Home Page › Forums › Art Showcase › My first posted render
Tagged: Techsnoop renders
- This topic has 13 replies, 7 voices, and was last updated 8 years, 6 months ago by
Yan.
-
AuthorPosts
-
April 6, 2018 at 9:45 PM #560305
OK so I think I'm beginning to get the hang of postwork in PS afteer Daz. My machine is not powerful so my renders are not at 100%. But I would appreciate any feedback or suggestions.
Thanks
April 6, 2018 at 9:49 PM #560312Yeah...you know, you need to put more twelve characters from font factory. Your PS cant be that bad it cant render 1 character from daz.
April 6, 2018 at 10:16 PM #560342Is the entire scene 3d? You could try rendering it in parts, render the background, save the png, render the mid objects, save the png, render the forground and characters with shadows, save the png, then take to photoshop to layer them together.
IRAY or 3Delight?
I would render the background, save as png or jpg, that shouldn't matter at that point, then Render the buildings with no background, no environment, so you can layer that on top of the background.
Finally Render the Foreground, Rocks, Droid, Characters in the same fashion, hiding everything else. That way, when you make changes to the characters, you are only going re render that layer. The scene you created is a perfect layout for that, since you can chunk out Back, Mid and Fore objects. You might even be able to render the buildings and background as one, the do the rocks,drone and characters as a second.
It's what i would do when i was running on my 760 GTX, and still sometimes do on my 1080TI, depending on the complexity of the scene.
April 6, 2018 at 10:30 PM #560381Looks great! I like the expression of the girl in the foreground. Like she’s on high alert for any threat that might come her way.
April 7, 2018 at 8:21 AM #560964Welcome aboard Tech. Personal view here, and you don't need to worry too much on the technical perfection of your renders, at least not early on. That will come in due time, as you grow more familiar with the workflow, UI, and assets, and you'll try experimenting. What's as important, if not more, is what you want to do with your render. And, in that aspect, you look like you got the hand of that already.
April 7, 2018 at 12:55 PM #561203Welcome, @techsnoop! My only comment is that the foot of the middle girl is behind a rock and it seems truncated. Other than that, I agree with Yan. Experiment, experiment and experiment again. Also you can watch a lot of tutorials on Youtube, it helped me a lot.
April 7, 2018 at 1:40 PM #561252Anonymous
InactiveRank:
The composition looks fine to me, the image could possibly use a bit more depth of field to help things in the foreground pop a bit more.
Daz can be a pain but it's all trial and error, we learn as we go 🙂
Gotta love Photoshop!Keep up the good work!
April 9, 2018 at 6:19 PM #564209I have no idea what your statement means.
April 9, 2018 at 6:20 PM #564210Thanks @tomsk, I haven't started trying any depth of field yet. Still trying to get posing and lighting to work. 🙂
April 9, 2018 at 6:23 PM #564215Thanks @maiarka and @yanmartyn. The foot was an issue due to the planet teros ground being really uneven. I think I'll turn off some of the rocks if I can.
April 9, 2018 at 6:25 PM #564217@aftyrbyrn The background is photoshop, sky & planet. I'm going to be using a different comp of buildings and I think I'll do the separate render thing to conserve.
April 9, 2018 at 7:04 PM #564237About the foot part... it's something I am often doing in Poser, but, you can also do in DAZ. Set your main camera, get the angle right, and, leave it alone. Then, create another camera, and use that to orbit around, and look for imperfections. Like a foot stuck inside the ground, for example... you can always translate (a.k.a. move) a figure, by going up in the Y axis translation dial, with small steps until you get the foot at the proper height, without messing the pose and outfit. Also, if DoF (Depth of Field) is giving you slow render speeds, you can "tell" DAZ to make a Z-depth map during the render... or Depth map in general, in the canvases menu, in render settings.
Then, you can use that (it will be a black and white sillhouette image) in photoshop, to generate a considerably faster lens blur effect. There are plugins to automate this, but there's also a manual method. Copy/paste the depth map over your render, after you're done post-working it, then ctrl-A to select all, then ctrl-C to copy. Then, go to channels, and hit the mask icon. It will generate an alpha channel. click the white box on it, then ctrl-V to paste. click away the eye icon (you don't need it visible anyway), return to the layers, delete the depth layer you pasted before. The alpha channel will remain. Then, go to blur-lens blur, and toy around with the settings. And, click on the figure, or element you want to be the focus center. Everything else will blur in relation to their level of brightness in that outline silhouette map, you used an an alpha channel. The darker the gradient, the more the blur, unless you tick the "reverse" tickbox on lens blur settings.April 10, 2018 at 1:24 PM #565072@yanmartyn, what a useful tip! I didn't know about this one.
April 10, 2018 at 3:08 PM #565153 -
AuthorPosts
- You must be logged in to reply to this topic.



