Gosh, you lot have been busy overnight! I only had a quick scan, or I'll be late for work!
I've tweaked mine slightly, changing the shaders on the girl's outfit, and adding a pin spot just to highlight her face better. Oh, and I changed some of the gags on the cover ;)
I feel sorry for the girl ...but the car ...?
Perhaps you have in mind an other cover for that?
Anyway, very nice work!
Here's some quick test renders from me, Cyborg Vicki and her son Ralph
The three without text are straight renders. The Vickii Cyborg is made with two Vickis, one inside the other, the innards exposed by transmaps on the top Vicki
I may not sleep tonight, and it's all your fault. :)
Wow! It's always such a feast for the eyes here! I truly need to do a search for better exclamation words because I don't have enough at my command to do it all justice.
evilproducer The alien vegetation is so lush and interesting. I feel your pain with the lighting not cooperating. I can't seem to light the liquid around my robot's feet to save my life.
dustrider The skin and clothing materials are beautiful! I hope that poke through doesn't cause the whole forum to go insane.;-P I have Stonemason's enchanted forest , but can't seem to ever get it to look good. It looks great in your scene. Did you have to do anything special with the textures for your setup?
Stezza Love the martian and raygun! Trust us! Ak Ak Ak
booksbydavid Holy @#$%! That's some breathtaking modeling and postwork! Great explanation of your postwork process too! Pssst! The editor says the magazine title is DAZling Stories, but no worries if it's too late to change.
Tim_A Wow! Your minor tweaks make a big difference. The woman really comes alive!…if, sadly, for only a short time longer.:ohh:
head wax That's beautiful and frightening at the same time. The layering and multimedia effects are amazing. Love the transmapped skin layer effect! Can you pretty please show more how you did that? I've so wanted to do something like that for another project. My avatar has V4 and V4 skeleton together then I eroded layers with photoshop masks, but I think yours looks so much better because it has the 3D peeling effect. Pssst! Editor says it's DAZling Stories for you too. :)
I am not even going to pretend that I can keep up. I am just loving and am in awe of the WIPs being posted here, and feel like I am the only one stuck, while everyone is moving ahead at the lightning speed...
booksbydavid, I love the last postworked image. Absolute stylistic perfection.
Tim_A, I like the updated colors.
head wax, great cyborg idea and technique, but I am slightly confused by the image with the title. I can't quite figure out what's going on there.
dustrider I deeply admire your perfectionism :) I hardly noticed that poke through even in close-up... :) But why the insistence on no postwork? Is there a special purpose for it?
Stezza, your Martian is adorable! :) It also inspired one of my titles :)
Antara Thanks so much for the lighting advice! Underwater scenes are so challenging. I'll definitely try out the additional spots and the filter effects vs glow render pass to see which works better. I'm not sure I follow on the torus and the anything glows, but I'll check it out. Do you think I need to bring the fog closer to the main subject? Also, I think the scene is a bit too crisp in the foreground. Any suggestions on what might make it read as watery there. Air bubbles and some subtle wavy distortion maybe?
I know it's just your scene setup, but all the cameras trained on the figures look neat. If this were an actual image, it would look as if you were making a strong statement about something like the intrusion of media or social media into private moments of life and death.
sukyL, about the foreground: in addition to bubbles or just some tiny replicated debris, you can also add murkiness in post by using the distance pass as your mask for various fogs (I would not recommend adding more fog to the scene unless you really have to, because that would slow down your render. If render time is not much of an issue, then adding scaled down clouds might be a good idea to add murkiness that's not too even. I found clouds look better than just fog with turbulence added in).
About my cameras: I actually had this idea initially of placing floating robot cameras all round these figures, but then I figured that with the addition of text it would make the scene too busy and harder to read, so I decided to go with simplicity instead, but yeah, it would add a whole new layer of meaning to the scene, and I really love that idea.
Now to my own progress: here is my first post-worked cover (Camera 6), before and after post. (The stories titles might change if i come up with anything better.)
The angle is kinda the combination of my two original extremes. I might have killed the best points of each and ended up with something which is just 'meh', so if that's the case, please tell me.
(Here is the original POV dilemma: http://www.daz3d.com/forums/viewreply/694113/
and here is another alternative I am considering: http://www.daz3d.com/forums/viewreply/694225/ )
I still need help, though. If you could give me feedback about my 1 - shaders (do fabrics look like fabrics, and metals look like metals, and skin looks like skin?)
2 - postwork (especially whether I managed to make the metal on the robot look less flat and boring?),
3 - titles (looks-, content-, and placement-wise?),
4 - page layout overall, as well as
5 - how does this look compositionally in comparison to my other views?,
I'd really appreciate it.
Side notes: I was pleasantly surprised to find out that my full size render with best quality settings is taking about 4 hours to render (no transparencies or lights through transparency at all, no bump), even on my slow machine. I was expecting much worse...
Another postworked angle for comparison. (For reference, this is Camera 3) And I am setting up 2 more to render overnight, so I will post them tomorrow. ...by which time this thread will have 7 more pages :) and everyone will forget what these images looked like...
