Seed Values Decoded: Elevate Your Stable Diffusion Techniques

[@jmagick7281]

I really do hope your channel takes off. Massively good information this video! Subscribed and liked!

[@PromptingPixels]

Dude, thanks so much for the kind words! Made my day! Here's a video response here: kzbin.info/www/bejne/p4uknmeMjc-taLssi=L2LhL9XNQb49e_ZA&t=412

[@kantube07]

Hello and thank you for your video. I'm trying to figure out Variation Seed right now. And I blended the dogs from your example. But then I tried to blend more complex prompt, and it did not work. I used the same model as you are, the first image was "Cyborg, night, Tokyo" the second one is "Samurai, bamboo forest" No blend. What I'm doing wrong?

[@MrChamesy]

Oh this is super interesting thank you

[@wzx1229]

Awesome! It blows my mind! Thank you, Shawn

[@PmXism]

Great video, thank you. I thought I knew all about seeds but I had no idea you could blend them like that!

[@PromptingPixels]

Thanks so much for the kind words. Full response here: kzbin.info/www/bejne/ooaQo3-cgpqJmMUfeature=shared&t=280

[@glitchsandwich]

🙌Great video, thanks!

[@PromptingPixels]

Np - appreciate the kind words!

[@RoopeBb]

Well laid out video full of information and easy to follow even for a person like me who's just starting out. Loving your channel! What kind of Mac are you using? Looks like you're able to generate much quicker than me on my Mac Studio Max M1 machine. :)

[@PromptingPixels]

Hey there - thank you so much for the kind words! Just did a short video explaining the setup: kzbin.info/www/bejne/ooaQo3-cgpqJmMUfeature=shared&t=26

[@RoopeBb]

@@PromptingPixels Thanks for the answer and another great video!

[@RoopeBb]

I had to test how many it/s I'm currently getting and mine is quite a bit slower compared to your 6it/s. SDXL 1.3it/s sd_xl_bas_1.0 512x512, steps 20, Euler a SD - 1.92it/s v1-5-pruned-emaonly, 512x512, steps 20, Euler a

[@AndrewTateNewsOfficial]

can i make an character with it and use the same character over and over again with the same seed so only the poses and environment changes?

[@PromptingPixels]

Honestly, I am not too confident in character design skills. However, I elaborate on this in a full response here: kzbin.info/www/bejne/ooaQo3-cgpqJmMUfeature=shared&t=504 Also within the linked video above, the description has some resources that I added that might be worth checking out that elaborate on character design quite a bit and a few different approaches you can take (ControlNet, LoRAs, Dreambooth, etc.).

[@Beauty.and.FashionPhotographer]

what you are showing with your test with the X/Y/Z plot are sequential numberings , not random numberings. but what would be actually of amazing value would be to figure out, how to generate grids with "actual" real random variations seeds.... not sequential ones. reason being: the pose of the dog or person would dramatically change, if there were a way to create seeds with total randomize numbers... for example 345..then 873487567,...then 21...then 10034354355,... etc etc,.. only this would show different shooting angles as well, that are so dramatically different that they can be used really well ... i am still searching for someone who has done this.... until then the option "Var. Seed", is actual a sequential seed, and not at all a "variation" seed .... and should be renamed as "Seq.Seed"...