Chosen Few

Glad it worked out,, Madeliefste. Thanks for following up. I use Illustrator curves in Maya, sometimes, too. It's handy. I didn't know C4D also had that capability. Good to know. So you know, with a simple image like this, you may want to consider using Illustrator's Trace Image feature. It could save a you a few steps.

The selection will never be 100% identical to the path, since the selection is raster, and the path is vector. However, it can certainly get a lot closer than what you've got in your example, without too much fuss. Simply set the tolerance lower, when you create the path. Here's how: 1. Form your selection. 2. In Paths panel, click on the panel options button (upper right corner). The panel options menu will open. 3. In the panel options menu, click on Make Work Path... The Make Work Path dialog will open. 3. In the Make Work Path dialog, set the tolerance as low as it will go, which is 0.5 pixels, and click OK. The path will now be as close to the selection as Photoshop can do automatically. If you want to further refine it, you'll need to do it by hand. You can add, subtract, and convert points with the pen tools, and you can move points and handles around with the Direct Selection tool. If you already had the tolerance at 0.5, and the path still didn't come out very close, then the image is simply too small for what you're trying to do. In other words, there aren't enough pixels in the whole image for half a pixel to be relatively small enough. If that's the case, then you'll need to work with a larger image, or else do a lot more work by hand to adjust the path the way you want it. Now that the above has been said, might I ask why you need to form that path? If your goal is to extract the subject from its background, I'd recommend using a channel mask instead of a path. Channel masks are extremely handy for stuff like this, and can be made in seconds. Here's what to do: 1. Go to the Channels panel, take a look at the individual channels, and pick out the one that has the most contrast in it. In this particular case, you'll find it's the blue channel. 2. Copy the channel you picked out. This new channel you just created is what will become your selection mask. For the following steps, you'll be working with the copy, NOT any of the originals. So, make sure the copy, and only the copy is visible. 3. Right now, you'll see that the leaf is mostly black, and the background is white. Invert the colors (ctrl-I). The leaf will now be mostlyh white, and the background will be black. Notice what this is starting to look like? That's right, it's an alpha channel. 4. The leaf area is not yet white enough to generate a useful selection. To fix that, adjust the levels (ctr-L). Drag the sliders untul the leaf is as white as you can get it. If you go too far, you'll end up aliasing the edges, so keep an eye on that. 5. If any stubborn parts within the leaf area remain gray or black, simply paint them white with the paint brush. Likewise, if any parts of the background turned gray or white, paint them black. 6. If you did the above steps properly, you now should be looking at a solid white leaf, against a solid black background. If you simply want the background to be transparent, save the image as 32-bit TGA, and you're done. If you want to extract the leaf, so you can put it against a different background, you'll need to copy the channel to a layer mask. To do that, simply ctrl-click on the channel's thumbnail, to form a selection from it. Then go to the Layers palette, click on the layer's name, to return to normal (full color) view, and then click the button at the bottom of the panel, to create a mask from the selection. The white background will disappear, and you'll now be looking at the leaf against the checkerboard. From here, you can put any background you want behind it. The channel mask process may seem slightly complicated, all written out like this, but really, it's just a few seconds worth of work. If the whole thing takes you ten seconds, you're doing it slowly. You'll find that this technique works really well, not just for leaves, but also for hair, fur, or anything else with fuzzy edges that can be difficult to select by hand. Generally speaking, it's the fasted method there is for forming selections, and it also produces the best results. If extraction wasn't your goal, and I just wrote out that channel mask tutorial for no reason, then consider it a free bonus, and thanks for playing. I'd be curious what else you might need leaf-shaped path for, if you wouldn't mind sharing.

I've got one I had been hosting for years, along with a bunch of other SL<->Maya stuff. However, I forgot to pay my hosting bill a few months ago, and my site got nuked. I haven't had time to rebuild. Not to worry, I threw it in my Dropbox, just now for you. Here's the direct link: https://www.dropbox.com/s/3nkuzc866ojene6/bvhExport.mel It was written by Alexia Mechanique, for Maya 8.5. I've confirmed it works with subsequent versions up through 2009. I'm still using 2009, so I haven't been able to test if it works with newer versions or not. Let me know how it goes for you with 2011. There is one minor flaw with it, so you know. It can't always handle backward pelvis rotation very well. So, if you want the avatar to do backflips, you may have to find a different solution.

