ai:start

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision		 Previous revision
ai:start [2026/04/01 23:19] – [Atlas] mhai:start [2026/05/27 17:28] (current) – [Style] mh
Line 14: Line 14:
  
 <code txt> <code txt>
-an old worn book with a thick leather cover, slightly weathered edges and visible page layers, simple embossed details on the cover, resting on a flat surface, closed, low angle shot, soft ambient light, clean composition, neutral background, realistic materials, balanced and readable structure+an old worn book with a thick leather cover, slightly weathered edges and visible page layers, simple embossed details on the cover, resting on a flat surface, closed, low angle shot, realistic materials,  
 +</code> 
 +<code txt> 
 +soft ambient light, 
 +clean composition, 
 +balanced and readable structure, 
 +neutral background,
 </code> </code>
  
Line 74: Line 80:
     * //Tests whether layout instructions override entropy.//     * //Tests whether layout instructions override entropy.//
   * **Visual hierarchy conflict :** dense environment, subject remains dominant, reduced emphasis on subject   * **Visual hierarchy conflict :** dense environment, subject remains dominant, reduced emphasis on subject
 +
 +  * **[[ai:atlas-prompts|Atlas Prompts]]**
 ==== Model List ==== ==== Model List ====
  
   * **[[ai:nanobanana|nanoBanana 2]]**   * **[[ai:nanobanana|nanoBanana 2]]**
  
-===== Model specifics =====+  * **[[ai:flux2|Flux Pro 2]]**
  
-  ***[[ai:firefly|Firefly]]**+  * **[[ai:firefly5|Firefly 5]]** 
 + 
 +  * **[[ai:runway4|Runway Gen4]]** 
 + 
 +  * **[[ai:gptimage1p5|GPT Image 1.5]]** 
 + 
 +  * **[[ai:gptimage2|GPT Image 2]]** 
 + 
 +  * **[[ai:ideogram3|Ideogram 3]]** 
 + 
 +  * **[[ai:midjourne|MidJourney]]** 
 + 
 +===== Model specific Workflows =====
  
   ***[[ai:midjourney|MidJourney]]**   ***[[ai:midjourney|MidJourney]]**
  
  
-===== Tools =====+===== Other Models =====
  
   ***[[https://runwayml.com/|Runway]]** : Video generation and Manipulation   ***[[https://runwayml.com/|Runway]]** : Video generation and Manipulation
Line 93: Line 113:
   ***[[https://www.openai.fm/|OpenAI FM]]** : Text to speech+   ***[[https://www.openai.fm/|OpenAI FM]]** : Text to speech+
  
 +  ***[[https://lumalabs.ai/|Luma AI]]** : video + 3D + multimodal creation
 +
 +===== AI Services =====
 +
 +  ***[[https://lambda.ai/|Lambda AI]]** : Pay-per-use AI Infrastructure (GPU superpower)
 ===== Image Generating ===== ===== Image Generating =====
  
Line 110: Line 135:
  
   * **S23 Ultra:** 1440x3088: 9:19.5 or 9:20   * **S23 Ultra:** 1440x3088: 9:19.5 or 9:20
 +
 +==== Common choices ====
 +
 +  * 4:5
 +  * 9:16
 +  * 21:9
 +  * 1:1
 +  * 2:3
 +  * 4:3
  
 ===== Camera Views ===== ===== Camera Views =====
Line 133: Line 167:
 |  //Less vertical than bird’s-eye, more **cinematic landscape**//  |  //Works **surprisingly** well even in stills//  |    |    | |  //Less vertical than bird’s-eye, more **cinematic landscape**//  |  //Works **surprisingly** well even in stills//  |    |    |
 ===== Style ===== ===== Style =====
 +
 +The style list can be divided into two complementary layers: 
 +
 +  * **medium styles**, which describe ''how'' an image is rendered
 +  * **aesthetic / genre styles**, which describe ''what kind of visual world'' or artistic direction the image tends toward. 
 +
 +This split is practical rather than absolute, but it helps clarify what exactly is being tested.
 +
 +==== Medium styles ====
  
   * Photographic / Photo realistic   * Photographic / Photo realistic
-  * Anime 
-  * Neon Punk 
   * Oil Paint   * Oil Paint
   * Watercolor   * Watercolor
-  * Fantasy Art 
   * Pencil Sketch   * Pencil Sketch
   * Chalk Pastel   * Chalk Pastel
-  * Digital Art+  * Digital Painting 
 +  * 3D Render / CGI 
 +  * Vector Illustration 
 +  * Ink / Comic Illustration 
 + 
 +==== Aesthetic / Genre styles ==== 
 + 
 +  * Anime 
 +  * Neon Punk 
 +  * Fantasy Art 
 +  * Cinematic / Film Still 
 +  * Sci-Fi Concept Art 
 +  * Storybook Illustration 
 + 
 +==== Notorious medium + aesthetic / genre mixes ==== 
 + 
 +  * **Digital Painting + Fantasy Art** 
 +    * produces polished concept-art imagery 
 +    * useful for fantasy characters, environments, props, and worldbuilding visuals 
 + 
 +  * **Photographic / Photo realistic + Cinematic / Film Still** 
 +    * produces movie-like staging, lighting, and color grading 
 +    * useful for realistic scenes with strong atmosphere and narrative framing 
 + 
 +  * **3D Render / CGI + Sci-Fi Concept Art** 
 +    * produces hard-surface futuristic visuals 
 +    * useful for vehicles, robots, environments, props, and technology-heavy scenes 
 + 
 +  * **Watercolor + Storybook Illustration** 
 +    * produces soft, gentle illustrated scenes 
 +    * useful for fairytale imagery, children’s-book aesthetics, and whimsical environments 
 + 
 +  * **Ink / Comic Illustration + Neon Punk** 
 +    * produces bold graphic cyberpunk imagery 
 +    * useful for stylized characters, posters, high-contrast scenes, and urban neon compositions 
 + 
 +  * These mixes are useful because they reveal: 
 +    * whether a model can imitate a style 
 +    * whether it can combine rendering language and visual genre coherently
  
 ---- ----
  • ai/start.1775078349.txt.gz
  • Last modified: 2026/04/01 23:19
  • by mh