Skip to content

Author / Systems & Technical Editor / 247 articles

Marcus Thorne

Marcus writes Forge PC's step-by-step build, troubleshooting and system optimisation guides. His background spans Windows kernel and power-state diagnostics, UEFI/BIOS configuration, NVMe storage arrays, ATX 3.1 power delivery and running local AI and LLM workloads on consumer graphics cards.

  • How to Enable XMP or EXPO on a DDR5 PC

    How-To

    How to Enable XMP or EXPO on a DDR5 PC

    How to turn on Intel XMP or AMD EXPO so a DDR5 kit actually runs at its rated speed: official steps from G.SKILL and ASUS, how to check the speed in Windows, and what to do if the PC will not boot.

    Marcus Thorne4 min

  • How to Use LM Studio as a Local API Server

    How-To

    How to Use LM Studio as a Local API Server

    How to turn LM Studio into a local LLM API server on http://localhost:1234: start the server from the Developer tab or with lms server start, then send OpenAI-compatible chat requests, using LM Studio's official developer docs.

    Marcus Thorne4 min

  • How to Use Continue in VS Code with a Local Ollama Model

    How-To

    How to Use Continue in VS Code with a Local Ollama Model

    How to connect the Continue extension in VS Code to a local Ollama model for AI coding help without sending your code to a cloud API: official prerequisites, ollama pull, config.yaml, Autodetect, and common errors from Continue's documentation.

    Marcus Thorne4 min

  • How to Use Ollama's OpenAI-Compatible API

    How-To

    How to Use Ollama's OpenAI-Compatible API

    How to call a local Ollama model from curl, Python or JavaScript using Ollama's official OpenAI-compatible API on http://localhost:11434/v1, including which endpoints are supported and how to pull a model first.

    Marcus Thorne4 min

  • How to Back Up a Windows 11 PC

    How-To

    How to Back Up a Windows 11 PC

    How to back up a Windows 11 PC using Microsoft's official tools: the Windows Backup app and OneDrive for files and settings, File History for previous versions on an external drive, and why System Restore and a recovery drive are not a substitute for a file backup.

    Marcus Thorne5 min

  • How to Install Open WebUI with Ollama on Windows

    How-To

    How to Install Open WebUI with Ollama on Windows

    How to give Ollama a browser-based chat interface with Open WebUI on Windows: install Docker Desktop with the WSL 2 backend, run the official container, set a secret key, and connect to Ollama on port 11434.

    Marcus Thorne5 min

  • How to Choose and Load AI Models in LM Studio

    How-To

    How to Choose and Load AI Models in LM Studio

    How to find, download and load a local LLM in LM Studio: the Discover tab, what the Q4 and Q8 file names mean, official Windows requirements, and how to set context length and GPU offload.

    Marcus Thorne5 min

  • Windows 11 Gaming Settings That Actually Matter

    How-To

    Windows 11 Gaming Settings That Actually Matter

    Which Windows 11 settings Microsoft says are worth checking for gaming: Game Mode, hardware-accelerated GPU scheduling, windowed-game optimisations, Auto HDR, Game Bar and DirectStorage, and where to find them.

    Marcus Thorne5 min

  • How to Install ComfyUI on Windows for Local AI Images

    How-To

    How to Install ComfyUI on Windows for Local AI Images

    How to install ComfyUI on Windows for local AI image generation in 2026: Comfy Desktop, the Portable build or a manual install, with official requirements for NVIDIA, AMD ROCm and Intel Arc graphics.

    Marcus Thorne4 min

  • How to Clone Windows to a New SSD (Free Tools, 2026)

    How-To

    How to Clone Windows to a New SSD (Free Tools, 2026)

    How to clone Windows, apps and files to a new SSD with free tools in 2026: Samsung Magician for Samsung drives, SanDisk's Acronis edition for WD_BLACK, and Clonezilla for any brand, plus why Macrium Reflect Free is now end of life.

    Marcus Thorne5 min

  • Buying a Used GPU for Local AI in the UK

    How-To

    Buying a Used GPU for Local AI in the UK

    Used graphics cards can add cheap VRAM for local AI, but only if the used price is actually below a new card. We compare UK new and used figures, including a 24GB RTX 3090, from mid-November 2026.

    Marcus Thorne3 min

  • LLM Quantization and GGUF Explained

    How-To

    LLM Quantization and GGUF Explained

    Quantization shrinks a local AI model's weights so they fit in VRAM. We explain GGUF, llama.cpp's Q4_K_M example, and llamaperf's published memory figures for common model sizes.

    Marcus Thorne4 min

  • How to Stop a Gaming PC Overheating

    How-To

    How to Stop a Gaming PC Overheating

    How to confirm a gaming PC is overheating, clean dust, check airflow and reseat a CPU cooler, using official temperature limits and warranty caveats rather than invented safe numbers.

    Marcus Thorne4 min

  • DLSS, FSR and Frame Generation Explained

    How-To

    DLSS, FSR and Frame Generation Explained

    DLSS 4.5, FSR Redstone and frame generation are not the same feature. We explain what NVIDIA and AMD officially offer, which GPUs each mode needs, and how to turn them on.

    Marcus Thorne5 min

  • How Much RAM Do You Need to Run AI Models Locally in 2026?

    How-To

    How Much RAM Do You Need to Run AI Models Locally in 2026?

    Video memory decides which AI models load on a graphics card, but system RAM still matters. We explain how much you need with a GPU, without one, and when spending on 64GB is worth it at 13 November 2026 prices.

    Marcus Thorne5 min

  • How to Install Windows 11 on a New PC

    How-To

    How to Install Windows 11 on a New PC

    A clean Windows 11 install on a new build is straightforward if you start with official media and the right firmware settings. We walk through Microsoft's Media Creation Tool, USB boot, TPM and a Microsoft account.

    Marcus Thorne5 min