- Jan 10, 2020
- 16
- 162
the fabled alchemy Chasm
- Jan 31, 2019
- 3,868
- 14,176
I've Never been with a Human before, so ... You'll be gentle with me, Right?
- Jul 7, 2017
- 80
- 772
- Jul 31, 2017
- 1,079
- 20,841
- Oct 13, 2016
- 2,757
- 9,160
After going thru 104 pages lol, since my last visit, I collected a lot of renders I liked, and now I'm quoting those and making a comment about it, even if it's just a one word, is my way to appreciate it's creator and to let them know so.
Some think it's pointless to comment on pics that were posted literally a month ago, but while I LOVE this thread, I'm not visiting it all that often, and people post new renders daily and in quantities (which is awesome btw)
So, when I do visit the thread, I go thru every single page to see what's new.
Also, due to the limit in the amount of reactions you can make daily, I can't let each user know that I liked their work unless I quote them. So, that's that.
Some think it's pointless to comment on pics that were posted literally a month ago, but while I LOVE this thread, I'm not visiting it all that often, and people post new renders daily and in quantities (which is awesome btw)
So, when I do visit the thread, I go thru every single page to see what's new.
Also, due to the limit in the amount of reactions you can make daily, I can't let each user know that I liked their work unless I quote them. So, that's that.
You don't have permission to view the spoiler content.
Log in or register now.
- Jan 16, 2018
- 531
- 11,647
- May 31, 2020
- 160
- 1,219
- Jan 31, 2019
- 3,868
- 14,176
I've Always had a thing for She-Hulk, So I thought, Why not make a Pic?
- Oct 18, 2019
- 350
- 5,164
You don't have permission to view the spoiler content.
Log in or register now.
- Jan 31, 2019
- 3,868
- 14,176
- Mar 25, 2020
- 129
- 803
Image is a bit dark, so hard to make full determination.
She's got good curves inside her clothes. The lighting is nice and interesting, but you're hiding a lot by the use of light. It's fine to do it that way, but then it's hard to make judgment on the figure because of the lighting choices.
- Aug 13, 2019
- 1,075
- 4,421
- Jan 31, 2019
- 3,868
- 14,176
- Jul 5, 2017
- 886
- 3,118
Remake of a pic that I did some days ago and shared here yesterday. After almost 10k of interations and four and a half hours of rendering, I decided that I need something better than my GTX 1060 to continue my game.
Also the props of this basement sucks.
Also the props of this basement sucks.
- Aug 18, 2017
- 289
- 9,549
I've been playing around with light sources at the opposite ends of the spectrum. Post was done in Lightroom to give the "Playboy" type effect.
- Mar 25, 2020
- 129
- 803
To me, anyone can throw more brute force at a render, but this is not art. This is simply throwing hardware at a problem.
Take your scene apart, down to the component parts. Understand what it is you want, and what the pieces are doing to your engine. The grain of the leather, for instance. This comes at a high cost. The bumps, the noise, the sheen of the reflection. This is all part of the sim your are trying to force through the engine.
The mesh light on the laptop. What is it adding? Why? Why use it in that format? Mesh lights are heavier and more intensive than points, but they are good for some things... but in that case, why? Why use it? Turn the laptop away, turn off the light. Add a point light if you want the glow, for whatever minor addition it gives you to the scene. Or, better yet, do it 100% in post.
What is the background beyond the sofa adding? Render it separately at lower res and composite it back in. If you were using DOF, I'd also say to simulate it in post because it's heavy on the engine, but it doesn't look like you're doing it.
Look at the scene, decompose it. Figure out what's in front, what can be rendered separately, then put it all back together. A one-button ability to make art? It's a lie. That does not exist. Rendering like that is crazy IMO.
Look at your scene and break it down, then reassemble it. Maybe you need hardware, maybe you don't. To me, it's a mistake to just think that hardware solves the problem. It never does. The scene will always eclipse your hardware if you don't fix the problem, which is scene assembly and understanding the medium.
- Mar 25, 2020
- 129
- 803
This scene, I rendered while asleep, maybe 5-6 hours? Maybe less? I forget when I hit the button. I have near potato power in this day & age, an older Ryzen. No gfx card.
That was too long for that image. Too much fucking time for that.
So I broke the scene down for all future work based on this setup.
The wall? Takes too long. The bumps, the mottled surface. It's shitty on the engine. So what to do? Take it out, render separately from the girl and her dog. Then all scenes based on this camera angle do not have to recalculate the wall in the sim. I just composite it later. Render once, use multiple times later.
The same goes for the coffee table with the vodka bottle and the pills. I separated it from the scene, rendered it separately. That overhead is now gone. I can add it back in afterwards, no problem.
NOW, when I render the pair on the sofa on their own?
An hour maximum. Probably more like 45 minutes to get a usable render/resolution, especially if I downscale to 1080p rather than leave it at the render resolution (the intent is to do final work in 1080p). And that's a complex dog. dForce hairs, not a flat surface... harder on the engine than a person would be.
It would be faster if I had real hardware to do this, but whatever. I am a busy person. I have things to do. I can give up 45 minutes while I do other shit.
Why do all that when you can optimize a render to be perfect at one button push (after multiple multiple iterations)? Well, at this point, I have a production model to make an image every 45 minutes to an hour at 1080p whereas your one-button art render philosophy takes many multiples of that before you iterate 1 useful image. This is simple enough to manage for a VN, I think. Even faster for collaborative review process... tweaking tone and lighting can be done in minutes, seconds in post rather than hammering it through an engine.
Is it a perfect image? No.
I'm just illustrating one process, one mental exercise for breaking down a scene and making it better to render.