# Few tips I learned as a total noob with minimax h3 — retained record

> Retained reddit record from reddit:StableDiffusion.

Canonical: https://highsignal.app/data/records/130c406771d723bd

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:StableDiffusion)  published 2026-11-01 17:46 ingested 2026-11-02 02:40

[original source ↗](https://www.reddit.com/r/StableDiffusion/comments/1xl9epi/few_tips_i_learned_as_a_total_noob_with_minimax_h3/)

## retained text



```
I'm finally kind of happy about my last video ( french ). Here's a few tips that are obvious but didn't know in order to avoid spending hours doing reshoots:

Work the audio first. Don't try to squeeze a line so that it fits a shot, it's the other way around.

Use distinguishable characters: in this video, I purposely added a hood to the male cyber scavenger as I was getting character swaps in the prompts with two people "having fluo shirts". Also, the scene having a red neon side and a green neon side helps for character placement.

Use a skill for prompt on top of the h3 prompting skill and add the rules you like ( don't look at the camera, use shallow depth of field, etc.)

This is ref2va, avoid using more than 3 ref passed on each shot. It dilutes attention and generates more errors 

That is all I can think of for now 😄
```



document fetched 2026-11-02 00:29 · canonical https://reddit.com/r/StableDiffusion/comments/1xl9epi/few_tips_i_learned_as_a_total_noob_with_minimax_h3
