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
Right... puzzled me for a bit, once that pane was enlarged like that. (thanks for that, but then I happened to glance over to the left and saw the other tab, the one that says 'Browser'.
and now I found all those lighting sets - neat!
but I have another question, how can I move the 3 axis pointer/controller downwards without moding the model down?
(This is so that I can be close enough to see the characters' feet and move the rooms up, down or sideways so that their feet are on, rather than IN the floor) If I'm close enough to see the feet easily, then the controller is out of view - totally.
In C4D you click the 2nd icon - move the controller and click the 1st icon again (to sort of re-attach the controller to the mesh) and then the controller is where you need it to be, and you can move the mesh again.
How do you do that in Carrara?
Plus how can I move the view in Left, Right, back or front views? None of the movement buttons work in those views.
Caps Lock to move the Hot Point
Alt + Mouse Drag (Right mouse = Pan, Left Mouse = Rotate), but you cannot rotate while in those views, just Pan.
HI :)
It's called the Hot spot,. and you can Lock/Unlock it using the CapsLock key on your keyboard.
Use the Roller button on your mouse to zoom in out of the selected view.
Or,..
Use the Zoom tool (Magnifying glass) to zoom in (ALT+Click) to zoom out.
You can also hold (ALT) and hold the Middle mouse (roller button) down and move your mouse to zoom in / out.
Use (ALT) and hold the Right mouse button, and move your mouse around, to Pan in the view.
In the 3D perspective views, you can Move around, Zoom and Pan by holding down the ALT key and click/dragging in the empty space of your scene
ALT plus your mouse buttons = Easy navigation.
Hope it helps :)
Brilliant, worked beautifully. Phew!
The mention of "only the visible faces" got me ..
All models in All applications need to have UV mapping,.
some basic UV mapping is normally added when you create your model, and it's also given a shading domain.,.
Your floor should ideally be a Plane. rather than a Primitive object like a cube,
Oh? Why?
For a floor - well okay, but for a wall??? Surely a cube is better? Makes it easier excluding unwanted external light, and you can have the wallpaper on the inside and bricks on the outside. Why not?
Also I find lining the feet up with a cube much much easier than with a thin plane that's hardly wide enough to see...
I have some floors that are cubes... but for necessary reasons. Walls always benefit from the depth of a cube as well. Yeah, I think he meant just the floor.
HI :)
If you look at the majority of "premade" sets available,. they're using Flat surfaces for walls,. Internal walls are normally separate objects, from the external walls. and the exterior is normally more decoratively modelled than the inside structures.
A room made of planes will block light just as effectively as a room made of Cubes.
Easier UV mapping,..and efficient modelling,.. are the why's
You still have another four sides to deal with, what texture do those faces have !.
I'm not trying to say that you can never do this,... if you want to do it that way for your own use,..,.. then do it,..
but be aware that it's not a good method from a Modelling, or Mapping point of view.
If you want to learn,. learn the right way,.. don't learn the bad ways,. :)
From a potential "Product" and customer point of view,. Why would I buy something that I can easily make myself ?
Also,.. you mentioned that you didn't make your Character's legs by extrusion,. can you elaborate on that..
It may be important for rigging, or from a potential product point of view.
Just to add..
On the issue of placing figures on a floor, or any other objects..
Carrara has an function to detect collisions, which can be switched on and off as needed. ..leave this switched Off. for most of the time you're working as it'll be constantly trying to detect any collisions, and that can slow things down.
Yes. (Hangs head in shame,) I'm sure it would be better to learn the 'right' way.
No, I wouldn't do it that way for a product, but these characters and interiors were not intended as products. They are just for the animation.
Erm... the legs are 2 separate meshes, the body is another mesh, the eyes, mouth and icicles are all separate meshes as is the icecap.
But I have managed to animate this rigged mesh in C4D, so even with the legs etc as separate meshes, rigging, weight painting and animating are still possible.
I will just say that although I set the materials to import as separate groups I think when I bound the skeleton and controls to the mesh I got just one single skin object. This is all in C4D of course, and terms and methods of doing all this will be different in Carrara.
Here's a link to the (done too fast and very badly just this evening) resultant animation.
FrozenPeaWalkTest1
I do know it's not the way to do it but I also knew I had 12 or more characters to create, and I knew what a nightmare they'd be to UVMap as one single mesh, so I thought - try separate meshes, - if you can't animate with the results then you'll just have to go back and re do it.