Another postworked angle for comparison. (For reference, this is Camera 3) And I am setting up 2 more to render overnight, so I will post them tomorrow. ...by which time this thread will have 7 more pages :) and everyone will forget what these images looked like...
Hi Antara
this camera is fantastic!
I like the scene and the texture (Great reflection on the robot surface)
I love the title “Robot and Jiuliette”. Great!
Another postworked angle for comparison. (For reference, this is Camera 3) And I am setting up 2 more to render overnight, so I will post them tomorrow. ...by which time this thread will have 7 more pages :) and everyone will forget what these images looked like...
Looks great Antara! A really nice job on adding emotion to the robot through pose alone.
Another postworked angle for comparison. (For reference, this is Camera 3) And I am setting up 2 more to render overnight, so I will post them tomorrow. ...by which time this thread will have 7 more pages :) and everyone will forget what these images looked like...
Antara I totally agree with pimpy. Camera angle 3 is fantastic! It places emphasis on the robot to show its emotion and the clasped hands very well. Robot and Juliet is a great title. Romeo and Juliet is exactly what came to my mind when I saw your first render. The shape of the DAZling stories text and general text layout complement the scene very well. The hot and cool colors are very effective and metallic texture on the robot is lovely. Looking forward to checking out your next renders as well.
This month's theme is Rankin and Bass claymation, right? :red:
I'm glad there is still a lot of time left.
I don't think your cover is bad at all. I do think a couple more dynamic poses along with some work on the camera angles could help sell the drama.
For poses, if you look at the collage in your post, the super hero picture has the woman's leg forward and the knee bent. I think if you did something like that for the femme fatale, that may help loosen up the stiffness of the pose a bit.
Looking slightly up at the subject may help a bit as well. I would also refer to some of the other pulp fiction covers posted here and try and add a dramatic background. It does not need to be complicated. It's film noire inspired, so maybe the background is a dark color with a circle, or oval shape suggesting a spotlight, or a rectangle with a blind pattern or bars. You could even add it in post as the 'toon scene effect can use the alpha channel.
Here's a quick example of what I mean. Made a layer with a circular gradient, then another layer with the window shape. Took less than five minutes. Most of the time was fiddling with the gradient.
Here's some quick test renders from me, Cyborg Vicki and her son Ralph
The three without text are straight renders. The Vickii Cyborg is made with two Vickis, one inside the other, the innards exposed by transmaps on the top Vicki
I may not sleep tonight, and it's all your fault. :)
I have a spotlight shining onto my moss which is hair. It is supposed to fake the light of the glowing stalks. Lights grouped with the stalks looked great but were taking forever to render. As you can see, I have the spot restricted to the moss hair models. The light on the astronaut and surrounding rocks and plants are the lights grouped with the replicated stalks. They are set to exclude the moss hair. As you can see, the moss has no red light. I've tried deleting the lights and adding new ones, still nothing.
On the plus side, the astronaut's light is back to where it is supposed to be. Or at least close enough for my purposes. Widening the half angle a couple degrees did the trick.
I have a spotlight shining onto my moss which is hair. It is supposed to fake the light of the glowing stalks. Lights grouped with the stalks looked great but were taking forever to render. As you can see, I have the spot restricted to the moss hair models. The light on the astronaut and surrounding rocks and plants are the lights grouped with the replicated stalks. They are set to exclude the moss hair. As you can see, the moss has no red light. I've tried deleting the lights and adding new ones, still nothing.
On the plus side, the astronaut's light is back to where it is supposed to be. Or at least close enough for my purposes. Widening the half angle a couple degrees did the trick.
Hi EP
I like the scene (is quite dark but as a test is ok) but...could be an idea try with a light dome to create a diffuse atmosphere and put the red lights at the base of the plants?
head wax That’s beautiful and frightening at the same time. The layering and multimedia effects are amazing. Love the transmapped skin layer effect! Can you pretty please show more how you did that? I’ve so wanted to do something like that for another project. My avatar has V4 and V4 skeleton together then I eroded layers with photoshop masks, but I think yours looks so much better because it has the 3D peeling effect. Pssst! Editor says it’s DAZling Stories for you too. smile
Oh, I made a few renders before work this morning - will post them later. The peeling effect is from one of Ron's Cyborg Brushes. I did this morning's renders on a sphere and came to the conclusion you are better off painting the alpha channel with a 3d painter .
Basic method. Have one Vicki posed as you like. Then duplicate her with Fenric's tool/plugin. Shrink the second one doan a little.
Take the first one's face texture into photoshop, make a new layer, paint your alpha channel black where you want it to be clear and fill in the rest with white. Save this. Use this on the outer most vicki in the alpha map section.
Take this alpha map. Save it as new image. Select the black parts and delete.
Make a new layer underneath and paste your electronics etc. Make the top layer a multiply parameter and add a drop shadow so the electronics layer has a shadow on it.
Make this the texture map for the inner most Vicki, use the original alpha map on her but invert it.