Codewarrior Congrejo wrote:. Another thing to make sure when uploading images to SL is to choose a correct format. I.e. PNG for lossless compression. Or create a JPG with the right settings of compression and quality. Code, usually I agree with most all of your comments, but I have to disagree on this. JPG is almost never a good option for texturing. It's lossy, low quality. It's a relic from 25+ years ago, when storage space was so limited, a single uncompressed image could fill an entire hard drive. Back then, quality had to take a back seat to practical file size considerations. Nowadays, however, with storage space being dirt cheap and practically unlimited, it's rarely justifiable to use lossy compression, for much of anything. For texturing, there's no justification at all. With the exception of people who sell stuff on the DAZ market (just because DAZ is inexplicably weird like that), I don't know a single texture artist who touches JPG with a 10-foot pole, myself included. With SL in particular, the quality loss problems from JPG get compounded. Internally, SL uses a lossy* implementation of JPEG2000 for textures. So, when you start with a lossy JPG image to begin with, and then it gets converted to a lossy JPEG2000, you end up losing quality not just once, but twice. The compounded artifacts can be noticeable. When you start with a lossless image, however, the one-time loss that occurs upon upload is relatively negligible, much harder to notice. All the other formats SL can import (TGA, PNG, and BMP) are lossless. JPG is the only one that isn't. With 75% of your options being of good quality, you kind of have to go out of your way to use the one option that is not. Further, because SL converts everything to JPEG2000 at the moment of upload, there's no need to be concerned with file size at all. All other things being equal, no matter what kind of file you start with, you end up with the same sized JPEG2000 file when it's uploaded. Add all that up, and the scales are tipped against JPG, 100%. The basket of reasons not to use it is filled to the brim, while the basket on the other side is completely empty. TGA is the most common industry standard format for texturing, and it's what I recommend all SL texture artists use. PNG is also popular, but it can be somewhat problematic for SL. Since PNG supports multiple types of transparency, SL sometimes has trouble determining proper bit depth when converting it to JPEG2000. It's not uncommon with PNG to accidentally end up with a 32-bit texture where a 24-bit one was the intent. With TGA, that simply can't happen. Either you put an alpha channel in the thing on purpose, or you didn't. There's virtually no possibility of error. Where JPG does dominate is on the Web, obviously. Web pages are one of the last environments where file size is still more important than quality. JPG is also common with cell phone cameras, and other cheap digital cameras, which don't have much storage space in them, and are not typically built with the ability to write other formats. None of that has anything to do with texturing, though, of course. * SL's "lossless upload" option is meant for sculpt maps, and is only applicable to very small images. For medium and large images, the setting is ignored, and the upload is always lossy.

As long as it's got the same weighting as the area around it, it should move the same way, whther it's the same mesh or separate. Assuming the button is in the middle, near the navel, I'm not sure why you would have weighted it to the hips. That area doesn't move much, which means it should be weighted primarily to the pelvis. In any case, buttons can be tricky. They're rigid, while the fabric around them is flexible. To keep the button from deforming, it has to be weighted 100% to just one joint. That means the fabric area around the button it also has to be weighted 100% to that same joint, so the two don't separate as they move. That, in turn, means you have to be careful where you place your buttons. Certain locations simply aren't workable.

I didn't realize you were already familiar with the basics, Tenley. You're right that there is more than one way to skin a cat, and it certainly can be useful to see different people's various approaches to any given problem. Sorry I don't have a specific link for you. I would suggest you search the usual sources (YouTube, Digital Tutors, Gnomon, etc.) for general tutoraisl about modeling game ready clothing. Don't limit your search criterea to Blender, and absolutely don't limit it to SL. While the particular buttons to push will vary from program to program, the modeling techniques are universal, and that's what it sounds to me like you're after. Assuming you do have a solid command of the basics, you'll find that this isn't necessarily the kind of thing where a detailed step-by-step accounting is needed. It's more about the beginning, middle, and end. Once you see what to start with, and how it should end up, the intermediate steps become fairly self evident. If I can find a little spare time over the next few days, I'll whip up a T-shirt from scratch, and either record a video or just take some screen shots at key stages. I can't promise I'll have the time, but if I do, I will, and I'll post it here for you.

