Couple of questions about scenes and renders
3dOutlaw
Posts: 2,471
Is carrara capable of making hi-definition images that can be used in Poser for backgrounds? If so, how about equirectangular style images for use in an Env Sphere?
Also, is it OK to offere for freebie, a preset scene that you make in Carrara, if you do not give any of the assets (if that is possible)?
If so, has anyone made a beach scene?
Thanks!
Daz 3D is part of 
Connect
DAZ Productions, Inc.
7533 S Center View Ct #4664
West Jordan, UT 84084
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.
Comments
Use a sphere cam and set your output render to a ratio of 2:1. Like 4000x2000 or even higher.
I don't under stand the question. Are you talking about giving away a Carrara scene or a rendered background?
There are a few beach scenes in the native content I believe...
Yea, sorry, giving away a Carrara scene file that is already made. So for example, if I made a nice mountain or beach or other scene using Carrara assets, saving the scene, and giving away the scene file as a freebie. (i.e. they need Carrara to load it)
(ill try the sphere cam here, and see how it works, thanks)
If you are not using anyone else's models or textures...., just using Carrara tools like terrain, plants, skies, or your own models/textures then you can give away the whole scene as a freebie. There use to be a Carrara Freebies forum.... You can also check ShareCG to see what Carrara freebies are usually given away....
Thanks Holly :)
OK, so those Spreical images come out a little distorted on an Env Sphere...? Is that for a Dome or a Sphere?
(unless my Zoom and distance to near plane settings are wrong...not sure about those)
Compare attached carrara photo to this photo: Equirectangular beach
The linked one has two swells, versus the single lump, if you will
Without knowing the available options (haven't touched Carrara in a very long time...and never did get that far into it), I'd say it's likely, just by looking at the way it's distorting.
It looks to me that your spherical camera is not quite horizontal. Check the "motion" parameters tab and make sure that your camera points to the horizon with an X rotation of 90 degrees exactly and zero for the other two rotations (or is it -90 - one or the other!) and see if that works.
The OP keeps asking about equirectangular. Not sure if Carrara does that. I know it does spherical renders, as I've done a couple myself for backgrounds.
For the Spherical renders, what did you apply them to? I am trying to apply them to the inside of a large sphere (Bagginsbill's Environment Sphere to be exact). Then I make the ground plane a shadow catcher and ta-da!
Anyway, here is where his Sphere is at, but it call for Equirectangular...which seems to be different than the "spherical" that Carrara is producing.
Ugh! Sounds overly complicated and maybe unnecessary. You can create a sphere primitive or create a vertex sphere and use that for your image map/shader, and then scale the sphere to whatever size you want. If you want IBL lighting, like an HDRI, then you can put an image, HDRI, color, color gradient, etc. in Carrara's Scene background, and it will wrap it around the virtual universe, then turn on Skylight under GI in the render room and done.
Here's a spherical render I made in Carrara 7.2 Pro. It uses Dystopia City Blocks and some customized shaders. Try it in the scene's background. For extra fun, use it as an IBL by turning Skylight on. I also have a daylight version with a sucky mountain in the background.
http://www.sharecg.com/v/64485/view/6/Texture/Spherical-Dystopia-night-with-clouds
http://www.sharecg.com/v/51297/gallery/6/Texture/Dystopia-Spherical-background-image-map
Yea, see those 2 spherical renders you made there are correct. I am not getting the same?
Other than an OBJ boat in there, its all Carrara 7.2 pro. I got a background sky map
Here is the basic setup I have.
...and here is the render. Its nothing like the spherical type you have there in your sharecg stuff. I must be missing something. ????
OK, I may have found the problem. My focal length on my camera was set to like 75, and it should have been around 15mm or so. This may be workin.
The trick to this is to place your spherecam at the same height as the camera in the other scene. That way their "eyes" match and the ground planes will line up.
If you are making a 360 for viewing (as if through the character's eyes) then the camera should be at about 5 ft high (or a little less seems to be the average tv show camera height...).