I then took the electronics layer and used the free Nvidia normal map maker to make a normal map and used that in the bump channel.
You need to play with the edges of the alpha map to decide whether you want it soft or hard. Hard will give more of a peel effect.
Also for the electronics Viiki you can use the electronics texture in the glow channel then add aura/glare in the Scene parameters
I have a spotlight shining onto my moss which is hair. It is supposed to fake the light of the glowing stalks. Lights grouped with the stalks looked great but were taking forever to render. As you can see, I have the spot restricted to the moss hair models. The light on the astronaut and surrounding rocks and plants are the lights grouped with the replicated stalks. They are set to exclude the moss hair. As you can see, the moss has no red light. I've tried deleting the lights and adding new ones, still nothing.
On the plus side, the astronaut's light is back to where it is supposed to be. Or at least close enough for my purposes. Widening the half angle a couple degrees did the trick.
The rule with faster dynamic hair lighting that I discovered recently: Duplicate your lights, have one set affect ONLY the hair, have the other set affect everything EXCEPT hair. Changing the hair settings to not cast shadows might also help, but might flatten the look of them too.
But in your case it might not be enough. Since you want multiple light that you need to shine on the hair = longer render time.
- One way to avoid it would be to fake it with a gelled single light on the hair moss alone and nothing else. After all your glow from the stalks is not going to be all that precise anyway.
- Another way to fake it would be to partition your hair into multiple objects and have one light per one object ratio. Why? to avoid multiple lights interacting with too many hairs and affecting each other, thus increasing computation intensity.
Also, I would recommend adding shading glow to the stalks and adding a 3d Aura filter to your scene or the Aura Effect in your stalk's object's Effects settings. This will help sell the red lights on the moss.
Thank you, sukyL, evilproducer and pimpy for your comments!
So far it looks like Camera 3 is the winner :). I'll see whether I like the renders I get in the morning enough to mess with them and post them too.
I'd love to hear more opinions about Camera 3 (vs Camera 6 and vs original POVs) from people who originally liked the second image (the one on the right in my original POV dilemma post) better.
I have a spotlight shining onto my moss which is hair. It is supposed to fake the light of the glowing stalks. Lights grouped with the stalks looked great but were taking forever to render. As you can see, I have the spot restricted to the moss hair models. The light on the astronaut and surrounding rocks and plants are the lights grouped with the replicated stalks. They are set to exclude the moss hair. As you can see, the moss has no red light. I've tried deleting the lights and adding new ones, still nothing.
On the plus side, the astronaut's light is back to where it is supposed to be. Or at least close enough for my purposes. Widening the half angle a couple degrees did the trick.
evil, your scene is really looking good, I'm sorry I haven't had a chance to comment on it, the thread moves so fast.
and thank you for sharing so much about your method
any reason you aren't adding something to the glow channel of the moss stalks?
for me, as alover of post work, I would render out an object pass and use that to select the stalks in ps and add the glow there ;) lazy bugger that I am :)
I have a spotlight shining onto my moss which is hair. It is supposed to fake the light of the glowing stalks. Lights grouped with the stalks looked great but were taking forever to render. As you can see, I have the spot restricted to the moss hair models. The light on the astronaut and surrounding rocks and plants are the lights grouped with the replicated stalks. They are set to exclude the moss hair. As you can see, the moss has no red light. I've tried deleting the lights and adding new ones, still nothing.
On the plus side, the astronaut's light is back to where it is supposed to be. Or at least close enough for my purposes. Widening the half angle a couple degrees did the trick.
The rule with faster dynamic hair lighting that I discovered recently: Duplicate your lights, have one set affect ONLY the hair, have the other set affect everything EXCEPT hair. Changing the hair settings to not cast shadows might also help, but might flatten the look of them too.
But in your case it might not be enough. Since you want multiple light that you need to shine on the hair = longer render time.
- One way to avoid it would be to fake it with a gelled single light on the hair moss alone and nothing else. After all your glow from the stalks is not going to be all that precise anyway.
- Another way to fake it would be to partition your hair into multiple objects and have one light per one object ratio. Why? to avoid multiple lights interacting with too many hairs and affecting each other, thus increasing computation intensity.
Also, I would recommend adding shading glow to the stalks and adding a 3d Aura filter to your scene or the Aura Effect in your stalk's object's Effects settings. This will help sell the red lights on the moss.
Hope this helps.
Did all that. ;-) My intention is to post my light settings when I have chance. You'll see that I use light linking, gels, etc.
My problem was that I couldn't get a single light linked to the hairs to illuminate them for some reason. Maybe I've hit some kind of light linking limit? Each moss clump also has it's own hair applied. You are right about turning off shadows for the hair- One that it would save render times, and two that it would look flat. I would rather have long render times than flat hair. I was going to try a glow in postwork as it would be faster, and since Carrara's aura is a post render filter, I think it would be more efficient to do it in Photoshop.
