So this one will be again, the call for a new tool . Now it might very well be that
someone of you know some work around or some way to achieve this, if so please share with us your knowledge. So its all about texturing and laying out uv maps. Lets say I have a character which I am planing to texture using unwrap uvw modifier and say a pelt mapper. Now my model has a separate model for head, torso, cloths, hands and so on. After I peel each of them, the sizes of my characters head texture, hand texture and everything come in same size. they all are fitted to this one square.. Now if I want to hand paint textures things like the width of the line in a texture matters. so in the case where I have, lets say a giant head of the character and tiny tiny eyes, the unwraper will make them same in a size. And that gives me a problem. If I apply same texture say with a dots, the dots on a head will be huge compared to the ones on the eyes…
naturally in a unwraper I can resize everything, but as far as I know its all done by hand. So the question remains, how to make many pelt maps from many separate objects so the sizes would be very simar to these sizes of actual geometry?
————————–Ok! seems I have an answer myself :) I found this grait tool – Unwrella! from now on its sort of advertising, but i found it solves exactly this problem – it makes the sizes of pelt uv’s all correct sizes relative to each other! text from their website:
Unwrella is an exact unwrapping plug-in for Autodesk 3DSMAX and Autodesk Maya. It is a single click solution which allows you to automatically unfold your 3D models with exact pixel to model surface aspect ratio, speeding up texture baking UV map production significantly.
Automatic one-click solution – Just apply the Unwrella modifier
Precise – Preserves user created UV Seams
Smart – Reduces texture mapping seams almost completely and minimizes surface stretching
Efficient – Chunks are kept large and are arranged on the UV surface with maximal use of available space
User-friendly – User defined pixel based padding between UV chunks
Excellent for all kinds of models (organic, human, industrial)
So another minor complain on 3dmax’s UI. Its sort of very minor, and in essence is just a tiny thing, but still I personally think it could, and I think should be improved. So it’s about camera (not exactly camera, just viewport) position when you are modeling. So I guess many of us, max users are using “zoom to selected” viewport tool. I use it a lot, and especially when modeling. So the situation is this. One is modeling something, and selects a couple of vertexes, which happen to be in same position. Now we can’t really see anything cos now we look at whole model. so we hit zoom to selected, and bam most likely we see.. nothing. Or just some part of our object, which has suddenly lost its shape. Its viewport clipping I guess. The zoom tool zoomed so close to selected vertexes, that max decided it’s too close and just clipped the model. Or that’s what I think that happened. Look at the pictures, or try it yourself. Not sure if it is exactly what happens, but it sure looks so. Why does max do this?
(usually this happens when we model say a character, or whatever, and we work only on one half of the model. Other half we minor, and later on we want to weld vertexes which duplicate in the original and mirrored object.. at least that is one of situation where this occurs..)
So what do we do? just scroll back a bit, bam, the model is back, and we are close to our selected elements to work with. But, if you are unexperienced, u might go … Ouch, what happened??? or just scroll back, like you did 1 million times. But can’t we make a tweak in viewport camera positioning and zoom to selected tool’s code, so it doesn’t go to such extremes where model gets clipped? Anyone? So it’s just another small complain. Yes, I know there is million other things that are more important to improve. And this is just a tiny annoyance.. but if it would be just a couple of lines of code, might be it would be worth while writing them down? sins I suppose it should be quite straight forward thing, it’s not like one has to write Lagoa for max or anything like that… (Yes Lagoa would be cooler to get… :) Am I right? am I wrong?
1. (in image) The clipped area that is not shown….
One proposal would be very simple.
It would be to enhance existing uvw map modifier by adding ability to apply uv
mapping not for a single mesh, but to each element of the mesh.
I know its possible to achieve same result by selecting each element in mesh edit or poly edit mode
detach these elements from single mesh, and as a result, to have lots of objects to which u can individually apply uvw map modifier.
but in some cases, when u have a complex mesh, which consists of lots of elements, that could be very time
consuming work. These complex meshes are often result of importing geometry from other 3d applications.
So the idea is to apply, say a spherical or box or whatever type of uv projection method to each separate mesh element without a need to divide mesh and detach parts of it.