Your land won't know the difference. Bumpiness is a client side rendering effect. It cannot affect the server in any way at all. If you're asking if people viewing your builds might experience a drop in FPS, the answer is potentially yes. Every visual effect requires some amount of processing time. However, if you're using bump in place of actual geometry, it can potentially be beneficial to FPS. For example, if you use bumpiness to create a brick pattern, that would lkely be less costly than if you had modeled every single brick. That's exactly the reason bump mapping was invented in the first place. To summarize, all other things being equal, a smooth surface will perform better than a bumped surface, and a bumped surface low-poly surface will peform better than a fully modeled high-poly surface.

Looks like a web glitch. It seems to be be about 50/50 between getting a download dialog or getting a page full of illegible code. Try clicking the link a few more times, and it should work sooner or later.

As always, the answer is use only as many polygons as are necessary to convincingly portray the shapes you're attempting to represent, no more. There can be no "one size fits all" answer. 5000 is a nice round number, but without knowing what exactly you're trying to make, it's impossible to say whether it's high or low. If you can make the same jacket with less, than it's high. If you can't make it from that many, then it's low. If it's the least amount you need to get the job done, then it's just right. Keep in mind, whatever number you arrive at for your first draft, chances are excellent that you can lower it significantly upon doing it a second time. There are almost always more efficient ways to solve geometry problems that don't reveal themselves until after the model has been made the first time.

I know this isn't what you think you want to hear, but really, basic tutorials on how to use Blender itself are what you need, first and foremost. As I find myself saying over and over and over again in response to these kinds of questions, don't put the cart before the horse. Any effort to begin learning 3D modeling that starts with "I just want to know how to do _____" is destined for disaster, no matter what the blank happens to be. The ONLY way to learn this stuff that works is to start at the very beginning, just like every other person on this planet who has already learned it did, and to take the process one step at a time. Do that, and you'll never again find yourself stuck in a situation where you have to say, "The video said how to do this but not that." Because you'll have already learn all the basics first, you'll already know the "that", and it will never be an issue. Seriously, I can't stress this enough. A couple of weeks learning the fundamentals upfront will save literally years worth of frustration, when you add up all the hours you'd otherwise need to spend hunting for answers to questions you wouldn't even have needed to ask had you learned everything in the proper order. Some people get mad at me when I try to point this out. I never understand their reasoning. Whether those unfortunate people happen to like it or not, this is the only truth there is. I didn't invent it, and geting mad about it won't change it. It simply is what it is. I've been doing this for a long time, and I've never seen anyone succeed by trying to learn out of order. That just doesn't work, period. What ALWAYS does work is to go step by step, from the beginning, through the full process, so that all bases are covered. I've never seen anyone fail at that. Imagine if you saw someone trying to learn to read, just by trying to focusing on one particular sentence, without first having learned the alphabet, or how letters form words, or how words form sentences. I'm sure you'd want to help that person by explaining that that's just not how learning to read works. There's just no way to do it without starting with the ABC's, just like every other reader did. Well, just like every reader in the world began by learning the alphabet, every 3D modeler in the world also started with one set of basic fundamentals that everybody needs to know in order to make it work. Trying to skip over that doesn't do anyone any good, ever. Whatever it is you think you want to know how to do, forget all about it for the time being. Those things will reveal themselves along the way, naturally, as will a many other things you absolutely need to know, but don't yet realize it. By the inch, it's a cinch; by the mile it's a trial. In other words, learn modeling itself, and then you'll automatically know how to make that T-shirt, and that dress, and those pants, and whatver else you might ever want to do. But try to learn just those specific things without first covering the fundamentals, and you'll only experience frustration. The choice is yours.

Glad to hear you got it worked out.

It could be a UV problem, as Rahkis suggseted, but it could also be a number of other things. If the problem only started after weighting, than weighting is definitely the issue. One or more vertices is being pulled on by the wrong joint. If it started before the model was rigged, then it could be in the geometry, the UV'ing, or the texturing. Without seeing the wireframe, the UV map, and the texture, it's not possible to say for certain what's going on.