So here's a screen capture of the moss. remember, the little stalks are grouped with a single bulb and are replicated using a surface replicator. each moss clump (there are four) has its own hair and surface replicator for the stalk/bulb group, so a total of four hair objects, four stalks and four bulbs. Each stalk is replicated around 120 times.
EP: But if you are replicating the stalks/bulb groups, wouldn't the lights get replicated as well? therefore, wouldn't you have the 120x4 light-bulbs as a result?
I don't know how many hairs you've got and what the settings are, but I had the situation with hairs and multiple lights, in fact - multiple hairs and multiple lights and although it did slow down the render, it wasn't prohibitively so - maybe by something like 30-40 min for the image where the hair covered about 45% of the area. And I only have 4 slow cores and under 8Gb of RAM, so with what seems like a much more powerful system, you shouldn't be getting into serious trouble with this. Therefore I am trying to think what could be causing the slow down.
But if you are replicating the stalks/bulb groups, wouldn't the lights get replicated as well? therefore, wouldn't you have the 120x4 light-bulbs as a result?
Yes. That's why I tried a couple different non-replicated lights to light the moss, with just the astronaut, boulders and plants linked to the little bulb lights. The hair was excluded. I tried the shape light and spotlight linked to the hairs with various ranges, intensities etc. I just couldn't get them to light the moss/hair for some reason. The screen shot I posted was the last light I tried at full value. It should have looked like a flash bulb going off, but zippo!
It sucks for time, but the only way I could get the moss to look lit, was to include the moss hair in the replicated bulb lights.
Well, I think I may have the final render, maybe, I hope. I also added a screen grab of the render running in the Octane Render View Port so you can get an idea about the Octane side of the system resources used.
I think I've mentioned everything about the set up except the lighting. The scene is lit with two lights. One main mesh light in front on the right side of the camera, and one small mesh light behind the dinosaur to give a bit of light to the dark area. The only post work on the image was re-sampling to reduce the size from 2400 to 2000 pixels high, and a slight levels adjustment to brighten the image just a touch.
dustrider The skin and clothing materials are beautiful! I hope that poke through doesn't cause the whole forum to go insane.;-P I have Stonemason's enchanted forest , but can't seem to ever get it to look good. It looks great in your scene. Did you have to do anything special with the textures for your setup?
Actually I was wrong, the scene is Return to Enchanted Forest :red: The default materials for the trees (trunks) need to have the highlight and shininess adjusted, they are too shiny (specular in Octane). They also need some bump, as that is lost by Carrara. I just copied the diffuse map to the bump channel (node), and it helped a lot. Texture sampling also needs to be changed from "Fast Mip Map" to "Sampling", or you will be able to see the texture seams. One final adjustment for any sort of close up is you need to enable smoothing (in the modeling room), or you can see the hard angles on the trees and foliage. Hope this helps a bit.
dustrider I deeply admire your perfectionism :) I hardly noticed that poke through even in close-up... :) But why the insistence on no postwork? Is there a special purpose for it?.
lol - That's and easy one, I suck at post work :bug: It was also a bit of a personal challenge. After I saw the results of the first bit of text, I wanted to see how much I could do for this challenge directly in Carrara, even though logic would dictate that a fair amount of it should be done in post.
I typically don't do much post work. Other than not being very taented at the artistic aspects of post work, I'm a bit too logic oriented, which means if logically things like poke through "should" be corrected in the render (if possible), not in post, then I have to fix the original geometry. That, combined with a bit to much perfectionist and a dearth of artistic talent, means that I'm naturally inclined to push the render end of things to the best I can get, rather than relying on post work except for simple fixes.
EP: But if you are replicating the stalks/bulb groups, wouldn't the lights get replicated as well? therefore, wouldn't you have the 120x4 light-bulbs as a result?
I don't know how many hairs you've got and what the settings are, but I had the situation with hairs and multiple lights, in fact - multiple hairs and multiple lights and although it did slow down the render, it wasn't prohibitively so - maybe by something like 30-40 min for the image where the hair covered about 45% of the area. And I only have 4 slow cores and under 8Gb of RAM, so with what seems like a much more powerful system, you shouldn't be getting into serious trouble with this. Therefore I am trying to think what could be causing the slow down.
My computer is a single core 2003 PPC G5 1.8 Ghz with 4.5 GB of RAM. I do network with a Dual Processor MDD G4 and an Intel Core 2 Duo iMac running Mavericks. The C7.2 render node works on the iMac, but the stability to run the full app is just not there. by rights, I shouldn't be able to run C7 at all with Mavericks.
thanks, that's because nothing s going on ... yet :)
for your image, its' fantasticagorical ! Love cam 3, the sharper diagonal is very fast.
you asked about the cloth - some of the bends in the cloth are a bit square - doesnt really worry me though, just if you want nitpicking
it's a great render, everything sits so well
edit: I started this post half an hour ago, just got to post it - gah
Thank you! :)
And sorry for jumping the gun on the image with nothing going on yet :). I look forward to seeing what's going to happen there.
