Тёмный
No video :(

Game Design with Stable Diffusion Forge UI & Photoshop 

pixaroma
Подписаться 9 тыс.
Просмотров 16 тыс.
50% 1

In this tutorial, I'll show you how I design graphics for a Match-3 game called JunglePix using Stable Diffusion Forge UI and Photoshop. Learn how to create game elements like logos, loading bars, and backgrounds with these tools. Whether you're just starting out or already familiar with digital art, you'll find useful tips for using AI and Photoshop in your designs.
🔹 What You'll Learn:
- How to make a game logo in Photoshop.
- Setting up and using AI models like Juggernaut XL in Stable Diffusion Forge UI.
- Editing and finalizing game graphics in Photoshop.
🔹 Why Watch This Video?
- Get practical tips on using AI for game design.
- Learn simple Photoshop techniques for your projects.
- Find out how to handle common design challenges.
#GameDesign #PhotoshopTutorial #stablediffusion
Chapters:
00:00 Game Design Intro
00:14 Logo Design + ControlNet Settings
04:57 Loading Bar Design
06:13 Game Background Design
09:26 Grid Design
10:29 Button Design
11:56 Game Items and Icons
13:15 Mouse Cursor Design
Useful Resources
How to install Stable Diffusion Forge UI on Windows (Nvidia GPU)
• Stable Diffusion Forge...
Settings and Tips and Tricks for Forge UI
• Stable Diffusion Forge...
How to get 260+ Free Art Styles for Stable Diffusion A1111 and Forge UI (The styles.csv download link is on the pinned comment of that video)
• Instant Art Upgrade: 2...
In this video I am using the model: Juggernaut X RunDiffusion (version 10) from CivitAI
civitai.com/mo...
you download it and place it in the folder webui\models\Stable-diffusion
You can get more control net models from
huggingface.co...
In this video I am using the kohya_controllllite_xl_canny.safetensors
you download it and place it in the folder webui\models\ControlNet
Outpaint Tutorial for Forge UI
• Stable Diffusion Outpa...
Inpaint Tutorial for Forge UI
• Stable Diffusion Inpai...

Опубликовано:

 

15 авг 2024

Поделиться:

