Glossary

Definition and usage for key terms, concepts, and terminology used within the Pictor network.

Pictor Ecosystem

🔹 PICTOR Token

PICTOR Token ($PICTOR) is the native token of Pictor Network. It’s designed to foster the Pictor network’s decentralized ecosystem for tools, applications, and services. As a multifaceted asset, the $PICTOR plays a key role in liquidity, governance, incentivization, and network development. It serves the needs of several main groups: Workers, 3D Creators, AI Builders, Checkers, and $PICTOR Holders.

🔹Pictor Cloud

Pictor Cloud is the central access layer of the Pictor Network, providing a seamless web-based interface for all key participants — GPU Requesters, GPU Providers, and Checkers. Through Pictor Cloud, GPU Requesters can submit and manage rendering or AI jobs, GPU Providers can manage their Worker nodes and monitor performance, while Checkers can efficiently manage their Checker nodes to uphold network quality, fairness, and integrity.

🔹Pictor Worker

Workers, also known as GPU providers, are the supply side of the Pictor Network. They contribute GPU compute power, which fuels both AI workloads and 3D rendering jobs. Through the Pictor Worker App or Docker Image, they can easily connect their idle GPUs to the network and earn rewards in return.

🔹3D Rendering

Pictor’s core rendering service that helps 3D Creators render their 3D projects faster and more affordably, powered by decentralized GPU infrastructure.

🔹AI Training

Pictor’s core AI service enables AI Builders to train and fine-tune AI models, leveraging a global network of idle GPUs to deliver scalable, decentralized, and cost-effective compute power.

🔹AI Inference

Pictor’s core AI service enables AI Builders to run AI inference workloads on decentralized GPUs at scale and without the cloud overhead.

🔹AI Model Hosting

Pictor’s core AI service for deploying trained AI models on global, decentralized GPU infrastructure.

🔹Pictor Checker Node

Pictor’s validation nodes, responsible for inspecting and verifying the rendering outputs from Workers to ensure job accuracy, integrity, and overall network performance.

🔹Pictor Dashboard

A real-time statistics page that displays key metrics of the Pictor Network. It’s designed for full transparency and easy network monitoring.

Basic Terms

🔹3D Creator (GPU Requester)

Individuals and entities who hire GPU compute resources to render their 3D rendering projects.

🔹AI Builder (GPU Requester)

Individuals and entities who hire GPU compute resources for their AI workloads, such as training and fine-tuning AI models, running AI inference, and hosting trained AI models.

🔹Worker (GPU Provider)

Individuals and entities who rent out GPU compute resources to 3D Creator and/or AI Builders, and earn rewards for renting out their hardware.

🔹Checker

Individuals and entities who run the Pictor Checker Nodes to verify rendering results produced by Workers, and ensure the network’s service quality and integrity.

🔹Worker Node

A computer (PC or laptop) equipped with a GPU and other necessary hardware components (including CPU, RAM, storage drive, etc.) that is dedicated to executing 3D rendering tasks and AI workloads. The Worker Node is provided by Workers.

🔹Checker Node

Pictor’s validation nodes, responsible for inspecting and verifying the rendering outputs from Workers to ensure job accuracy, integrity, and overall network performance. The Checker Node is operated by the Checker.

🔹Pictor Checker Node NFT

Pictor's NFT (Non-Fungible Token) enables users to stake in Pictor Checker Nodes, earning rewards through verification jobs performed by the Nodes.

🔹VPS (Virtual Private Server)

A virtual machine that provides virtualized server resources on a physical server (machine) that is shared with other users. It uses virtualization technology to partition a physical server (machine) into multiple isolated virtual servers, each with its own dedicated resources (such as CPU, RAM, storage, and operating system). Each VPS operates independently, allowing users to host websites, applications, or services in an environment that mimics a dedicated server (machine).

🔹Computing

Computing refers to the process of performing calculations, such as addition, multiplication, or more complex mathematical functions. This term is closely associated with computers designed to perform computations rapidly and efficiently.

🔹Compute Hours

Compute hours are the measurable hours or time that your process is loaded and executed. Compute hours are one of the two main metrics that are used to determine costs.

🔹Connectivity Tier

It’s the speed or bandwidth of internet connectivity provided by an internet service provider (ISP) or telecommunications company. It represents the rate at which data can be transmitted over a network connection, typically measured in megabits per second (Mbps) or gigabits per second (Gbps).

