[email protected]
Twitter
LinkedIn
YouTube
GitHub
  • Serviços
  • Blog
  • Repositórios
  • GitHub
  • Currículo
    • LinkedIn
    • PDF
  • Contato
Produto foi adicionado ao seu carrinho

Carrinho

Curso de Babylon.js

Maio 1, 2015Ferramentasjfadev

Babylon.js É um framework completo para a criação de jogos 3D com a ajuda de HTML5 e WebGL basiado 100% em JavaScript, desenvolvido por programadores da Microsoft. Ele funciona corretamente no Firefox e Chrome e é compatível com as seguintes características:

  • Gráficas completas de cenas, luz, câmeras, materiais e texturas
  • Motor de Colisões
  • Seleção de cenas
  • Antialiasing
  • Motor de animações
  • Sistemas de partículas
  • Sprites e layers 2D
  • Mecanismos de otimização
  • Materiais padrão no nível do pixel
  • Nevoeiro
  • Blending alpha
  • Teste alfa
  • Billboarding
  • Modo de tela cheia
  • mapas de sombra e mapas de variação das sombras
  • Renderização de texturas
  • Texturas dinâmicas (canvas)
  • Texturas de vídeo
  • Câmeras (perspectivas e ortogonalidade)
  • Clonagem de malha
  • Malhas dinâmicas
  • Mapas de altura
  • As cenas da Babilônia podem tornar-se. OBJ, .FBX, .MXB
  • Exporta a Blender

 

Abaixo está um vídeo curso muito bem explicadinho em espanhol deOscar Uh Pérez (Develoteca) Isso nos dá todas as habilidades necessárias para colocar as mãos na massa.

 

 

Também deixo alguns tutoriais interessantes de Julian Chenard para começar com Balylon.js no link a seguir: http://www.pixelcodr.com/

 

Mais tutoriais: http://www.babylonjs-playground.com/

 

Repositório

Babylon.js é um poderoso, lindo, simples, e jogo aberto e motor de renderização embalado em um framework JavaScript amigável.
https://github.com/BabylonJS/Babylon.js
3,686 forks.
26,085 estrelas.
22 questões em aberto.

