Concept art using AI generation

Hi! If you're new to my Patreon, please know that I usually upload rewards between the 18th - 26th of the Month. This Patreon page charges per creation whenever you see the ✅ icon in the title of the post. Thank you for your support to continue to create exclusive content regarding anime creation in Blender. 

I've been wandering through other (Patreon) pages, and I guess our community gets a lot of rewards in technical and artistic terms in this Patreon page. Even if you don't pledge as a new supporter, you still have access to a lot of previous rigs and shaders I've built. It's mind blowing how many people come into this Patreon and then delete their pledges between days 10-18th of each month (but they still have access to the posts!). It is not fair for everyone else who pledges; which is why starting November, I will email the monthly pledged rewards, individually. Your support keeps amazing tutorials, exclusive video, models and shaders for Blender being made for you.

IF YOU KNOW, I started writing the Fixy universe back on this post. Today, I wanted to share these ai images generated from Midjourney (I quit Netflix's $13 a month, to get Midjourney $10 a month 😁). 

I applied for Dalle1,2 but so far all I've been able to generate is unusable images.
I've been trying StableDiffusion GUI locally with my PC and it works "well"
I've registered with MidJouney in their Discord server (you have 50 free images to start) and I can say it's a well-stablished community and service.

"Blue girl full body electronic dress and white short hair, dystopian Blederunner city brutalism::4, small red lights, night, snow::4, foggy, glowy, --ar 16:9"

In all ai services, you'll use keywords which are called "prompts". These prompts give the image generation parameters for the "weights" (aka tensors) to dominate the image. When you use a double colon, you can "override" the weight of the trait you want to use with your image like "snow::4" to give it more priority than other promps in the same sentence, that way the "model" (aka training file) knows what to create. Usually the ai will decide the order of the elements by the arrangement you use. The farthest prompts from the first words will usually generate "weaker" results. The ai doesn't know what to "prioritize" and it will need you to feed the information and their weights.

I will create a full blog post with more details about ai generation services, word prompts and how to be proficient when generating ai art. I'll post a full article at https://3dcinetv.com/blog so subscribe there to get automatic notifications.

As always, if you have any questions, please let me know in the comment section below.

-Cheers!

Pierre.