🔹DePIN

DePIN stands for Decentralized Physical Infrastructure Networks. It leverages blockchains, IoT, and the greater Web3 ecosystem to create, operate, and maintain real-world physical infrastructure. These networks leverage token incentives to coordinate, reward, and safeguard members of the network.

🔹Rendering Script File

A rendering script file contains a set of instructions or commands that control the behavior of a 3D software, rendering engine, or plugin. Script files are used to automate tasks, customize workflows, or create custom tools and plugins. Rendering script files typically contain code written in a programming language, such as Python, JavaScript, or C++, and are executed by the 3D software or rendering engine to perform specific tasks.

For example, Blender and Cinema 4D use the “.py” (Python Script File).

🔹Rendering Job

A rendering Job refers to a specific task or set of tasks submitted by a user (often referred to as the “Creator”) that involves rendering a 3D scene, animation, or visual effect. This job is processed within the Pictor Network, leveraging the distributed computing resources available across the network to complete the rendering process efficiently.

🔹CPU

CPU stands for Central Processing Unit, also called a central processor, main processor, or just processor. It is the primary component of any computer or electronic device, responsible for carrying out all the instructions that make your computer work. It reads and interprets commands from software programs and uses them to control other components within the computer.

🔹GPU

GPU stands for Graphics Processing Unit. It is an electronic circuit designed to speed up computer graphics and image processing on a variety of devices, including video cards, motherboards, mobile phones, and personal computers (PCs). It can perform mathematical calculations at high speed. Computing tasks like graphics rendering, machine learning (ML), and video editing require the application of similar mathematical operations on a large dataset. A GPU’s design allows it to perform the same operation on multiple data values in parallel.

🔹RAM

RAM stands for Random Access Memory. It is the hardware in a computer that provides temporary storage for the operating system (OS), software programs, and any other data in current use so they’re quickly available to the device’s CPU. RAM is often referred to as a computer’s main memory, as opposed to the CPU cache or other memory types.

🔹VRAM

VRAM, also called Video Memory, stands for Video Random Access Memory. It is a dedicated memory on a computer’s graphics card or graphics processing unit (GPU) that stores and manages data related to graphics and video processing. Video memory directly affects graphics performance by determining the amount of data that can be stored and accessed by the GPU.

🔹HDD

HDD is a hard drive or hard disk drive. It is a type of data storage device that is used in laptops and desktop computers. An HDD is a “non-volatile” storage drive, which means it can retain the stored data even when no power is supplied to the device. Operating systems (OS) tell the HDD to read and write data as needed by programs. The speed at which the drive reads and writes this data is solely dependent on the drive itself.

🔹SSD

SSD is a solid-state drive. It is a semiconductor-based storage device that typically uses flash memory to save persistent data. An SSD is a non-volatile memory computer hardware that stores data without moving parts. SSDs feature swift read, write, and boot times on modern machines that are unparalleled by traditional hard disks (HDDs).

🔹NVMe

NVMe is Non-Volatile Memory Express. It is a logical device interface protocol for accessing a computer’s non-volatile storage media. NVMe is a standard specification that is used on SSD drives so that each manufacturer does not have a unique device driver.

🔹HIP

HIP (Heterogeneous Interface for Portability): HIP is a C++ runtime API and kernel language that enables developers to write portable code for both AMD and NVIDIA GPUs. It allows for the conversion of CUDA code to run on AMD hardware, promoting cross-platform compatibility.

🔹OpenCL

OpenCL (Open Computing Language): OpenCL is an open standard for parallel programming across heterogeneous platforms, including CPUs, GPUs, and other processors. Supported by AMD, it enables developers to write code that can execute on various hardware architectures, fostering flexibility and portability.

🔹CUDA

CUDA (Compute Unified Device Architecture) is a parallel computing platform and programming model developed by NVIDIA for general computing on graphical processing units (GPUs). With CUDA, developers can dramatically speed up computing applications by harnessing the power of GPUs.

🔹Rendering/3D Rendering

3D rendering is the process of generating a 2D image from a 3D model using computer software. This process involves converting the mathematical and geometrical representation of a 3D object (stored in a 3D model) into a realistic or stylized image. The resulting image is known as a “render.” Rendering can be done in real-time (as in video games, where the images are rendered on-the-fly as the player interacts with the game) or offline (as in movies, where each frame is carefully rendered over a longer period to achieve higher quality).

