The most professional SD explainers out there... I feel like I am watching the program "How's Its Made" With videos like this I can finally retire from making AI explainer videos and get back to staring at my invisible children while they dream about careers in the tiktok or finally making it to head of the Text Prompt Engineer firm alongside other fictional professions of the future...
@@fictitiousness my hypothesis is universal income inevitability to lessen the strain on humans achieving obsolete status to the point where a reward system will need to be initiated to feel relevant in a society illusion laced with instant gratifications and cerebral devolution due to chaotic stimulus overload... Run-on sentences will finally become acceptable grammar so I will still be relevant at the Amazon consumption factory... hahaha...
Part 3 please!! This is hands down the best tutorial series I've seen on SD! I've been experimenting all by myself, this has helped me break down so many of the parameters. So excited for seeing your breakdown of math in prompts!
Honestly, if you enjoy making these kinds of tutorials - you may have found your calling. Or at least something that you're casually better at than 95% of the people trying to do it. Truly great work on these tutorials. It leaves me truly excited to see what else you share.
Ти найкращій, я вже 5 днів сам тести проводжу, а тут твій канал, привіт тобі з України You're the best, I've been doing tests myself for 5 days now, and here's your channel, greetings from Ukraine
Thank you so much for your input with your first two tutorials. Looking forward to part 3 and perhaps part 4 describing the video input settings. Much love!
I am new to Deforum, and little experience with diffusion apart from text to image, for me this is brilliant. Clear examples and explanations of each setting. Thank you
Thank you again. In part 3 I would love to see how to morph between predetermined seeds. I have been trying to use seed schedule to guide a video through images I already know look good. I'm just not smart enough to make it work. Thanks! You are doing a great job!
Hey, i see a year ago you mentoined part 3 is soming. Still working on it? I have been looking for such a tutorial as yours for days. So happy I found these. I've read trough all the comments to see if anything new has been said by you about part 3. Very curious if it's still coming. Thanks for your work!
@@iamYork_ I learned alot from you all and we need more generous people like you. I look forward to see your art. Im still trying to get the hang of it, the first frame is always sharp and then it turns blurry haha Ill figure it out though👍
@@neokortexproductions3311 Thank you... As one of the first channels to make Stable Diffusion Tutorials it is nice to see so many people making them now... Especially high end videos like this one... I just made them to help the community and have a bit of fun... I never expected so much positive feedback... I will possibly still make more but Im definitely planning on taking my channel back to more creative pieces with a tutorial here and there when I can...
Amazing, just amazing. Thank you. You might changing people's life here with the incredible quality level of your education. Appreciate that. Respect !
This is so helpful. This puts so much perspective on what's happening is such a visually helpful way. Thank you for putting so much time into explaining this!
I hardly ever gives thumbs up on youtube vids. But yours (OP) is so very well made. If you have the energy to continue the series then please do. I know there is lots of other (new) toys to play around with. heh - keep up the good work.
Super helpful video, thanks @fictitiousness! Do you know what has changed with motion values recently? My deforum extension is up to date, but the movement easing is not working as it was for me. Instead of easy ease movement change I got sharp shift when it switches from one value to another.
This Deforum series is fantastic. Why did you stop? Is it because the technology is leaping forward and changes are coming so fast that you feel that your efforts will be outdated too soon? If that's the reason, I can understand. But even with that fact, these videos are still super useful!
Hello, how can I decrease the zoom and then put the angle of view from top to bottom?
ปีที่แล้ว +1
Really great tutorial! I have two more simplistic questions: 1. how can I calculate in advance how many seconds an animation will last? I would like to put music, text with a given time span underneath. 2. basically, how should I proceed if I want the animation to have as little flicker and bounce as possible? I see that eye-catching vibration is a common problem. Thank you!
Hey, did you watch the first video? Both these topics are covered there :) Check it out and let me know if you still have questions after
ปีที่แล้ว +1
@@fictitiousness I must be clumsy, I watched both videos :-( . But English is not my mother tongue, and I am a beginner in this field. It would be nice of you to write the answer or tell me at what seconds I get the answer, I will try to understand it from your speech.
No Problem, the length part is at minute 7:28 and the "flickering" part starts from minute 8:46 but its part of the animation and you can hardly avoid it
Probably next week, wanted to finish it much earlier but kind of busy with important stuff right now. Music syncing will be covered in another part as well :)
Amazing vids! Maybe i missed something, how would you use a picture i have to start with for deforum stable diffusion? Is it just by like pasting a link within the prompt of what i want it to look like?