About square folds - I am guessing you are talking about the blanket, not the dress? Or were you talking about the pillows too? The pillows could use some work, but I probably won't re-render the whole thing just of them. With the blanket I was hoping for a not-very-flexible-fabric look, so I didn't mind the squarish folds. If you are talking about the dress, then it's a problem, but I am not seeing it. Is there a problem area somewhere?
I have a spotlight shining onto my moss which is hair. It is supposed to fake the light of the glowing stalks. Lights grouped with the stalks looked great but were taking forever to render. As you can see, I have the spot restricted to the moss hair models. The light on the astronaut and surrounding rocks and plants are the lights grouped with the replicated stalks. They are set to exclude the moss hair. As you can see, the moss has no red light. I've tried deleting the lights and adding new ones, still nothing.
On the plus side, the astronaut's light is back to where it is supposed to be. Or at least close enough for my purposes. Widening the half angle a couple degrees did the trick.
evil, your scene is really looking good, I'm sorry I haven't had a chance to comment on it, the thread moves so fast.
and thank you for sharing so much about your method
any reason you aren't adding something to the glow channel of the moss stalks?
for me, as alover of post work, I would render out an object pass and use that to select the stalks in ps and add the glow there ;) lazy bugger that I am :)
Hi Andrew,
I'm rendering a glow pass, a volume primitive pass, a volume effect pass and a depth pass. For the glow, I'm hoping to blur it and adjust the fill or opacity, or blend method or something. I'll know better when I get in PS. I don't quite know what to expect from it, as I don't recall rendering a glow pass before.
I did add a glow to the moss stalks. I copied the color gradient and elevation shader from the color channel and pasted them into the glow channel. I wanted to have the red be the main glow, so I changed the green key in the gradient to black. I'll post all that info as well. I may do it in bits and pieces as I have my play this weekend.
BTW, should I leave tickets at the door for you? How many in your party? Will you be parking in Wisconsin or Australia? Since you're from Australia, what kind of special accommodations do you need for your sheep? ;-P
I still need help, though. If you could give me feedback about my 1 - shaders (do fabrics look like fabrics, and metals look like metals, and skin looks like skin?)
2 - postwork (especially whether I managed to make the metal on the robot look less flat and boring?),
3 - titles (looks-, content-, and placement-wise?),
4 - page layout overall, as well as
5 - how does this look compositionally in comparison to my other views?,
I'd really appreciate it.
Side notes: I was pleasantly surprised to find out that my full size render with best quality settings is taking about 4 hours to render (no transparencies or lights through transparency at all, no bump), even on my slow machine. I was expecting much worse...
I like Camera 3 too.
For the post work, yes you definitely made the materials look better! The titles look fantastic, and the page layout is great.
dustrider, that is a superb render. To me it's looks like a real render of a diorama that has been set up in real life , everything s so real!
For the enchanted forest, it's a pitb, Fenric's shader fixer fixes it with two clicks- knocks down the highlight and transfers the diffuse texture to the bump channel if there is nothing in the bump
And sorry for jumping the gun on the image with nothing going on yet smile. I look forward to seeing what’s going to happen there.
About square folds - I am guessing you are talking about the blanket, not the dress? Or were you talking about the pillows too? The pillows could use some work, but I probably won’t re-render the whole thing just of them. With the blanket I was hoping for a not-very-flexible-fabric look, so I didn’t mind the squarish folds. If you are talking about the dress, then it’s a problem, but I am not seeing it. Is there a problem area somewhere?
Hi Antara, no there wasn't a Gun to jump! I just wanted to post something :) that's all.
Yes the square folds are just in the blanket - I wouldn't have even noticed them except you asked and I went looking ;)
If you were going to re-render something like the pillows (not saying you should) , just use a vertex plane with holes cut out as a mask so you only have to render one part - but I guess you already know that !!!! Major part of my work flow that mask trick is
Comments
I feel sorry for the girl ...but the car ...?
Perhaps you have in mind an other cover for that?
Anyway, very nice work!
I may not sleep tonight, and it's all your fault. :)
Very creative. Always enjoy your creative pov.
Wow! It's always such a feast for the eyes here! I truly need to do a search for better exclamation words because I don't have enough at my command to do it all justice.
evilproducer The alien vegetation is so lush and interesting. I feel your pain with the lighting not cooperating. I can't seem to light the liquid around my robot's feet to save my life.
dustrider The skin and clothing materials are beautiful! I hope that poke through doesn't cause the whole forum to go insane.;-P I have Stonemason's enchanted forest , but can't seem to ever get it to look good. It looks great in your scene. Did you have to do anything special with the textures for your setup?
Stezza Love the martian and raygun! Trust us! Ak Ak Ak
booksbydavid Holy @#$%! That's some breathtaking modeling and postwork! Great explanation of your postwork process too! Pssst! The editor says the magazine title is DAZling Stories, but no worries if it's too late to change.