The results of this render, the 2:1 image you have posted is called Equirectangular. This image is ready to wrap onto a sphere. In Carrara you can load it as the Background.
OK, I'll try that on the next one. I just switched to a large non-spherical render and dropped it in as a background via postwork instead, since I could not get it to work properly. (result)
Thanks for the tip!
OK, I'll try that on the next one. I just switched to a large non-spherical render and dropped it in as a background via postwork instead, since I could not get it to work properly. (result)
Thanks for the tip!
Nice picture! At the risk of going off-topic, if you applied a surface replicator to the suit and populated it with little spheres with an appropriate water shader, you could simulate a wet swim suit.
I guess I could, but I rendered her in Poser. I find it much easier to get a better "character" render than in Carrara. Because the focus on Carrara had stalled it seems like, I do most stuff in Poser now, and probably am gonna pick up Vue for Christmas. I was giving Carrara a last shot at backgrounds here. I still like it, but sort of a shame. :(
There are some methods in Poser to get a wet look, but since I did not have wet hair.
I suppose she just zipped over on an inflatable mini-boat from the yacht and wore her wetsuit for the underground river that she is traveling to... :)
It's your picture, and it looks really nice. If you don't want a wet look that's cool. I wasn't trying to imply it was lacking. I've just had good luck in the past with that effect.
As to the spherical backdrop, I don't see that it's that hard or inferior, especially compared to the link you posted about the pre-modeled Poser sphere with all it's caveats and size limitations. There is a learning curve with Carrara to be sure, but then again it's a more feature rich program than DAZ Studio, Poser or even Vue. They all have their place, and Vue is exceptional as a terrain renderer, so maybe that's a better way to go for backgrounds for you. Personally, not sure I'd even bother with Poser except for maybe the walk generator, but that's just a personal opinion and is not meant to reflect on anybody else' preferences.
NP, it's definitely got more features! I think Poser has got it beat in render engine now, but that's just my opinion as well. Seen some real good stuff come out of Carrara.
Anyway, got a little OT...thanks for the suggestions!
OK, back OT, I am trying to get this equirectangular thing working again.
Made an infinite plane with a checkerboard, and a spherical camera. Set it up for 78.5 feet off the ground (???) and get the first pics, then raised it to 1000 feet off, and get the 2nd pic. The second looks closer, but I have no idea the relationship between settings, height and things here? Anyone ever make any of these in Carrara and have some tips?
What scale is the checkerboard floor suppose to be? 1ft per square? 50ft per square? 1 mile per square?
You act shocked that these numbers seem completely arbitrary, but without anything in the image to supply a sense of scale I don't see how it is suppose to look "correct". The objects will distort the closer they are to the camera. That is normal since it is essentially an extreme fisheye lens.
Try adding something with a "known" scale, like a simple plane scaled to 10ft x 10ft. Then set the checker shader to show 10 squares horizontal and vertical. That will give you a "kitchen floor" that lets you see exactly how far the ground plane is from the camera.
Good idea, I will give that a whirl, thanks! I will post results and measurements. Gotta run to start vacation, without internet (yes, it can be done!)...but will post back later.
My 1st goal is a terrain (such as desert or grass field) with a distant mountain range or forest, that I can add a figure, a shadow catcher plane and render in other software.
From how I understand it, if your spherical camera is 50' above ground when you shoot the render, your render camera will give better results if it is 50' above ground when using that spherical image as a background in that image, with a shadow catcher adding shadows from what it now in this scene onto the background.
Less wordy version:
Say you want to render a car coming from a distant edge and zooming up through the view of the render camera in the final animation.
Put your car in the scene and set up the camera how you'd like it. While you're at it, envision exactly what your background should be. Follow me so far?
Now select the camera that you've just aimed and go to the motion tab and look at the number in the height (z, or blue) box. Rotation and stuff won't matter so much - as you'll be using a spherical image. The height that this camera is at is the height that your spherical camera should be at in order for a shadow catcher plane to work properly between the two. You'll likely be three to seven feet off the ground for this example.