Now if we have a complex model, say a tree, or dono, anything actually, and we are not planing to have perfect uv’s by pelt mapping or selecting separate polygons and applying separate planar maps to parts of a mesh, we would want to have some simple method to get uv coordinates fast, and semi accurate. in current situation we can simply apply uv map modifier to whole object,
and expect that box or cylindrical map projection will do fine. and in some cases it does.
say model is far away from camera, and we have no time to produce good uv coordinates. But imagine we would have some middle quality solution. that is something between creating precise uv maps for separate polygon selections, or having to use pelt mapping, and between simply applying say cylindrical projection
to a whole mesh.
I propose to have an ability in a uvw map modifier to have a “per element” button which would apply
selected map projection method not to whole object but to each separate mesh element.
and also we would have an ability to manipulate (scale, rotate, move) uv projections within a single interface for each single element in one mesh,
at the same time without a need to apply hundreds, or be it tens of mesh select modifiers, and uv map modifiers, and struggling and getting lost in huge modifier stack.
lets look at the images below. 1. cylindrical map projection applied to a single mesh.
2. lots of uvw map and polygon select modifiers to achieve accurate projections for whole model.
3. proposed projection per Element mode
4. same, but with floating toolbox for selecting each element and having ability to apply different (cylindrical, box, etc.) projection modes.
Some additional thoughts.
Why do we think only about elements of mesh, can we use mat id’s for same porpoise? can we
apply separate uvw projection methods for mat id’s also? That could come in hand too..? Next question would be, can we use both mat ID’s and Mesh elements? Or is there an easy and fast (one button solution) way to convert separate mesh elements to mat ID’s. Or can we easily convert Mat ID’s to mesh elements? And if we where to write such tool, where would we place it?
Should it be a part of edit poly modifier? edit mesh? or uvw modifier?
somehow i cant finish my tiling/ rotating/ mirroring article. so while i am struggling with part 3 of tiling posts, something else….
Alzo, one very small remark. I noticed it long time ago,
its such a small detail, not worth writing a post. But it happen so many times,
Its my love and a curse. Well as usual I am talking about 3d max’s user interface.
Its nothing less then right mouse button. Its a bless and a course.
Imagine u selected something, and now u are dragging that stuff across the screen to mech
something else… U are so concentrated, the sweat is running on your back, u cant blink,
it must be so precise..
and u push your mouse so hard. And then it happens. u accidentally click right mouse button.
last thing u did is gone. Dam it!
On other hand sometimes u see u are selecting stuff wrong. U click right mouse button and last operation is gon.
So nice. Not like in autocad (right?)
Ok it shouldnt have been a post, but whatever :)
Ok sins its such an uninformative post well add some more.
Some complaining ofcourse :)
Ok what do I find annoying about max? Its mirror, and clone tools. Whats a problem u ask?
Say I have to clone (and move) one element many times. I want to see it from very close distance to move it exactly the right amount.
So I enlarge it to fill whole wievport. I hit shift, i move it exactly the right distance,
And then I need to specific a number of copies. But it happens a lot that for that I need to see
all the scene, not one object, to count how many copies i need. Ok I am not sure one can understand what I mean.
So to put it in other words, after we see clone or mirror dialog boxes, we are no more able to
navigate in viewport. and that sucks. Look at all extrude. bevel and other tools under edit poly,
if we use these we can both have dialog box open and navigate in viewport. that saves time,
and quite much id say.
So something to consider for max UI developers.
Ok, this line was just for fun, but.. There used to be such cool procedural map called planet. Ok it had its limitations. I would have liked to be able to add complicated maps inside each color slot of planet procedural map. But still, it was very good map. To me it was as important as noise and smoke.
I used to use three of them to achieve very nice random organic shaders. Ofcourse it was written to simulate planet surface, but that’s, to me at least, is only one way to use it. I used it in any situation.. But now I have 3D max 2010 and I see no Planet!!??
What happened, am I blind? Whats so good about it? I will add some pictures here:
Ok. there must be a reason why its gone. Might be its old code, somehow not compatible with some other newer elements.. So why not to rewrite it?
But if someone decides to do so, lets add some more functions to it.
In original Planet we have function called Iceland factor. It was probably the reason I used planet.
But one can increase or decrease island factor to whole generated procedural . To all its elements.
Why not to separate this effect, so we could apply it to separate elements of planet map, called colors. Now here its a bit hard to explain. Pls look at previous image and next ones.