Tim_A Wow! Your minor tweaks make a big difference. The woman really comes alive!…if, sadly, for only a short time longer.:ohh:
head wax That's beautiful and frightening at the same time. The layering and multimedia effects are amazing. Love the transmapped skin layer effect! Can you pretty please show more how you did that? I've so wanted to do something like that for another project. My avatar has V4 and V4 skeleton together then I eroded layers with photoshop masks, but I think yours looks so much better because it has the 3D peeling effect. Pssst! Editor says it's DAZling Stories for you too. :)
Ladies and gentlemen, I'm officially in awe of y'all. What's going on here is nothing but magic. Signed- A fan.
I am not even going to pretend that I can keep up. I am just loving and am in awe of the WIPs being posted here, and feel like I am the only one stuck, while everyone is moving ahead at the lightning speed...
booksbydavid, I love the last postworked image. Absolute stylistic perfection.
Tim_A, I like the updated colors.
head wax, great cyborg idea and technique, but I am slightly confused by the image with the title. I can't quite figure out what's going on there.
dustrider I deeply admire your perfectionism :) I hardly noticed that poke through even in close-up... :) But why the insistence on no postwork? Is there a special purpose for it?
Stezza, your Martian is adorable! :) It also inspired one of my titles :)
sukyL, about the foreground: in addition to bubbles or just some tiny replicated debris, you can also add murkiness in post by using the distance pass as your mask for various fogs (I would not recommend adding more fog to the scene unless you really have to, because that would slow down your render. If render time is not much of an issue, then adding scaled down clouds might be a good idea to add murkiness that's not too even. I found clouds look better than just fog with turbulence added in).
About my cameras: I actually had this idea initially of placing floating robot cameras all round these figures, but then I figured that with the addition of text it would make the scene too busy and harder to read, so I decided to go with simplicity instead, but yeah, it would add a whole new layer of meaning to the scene, and I really love that idea.
Now to my own progress: here is my first post-worked cover (Camera 6), before and after post. (The stories titles might change if i come up with anything better.)
The angle is kinda the combination of my two original extremes. I might have killed the best points of each and ended up with something which is just 'meh', so if that's the case, please tell me.
(Here is the original POV dilemma: http://www.daz3d.com/forums/viewreply/694113/
and here is another alternative I am considering: http://www.daz3d.com/forums/viewreply/694225/ )
I still need help, though. If you could give me feedback about my
1 - shaders (do fabrics look like fabrics, and metals look like metals, and skin looks like skin?)
2 - postwork (especially whether I managed to make the metal on the robot look less flat and boring?),
3 - titles (looks-, content-, and placement-wise?),
4 - page layout overall, as well as
5 - how does this look compositionally in comparison to my other views?,
I'd really appreciate it.
Side notes: I was pleasantly surprised to find out that my full size render with best quality settings is taking about 4 hours to render (no transparencies or lights through transparency at all, no bump), even on my slow machine. I was expecting much worse...
Another postworked angle for comparison. (For reference, this is Camera 3) And I am setting up 2 more to render overnight, so I will post them tomorrow. ...by which time this thread will have 7 more pages :) and everyone will forget what these images looked like...
Hi Antara
this camera is fantastic!
I like the scene and the texture (Great reflection on the robot surface)
I love the title “Robot and Jiuliette”. Great!
-
Looks great Antara! A really nice job on adding emotion to the robot through pose alone.
The metal looks much better.
Antara I totally agree with pimpy. Camera angle 3 is fantastic! It places emphasis on the robot to show its emotion and the clasped hands very well. Robot and Juliet is a great title. Romeo and Juliet is exactly what came to my mind when I saw your first render. The shape of the DAZling stories text and general text layout complement the scene very well. The hot and cool colors are very effective and metallic texture on the robot is lovely. Looking forward to checking out your next renders as well.
I don't think your cover is bad at all. I do think a couple more dynamic poses along with some work on the camera angles could help sell the drama.
For poses, if you look at the collage in your post, the super hero picture has the woman's leg forward and the knee bent. I think if you did something like that for the femme fatale, that may help loosen up the stiffness of the pose a bit.
Looking slightly up at the subject may help a bit as well. I would also refer to some of the other pulp fiction covers posted here and try and add a dramatic background. It does not need to be complicated. It's film noire inspired, so maybe the background is a dark color with a circle, or oval shape suggesting a spotlight, or a rectangle with a blind pattern or bars. You could even add it in post as the 'toon scene effect can use the alpha channel.
Here's a quick example of what I mean. Made a layer with a circular gradient, then another layer with the window shape. Took less than five minutes. Most of the time was fiddling with the gradient.
I may not sleep tonight, and it's all your fault. :)
Very creative. Always enjoy your creative pov.
What? You sleep ? :) ! ;)
I'm having a bit of dilemma with my scene.
I have a spotlight shining onto my moss which is hair. It is supposed to fake the light of the glowing stalks. Lights grouped with the stalks looked great but were taking forever to render. As you can see, I have the spot restricted to the moss hair models. The light on the astronaut and surrounding rocks and plants are the lights grouped with the replicated stalks. They are set to exclude the moss hair. As you can see, the moss has no red light. I've tried deleting the lights and adding new ones, still nothing.