Unreal Engine Real-time Rendering vs. Offline Rendering

🔹Offline Rendering/Pre-rendering

Offline rendering, also known as pre-rendering, is a rendering process where the images or animations are generated and processed ahead of time, rather than in real-time. In pre-rendering, the 3D scene is meticulously designed with high levels of detail, textures, and photorealistic effects. Once the design is complete, the scene is then rendered into a 2D image or sequence of images using the processing power of a CPU, GPU, or both. The key aspect of pre-rendering is that the rendering process happens before the image is displayed, hence the term “pre-rendering.”

Some of the most commonly used offline rendering software include Blender (Cycles render engine), Maxon Redshift, Otoy OctaneRender, Chaos V-Ray, Autodesk Arnold, and Pixar RenderMan.

🔹Real-time Rendering

Real-time rendering is the process of generating and displaying images in real-time, typically at a rate of 30-60 frames per second, to create the illusion of motion and interactivity. In real-time rendering, the rendering engine calculates and displays the image as the user interacts with the environment. Real-time rendering is used in applications that require fast and responsive graphics, such as video games, Virtual reality (VR), Augmented reality (AR), Simulations, etc.

Some of the most commonly used real-time rendering software include Epic Games' Unreal Engine, Unity, Blender (Eevee render engine), and NVIDIA Omniverse.

🔹GPU Rendering

GPU rendering is a technique that renders images solely using the GPU. The CPU handles the complex mathematical calculations required to generate 2D images from 3D models. Generally, in CPU Rendering, the CPU processes the rendering tasks sequentially, one instruction at a time.

Some popular GPU rendering engines include Cycles (GPU mode), Redshift (GPU mode), OctaneRender, Unreal Engine, Unity, V-Ray (GPU mode), Arnold (GPU mode), etc.

🔹CPU Rendering

CPU rendering is a technique that renders images solely using the CPU. In GPU Rendering, the GPU processes rendering tasks in parallel, using thousands of cores to handle complex mathematical calculations simultaneously.

Some popular CPU rendering engines include Corona, RenderMan, V-Ray (CPU mode), Redshift (CPU mode), Cinema 4D’s Standard and Physical renderers, etc.

🔹Hybrid Rendering (XPU Rendering)

Hybrid rendering (XPU rendering) is a rendering technique that uses both CPU and GPU to render images.

Some popular rendering engines that support hybrid rendering include Cycles, Redshift, V-Ray, Arnold, Karma, etc.

🔹3D Production Pipeline

A 3D production pipeline refers to the systematic sequence of steps involved in creating 3D visual content, typically used in animation, film, video games, and visual effects. Typically, the pipeline organizes the process into three distinct phases (Pre-production, Production, and Post-production), each with specialized tasks that contribute to the final product.

Note: The 3D production pipeline can vary depending on the specific project, studio, or industry.

🔹3D Software

3D software, also known as 3D modeling software, is a type of computer program that enables users to create, edit, and manipulate three-dimensional (3D) models, objects, and scenes. These software applications are used in various industries such as film, animation, architecture, product design, games, etc. Some popular 3D software includes Blender, Cinema 4D, Maya, 3ds Max, Houdini, etc.

Some 3D software packages come with their built-in rendering engines, allowing users to both model and render within the same application. Others are primarily focused on modeling and require external rendering engines to produce final images.

For example:

• 3D software with built-in rendering engines: Blender (Cycles and Eevee), Cinema 4D (Redshift, Physical, and Standard), Maya (Arnold), Houdini (Mantra)

• 3D software without built-in rendering engines: SketchUp, Rhino, Fusion 360

In a 3D production pipeline, 3D software can be used in various steps such as Modeling, Texturing and Shading, Layout, Rigging, Animation, Lighting, and Rendering.

🔹Rendering Engine/Renderer

A rendering engine is a specialized software or component within a 3D software that converts 3D models into 2D images. It takes the 3D data, such as geometry, materials, lighting, and camera information, and uses complex algorithms to simulate the way light interacts with the scene. The goal of a rendering engine is to produce a final image that can range from a simple visual representation to a highly detailed, photorealistic image.

Some rendering engines are separate software (standalone) applications that function as plugins for 3D modeling software, while others are integrated directly into the 3D modeling software as built-in rendering engines.

• Separate Software (Plugins): V-Ray, Redshift, OctaneRender, Arnold (for some 3D software), Corona Renderer.

• Built-in Rendering Engines: Cycles (Blender), Eevee (Blender), Mantra (Houdini), Physical Render (Cinema 4D), Redshift (Cinema 4D).

In a 3D production pipeline, a rendering engine is typically used in the final step (Rendering step) of the Production phase.