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист
Посмотреть позже
Комментарии : 83   
@pixaroma
@pixaroma 2 месяца назад
If you have any questions you can post them in Pixaroma Community Group facebook.com/groups/pixaromacrafts/ or Pixaroma Discord Server discord.gg/a8ZM7Qtsqq
@uniyuki8712
@uniyuki8712 3 месяца назад
Its always nice to see people actually using ai the way it should be used, like any other art tool!
@SyntheticSurrealism
@SyntheticSurrealism 3 месяца назад
Thanks for going beyond the usual Stable Diffusion tutorials and showing how AI is helpful for designers. Honestly, AI is just another tool in a our toolbox. Also showing the problems and errors you had is just as educational as the technique itself! As a game dev I found this really interesting, and while I do use some of these techniques, you've taken it further and given me something to think about. Please give us more vids like this!
@johnlin6121
@johnlin6121 3 месяца назад
This is ridiculously useful. Haven't seen a video that valuable for a while.
@natalietarson
@natalietarson 2 месяца назад
This is AWESOME! Thank you!
@domovoi_0
@domovoi_0 3 месяца назад
Legend. Thank you. Love and blessings!
@RonMar
@RonMar Месяц назад
I think this is a fantastic idea! I'm new to the SD world and have been so lost in the weeds that it never even occurred to me to use SD to make menu assets. Clearly, SD has much to offer here.
@rundiffusion
@rundiffusion 3 месяца назад
We're glad you enjoy our model Juggernaut! Great video! (Edited out my comment from earlier! Sorry about that! You’re awesome!)
@pixaroma
@pixaroma 3 месяца назад
Hi, yes is a very good model, as for renaming I didnt change the name, is the same name you get when you download it from civitai civitai.com/models/133005?modelVersionId=456194 click download and you will see it has the name juggernautXL_juggernautX.safetensors so maybe who uploaded it on civit should rename it, and I gave the link from where is download and there is mentioned rundiffusion in the title and description. I mean I didnt think that I should rename the models what is downloaded I just keep them as it is so people who follow tutorial can check if they have the same model file even if I dont mention. Hope that make sense. Thanks
@rundiffusion
@rundiffusion 3 месяца назад
@@pixaroma woah! I’m so so sorry! Civitai changed how they automatically name the safetensors file. I’ll reach out to them. I should have double checked. I’m so sorry for accusing you. I must look like a total jerk. Please forgive me!
@rundiffusion
@rundiffusion 3 месяца назад
@@pixaroma edited my comment. Again I’m so sorry. I’m in contact with Civitai to see why they’re renaming files.
@pixaroma
@pixaroma 3 месяца назад
@rundiffusion all good don't worry about ☺️
@rundiffusion
@rundiffusion 3 месяца назад
@@pixaroma Mind if we touch base and see if we can partner on anything? We offer Auto1111 in the cloud and we have contacts with businesses and gaming companies that would love to learn from you! Obviously there would be compensation involved. I really feel like we got off on the wrong foot. I hope I can make it better.
@An0therR0gue
@An0therR0gue 2 месяца назад
just came across your channel today. i am loving it. i love the great practical example of how to use SD for a specific purpose (game design).
@AIGrasp
@AIGrasp 3 месяца назад
Beyond the Forge and Stable Diffusion, you are talented.🎨
@pedroserapio8075
@pedroserapio8075 3 месяца назад
Really super cool to see AI tools being used together with "traditional" tools to get the job done.
@fotocube
@fotocube 3 месяца назад
Oh my!! Thank you for sharing, you just open a hole new world for me❤
@nuwanchandrasirigamage8130
@nuwanchandrasirigamage8130 3 месяца назад
Excellent tips 💯
@valentine_shirin
@valentine_shirin 3 месяца назад
Cool! That's so powerfull lesson. Thanks you!
@GuysLightburnLaserGuides
@GuysLightburnLaserGuides 3 месяца назад
Bro you just blow my mind.
@StanyBodson
@StanyBodson 3 месяца назад
Trop sympathique de ta part. Ce tutoriel est un très bon résumer compréhensible, d'un travail qui auparavant prenait des heures à être réalisé. Merci. Demain (dans le futur) il y aura peut-être un tuto qui expliquera comment faire un résumé vidéo comme celui-ci, en 3 minutes. ;) So nice of you. This tutorial is a very good, understandable summary of a job that used to take hours to do. Thank you. Tomorrow (in the future) there may be a tutorial explaining how to make a video summary like this, in 3 minutes ;)
@edge-rps
@edge-rps Месяц назад
this is insanely cool, slot machine owners are foaming at the mouth with rabies at the sight of this.
@southcoastinventors6583
@southcoastinventors6583 3 месяца назад
I used a similar work flow to you but I started with a logo generated from ideogram as it can do text logos really well in 2-3 shots plus you get 25 prompts of 4 image outputs free a day. Good Video
@CyborgMedia
@CyborgMedia 3 месяца назад