On the plus side, the astronaut's light is back to where it is supposed to be. Or at least close enough for my purposes. Widening the half angle a couple degrees did the trick.
I think I'm just going to suck it up and use the replicated bulbs. It added a dynamism to the scene that faking it couldn't do.
Hi EP
I like the scene (is quite dark but as a test is ok) but...could be an idea try with a light dome to create a diffuse atmosphere and put the red lights at the base of the plants?
sukyL
Oh, I made a few renders before work this morning - will post them later. The peeling effect is from one of Ron's Cyborg Brushes. I did this morning's renders on a sphere and came to the conclusion you are better off painting the alpha channel with a 3d painter .
Basic method. Have one Vicki posed as you like. Then duplicate her with Fenric's tool/plugin. Shrink the second one doan a little.
Take the first one's face texture into photoshop, make a new layer, paint your alpha channel black where you want it to be clear and fill in the rest with white. Save this. Use this on the outer most vicki in the alpha map section.
Take this alpha map. Save it as new image. Select the black parts and delete.
Make a new layer underneath and paste your electronics etc. Make the top layer a multiply parameter and add a drop shadow so the electronics layer has a shadow on it.
Make this the texture map for the inner most Vicki, use the original alpha map on her but invert it.
I then took the electronics layer and used the free Nvidia normal map maker to make a normal map and used that in the bump channel.
You need to play with the edges of the alpha map to decide whether you want it soft or hard. Hard will give more of a peel effect.
Also for the electronics Viiki you can use the electronics texture in the glow channel then add aura/glare in the Scene parameters
i'll post pics later
The rule with faster dynamic hair lighting that I discovered recently: Duplicate your lights, have one set affect ONLY the hair, have the other set affect everything EXCEPT hair. Changing the hair settings to not cast shadows might also help, but might flatten the look of them too.
But in your case it might not be enough. Since you want multiple light that you need to shine on the hair = longer render time.
- One way to avoid it would be to fake it with a gelled single light on the hair moss alone and nothing else. After all your glow from the stalks is not going to be all that precise anyway.
- Another way to fake it would be to partition your hair into multiple objects and have one light per one object ratio. Why? to avoid multiple lights interacting with too many hairs and affecting each other, thus increasing computation intensity.
Also, I would recommend adding shading glow to the stalks and adding a 3d Aura filter to your scene or the Aura Effect in your stalk's object's Effects settings. This will help sell the red lights on the moss.
Hope this helps.
Thank you, sukyL, evilproducer and pimpy for your comments!
So far it looks like Camera 3 is the winner :). I'll see whether I like the renders I get in the morning enough to mess with them and post them too.
I'd love to hear more opinions about Camera 3 (vs Camera 6 and vs original POVs) from people who originally liked the second image (the one on the right in my original POV dilemma post) better.
hello Antara :)
thanks, that's because nothing s going on ... yet :)
for your image, its' fantasticagorical ! Love cam 3, the sharper diagonal is very fast.
you asked about the cloth - some of the bends in the cloth are a bit square - doesnt really worry me though, just if you want nitpicking
it's a great render, everything sits so well
edit: I started this post half an hour ago, just got to post it - gah
evil, your scene is really looking good, I'm sorry I haven't had a chance to comment on it, the thread moves so fast.
and thank you for sharing so much about your method
any reason you aren't adding something to the glow channel of the moss stalks?
for me, as alover of post work, I would render out an object pass and use that to select the stalks in ps and add the glow there ;) lazy bugger that I am :)
The rule with faster dynamic hair lighting that I discovered recently: Duplicate your lights, have one set affect ONLY the hair, have the other set affect everything EXCEPT hair. Changing the hair settings to not cast shadows might also help, but might flatten the look of them too.
But in your case it might not be enough. Since you want multiple light that you need to shine on the hair = longer render time.
- One way to avoid it would be to fake it with a gelled single light on the hair moss alone and nothing else. After all your glow from the stalks is not going to be all that precise anyway.
- Another way to fake it would be to partition your hair into multiple objects and have one light per one object ratio. Why? to avoid multiple lights interacting with too many hairs and affecting each other, thus increasing computation intensity.
Also, I would recommend adding shading glow to the stalks and adding a 3d Aura filter to your scene or the Aura Effect in your stalk's object's Effects settings. This will help sell the red lights on the moss.
Hope this helps.
Did all that. ;-) My intention is to post my light settings when I have chance. You'll see that I use light linking, gels, etc.
My problem was that I couldn't get a single light linked to the hairs to illuminate them for some reason. Maybe I've hit some kind of light linking limit? Each moss clump also has it's own hair applied. You are right about turning off shadows for the hair- One that it would save render times, and two that it would look flat. I would rather have long render times than flat hair. I was going to try a glow in postwork as it would be faster, and since Carrara's aura is a post render filter, I think it would be more efficient to do it in Photoshop.
So here's a screen capture of the moss. remember, the little stalks are grouped with a single bulb and are replicated using a surface replicator. each moss clump (there are four) has its own hair and surface replicator for the stalk/bulb group, so a total of four hair objects, four stalks and four bulbs. Each stalk is replicated around 120 times.