Or, even cooler, might be we could add a map for each color slot, as how and where island factor should be high or low. What I mean is, if we change value per color slot, it would affect whole contour of that color slot, what if we could add noise or smoke material there… (i should update last pictures adding map selection button to each noise factor, i guess… it simply would work as either value (0-1) or as a map.)
We also have blend water and land switcher. As far as I understand if it’s on, max generates gradient transitions between colors. Again, why not to have this option separately for each element, or color. And here is my proposal for Planet Pro :)
Ok, continent size and water size hardly make sense, but, common, that’s just an idea, not a finished procedural. What you think of it? Could we have water regions and land regions like well… like a groups of colors? so we could apply some stuff to a bigger regions compared to one color?
like this might be:
Here is a text from 3d max 9 help file, describing planet material.
Here they say it’s not good for bump maps, but actually I loved it for rusty, heavily corroded surfaces.
Planet map used to create another world
Planet is a 3D map that uses fractal math to simulate the colors on the surfaces of a planet. You can control the size of continents, percent of ocean coverage, and so on. This map is meant to be used as a diffuse map. It does not work well as a bump map.
Tip: The Material Editor's sample slot doesn't show the planet effect very clearly. To help get the effect you want, double-click the sample slot to get a larger sample slot, or assign the map to geometry and render the scene. Another way to preview the planet map is to use the Material Editor Options dialog to set the 3D Map Sample Scale to equal a main dimension of the object you are applying the map to. For example, if you want to use the planet map on a sphere with a radius of 20, change the map scale from 100 (the default) to 20.
Continent Size—Sets the size of the fractal noise pattern used to generate the continents. Can range from 0 to 100. The higher the value, the larger the continents. Default=40.
Island Factor—Sets the size of the fractal noise pattern used to generate islands and mountains. Can range from 0 to 100. At 0, the geography is very low. Higher settings create a more rugged landscape. Default=0.5.
Ocean %—Sets the percentage of the planet's surface that is covered by water. Default=60.0.
Random Seed—Sets the seed for pseudo-random generation of the pattern. Changing this number can completely change the pattern, even if other settings remain the same. On the other hand, a different Planet map with the same settings including the same Random Seed will appear the same.
Water Colors group
The colors in these three swatches are applied to the water areas of the planet surface.
Water colors—Click a swatch to display the Color Selector and change the color. Color #1 is the "center" area of the water mass, Color #2 surrounds Color #1, and Color #3 surrounds Color #2, meeting the land.
Land Colors group
The colors in these five swatches are applied to the land areas of the planet surface. Their arrangement continues that of the water colors.
Land colors—Click a swatch to display the Color Selector and change the color. Color #4 is the shoreline of the land, meeting the water; Color #5 comes next, working toward the center of the land mass. Color # 8 is at the center of the land mass.
Blend Water/Land—When on, the boundary between water and land is blended, giving a hazy appearance. When off, the boundary between water and land is sharp.
One thing that takes long time while working is related to procedural maps and viewports.
If you have an empty scene, and one object, its not so hard to create a shader using procedural maps for it. Well it could still take time, What I what to say here is, It doesn’t take time or effort to apply a noise map in diffuse Chanel, render it all, look and continue adjusting noise size and other parameters. Each time rendering result to see the effect of your adjustments.
Now its a very different story when, A. you have very complicated map. Noise, or other procedural map represent only some very small and hardly noticeable part of general diffuse map.
Or B. your scene is full of objects, it has caustics and indirect illumination on, some volume fog and whatever else. In both cases it becomes very time consuming to isolate your procedural map to generate test render. That’s why one would click ”show standard map in viewport”
in material editor. And it works ! :) Quide good for texture maps (bitmaps). And we do not need perfect quality here, its just a viewport, just a guideline. But look what happens with procedural maps:
Well, it shows idea of noise. But size is all wrong. Which is the worst, I think. It has other problems, some cutting lines are visible.. but noise is noise, not such a big deal… but if we are using planet (which no longer exist in max 2010) or say Perlin Marble… –
Here is one example, say I have Utah teapot. And I what to add marble pastern, so its veins would go horizontally.. or, even more complicated, one vein should go around the opening of pot, where one adds lid..
here, u see picture (1) from viewport followed by rendered final.
And 2, what I had to change to have correct alignment of marble veins.. how could I do that without rendering? Here its simple, one object, only one texture… but.. If I have a situation described in the beginning of post?