RU-vid Algorithm, please, show this channel's videos to more people. This person deserves it!!! One of the thing I like about your videos is that you always show how we can use AI in a real scenario. Please, keep that "context" in every video you make. PS. You could also add videos on how to use Photoshop like you do it. Not a tut, but how to use the tools you did in a video. Maybe adding a link to the description so we can learn that part too. I know you do this as a hobby and maybe I'm asking for too much, but don't blame me for trying. :)
@pixaroma
@pixaroma 3 месяца назад
A photoshop tutorial is older, with not so good voice, for removing the backgrounds in photoshop ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-mbKiExIMDLo.html and for more photoshop related i have some really old videos on other channel that I dont use anymore, 7 years old but technique is the same ru-vid.com/group/PLhK9RlyNiUzPm5a3VJ-60Vi6Mg60vhIB1 ru-vid.com/group/PLhK9RlyNiUzNU0m7rQt4lQ0gt5WoXkt8f and some for vectors ru-vid.com/group/PLhK9RlyNiUzOLgDfOEPJBLn5j9LjkKUro
@CyborgMedia
@CyborgMedia 3 месяца назад
@@pixaroma Thanks!
@NAKOOT
@NAKOOT 3 месяца назад
It's insane! Thanks dude!
@несмотринаменятак
@несмотринаменятак 3 месяца назад
wow, this is amazing! I really liked these gems 😍 12:23
@caseyj789456
@caseyj789456 3 месяца назад
Very clever and inspiring to watch. i prefer comfyui but it same topic. Thx !
@SumoBundle
@SumoBundle 3 месяца назад
Very nice. I like the graphics.
@ayaz1979
@ayaz1979 3 месяца назад
Amazing, thanks
@konnstantinc
@konnstantinc 3 месяца назад
Best tut so far :D
@crypt0w1tch
@crypt0w1tch 3 месяца назад
This is incredible! You definitely just blew my mind and opened up so many possibilities. Thank you. 🙏
@The_MegaClan
@The_MegaClan 3 месяца назад
beautiful
@Mixogen
@Mixogen 3 месяца назад
WOW. Cool
@reliquiadomilenio
@reliquiadomilenio 14 дней назад
🤯
@saqibahmadofficial
@saqibahmadofficial 3 месяца назад
Nice!!
@manolomaru
@manolomaru 3 месяца назад
✨👌😮😮😮👍✨
@voyageroftheworlds
@voyageroftheworlds 3 месяца назад
Here before 100k ✨
@chucktobbes
@chucktobbes 3 месяца назад
You are very skilled with forge. I have a challenge please try to generate pixel art game ready assets in the style of the game stoneshard (top down dark fantasy). I have tried myself with mixing some loras but can't quite get it right.
@pixaroma
@pixaroma 3 месяца назад
I think is harder to get a specific art styles if is not trained on that styles, probably could get something if a lora is trained on that specific style. And from my test ia hard to get a consistent pixel art style only from prompts
@chucktobbes
@chucktobbes 3 месяца назад
@@pixaroma yes, with just prompting its impossible to get good pixel art from SD. You need to use some kind of pixel art lora and the sd-palettize script from Astropulse for post-processing.
@phat80
@phat80 3 месяца назад
He’s not skilled because he doesn’t know how to use control net. He’s definitely not aware of the settings.
@pixaroma
@pixaroma 3 месяца назад
I am aware of the settings 😂 this wasn't a tutorial for control net,i just use what works for me, and default worked for me most of the time, and for preprocessor i prefer canny, even the one with inverted BW can work i didnt want the strokes that something gives, the power of control net i use only when Denoise strength doesn't do a good job, i mean you can do the same things in different modes, what matters is the results and I got what i need it ☺️ you can combine different control net models and can put when to start and when to end but it wasn't the case. So just because i didn't use all the settings doesn't mean i am not aware of them
@phat80
@phat80 3 месяца назад
@@pixaroma You don’t need the preprocessor, that’s the point! Your drawing is a canny itself. Canny is an outline representation of the image. You draw it yourself. Preprocessors in control net are used for extracting data from images. For example, if you have a photo and want to use canny model you need to enable the preprocessor. Otherwise when you provide an appropriate image you should set None for the preprocessor.
@hookerokx
@hookerokx 3 месяца назад
good job!
@Arrrr-gj6kk
@Arrrr-gj6kk 2 месяца назад
4:51 could you please share the prompt that generated that text?
@pixaroma
@pixaroma 2 месяца назад
I dont usually save prompts since I just talk with chatgpt and explain what I want. So what you can do is to describe what game you want to design, and ask chat gpt to give a list of everything you need to design for that game, all the screens, buttons, menus, and usually it give you something, after it responds you can ask more, or ask to change something, guide it until you get what you want
@Arrrr-gj6kk
@Arrrr-gj6kk 2 месяца назад
​@@pixaroma I got you, thanks for awesome videos you're making for us!
@overskills8175
@overskills8175 3 месяца назад
I'm game designer to respect pro
@shawnnosaurus
@shawnnosaurus 3 месяца назад
How did you create the animations for the menu and level startup with the gems coming into place?
@pixaroma
@pixaroma 3 месяца назад
Is video editing in capcut, you can do some manually with key frames but you can also select an image and to to animation and you have animations for in and out, so i can select a style for intro animation on how to appear on screen and how to exit. Caput for desktop is free, there is a pro version that offers a few more extra versions and animations