So of course I'm hoping I won't have to do that.
You can still rig it. It may mean a little more work on the weighting, but I've rigged groups in the Assembly room that consisted of different model types, such as vertex and spline models in the same group. I wouldn't recommend that as it was kind of sloppy.
HI All :)
Sorry,.. I didn't meant to slap or bully anyone,.. just trying to point things in the right way.
On the Characters:
You shouldn't need to rebuild anything, as long as you have each character as an Object, even if it has separate parts.
Separate parts are, as you say, easier to map.
Most of the Daz3D figures are made of separate poly-meshes, so, no problems there.
I was checking because "product" was mentioned,. and some meshes you cant just copy bit's of and use in your own models.
on Mapping :
Yes, cube's are easy to map, since it's a standard primitive shape, it also has a standard UV Map template (Box mapping)
However,. there are different ways to handle how each face of the Box is UV mapped.
You could create separate shading domains on each face,. and apply a different texture map to each.
you could use a single texture on all sides, and have a single map which has areas of "outside wall" and "inside wall" textures.
The first method needs different shaders for each domain.
the second method needs only one, but the drawback is that the areas of texture are smaller, since the layout is different.
With a flat wall,.. you only need to deal with the flat wall, (planar mapping) and not the parts you won't see.
I suppose it's really up to the user, the scene they're building, and what the camera is going to see. that determines how they approach making their models,. also, whether it's for your own use, or to share.
Cubes and other primitives are a quick way to build,.. if you don't know how to model, but it's better to learn how to model.
for example,. an interior room wall, could have a skirting board at the bottom, and a cornice at the top,. and modelling those is the only way to do it,.
I'm out for today,. be back tomorrow
:)
Yes, for instance if one part needs transparency, or to be highly reflective, while the rest is... brick, leather or rust, etc. I find that if I need some areas to end up as glass, then I either give those parts of the model a separate material – with reduced opacity, or just a differently named grey would do. Since I'd then apply either reflection or transparency etc in the rendering program.
But that's because I don't render direct from Wings but import models to Bryce or Cinema4d for rendering, although you can use a PovRay, Yafray, or Kerkythea plugIn to render direct from Wings, I don’t bother since most of my models become part of a greater scene in Bryce or C4D.
Otherwise I might simply model a part that needs a different type of texture, separately – like the glass in a single window could become a single cube. Or each pane has the lower opacity while the frame has a different coloured material applied. The cube method works better if you need to show renders from both inside and outside that same window in the same animation.
Things may be different when modelling, UVMapping, texturing AND rendering all in the one program. It's why I'm interested in what you do in Carrara...
The first method needs different shaders for each domain.
the second method needs only one, but the drawback is that the areas of texture are smaller, since the layout is different.
If it's for a small scene – i.e. not many polys in it, then you could increase the size of the map... But unless there's a reason why you can't change the camera's angle during an animation (like it needs to be one long continous shot) then a change of camera angle can aloso incorporate a total change of scene – like now all the buildings we just passed in the train, now only have texture maps of say... 512x512 – no need for bigger as we are moving rapidly away from them anyway.
I suppose it's really up to the user, the scene they're building, and what the camera is going to see. that determines how they approach making their models,. also, whether it's for your own use, or to share.
Cubes and other primitives are a quick way to build,.. if you don't know how to model, but it's better to learn how to model.
for example,. an interior room wall, could have a skirting board at the bottom, and a cornice at the top,. and modelling those is the only way to do it,.
I'm out for today,. be back tomorrow
:)
Yup... I don't see how else you'd do it... unless you're thinking of painting an image of a skirting board onto the texture, a sort of tromp l'oile effect – but you'd have to be careful with that depending on the camera angle it will be seen from and you'd need to paint in the shadows to match the rest of the scene... much easier to just make a separate skirting board and let the software worry about shadows & highlights etc.
You mentioned my bringing up 'products'. Yes I did, but nothing I've posted yet has had anything from the range I wanted to sell in it. Not deliberately anyway. These are what I want to sell:
Carriage Lantern Antiqued
SpiderLantern
and this:
PoliceLantern
A different texture set for the carriage lantern:
Weathered Carriage Lantern
And here's a render of the whole set
LanternsAll
– at least one example of each anyway, the square stained glass lantern has 3 different designs for the glass, and 3 different types, Hand, Desk and Standard Lamp, and 3 different texture sets for the metal: Plain Brass, Verdigris and Copper with Verdigris.