Thanks! There's different approaches, either with a source image and prompting or just through prompting on its own. Will cover Prompting in the upcoming video.
excellent :)!! thanks for making such great informative video's. If you are able to, could you please go over how to properly use the seed scheduling ? I have not been able to get that to work. thanks again :))
Sure, as explained in the previous video, you need to set the seed_behavior to "schedule" and can basically determine which seed to use as a fixed seed at any point of the video. For example: 0 (555), 10: (777), 50 (888) will use seed 555 from frame 0-9, 777 from seed 10-49 and seed 888 from seed 50 till the end of the animation. Hope that's more clear :)
Thank you for your deep and clear tutorials. I have a question: I use Mac and when I generate FFMPEG video it says "Error : FFmpeg not found. Please make sure you have a working FFmpeg path under “FFmpeg location” parameter.” How can i fix that?
@@fictitiousness I’ll say this those math functions got me stumped so I’ll be waiting for this next one!Appreciate the effort of these videos there great!
I am interested in car body design and I need to produce orthogonal views of a vehicle (front, side, rear and top). Do you know if there is any Stable Diffusion extension that allows me to generate these views/images based on a render I already generated? My idea is to use these four views as a blueprint to make the 3D CAD model in Solidworks.
Hey great video!! Can you please tell me why in 3d mode, I haven't the rotation_3d_x, rotation_3d_y parameters ecc.. I just have Translation X and Translation Y, and when I try to start, give me an error. Thank you anyway
I was wondering if you can share where to put video init files when running this stuff locally on a Mac? I don't know where the content folder is, or where to put my init video files that I want to use:(
Sure, just create a folder in your root sd-webui called "input" and name your video "input.mp4". Then you just enter the path in the "video_init_path" field like this: input/input.mp4. This should work, i think the main issue is the first "/" in the default path example which will give you the error. This is how it works on Windows but must be similar on Mac.
@@fictitiousness do you offer any 1 on 1 help over discord by chance? I’ve been at this for days trying various promps and settings and it’s just not working out that well for some reason. Seems like when I don’t have a Zoom, the images corrupt themselves. And also the init video or init image also doesn’t seem to stick :(
Thank you so much! Has anyone else noticed a difference with the latest SD 2.0? I can't get any smooth transition between frames at 768X768. Even the default prompt settings are generating random images between frames.
No, that's a Windows/Linux Application and you need a Nvidia Graphics card for that to work. As a Mac user i would suggest using a cloud GPU from vast.ai or runpod.io, you can find lots of tutorials on other channels on how to get started.
hey guys.. i want to add an Image Init to Deforum in A1111 but i get always this error: Error: 'Init image path or mask image path is not valid'. Check your schedules/ init values please. Also make sure you don't have a backwards slash in any of your PATHs - use / instead of \. Full error message is in your terminal/ cli. Image path is: C:/webui/images/01.png i tried it with backslash, upload the image on the web and use a web link, tried another image, another path.. nothing work :(
yes, with a lot of testing its possible, or at least you can come close. I've seen some examples in the discord channel but that was some months ago, impossible to find :)
they should be saved in the same folder, where your frames are saved: "outputs/img2img" and depending on what you have set in your settings either a gif or an mp4
![[UNCUT]"ทนายตั้ม" แฉยับ! ความลับสวรรค์ ใครคือ "บิ๊กเทวดา" ของบอสพอล l คนดังนั่งเคลียร์ l 15 ต.ค. 67](http://i.ytimg.com/vi/ru9fpSaUwdA/mqdefault.jpg)
Part 3 please!!! Really great info!! Thank you so much!
hope to see Part 3 ❤
Thank you for the information.
The most professional SD explainers out there... I feel like I am watching the program "How's Its Made" With videos like this I can finally retire from making AI explainer videos and get back to staring at my invisible children while they dream about careers in the tiktok or finally making it to head of the Text Prompt Engineer firm alongside other fictional professions of the future...
Lmao 😂 That's so dystopian but not very far from reality
@@fictitiousness my hypothesis is universal income inevitability to lessen the strain on humans achieving obsolete status to the point where a reward system will need to be initiated to feel relevant in a society illusion laced with instant gratifications and cerebral devolution due to chaotic stimulus overload... Run-on sentences will finally become acceptable grammar so I will still be relevant at the Amazon consumption factory... hahaha...
