: Intel Core i7-4770K (Haswell @ 4.3 GHz(OC),AVX2), Intel HD 4600 iGPU (OpenCL 1.2), NVIDIA GTX 1050. Scenario: 500 Zombies active, 32 Chunks loaded.

Performance Metric	Vanilla Minecraft	Laminar + Sodium + Lithium
Server Tick Rate	19.35 TPS (Lag)	20.00 TPS (Locked)
Median MSPT	31.1 ms	14.7 ms (52% Faster)
95th% MSPT	50.5 ms (Stutter)	32.6 ms (Ultra-Flat)
Worst-Case Spike	253 ms	104 ms (Slashed by 60%)
CPU Usage	36.12%	21.19% (Highly Efficient)

---

🔌 Compatibility & Requirements

• Fabric Loader: v0.19.2+ (Minecraft 26.1.2)
• Hardware: A dual-GPU setup (CPU + Integrated GPU) or any APU.
• OpenCL: Your iGPU must support OpenCL 1.2 or higher.
• Java: Java 25 (GraalVM highly recommended).

🟢 Highly Recommended (Synergy)

Laminar is designed to be the ultimate companion to render-optimizing mods.

• Works flawlessly with Sodium, Lithium, FerriteCore, and ImmediatelyFast.
• VulkanMod Compatible: Seamlessly runs alongside Vulkan rendering by bypassing GPU memory conflicts via specialized CPU host-mapped memory handoffs.

🔴 Incompatible

• CPU-Bound Sound Physics Mods (Sound Physics Remastered, Acoustic Overhaul). Laminar's Project Echo already replaces these with hardware-accelerated equivalents.

---

❓ FAQ (Frequently Asked Questions)

Q: Will this work on a Dedicated Server (like PebbleHost/Apex)?

A: Most enterprise server CPUs (Xeon/EPYC) do not have an iGPU. Laminar will safely detect this and bypass itself without crashing. Laminar shines best on Single-Player, LAN, or Self-Hosted servers running on standard desktop/laptop processors (Intel Core / AMD Ryzen).

Q: Do I need to configure anything?

A: Laminar works out of the box!.

Q: Can I use this if I only have one GPU (No dedicated graphics card)?

A: Yes! If you are playing purely on an APU (like AMD Ryzen G-series or Intel HD Graphics), Laminar will still utilize OpenCL compute queues to run background tasks asynchronously alongside your OpenGL rendering.

---

❤️ Support the Project

Laminar is a passion project built to squeeze every last drop of performance out of mid-range hardware. If this mod saved your framerate or your server TPS, consider supporting the development!