Then there's the Police Lantern. Just has a single texture set. Wrought iron metal with a gold painted crown but with 2 slightly different textures for the glass and a few gels for the light.
Then the Spider Lantern. 3 texture sets: A pale copper with some verdigris, A bright copper with more verdigris, and a cast iron texture. Plus a plain colour for the glass and a web texture also for the glass. (includes a metal black widow spider and a web design wall bracket.)
Then the Carriage Lantern, with 3 texture sets: Brassy, Antiqued and Badly Weathered.
Very nice... I like them!
Aw... I love the motion you had on that little guy!
The fact that the body held so still, and those little legs did all the work... I loved it. Had to watch it over a few times :)
Rofl! Actually it's a little girl. I'm quite surprised you think the body held still, since I actually tried to get that effect. Of course most walks have an undulating motion on the body.
Um, I suppose you don't know of anywhere you can get or see videos of walk motions do you?
I want a slowish relaxed sort of saunter, if I can find one.
Not off hand, no. When I go out in public, I watch people! LOL
I do. I pay attention to all sorts of stuff that people do, and try my best to log it into memory for animations. Then I key frame them in!
Strange, eh? But that's how I get my motion ideas.
Hi Fran, when you rig an object, even if it is spline object, once the rig is attached it is converted to a vertex model. Depending on the spline model, this can create a somewhat messy underlying vertex mesh. The main issue for me though is that if you have a group of objects in the assembly room and then rig the group, each object in the group will need to be selected individually to work on the weighting. It makes it difficult to see what you're doing in relation to the other parts.
What I'm talking about is different than an object with multiple polymeshes in the vertex modeler. That will show everything in its proper relationship in the vertex modeler.
A few years ago I made a broom figure in homage to Disney's Sorcerer's Apprentice. It is made up of a combination of spline objects and simple vertex objects that I kit-bashed from Carrara's native content and some of my own elements. I grouped the separate parts together in the Assembly room and then rigged it and attached the skeleton.
You can see in the images below that I selected an arm piece, opened it in the VM and selected the Animation mode to weight paint. Notice that not only is the arm section the only part visible to paint, but the arm is in the orientation it was in when it was made. This throws off the camera perspective in relation to the rig in the vertex modeler. You can get around it somewhat by modeling in the Assembly room, but you're still stuck with editing one piece at a time.
No, it's not strange at all. But I have a very poor memory, plus there are often bits you can't see - like Eduard Muybridge and the horses - you could only actually see that all 4 feet left the ground in a gallop - when you paused the video - or for him back then, when you looked at each still frame of the film.
You can only look at one thing at a time - you're eyes can't focus on two different feet when you watch a cat walking, or I can't. I found I couldn't tell what the front paw was doing when I was focusing on a back paw, and I needed to see if they were both on the ground at the same time - or one was slightly ahead of the other... and which part of the paw did what... And I just couldn't tell until I managed to get some video of a cat walking and could pause the video. I ended up being able to Rotoscope the movement using the video.
Also of course is the problem that I don't want motion ideas, I've already got my own ideas - I want to see a particular kind of walk that I already have in my head, and unfortunately I don't know any actors - when I was at uni, I once asked a flat mate to walk for me - she wasn't an actress and the result was terrible - she couldn't take direction. In that you asked her to walk in a particular way and she just kept on doing the exact same thing as if she was ignoring the request... she wasn't, but she just couldn't act.
I once directed a play and the guy playing Buttons was a dream to work with, we decided he could do one of the songs to a broom as if it was a microphone - he did Freddie Mercury - and won an award for it!
You know the walk you want, so maybe set a video camera on a tripod and tape yourself. Get it from the side and front and maybe the back. You only should need a few steps to set up a cycle, so a really large field of view wouldn't be needed.
There are some walk tutorials on-line that are free so maybe do a search for those. Here's a couple I found on youtube:
https://www.youtube.com/watch?v=OdhuDGOyAXc
https://www.youtube.com/watch?v=qzhjNele4_A
I see, but what is the advantage to rigging a group rather than merging the objects and then rigging an entire single object?
I did try rigging the character with each piece as a separate object – and it just didn't work at all, things went haywire. But when I combined the objects in Wings3d and then imported that then single object into C4D, I could rig and weight paint with no problems at all.
I did try rigging the character with each piece as a separate object – and it just didn't work at all, things went haywire. But when I combined the objects in Wings3d and then imported that then single object into C4D, I could rig and weight paint with no problems at all.