You deserve tons of viewers. the quality of this video is impressive.
Hope to see a part 3 from you man, great work!!
Part 3 🏆🏆🏆
Holy crap, a youtube video on a.i that actually gives real information
PART 3 PLEASE! THIS IS SO AMAZING
very well explained,few people got this talent to deliver complex info into kids language
The best tutorial I saw so far! You shouldn’t stop 😭😇
haha i won't :)
always come back to this video when ive forgotten some info about deforum. ty fam
How it's Made: Deforum Animations
Lovin the classic How it's Made narration vibe. Absolute legend. Let's get that part 3 bumpin
Part 3 please!! This is hands down the best tutorial series I've seen on SD! I've been experimenting all by myself, this has helped me break down so many of the parameters.
So excited for seeing your breakdown of math in prompts!
I seriously can't wait for part 3, this is a fantastic series with amazing editing quality. Thank you for your work!
I seriously can't wait for part 3
Honestly, if you enjoy making these kinds of tutorials - you may have found your calling. Or at least something that you're casually better at than 95% of the people trying to do it.
Truly great work on these tutorials.
It leaves me truly excited to see what else you share.
Ти найкращій, я вже 5 днів сам тести проводжу, а тут твій канал, привіт тобі з України
You're the best, I've been doing tests myself for 5 days now, and here's your channel, greetings from Ukraine
Videos of unprecedented professionalism! If you can, please part 3! 🙂
i think we are ready for part 3 :)
I totally need part 3 right now!!!!
Thank you so much for your input with your first two tutorials. Looking forward to part 3 and perhaps part 4 describing the video input settings. Much love!
I am new to Deforum, and little experience with diffusion apart from text to image, for me this is brilliant. Clear examples and explanations of each setting. Thank you
this was an awesome series.. cant wait for part 3 🙌🏿
Part 3 broo! Your explanation is great!
dude, you rock! best tutorial for deforum hands down. Please keep it going
Thank you again. In part 3 I would love to see how to morph between predetermined seeds. I have been trying to use seed schedule to guide a video through images I already know look good. I'm just not smart enough to make it work. Thanks! You are doing a great job!
You've been more helpful than my high school math teacher
I am looking forward to part 3!
Hey, i see a year ago you mentoined part 3 is soming. Still working on it? I have been looking for such a tutorial as yours for days. So happy I found these. I've read trough all the comments to see if anything new has been said by you about part 3. Very curious if it's still coming. Thanks for your work!
We all appreciate you! The quality of videos you put out are extremely impressive!🙏🏼
Thank Odin... with these sort of videos out there I can finally retire and get back to making creative videos again...
@@iamYork_ I learned alot from you all and we need more generous people like you. I look forward to see your art.
Im still trying to get the hang of it, the first frame is always sharp and then it turns blurry haha Ill figure it out though👍
@@neokortexproductions3311 Thank you... As one of the first channels to make Stable Diffusion Tutorials it is nice to see so many people making them now... Especially high end videos like this one... I just made them to help the community and have a bit of fun... I never expected so much positive feedback... I will possibly still make more but Im definitely planning on taking my channel back to more creative pieces with a tutorial here and there when I can...
Amazing, just amazing. Thank you. You might changing people's life here with the incredible quality level of your education. Appreciate that. Respect !
I need the part 3 this is amazing
Thank u :)
Thanks 😀 I finally found something I could use. Please make more tutorials. Maybe about prompts?
Will do, just need to prepare some stuff :)
I came up with everything myself, but this video is great. Clear and cool, thank you.
This is so helpful. This puts so much perspective on what's happening is such a visually helpful way. Thank you for putting so much time into explaining this!
Great video, Helped me a lot, thank you brother.
Man, this is so well explained! thanks for taking all that time making that video. helped me a lot!
Awesome tutorials, part 3 please. Thy:)
Amazing thanks for this!! Definitely the best tutorial out there! When is the next video coming out?
Thank you so much! looking forward to the next one.
I'm getting a Error: '"log_vml_cpu" not implemented for 'Half''. Any ideas on how to fix?
Mannnn we need third part!
It's coming, needed to take a break on holiday season, but back to work now :)
nice !! i like you have video examples how settings gonna look.
Thank you very much!!!! Yua are grate !!!! I wish you all the best!!!
