Because apparently that's what he wanted all along. Generic mobile game chicks that look "realistic" as opposed to the style that made him recognizable.
Collection Video Affect3D - miro Collection [2025-12-17] [Affect3D/affect3d_miro/affect3dx/miro]
- Thread starter 7767
- Start date
Tzios
Newbie
- Oct 4, 2017
- 58
- 358
- 301
Makes me so sad... girlfriends4ever was life changing for a young me.
BayouBill
Well-Known Member
- Jul 8, 2018
- 1,291
- 1,118
- 421
Evan Hunt
New Member
- Apr 8, 2018
- 3
- 0
- 11
actually found all there is i guess on kemono, thanks for mentioning it previously.
They are all by Miro and many are on rule34gen . com/miro
Evan Hunt
New Member
- Apr 8, 2018
- 3
- 0
- 11
Hey, thanks xDreamZ.
Didn't know that side, and other artists there, that deserve acknowledgment imo.
You're welcome. There's a lot of backlash happening (still) because it's the early days (still) of using AI as a tool, rather than a pretty picture generator with obvious failures. But the tech is moving exceptionally fast and I'm positive within the next year you'll start to see more and more artists start to use AI in their animation workflows. NOT 100%... but as a tool. Some already are besides Miro, and the results are all looking so very good! Gooners Rejoice!
People are struggling to understand that even an rtx3090 is considered an entry level GPU for creating ai video, currently its limit is around 8 seconds on 1080p. a 5090 will be around 12 seconds 1080p. Its entirely dependant on VRAM, it holds the hundreds/thousands of frames before it to maintain consistency. Smaller the VRAM less you can make and lower resolution. To put it simply its a monsterous cost, even a server full of 5090's would kick out a minute at most. VRAM usage is NON LINEAR, it compounds massively with duration.
You're partially right, but mostly wrong. Currently 30 seconds is common. Wan2.2 or Z-image for start/end frames hold character consistency and shot selection. You plan your short video with a storyboard, do the shots, and then glue them together with traditional software like Davinci Resolve or Premier, same as 3D animation. You didn't think that 3D (or any video) was one long render, did you?
Its almost like making real art is better and more cost efficient or something.
Shane498
Member
- Apr 27, 2017
- 253
- 275
- 340
so are they going to change the name in affectai-miro?
Last edited:
I don't mind they doing AI CG but god damned can't they generate video using LoRAs of their own work? They can essentially do what they did with minimal effort. Now their works are no different visually from any other Huanyuan/WAN futa videos
For example:
For example:
Last edited:
Bueno
Active Member
- Apr 26, 2017
- 737
- 1,038
- 353
I believe I used a Wan2.1 workflow that loops the last frame of a generate video to be put as the first frame of the next video, that way it can hypothetically generate forever. There was an idea of a prompt traveling node added to each generation so it's done like 10 slices of one scene with the next video continuing from where the last one left off.
Never tried it myself, I've only ever managed to get Wan2.1 to run locally using ComfyUI with shared memory, with a lot of the memory need off-loaded to my RAM instead of VRAM. It took me about 20 minutes to generate a 680x520p video for 10 seconds. So I always generate image with PonyXL first, than generate video using that image.
Haven't tried using Wan2.2 at all though.
dakamuto
Member
- Jan 14, 2019
- 220
- 379
- 302
In 2013, we had the legendary masterpiece "Girlfriends 4ever." In 2025, we have AI that's nothing but dog shit. It's the era of degradation, gentlemen.