Just know that, due to the enormity of what Carrara considers to be the entire surrounding spherical background, when people say that you need to be at 8000 x 4000 pixels to help cut down on distorted pixels toward the feet of the render camera's cameraman, that is about as minimal of a resolution as you'd want for this sort of example, which takes quite some time to render, if you've got a fairly detailed surround.
This is where you can help fudge the entire process by counting on using something as a ground to hide the bottommost portion of the projected spherical image in the BG, if the camera is to come anywhere near pointing down.
In your examples , above, you can increase the probability of having an undistorted image towards the "looking down" position by raising the spherical camera higher. But if you render that into a background spherical projection, you'll need to keep your camera above the subject in the subsequent scene or it will be floating above the spherical images ground - floating in the air.
So to help all of this along, we'll do two things. First, we'll use the measurement we took for the car driving towards us for that shot and we'll build our scene and set the spherical camera to that height and make sure that it is looking right-side up.
Before we call this done, let's load that car back in and set it along it's motion path, passed that final camera that we've used for our height measurement. This time, now that we have our completed scene, we can double check our measurement for best results.
But now that we have our car in motion, why don't we add a new camera and parent it to the car - setting it's view to a dramatic angle, changing ever so much over the course of the animation. Now you can render a backdrop image for use in that shot. Backdrop images are much easier to deal with in this regard because they only render what will be seen in the final animation. So these images are faster, even at high res, and can then be used to render animated backdrops - which is incredibly inefficient for something as giant as a spherical image.
Speaking of practicality, you may just want to render the car at the same time as that motion backdrop and be done with it. I just wanted to show that possibility - which can then be used for many shots that need the same motion in the background - like outdoors scenes with birds flying or kids playing for use on interior scenes - this is what's happening outside, etc.,
So the scene is built and we've checked the height of our final rendering camera against the height of the spherical camera. The huge benefit of this is that it doesn't need to be an animation - as it surround the entire scene with something - which will also assist in Global Illumination results, reflections, etc.,
8000 x 4000 is very large. Going a few numbers higher will make very subtle differences and going a lot larger begins to remove the benefit of speed - as it will take a very long time to create - for a super-detailed surround.
So just go for it and keep from aiming the camera towards your feet, without having something between the camera and the BG image! :)
Hi 3DOutlaw. :)
In Carrara, you don't need to use a Sky dome or any OBject,. to have a spherical 360 degree environment. or use HDRI.
All you need to do is select "Scene" in the instance list,. select "Background", and then choose Map, or HDRI from the menu,. and browse to your file.
if you just want the environment image,. use Map, and load a jpeg or whatever format.
If you want it to light the scene, as well as provide the background,. then use an HDRI, and enable "Sky light" in the rendering options.
As Holly mentioned,, having a scale, and a texture to your Floor,. will make a difference,.
The Spherical camera, should ideally be around a foot or two above the scene floor level, but can be placed wherever you want it,. but it will become the Centre of your world, or, the central point in the rendered 360 environment image.
this is a spherical environment I made in carrara, using an infinite plane with a texture, and the Realistic Sky Editor,. then converted to HDRI in Photoshop.
The car is the "Prince" (I think),, from the Daz store, with a shadowcatcher plane under it.
Pic 2 is the same, but with an infinite plane added, with my wet tarmac shader.
The height of the spherical camera should ideally be the same height as the render camera in Poser (or whatever), unless you want Vicky to appear to be 2ft tall or 1000ft tall.... Of course if your background scene has no evident scale it won't matter, but if you have trees or objects with a known scale in the background it will become obvious very quick that things are "off".
Dart, I think is possibly describing the pixel distortion backwards.... The areas at the feet and zenith actually have the most pixels squashed into the smallest field of view. It's the horizon (the middle band of the equirectangular render) that will most likely show the pixel artifacts in your final scene. Depending on the field of view of your Poser camera, you may need VERY high resolution. A wide-angle will show more of the background (hence the pixels will be smaller), a telephoto camera will "zoom in" on the background and reveal pixel artifacts....