Fantastic video fam. Thank you
I hardly ever gives thumbs up on youtube vids. But yours (OP) is so very well made. If you have the energy to continue the series then please do. I know there is lots of other (new) toys to play around with. heh - keep up the good work.
Check out some of my Daily Animation Challenge in here : th-cam.com/video/weMibSCA7Ag/w-d-xo.html&ab_channel=Ermando
Super helpful video, thanks @fictitiousness! Do you know what has changed with motion values recently? My deforum extension is up to date, but the movement easing is not working as it was for me. Instead of easy ease movement change I got sharp shift when it switches from one value to another.
This Deforum series is fantastic. Why did you stop? Is it because the technology is leaping forward and changes are coming so fast that you feel that your efforts will be outdated too soon? If that's the reason, I can understand. But even with that fact, these videos are still super useful!
Hey, thanks for your feedback! Part 3 is still in the making and it's taking some time, but it's coming :)
@@fictitiousness Awesome to hear! The quality and thoroughness of your videos is amazing!
Easy to understand, thanks...
Great explanations--appreciate your thoroughness.
Great tutorials and information, looking forward to the next one!
Hello, how can I decrease the zoom and then put the angle of view from top to bottom?
Really great tutorial! I have two more simplistic questions:
1. how can I calculate in advance how many seconds an animation will last? I would like to put music, text with a given time span underneath.
2. basically, how should I proceed if I want the animation to have as little flicker and bounce as possible? I see that eye-catching vibration is a common problem.
Thank you!
Hey, did you watch the first video? Both these topics are covered there :) Check it out and let me know if you still have questions after
@@fictitiousness I must be clumsy, I watched both videos :-( . But English is not my mother tongue, and I am a beginner in this field. It would be nice of you to write the answer or tell me at what seconds I get the answer, I will try to understand it from your speech.
No Problem, the length part is at minute 7:28 and the "flickering" part starts from minute 8:46 but its part of the animation and you can hardly avoid it
@@fictitiousness Thank you
Great tutorial! Thanks for your work!
Super helpful video. Thanks for this.
I'm really interested in part 3.
thanks for this!
Amazing!
I would love to get in touch with you for a project
Sure, send me a dm on discord AffenBrot#0292
@@fictitiousness Sent!
Part 3 PLEASE!
Loved this when is the next part coming? And could you do a video on music syncing with deforum?
Probably next week, wanted to finish it much earlier but kind of busy with important stuff right now. Music syncing will be covered in another part as well :)
@@fictitiousness Music syncing will be covered in another part as well :)
wooohooo
Amazing vids! Maybe i missed something, how would you use a picture i have to start with for deforum stable diffusion? Is it just by like pasting a link within the prompt of what i want it to look like?
Thanks! There's different approaches, either with a source image and prompting or just through prompting on its own. Will cover Prompting in the upcoming video.
This is great, but can you make a video of actually making an animation with the SD 1111 Deforum?
excellent :)!! thanks for making such great informative video's. If you are able to, could you please go over how to properly use the seed scheduling ? I have not been able to get that to work. thanks again :))
Sure, as explained in the previous video, you need to set the seed_behavior to "schedule" and can basically determine which seed to use as a fixed seed at any point of the video. For example: 0 (555), 10: (777), 50 (888) will use seed 555 from frame 0-9, 777 from seed 10-49 and seed 888 from seed 50 till the end of the animation. Hope that's more clear :)
YES thank you!
Thank you for your deep and clear tutorials.
I have a question:
I use Mac and when I generate FFMPEG video it says "Error : FFmpeg not found. Please make sure you have a working FFmpeg path under “FFmpeg location” parameter.”
How can i fix that?
Thanks! Sounds like you need to set FFMPEG to PATH, check out this guide here: phoenixnap.com/kb/ffmpeg-mac
Master Please make part 3 🙏
in the making :)
Ooo you going to sneak post this one on us and I got work tomorrow guess I’m staying up late 🤣 please tell you got video 3 locked and loaded
Not yet locked and loaded, but won't take as long to finish as this one :)
@@fictitiousness I’ll say this those math functions got me stumped so I’ll be waiting for this next one!Appreciate the effort of these videos there great!
1 year later, where's part 3? 🎁
I am interested in car body design and I need to produce orthogonal views of a vehicle (front, side, rear and top). Do you know if there is any Stable Diffusion extension that allows me to generate these views/images based on a render I already generated? My idea is to use these four views as a blueprint to make the 3D CAD model in Solidworks.
bravo!