🔹Benchmark/Rendering Benchmark

A rendering benchmark is a standardized test used to measure and evaluate the performance of a computer’s hardware (CPUs, GPUs, and other system components) when handling rendering tasks. These benchmarks are designed to simulate real-world rendering workloads, allowing users to assess how efficiently their system can handle various aspects of rendering, such as processing speed, memory usage, and graphics capabilities. Rendering benchmarks provide valuable insights into how well a system can handle complex 3D scenes and help users make informed decisions about hardware and software optimizations.

🔹Benchmark Score

Benchmark scores are numerical values that represent the performance of a computer system or specific components (like a CPU or GPU) when subjected to standardized tests or benchmarks. These scores are used to compare the efficiency and speed of different systems or components, providing a quantifiable measure of performance.

Typically, benchmark scores are expressed in points-based scores (i.e., 1500, 2000).

3D

🔶3D Software and Render Engine

🔹Blender

Blender is a free and open-source 3D creation suite. It supports the entirety of the 3D pipeline—modeling, rigging, animation, simulation, rendering, compositing, and motion tracking, even video editing and game creation.

🔹Cycles

Cycles, Blender’s built-in renderer, is a physically-based path tracer for production rendering. It is designed to provide physically based results out-of-the-box, with artistic control and flexible shading nodes for production needs.

🔹Eevee

EEVEE, Blender’s built-in renderer, is a real-time render engine focused on speed and interactivity while achieving the goal of rendering PBR materials. EEVEE can be used interactively in the 3D Viewport but also produces high-quality final renders.

🔹AMD Radeon ProRender

AMD Radeon ProRender is a powerful, physically-based rendering engine that enables creative professionals to produce stunningly photorealistic images.

Built on AMD high-performance Radeon Rays technology, AMD Radeon ProRender’s complete, scalable ray tracing engine uses open industry standards to harness GPU and CPU performance for swift, impressive results.

🔹Cinema 4D

Cinema 4D is a professional 3D modeling, animation, simulation, and rendering software developed by Maxon. It has a fast, powerful, flexible, and stable toolset that makes 3D workflows more accessible and efficient for design, motion graphics, VFX, AR/MR/VR, game development, and all types of visualization professionals.

🔹Redshift

Redshift is Cinema 4D’s built-in rendering engine. It is a powerful GPU-accelerated renderer, built to meet the specific demands of contemporary high-end production rendering. Tailored to support creative individuals and studios of every size, Redshift offers a suite of powerful features and integrates with industry-standard CG applications.

🔹OctaneRender

OctaneRender is an unbiased, spectrally-correct GPU render engine developed by Otoy. It delivers quality and speed unrivaled by any production renderer on the market. OctaneRender is not only a popular rendering plugin for Cinema 4D, Blender, as well as other 3D software, but it is also a stand-alone application that can be used independently.

🔹Unreal Engine

Unreal Engine, also called UE, is a powerful game engine developed by Epic Games. Unreal Engine is widely used for developing video games of all sizes and genres. It is known for its high-fidelity graphics and is also used in various industries for real-time 3D visualization, virtual production, and simulation.

🔹Unity

Unity is a cross-platform game engine developed by Unity Technologies. Unity’s real-time 3D development engine lets artists, designers, and developers collaborate to create amazing immersive and interactive experiences.

🔹Omniverse

NVIDIA Omniverse is a platform of APIs, services, and software development kits (SDKs) that enable developers to build generative AI-enabled tools, applications, and services for industrial digitalization workflows. Applications built on Omniverse core technologies fundamentally transform complex 3D workflows, allowing individuals and teams to build unified tools and data pipelines and simulate large-scale, physically accurate virtual worlds for industrial and scientific use cases.

🔶Additional Software

🔹AMD Driver

AMD drivers are software programs that facilitate communication between your AMD hardware (such as graphics cards) and the operating system. They are crucial as they ensure your hardware functions optimally, providing better performance and compatibility with applications and games.

🔹AMD Radeon Software

AMD Radeon Software is a suite of drivers and utility software designed to optimize and enhance the performance of AMD Radeon graphics cards. It serves as the primary interface for managing GPU settings, enabling users to configure, monitor, and optimize their systems for gaming, professional workloads, and multimedia applications.

AMD Radeon Software is available for free and supports both consumer GPUs (Radeon RX series) and professional GPUs (Radeon Pro series), ensuring tailored features for gaming enthusiasts and professionals alike.

🔹AMD Radeon PRO Software

Radeon PRO Software is AMD’s suite of drivers and utilities tailored for professional graphics cards, specifically the Radeon PRO series. Designed to meet the demands of professionals in fields such as computer-aided design (CAD), digital content creation, and scientific research, this software ensures optimal performance, stability, and reliability.