You answered your own questions about rigging a group of objects. It can be tricky. That's why I said it was a mess. As to why I didn't stick them into the VM as separate polymeshes of the same object, it was because it was few years ago, as I said. I wasn't experienced in the VM and the objects were a mixture of spline and VM objects. Since it was for my own use, I didn't care what the underlying structure looked like. I just wanted something articulated in the least amount of time. I could easily convert the spline objects to vertex objects and copy and paste them into the VM- now. Then, it would have taken me far longer to do.
Yeah, I don't try to catch everything at once either. When I'm noticing things that people do, I focus on pretty much one aspect. Sometimes I'll pick up on more, but it's often just one thing in particular that I'm noticing that catches the animator side of me.
I know what you mean about trying to direct someone to do something, and all you get is an epic fail. Totally.
Have you ever looked through the Cambridge University BVH files collection? It's been a long time for me. But I think that they had a bunch of different walk style mocaps.
I know you don't want to spend money, but I get a lot of mileage from GoFigure's aniBlock collections.
Even if they don't have exactly what I'm looking for, I can still use something similar as a base to start from. In an example of editing a particular walking style, I might use various combinations from these:
Walk Styles 1
Everyday Motions
Gesture Pack 1
as well as others.
Since the importing of the aniBlock often adds keyframes all along the timeline throughout the range of motion, I'll consolidate it down to specifics so that I can use what's there, but easily edit changes into it. So I'll select all of the keyframes that I can see without scrolling the sequencer window, often after I've dragged it up to fill most of the work space, and then use the shift key to deselect the first and last keys, and then keys that fall on quarter of a second, half a second, three quarters, and on the second. I'll go through the whole sequence this way, or just the part that I need and delete the rest.
So after I shift-click to deselect those specific spots in time, I just delete the rest of the selection. Then scroll the sequencer down and do it again for other body parts, whatever is needed.
This gives me the basics of the motion and translation, so that I can then go in and make changes to get an entirely different animation.
Often times I just delete all of the head, neck and arms through to the fingers, and animate those myself.
To help in all of this, I set the time to 1 second before importing the aniBlock. This way I still have my default pose to fall back on for body parts where I delete everything from the imported aniBlock.
================================================================================================
Anyways, I suppose you're wondering why I'd mention aniBlocks for your custom models. Well, if the naming conventions in the rig are the same, you could still use the motions. Often when we cross figure like this, some of the joint behavior won't match, so the feet are pointing in a funny direction, or worse. This is where the graph editor (Click the word "Sequencer" in the blue button, and change it to Graph Editor) comes in. Open up the hierarchy down to the individual rotations of the joint in question, select all of the keys for that rotation, and drag it up or down in the editor to alter the rotation uniformly across the whole animation. Works very well. Rinse and repeat with other rotation axis' and other joints as necessary.
Oh my... babbling again, huh... :shut:
I've got a great book that shows you stuff like that about cell animation, it's Richard Williams' The Animator's Survival Kit.
But that's not really what I was looking for. I've looked on YouTube and not really found what I wanted, I just thought you guys being animators might know of where to look.
I wanted video of real people walking, strolling, sauntering etc. A whole load of animations both 3d and drawn came up and a few of people, and 1 actual proper reference, even with a grid behind the person, very helpful. But still no strolling or sauntering.
I keep thinking of the song: "I was walking down the street one day, in the merry, merry month of May." And someone sauntering in time to that. I have an image in my head of Gene Kelly walking along to that – but that may be just my imagination.
I hope it doesn't devolve into it having to be video of me – 'cos the end animation will be of someone with back problems and dicky knees. Lol!
(although it will involve a stick!!!)
Oh yes? I didn't know aniblocks could be used in Carrara, (I don’t really know what they are, but I thought they were only for DazStudio? You keep hearing them mentioned in the DS forum.)
This gives me the basics of the motion and translation, so that I can then go in and make changes to get an entirely different animation.
Often times I just delete all of the head, neck and arms through to the fingers, and animate those myself.
To help in all of this, I set the time to 1 second before importing the aniBlock. This way I still have my default pose to fall back on for body parts where I delete everything from the imported aniBlock.