Can u do a tutorial on the video input
It feels like Translation-X moves the canvas while Translation-Y moves the camera. There's an invertination. ?
Hey great video!! Can you please tell me why in 3d mode, I haven't the rotation_3d_x, rotation_3d_y parameters ecc.. I just have Translation X and Translation Y, and when I try to start, give me an error. Thank you anyway
Hey, that's trange because they should be there if you turn on 3D mode (They're below the translation parameters). What's the error you're getting?
I was wondering if you can share where to put video init files when running this stuff locally on a Mac? I don't know where the content folder is, or where to put my init video files that I want to use:(
Sure, just create a folder in your root sd-webui called "input" and name your video "input.mp4". Then you just enter the path in the "video_init_path" field like this: input/input.mp4. This should work, i think the main issue is the first "/" in the default path example which will give you the error. This is how it works on Windows but must be similar on Mac.
@@fictitiousness do you offer any 1 on 1 help over discord by chance? I’ve been at this for days trying various promps and settings and it’s just not working out that well for some reason. Seems like when I don’t have a Zoom, the images corrupt themselves. And also the init video or init image also doesn’t seem to stick :(
@@BeatoxYT i can try to help you out, hit me up on discord AffenBrot#0292
@@fictitiousness Thanks, I'll add you now and see if you can help me out, I appreciate it!
Part 3 plz :)
Thank you so much! Has anyone else noticed a difference with the latest SD 2.0? I can't get any smooth transition between frames at 768X768. Even the default prompt settings are generating random images between frames.
Nope, staying away from 2.0 for now until i see it's been improved or even close to the quality of 1.5 :)
Part 3 please!!!
is it compatible with macintosh? and an application that I have to download or a website? thanks
No, that's a Windows/Linux Application and you need a Nvidia Graphics card for that to work. As a Mac user i would suggest using a cloud GPU from vast.ai or runpod.io, you can find lots of tutorials on other channels on how to get started.
@@fictitiousness 🙌
part 3 please :D
Where is the part 3:( I'm struggling on how to actually use prompt
Coming :)
hey guys.. i want to add an Image Init to Deforum in A1111 but i get always this error:
Error: 'Init image path or mask image path is not valid'. Check your schedules/ init values please. Also make sure you don't have a backwards slash in any of your PATHs - use / instead of \. Full error message is in your terminal/ cli.
Image path is: C:/webui/images/01.png
i tried it with backslash, upload the image on the web and use a web link, tried another image, another path.. nothing work :(
Can you sync audio beats to the 3d animation? I know you can keyframe the zoom for 2d but is there a way with 3d?
Yes, you can sync any parameter with a beat, you just need to check that the value outbreaks are right. Will cover that in the next video.
can you tell me if it is possible to use pictures in animation (morphing between pictures)? And how to do it? Thanks!
Not with deforum but there is a script for interpolation you can check in the Automatic1111 github repo
@@fictitiousness what is the script called? thank you!
@@ramilgr7467 Interpolation
Waiting on #part3
Is there a way to rotate in the 3d y axis but have the object centered?
yes, with a lot of testing its possible, or at least you can come close. I've seen some examples in the discord channel but that was some months ago, impossible to find :)
Where do the videos save to? I don't see where anything saves.
they should be saved in the same folder, where your frames are saved: "outputs/img2img" and depending on what you have set in your settings either a gif or an mp4
please can you tell how to add negativ prompt on Deforum promt ? --neg and can i add as many as I want with , ?
yes exactly, there's also a negative prompt input field to add to all of your prompts
@@fictitiousness --neg word one , word two , word three,
May i ask you why 3D doesn't work? only 2D...
3D gives me an error
Tnaks
What's the error exactly?
please upload part 3
i wating your video...
How o remove that left right camera shaking effect?
i think you still have the default math expression in one of the axis movement parameters. Just make sure to put them on 0:(0)
@@fictitiousness Yeah i just removed the sine expression and it worked :)) Thank you!
greaaaaaaat
Part 3 plz.
Brother.....i don't know who you are or where you are from, but I need the part 3......PLS :)
It's coming :)
Should I update?
[notice] A new release of pip available: 22.2.1 -> 23.0
[notice] To update, run: (etc...)
Sure, doesn't hurt to update once in a while :)
Where is part 3 or more?
WHERE IS PART 3!!! PLZ!!!
please teach me how to create negative prompts
Prompts are being covered in the next video :)