If you're looking for something pre-made, I'd suggest you post over in the Wanted forum. That's what it's there for. If you'd like help making your own, ask away, right here. As for what's faster, it could go either way. Consider the following factors. If you do it yourself, obviously you need to make the model, rig it, and upload it, which for a T-shirt might take you a total of about an hour, if you work at slow to moderate speed, or about 20 minutes if you're a faster worker. If you use something pre-existing, you can skip all that, but then you're dealing with somebody else's UV map, which might slow things down just as much, or more, if the map isn't particularly well made, or if you just plain don't like the logic of how that particular person laid things out. Also, if you make it yourself, you'll have the full power of your 3D program(s) at your disposal, to aid you in the texturing process. You can project imagery, bake shader effects, 3D-paint, etc. If you're using something that's already in-world, chances are you'll only be able to paint in 2D, on the provided UV map (unless the creator also provides you the model itself, which many will not do, for obvious reasons). If it were me, I'd just make it myself. A T-shirt is so quick and easy, I'd have it done in less time than it would take me to search for one to buy. That's just me, though. I prefer to make everyhting myself, so unless there's a really compelling reason to use something off the shelf (such as a tight deadline and an already huge workload), that's what I'll nearly always do. If you don't have that same bias toward DIY that I do, then do whatever you think is less hassle.

Sanya Bilavio wrote: In the instance that the eyes and wrist and elbows are having influence on the dress do you suggest i remove the weights for these? They have tiny bits of influence on the dress and i can see them moving the dress slightly. Do you suggest i just add more influence to those areas on joints that are suppose to have it in those spots instead of erasing them? I would suggest you open up the component editor, and take a look at the vertices in question. If there are influences from extraneous joints, go ahead and zero those out. Alternatively, you could paint over those areas with weight from the correct joints, with the painting tool in Replace mode. The component editor will likely be faster and more accurate, though. Sanya Bilavio wrote: Also for the pelivs, i was told once that i should paint the entire item due to when avatars sit down and do other extreme animations, is this true? I'm not sure exactly whatyou mean, but it is true that sitting will always be a problem for loose fitting skirts. If you makw the fabric follow the legs tightly enough that it bends underneath for sitting, it won't look right during walking. If it folows loosely enough for good walking, it will be too rigid for sitting. There's no way to do both with the kind of rigging that SL supports. Tight skirts are much easier than loose ones, which is why you see more of them around. Sanya Bilavio wrote: I think the worst areas for me are the pelvis, chest, torso and hips due to i have no general idea how to paint the weights lol Also because im using the color ramp its a bit confusing. The same logic I outlined before applies. Work from lower heirarchy to higher. Start with the chest, and weight the area around it 100% to it. Then do the same for the torso. Rotate the chest joint, and watch how the its part of the dress moves, relative to the torso's part. If the border area moves too much, give it some more weight from the torso. If it now moves too little, give it some more weight from the chest. Just as before, you'll arrive at a happy medium. Remember, the pelvis is the one joint in the skeleton that cannot rotate independently, since it's the root. Any part of the skin that you want to be rigid should be weighted 100% to the pelvis. As for the color ramp, if you find the it to be confusing, you don't have to use it. Turn it off, and the paint will display in the default black and white.