EP: But if you are replicating the stalks/bulb groups, wouldn't the lights get replicated as well? therefore, wouldn't you have the 120x4 light-bulbs as a result?
I don't know how many hairs you've got and what the settings are, but I had the situation with hairs and multiple lights, in fact - multiple hairs and multiple lights and although it did slow down the render, it wasn't prohibitively so - maybe by something like 30-40 min for the image where the hair covered about 45% of the area. And I only have 4 slow cores and under 8Gb of RAM, so with what seems like a much more powerful system, you shouldn't be getting into serious trouble with this. Therefore I am trying to think what could be causing the slow down.
Yes. That's why I tried a couple different non-replicated lights to light the moss, with just the astronaut, boulders and plants linked to the little bulb lights. The hair was excluded. I tried the shape light and spotlight linked to the hairs with various ranges, intensities etc. I just couldn't get them to light the moss/hair for some reason. The screen shot I posted was the last light I tried at full value. It should have looked like a flash bulb going off, but zippo!
It sucks for time, but the only way I could get the moss to look lit, was to include the moss hair in the replicated bulb lights.
Well, I think I may have the final render, maybe, I hope. I also added a screen grab of the render running in the Octane Render View Port so you can get an idea about the Octane side of the system resources used.
I think I've mentioned everything about the set up except the lighting. The scene is lit with two lights. One main mesh light in front on the right side of the camera, and one small mesh light behind the dinosaur to give a bit of light to the dark area. The only post work on the image was re-sampling to reduce the size from 2400 to 2000 pixels high, and a slight levels adjustment to brighten the image just a touch.
lol - That's and easy one, I suck at post work :bug: It was also a bit of a personal challenge. After I saw the results of the first bit of text, I wanted to see how much I could do for this challenge directly in Carrara, even though logic would dictate that a fair amount of it should be done in post.
I typically don't do much post work. Other than not being very taented at the artistic aspects of post work, I'm a bit too logic oriented, which means if logically things like poke through "should" be corrected in the render (if possible), not in post, then I have to fix the original geometry. That, combined with a bit to much perfectionist and a dearth of artistic talent, means that I'm naturally inclined to push the render end of things to the best I can get, rather than relying on post work except for simple fixes.
My computer is a single core 2003 PPC G5 1.8 Ghz with 4.5 GB of RAM. I do network with a Dual Processor MDD G4 and an Intel Core 2 Duo iMac running Mavericks. The C7.2 render node works on the iMac, but the stability to run the full app is just not there. by rights, I shouldn't be able to run C7 at all with Mavericks.
Thank you! :)
And sorry for jumping the gun on the image with nothing going on yet :). I look forward to seeing what's going to happen there.
About square folds - I am guessing you are talking about the blanket, not the dress? Or were you talking about the pillows too? The pillows could use some work, but I probably won't re-render the whole thing just of them. With the blanket I was hoping for a not-very-flexible-fabric look, so I didn't mind the squarish folds. If you are talking about the dress, then it's a problem, but I am not seeing it. Is there a problem area somewhere?
evil, your scene is really looking good, I'm sorry I haven't had a chance to comment on it, the thread moves so fast.
and thank you for sharing so much about your method
any reason you aren't adding something to the glow channel of the moss stalks?
for me, as alover of post work, I would render out an object pass and use that to select the stalks in ps and add the glow there ;) lazy bugger that I am :)
Hi Andrew,
I'm rendering a glow pass, a volume primitive pass, a volume effect pass and a depth pass. For the glow, I'm hoping to blur it and adjust the fill or opacity, or blend method or something. I'll know better when I get in PS. I don't quite know what to expect from it, as I don't recall rendering a glow pass before.
I did add a glow to the moss stalks. I copied the color gradient and elevation shader from the color channel and pasted them into the glow channel. I wanted to have the red be the main glow, so I changed the green key in the gradient to black. I'll post all that info as well. I may do it in bits and pieces as I have my play this weekend.
BTW, should I leave tickets at the door for you? How many in your party? Will you be parking in Wisconsin or Australia? Since you're from Australia, what kind of special accommodations do you need for your sheep? ;-P
I like Camera 3 too.
For the post work, yes you definitely made the materials look better! The titles look fantastic, and the page layout is great.
dustrider, that is a superb render. To me it's looks like a real render of a diorama that has been set up in real life , everything s so real!
For the enchanted forest, it's a pitb, Fenric's shader fixer fixes it with two clicks- knocks down the highlight and transfers the diffuse texture to the bump channel if there is nothing in the bump
Hi Antara, no there wasn't a Gun to jump! I just wanted to post something :) that's all.
Yes the square folds are just in the blanket - I wouldn't have even noticed them except you asked and I went looking ;)
If you were going to re-render something like the pillows (not saying you should) , just use a vertex plane with holes cut out as a mask so you only have to render one part - but I guess you already know that !!!! Major part of my work flow that mask trick is