@Sirkayne
@Sirkayne 3 месяца назад
The video is absolutely brilliant 👍👍👍👍👍😎 Thank you very, very much for that. Could you possibly make another video where you create a skybox with Automatic1111? Similar to how it works at Blockadelabs? That would be cool. 🙏
@pixaroma
@pixaroma 3 месяца назад
I dont have to much experience with skybox but Olivio Sarikas has a similar video maybe that can help, ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-utBH8ouGQJE.html
@alex.nolasco
@alex.nolasco 3 месяца назад
I can only wonder how many hours you saved... thanks for the demo
@centurionstrengthandfitnes3694
@centurionstrengthandfitnes3694 3 месяца назад
I didn't like this. I loved it! Great content, Pix.
@videosfeoscomotucara9038
@videosfeoscomotucara9038 3 месяца назад
Gold
@JarppaGuru
@JarppaGuru 3 месяца назад
7:48 lets change one line on python files code that is universal. works like before but works when this happen. line that should be there allready
@DevgutP
@DevgutP 11 дней назад
Hello! What version of SD do you use in the video? Thank you 🎉
@pixaroma
@pixaroma 11 дней назад
The interface is Forge UI a commit version that start with 29. The model is sdxl, more precise the juggernaut x version it has link in description
@DevgutP
@DevgutP 11 дней назад
Thanks a lot!
@pixaroma
@pixaroma 19 часов назад
check this video How to Install Forge UI & FLUX Models: The Ultimate Guide ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-BFSDsMz_uE0.html
@PlayGamesWithX
@PlayGamesWithX 3 месяца назад
Hi, will it work with Radeon Pro 580 Graphics 8GB or it only work with NVidia? What is your PC specs that you used in this video?
@pixaroma
@pixaroma 3 месяца назад
From what I know it search for an nvidia driver where is installed so I will say it doesnt work with amd. I have this: - CPU Intel Core i9-13900KF (3.0GHz, 36MB, LGA1700) box - GPU GIGABYTE AORUS GeForce RTX 4090 MASTER 24GB GDDR6X 384-bit - Motherboard GIGABYTE Z790 UD LGA 1700 Intel Socket LGA 1700 - 128 GB RAM Corsair Vengeance, DIMM, DDR5, 64GB (4x32gb), CL40, 5200Mhz - SSD Samsung 980 PRO, 2TB, M.2 - SSD WD Blue, 2TB, M2 2280 - Case ASUS TUF Gaming GT501 White Edition, Mid-Tower, White - Cooler Procesor Corsair iCUE H150i ELITE CAPELLIX Liquid - PSU Gigabyte AORUS P1200W 80+ PLATINUM MODULAR, 1200W
@PlayGamesWithX
@PlayGamesWithX 3 месяца назад
@@pixaroma thank you!
@PlayGamesWithX
@PlayGamesWithX 2 месяца назад
Hi, Is it possible to use the same workflow like this but not with Stable Diffusion, instead I will use Midjourney, because I don't have the powerful PC like that?
@pixaroma
@pixaroma 2 месяца назад
​@@PlayGamesWithX i didn't use mid journey in the last months, i know it had something with image weight but not sure if you transform your sketches, you can give it reference images of a button and items maybe and prompt something similar, and use maybe describe to get prompts but not sure if you get the same controll, you will get maybe somehow similar, hard to tell
@shabnova
@shabnova 16 дней назад
а как добавить стиль 3d game assets?
@pixaroma
@pixaroma 16 дней назад
Check this video for more art styles ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-UyBnkojQdtU.htmlsi=lK0y9mckZc2F4LQU
@evelnogueira3112
@evelnogueira3112 3 месяца назад
Do you know a way to use sdxl lora with Forge?
@pixaroma
@pixaroma 3 месяца назад
I used sdxl loras they work with sdxl models and you put it in the Lora folder. Then you use trigger words for that lora and select it. Check this video i used a Lora there ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-q5MgWzZdq9s.htmlsi=B5GDXrqzW9PekWKE
@evelnogueira3112
@evelnogueira3112 3 месяца назад
@pixaroma I get an error every time I use an SDXL Lora, I will watch the video to see what I'm doing wrong. Thanks.
@pixaroma
@pixaroma 3 месяца назад
Depends of the error i guess. Does it work if image width and height is 1024*1024 px or what error you get?
@evelnogueira3112
@evelnogueira3112 3 месяца назад
@pixaroma I'm generating a 1024x1024 image when I insert a lora in the prompt I get a none type error.
@evelnogueira3112
@evelnogueira3112 3 месяца назад
@@pixaroma I made this video showing the error: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-h-Ua-oPmJFU.html
@ZZWWYZ
@ZZWWYZ День назад
5:20 Amogus
@pixaroma
@pixaroma День назад
😃
@Lysdexi
@Lysdexi 3 месяца назад
bro even got the AI voice
@pixaroma
@pixaroma 3 месяца назад
Yeah, I kind of like this Ai voice ☺️
@jeronimolopes
@jeronimolopes 3 месяца назад
the designers are getting mad
@pixaroma
@pixaroma 3 месяца назад
Are both good things and bad things like with any technology, I am a designer too, I cannot say I am mad but definitely I was affected, the sales dropped a lot since AI came but I have to adapt and work with what I have
Далее
Будзек и рецепт🐝
00:25
Просмотров 130 тыс.
Stable Diffusion Prompt Guide
11:23
Просмотров 24 тыс.
I Learned Blender in 1000 Hours
8:51
Просмотров 18 тыс.
choosing a game engine is easy, actually
15:08
Просмотров 418 тыс.
Best Alternative to Every Adobe Program
18:57
Просмотров 230 тыс.
Будзек и рецепт🐝
00:25
Просмотров 130 тыс.