commits recentes:
  • Fix WGSL vPartIndex redefinition when combining Gaussian Splatting material plugins (#18931)## What & whyThe Gaussian Splatting material plugins —`GaussianSplattingSolidColorMaterialPlugin`,`GaussianSplattingGpuPickingMaterialPlugin`, and`GaussianSplattingDebugMaterialPlugin` — each inject their own `varyingvPartIndex` via `CUSTOM_VERTEX_DEFINITIONS` /`CUSTOM_FRAGMENT_DEFINITIONS`. Attaching more than one to the samematerial declares `vPartIndex` twice, which is a **fatal `redefinitionof 'vPartIndex'` error under WGSL** (the duplicate `@location` varyingsfail to compile, so splats stop rendering); GLSL has the same duplicatedeclaration.This happens in practice whenever, por exemplo, theobject-ID/solid-color plugin and the debug plugin are both attached toone compound splat mesh.## FixAdded a shared `GetPartIndexVaryingDeclaration(shaderLanguage)` helperin `gaussianSplattingMaterial` that emits the `vPartIndex` varyingwrapped in an include guard (`#if !defined(…)` / `#define`). All threeplugins now declare the varying through it, so it survives exactly onceno matter how many part-index plugins are attached.## TestsAdded `gaussianSplattingPartIndexVarying.test.ts`:- helper emits guarded declarations for WGSL and GLSL;- two guarded declarations collapse to one after preprocessing;- control: two unguarded declarations stay two (the regression the fixprevents);- solid-color + GPU-picking combined declare `vPartIndex` once (WGSL +GLSL, vertex + fragment);- debug plugin routes through the shared guard.All Gaussian Splatting unit tests pass (183).Co-autoria de: Raymond Fei <[email protected]>, GitHub
  • Atualização de versão 9.27.1, Plataforma Babylon.js
  • Acionador de patch, Plataforma Babylon.js
  • Add normalized occlusion query visibility API (#18919)> 🤖 *This PR was created by the create-pr skill.*## SummaryAdd a backend-neutral `AbstractEngine.isOcclusionQueryVisible(consulta)`API for callers that need occlusion visibility without depending onbackend-specific query-result magnitude.- implements the portable visibility contract once as`getQueryResult(consulta) > 0`- keeps `getQueryResult()` unchanged for existing raw-result consumers- updates Babylon's mesh occlusion path to use the normalized API- documents both the raw and normalized contracts- tests zero, several positive values, and inheritance across the WebGLand WebGPU engine abstractions## Why a separate APIWebGL `ANY_SAMPLES_PASSED(_CONSERVATIVE)` results are effectively GLuintbooleans and commonly appear as `0` or `1`. Babylon's WebGPUimplementation converts the backend `u64` query value with`Number(…)`, so a visible result may be a positive value greater than`1`.Normalizing `getQueryResult()` itself to `0`/`1` or changing its returntype to boolean could break callers that consume the existing rawbackend value. The new method is additive and gives callers one portablevisibility contract without changing that behavior.Callers that only need visibility can migrate from backend-specificresult handling to:“`tsconst isVisible = engine.isOcclusionQueryVisible(consulta);“`The method belongs on the shared `AbstractEngine` query abstraction, soWebGL and WebGPU inherit the same interpretation. Babylon's own meshpath uses it as the canonical consumer. This PR does not changerender-pass creation or `beginOcclusionQuery` sequencing.## Validation- focused occlusion-query unit tests: 5 passed- `@dev/core` TypeScript compilation passed- targeted Prettier and ESLint passed- repository `format:check` passed- repository `lint:check` passed- repository unit suite: 5,682 passed, 1 expected failure; one unrelatedUMD test hit its 5-second timeout and passed when rerun alone (5/5)- prior devhost validation confirmed runtime registration through thequery side-effect module———Co-autoria de: Copilot App <223556219+[email protected]>, GitHub
  • Stop persisting invalid Nx cache in CI (#18925)> 🤖 *This PR was created by the create-pr skill.*## ContextMonorepo CI jobs have been completing with a silent post-job failurewhile Azure Pipelines saves the Nx computation cache. In build[59703](https://dev.azure.com/babylonjs/ContinousIntegration/_build/results?buildId=59703),startup cleanup reduced disk usage from 84% para 44%, but `Cache@2` laterrestored and attempted to archive a 19.5 GB Nx cache. The temporaryuncompressed tar exhausted the hosted agent's 72 GB OS disk.Nx 22 stores its cache database metadata outside `NX_CACHE_DIRECTORY`and keys it to the machine. Azure restored only the artifact directoryon each new hosted agent, so Nx reported the restored content asunrecognized and could neither use nor prune it. The configured 4 GBlimit therefore did not constrain the restored payload.## Change- Stop restoring and saving `NX_CACHE_DIRECTORY` through Azure`Cache@2`.- Retain npm and Playwright pipeline caches.- Retain the existing hosted-agent disk cleanup.- Retain Nx's bounded 4 GB local cache within each job.- Remove the obsolete Nx cache template parameters from the Build, ES6,and ES6Tools jobs.This removes both the unusable 19.5 GB restore and the duplicatepost-job tar write that caused the disk failure.## Validation- Parsed all changed Azure Pipeline YAML files successfully.- Verified no `Cache Nx computation`, `nxCacheKey`, or `nx: true`pipeline configuration remains.- `npm run format:check`- `npm run lint:check`- `npm run test:unit` — 5,658 passed, 1 expected failure, 64 skipped———Co-autoria de: Copilot App <223556219+[email protected]>———Co-autoria de: Copilot App <223556219+[email protected]>, GitHub
: 3D, Babylon.js, Framework, Game Engine, HTML5, JS, WebGL

Serviços

  • Correção de bugs em seu site Wordpress $40.00 / hora
  • Bot Tok $45.00
  • Correção de bugs em seu aplicativo PHP Symfony $50.00 / hora
  • Automação de tarefas usando Node.js $50.00 / hora
  • Desenvolvimento de API REST com Symfony $50.00 / hora

Blog

  • Como pagar com um cartão bancário em Cryptomus
  • Guia completo para iniciantes para o Bot Tok: Comandos do terminal explicados
  • Melhor site para obter visualizações no TikTok
  • Jfa Whatsapp Chatbot
  • Bot do TikTok

Explorar

  • Livre 10 Curtidas no TikTok
  • Visualizações 2K gratuitas do TikTok
  • Livre 100 Favoritos do TikTok
  • Livre 300 Compartilhamentos do TikTok
  • Comprar visualizações do TikTok
  • Livre 100 Curtidas no Instagram
Twitter
LinkedIn
YouTube
GitHub

© 2013-2026 Jordi Fernandes Alves (@jfadev)