🔹NVIDIA Driver

An NVIDIA driver is a software program that enables communication between your computer and the NVIDIA graphics processor installed in your system. It is used to ensure that your hardware works as intended with the latest software, games, and applications.

🔹NVLink

NVLink is a high-speed, high-bandwidth interconnect technology developed by NVIDIA. It is designed to enable fast communication between multiple GPUs (graphics processing units) and between GPUs and CPUs. NVLink enables scalable performance and memory sharing between multiple GPUs, which is particularly important for complex rendering tasks that require massive parallel processing.

🔹Blender Benchmark

Blender Benchmark is a performance testing (benchmarking) tool developed by the Blender Foundation. It is designed to evaluate the capabilities of a computer system in handling 3D rendering tasks using Blender’s Cycles renderer. The benchmark is used by individuals, professionals, and hardware reviewers to measure and compare the rendering performance of different hardware configurations, such as CPUs and GPUs.

🔹CineBench

Cinebench is an industry-standard benchmarking software developed by Maxon based on cutting-edge technology. Cinebench 2024 utilizes the power of Redshift, Cinema 4D’s default rendering engine, to evaluate a computer’s CPU and GPU capabilities.

🔹Redshift Benchmark

Redshift Benchmark is a command-line tool that can load a scene, render it, and measure the time it takes to render, excluding certain CPU operations such as loading the scene or textures from disk. It prints out certain information about the GPU configuration on the screen and also as a watermark in the resulting rendered image, which is saved on disk.

🔹OctaneBench

OctaneBench is a benchmarking tool developed by OTOY specifically designed to measure the performance of GPUs (Graphics Processing Units) when rendering 3D scenes using OctaneRender. It provides a comprehensive score that allows users to compare the rendering capabilities of different GPUs, including those in multi-GPU setups.

AI

🔹Artificial Intelligence (AI)

Artificial intelligence (AI) is technology that enables computers and machines to simulate human learning, comprehension, problem-solving, decision-making, creativity, and autonomy.

🔹Machine Learning (ML)

Machine learning (ML) is a branch of artificial intelligence (AI) focused on enabling computers and machines to imitate the way that humans learn, to perform tasks autonomously, and to improve their performance and accuracy through experience and exposure to more data.

🔹Large language models (LLMs)

Large language models (LLMs) are a category of foundation models trained on immense amounts of data, making them capable of understanding and generating natural language and other types of content to perform a wide range of tasks.

🔹AI Model

An AI model is a program that has been trained on a set of data to recognize certain patterns or make certain decisions without further human intervention. Artificial intelligence models apply different algorithms to relevant data inputs to achieve the tasks or outputs they’ve been programmed for.

🔹AI Training

AI training is the process of teaching an AI model to perform specific tasks by exposing it to data. This data acts as a learning material, enabling the AI to identify patterns, make predictions, and ultimately improve its performance over time.

🔹AI Inference

Artificial intelligence (AI) inference is the ability of trained AI models to recognize patterns and draw conclusions from information that they haven’t seen before. AI inference is critical to the advancement of AI technologies and underpins its most exciting applications, such as generative AI, the capability that powers the popular ChatGPT application. AI models rely on AI inference to imitate the way people think, reason, and respond to prompts.

🔹AI Model Hosting

Once your AI model is trained and ready, you must host or “deploy” it. AI model hosting is the process of making your AI model accessible to other users or applications. Typically, the model exposes an API (application programming interface) that authorized users and software systems can use to communicate with your model via code.

Security

🔹E2EE

End-to-end Encryption (E2EE) is a method of secure communication that prevents third parties from accessing data while it’s transferred from one end system or device to another.

In E2EE, the data is encrypted on the sender’s system or device, and only the intended recipient can decrypt it. As it travels to its destination, the message cannot be read or tampered with by an internet service provider (ISP), application service provider, hacker, or any other entity or service.

🔹SOC2/HIPAA

SOC 2 and HIPAA are compliance frameworks that address different aspects of data security and privacy. SOC 2 focuses on assessing the controls implemented by service organizations to protect customer data, while HIPAA sets standards for protecting sensitive personal information.

🔹SSL

SSL, or Secure Sockets Layer, is an encryption-based Internet security protocol. SSL ensures privacy, authentication, and data integrity in Internet communications. To provide a high degree of privacy, SSL encrypts data that is transmitted across the web. This means that anyone who tries to intercept this data will only see a garbled mix of characters that is nearly impossible to decrypt.