================================================================================================
Anyways, I suppose you're wondering why I'd mention aniBlocks for your custom models. Well, if the naming conventions in the rig are the same, you could still use the motions. Often when we cross figure like this, some of the joint behavior won't match, so the feet are pointing in a funny direction, or worse. This is where the graph editor (Click the word "Sequencer" in the blue button, and change it to Graph Editor) comes in. Open up the hierarchy down to the individual rotations of the joint in question, select all of the keys for that rotation, and drag it up or down in the editor to alter the rotation uniformly across the whole animation. Works very well. Rinse and repeat with other rotation axis' and other joints as necessary.
Oh my... babbling again, huh... :shut:
No no, once I find out what all these things are (aniblocks, BVH's etc) your 'babble' could come in really handy.
Oh I see. Thanks for clarifying that.
Um... how do you convert a spline object into a vertex object? Is it complex?
Oh I see. Thanks for clarifying that.
Um... how do you convert a spline object into a vertex object? Is it complex?
The easy way is to go to Edit--> Convert To Other Modeler. A dialogue will open showing the native format and your other options, There's a slider on the bottom to adjust the fidelity of the conversion. Using a high value to convert a spline object will result in a very dense mesh. I would try a lower setting and see what smoothing does for you first. The conversion process will produce a vertex object consisting of triangles. The standard wisdom is that quads are better.
I've got a great book that shows you stuff like that about cell animation, it's Richard Williams' The Animator's Survival Kit.
But that's not really what I was looking for. I've looked on YouTube and not really found what I wanted, I just thought you guys being animators might know of where to look.
I wanted video of real people walking, strolling, sauntering etc. A whole load of animations both 3d and drawn came up and a few of people, and 1 actual proper reference, even with a grid behind the person, very helpful. But still no strolling or sauntering.
I keep thinking of the song: "I was walking down the street one day, in the merry, merry month of May." And someone sauntering in time to that. I have an image in my head of Gene Kelly walking along to that – but that may be just my imagination.
I hope it doesn't devolve into it having to be video of me – 'cos the end animation will be of someone with back problems and dicky knees. Lol!
(although it will involve a stick!!!)
It really is about observation when it comes to any animation. The tutorials will show you the tools to use. Once you've followed the tutorial and understood the mechanics of it, that's when you try and apply character to the movements. You mentioned sauntering. Perhaps look up old cartoons. That's where walk cycles originated after all. You'll more than likely find a side scrolling shot in some of those. Maybe even a front shot.
Sorry, I was thinking of Carnegie. Yeah, bvh works in Carrara, and also Carnegie Mellon has some downloads already converted for Carrara NLA, I believe. Been a long time since I've played with that stuff.
HI :)
Yes, Carrara can load BVH files, ...Given that the Bone names in your figure, match the bone names in the BVH file.
You can create your own BVH files using Poser,. or using Fenrics PZ2 / BVH exporter from Carrara (very handy) ...(very cheap).
Of course, that's only handy if you're exporting BVH from Carrara to use in other software.
Carrara has it's own animation tools, and for a Custom figure like you have there, that's probably the best way to go.
Your best bet, is to try to create a simple Walk animation (put the strolling / sauntering walk on Hold)
Practice making a simple walk,. then the strolling walk could be made by simply adjusting the timing of some key-frames.
Any Key-framed animation you create can also be converted to Carrara's NLA (Non Linear Animation) format, and saved in your browser.
NLA allows you to combine different animations clips, or "blocks" (like the Aniblocks from DS) and you can also Import Aniblocks from DS if you have the Aniblock Importer for Carrara.
Non Linear animation,. is a "block" of animation which you can place and move anywhere you need it in the timeline. but is still essentially Key-frames.
Carrara can also load the Keyframes from an NLA clip into the Timeline so that you could edit that animation,.. then save it as a New NLA clip.
:)
Just to add a bit about BVH...when it first became popular a few years ago, we were trying to get the CMU bvh's working in Carrara - even with serious editing in BVHacker, it was very much a hit-and-miss exercise. Carrara simply didn't play well with bvh.
After reading this discussion I went in search and found a very generous person who goes by mojoDalls converted the CMU bvh's to be Poser friendly - he has a library of them at ShareCG.
Tried some random ones in Carrara and they were absolute #@%. Tried in DS and they work just fine. Exported as bvh and used in Carrara - huge improvement, but still some nasty unwanted translations in the Z axis and in the neck - works OK with some editing.
Lastly, I exported from DS as pz2 - works just as well as it did in DS - joy-oh-joy!
So, if you want plenty of free, usable motions, get them from ShareCG, put through DS and export as pz2. Will probably work just as well via Poser, but my Poser Debut has gone mia, so I can't test.