Your videos are so refreshingly... different. Introductory AND deep, at the same time Both videos that I watched so far gave me answer to a number of questions that have been lingering in the back of my mind for quite some time. great stuff, I'm hooked!
Since there is so much to learn, I feel like I am kind of swimming, in so many respects. I kind of improvise my way through. I had noticed that comfy and forge produce different images, but I had no idea why. I have no idea where comfy actually gets its images from, when using the standard "Load image" node. (With "load image from path", it is clear.) For example, when I use images as reference, like when using controlnet - ideally, I would want to be able to reproduce the exact same output I got, just by throwing in the generated image. Somehow, at the moment, it seems to have no trouble finding those images. If you asked me now, how it does this, and what I would need to do to make sure it will still be able to this a year from now, I would have to say "I have no idea". The list goes on and on, I can use Comfy, and I absolutely love using it, but there are so many fundamental questions I don't know the answer to.
![BOWKYLION - วิงวอน (ex-change) [Official MV]](http://i.ytimg.com/vi/cLdnZWDaO-w/mqdefault.jpg)
Your videos are so refreshingly... different. Introductory AND deep, at the same time
Both videos that I watched so far gave me answer to a number of questions that have been lingering in the back of my mind for quite some time.
great stuff, I'm hooked!
Since there is so much to learn, I feel like I am kind of swimming, in so many respects.
I kind of improvise my way through.
I had noticed that comfy and forge produce different images, but I had no idea why.
I have no idea where comfy actually gets its images from, when using the standard "Load image" node. (With "load image from path", it is clear.)
For example, when I use images as reference, like when using controlnet - ideally, I would want to be able to reproduce the exact same output I got, just by throwing in the generated image.
Somehow, at the moment, it seems to have no trouble finding those images. If you asked me now, how it does this, and what I would need to do to make sure it will still be able to this a year from now, I would have to say "I have no idea".
The list goes on and on, I can use Comfy, and I absolutely love using it, but there are so many fundamental questions I don't know the answer to.
what is the minimum GPU VRAMs required to run ComfyUI bro ?
4-6gb Nvidia card, though it depends on the actual stuff you're going to do