3D maps are patterns generated procedurally in three dimensions. For example, Marble has a grain that goes through the assigned geometry. If you cut away part of an object with marble assigned as its texture, the grain in the cutaway portion matches the grain on the object’s exterior.
The map appears on objects assigned the material in all shaded viewports. Now when you adjust the map, the viewports update to display the adjustments.
Turning on Show Map In Viewport for one map automatically turns this button off for all other maps the material has.
Viewports can display 2D maps such as Checker and Bitmap.
Viewports can also display most kinds of 3D maps. The exceptions are Particle Age and Particle MBlur. Also, the appearance of the Falloff map in viewports give only a vague indication of how it will appear when rendered.
Show Map In Viewport is unavailable if the active map type cannot display in viewports.
Displaying mapped materials in a viewport can slow performance. If you don’t need to view the texture, turn off its viewport display.
So lets imagine we want to create a new material. We go to material editor, we select a type of material, and hurray, we have it. No we want to add noise, or non existing planet :) map to our diffuse canal. And we click button “none”. Nice we are in material/map browser.
So far so good. Now lets count how many clicks we need to apply say, speckle map. And here we go, its not in browser, or, unless your screen size is extremely huge, you need to scroll down. That’s normal. But faster you scroll, harder to see if you reached wanted map. (some people have no problem with this) Slower you scroll, well it takes more time. (some other get annoyed). So what do we do? Simple. Same as we do in modifier selection list.
So we go over menu with mouse pointer, we don’t even click as in many other programs (winch is cool innovation by max) And we hit a letter on a keyboard, and here you go, no more scrolling, you are almost at speckle map. (providing there are no other maps starting S) and there are, but still, it takes fraction of second to spot required map. Why cant we have this functionality in material/map browser?
There is such a cool thing that I would love to see say in adobe photoshop. So say we work with two image sets. One is my dogs photos to add to my first ever html website, and other is a diagrams I am preparing for tomorrows presentation. I work on both simultaneously. More or less. (I edit dogs photos when my boss is in next room).
So I am done with some and I am saving files.
Now my dogs pictures go under:
C:\Documents and Settings\Peter_the_best_employe\My Documents\My Pictures\my new website\test files\images\gallerys\my cute doggy\
Now say I save one dog picture, then one diagram, and again dog, diagram.. dog diagram..
After I saved first image my picture editing program remembers where I saved it. And next time I save, it suggest same location.
Now each time I switch from dogs to diagrams, I have to specify file location. Every time…. Now that can take years. And lots of patience. Especially looking at file destination addresses I made up above.
Now lets see how max saves the world:
Now you see, dialog box saves all locations where I saved things last time. Neat, eh?
yes that happens. And i understand that. more complicated program, more code, more chance for this to happen. But thats not what this post is about.
Its about what autodesk does about crashing max. Now I understand that sending a report is good for everyone, theoretically autodesk
finds a reason for a crash and in next version of max problem is fixed. But what if I dont have internet? why do I have to see this dialog box everyday, with nowhere
“click here not to see this message ever again” to be on. WHy? Each time max crashes I have to click couple of clicks, which are totally pointless…. why can’t we have “dont show me this again”
thingy???? ok, lets say Autodesk thinking is something among these lines ” user hates sending crash results. they cancel it. so we make it hard to cancel. we need all possible reports, even
if that means we are risking to annoy our costumer”.
But if I dont have internet???
Whats the point?
just a quick idea. It happens a lot that i have to wait for some operation in max. its shorter than rendering for 2 hours, but its way longer
for normal interaction with program. like lets say creating blob mesh, applying multires, substracting objects via booleans or subdivision.. or something like that. I have always task manager window opend and minimised in try bar,
so i see how much my processor is owerloaded. but I have to sit there and waite, staring at that small indicator, till I can finally move my mouse again. thats so annoying.
couldn’t we have a small beep? lets say max starts to use 100 cpu power, for longer then, dono, 20 s. our beeper program monitors this cpu usage. and after cpu usage for max goes down dono, 10 % lets say beeper still waits a bit to make sure its not a temporary
pose in cpu usage by max, and lets say after 5s. it beeps. meanwhile, user, has applied some heavy operation, and went to make cafe. and he doesn’t need to look at cpu usage indicator, all i care about
is my coffee, till i hear the beep.
i think it could be helpful, or? annoying? of course it should not be on by default, and user could turn it on only if he or she needs it…