Sanya Bilavio wrote: For some reason when i go to paint weights i cant erase the blue bits 100% it will erase bits and pieces and it takes forever, so i have to end up going with a huge brush and erase it all at once. erase issue - http://gyazo.com/d67e1a25299c9f334a8a5692119e5853 Don't ever try to erase when weight painting. Always work additively, rather than subtractively. If an area has too much influence from a wrong joints, add weight from a right joint to correct the problem. For best results, alwys start at the extremities, and work your way inward. For the upper part of your sleeveless dress, that means start with a shoulder, and then the spine. Repeat for the other shouder. Then do the neck, and then each spine joint, all the way to the pelvis. In terms of heiarachy, this means work your way up from child to parent to grandparent, etc. If you find yourself in the rare circumstance under which you truly do need to remove the influence of a joint, and painting from another joint won't do the trick, then go into the compinent editor, and zero out the incorrect influences. You shouldn't need to do that very often, though. When you do, it's usually due to bad UV mapping borking the weight maps, or screwy topology making painting difficult. Sanya Bilavio wrote: Along with the erasing issues when i believe to have painted an area correctly i get weird results like below http://gyazo.com/112096bafa01c5f7b050505287fbba8e http://gyazo.com/b2dc70ac8626f067c4dbc3178311d67e You can correct those issues pretty easily by following the above startegy. Start with the extrmities, and work inward. When an area misbehaves, it's generally because it doesn't have enough weight from the next most inward joint. In the first pic, it looks like the pinched area at the waist needs a bit more influence from the pelvis, to keep it from moving. The deformed areas under the breasts probably need more influence from the green mChest joint. That chest joint is one one of the most severe weaknesses in the SL skeleton, by the way. Better skeletons will often have a pair of rib bones, one on either side of the lower chest, to help maintain volume in the skin, and the arms will be attached to an upper spine bone, which the SL skeleton also does not have. The SL avatar's skeleton, however, just like its skin, is not so well put together. (This is the part where if I had the energy, I'd go on for a while about how inexcusable and pathetic it is that with SL about to celebrate its 10th birthday, it's still using the same crudely hacked together avatar model it had on day one. But I'm afraid I don't have that much energy today.) In the second pic, the weighting of the skirt is totally messed up. You've got wong joints pulling on wrong areas, all over the place. Complicated as the problem looks, the solution is easy Again, start with the extemites, and work your way inward. Here's the general procedure: Start with the right hip, and weight the right half skirt 100% to it. Do the same for the left hip, with the left half of the skirt. Now when you move the legs around, each leg will pull on its half of the skirt, without the other leg having any undue infuence. Don't worry that the the influences are too strong at this point. We'll correct that in the next step. The only thing to worry about right now is making sure that only the right leg affects the right half, and only the left leg affects the left half. With the legs done, now go to the pelvis, and start gently adding its influence. The more pelvis influence you add, the less anthing will move. The heaviest influence should down the middle, to help even out the stretching, as the two legs move. You'll want at least some degree of pelvis influence on every part of the skirt, since the skirt is so loose fitting. The legs shouldn't be able to move it a tremendous amount. If you end up painting too much pelvis inflence somewhere, simply add wright from the appropriate hip joint. If the hip now has too much influence, add a little more pelvis. Just keep adding and adding from each direction until you get it right. Never subtract. Eventually, you'll arrive at the correct happy medium. Sanya Bilavio wrote: Before the dress was smoothed out and at it's lowest verts possible i used the soft preview button 3 and rigged it as it was and it was moving perfectly. But as soon as i smoothed it out it started having these problems. Two things here. First, don't expect Maya's smooth display modes to provide an accurate preview of anything when doing polygonal modeling for realtime environments. That's just not what they're for. Always use display level 1 for poly modeling. Second, be acutely aware that the order of operations directly affects the success or failure of your project, always. The model's topology should be finalized BEFORE you start the rigging. Changing it afterward is not a good idea. It almost always causes problems. When you smoothed the model, you introduced new geometry, which Maya doesn't know how to weight, since it wasn't there before. The best it can do is guess, and in most cases, it's going to guess wrong. To fix the problem, you'll first need to delete non-deformer history from the model, and then you'll need to repaint all the weights, again starting at the extremities, and working your way inward, painting additively the whole way through.

The OBJ's from the SL downloads page work just fine in PS CS6, but the 3D interface is quite different from what you're used to from CS4. I'd be willing to bet the files you have are fine, and it's just a few aspects of the new interface that have been blocking your path so far. Getting used to the new version may be a little uncomfortable at first. Once you do get into it, you'll find it's leaps and bounds better than it was. They've taken a huge step forward toward making it behave like a typical 3D program, rather than as the somewhat clunky 2.5D hybrid that it was before. It's still not yet far enough along to rival the likes of Mudbox and Zbrush, but it's getting there. Here are the main things to know about the new interface: You don't need to touch the 3D menu anymore, if all you're doing is texture painting. The functions you need have been removed from the menu, and placed on panels, where they're much easier to access. Each 3D layer in your working document is now treated as a full blown 3D scene, complete with a ground plane (grid), multiple cameras, lights, selectable objects, the whole nine yards. When you're working on a 3D layer, The 3D panel now behaves as an outliner, just like you would have in any traditional 3D modeling program. It shows you all the objects in the scene, by hierarchy. The new Properties panel is great. It shows all the editable variables for every selected object. When you select the Move tool, the entire 3D scene becomes exposed. You'll be able to see the grid, the multiple view ports, object selection boxes, manipulators, camera controls, etc. Because of this, the camera controls and model manipulators that had previously been in the toolbox have been removed. Hopefully, after reading the above, the interface is beginning to make more sense to you now. Here's how to start texture painting: Open the OBJ file. You can do this in a number of ways. You can drag it into PS from Explorer or Bridge, you can use the File -> Open menu command, or you can go to it inside the new Mini Bridge panel in PS. In the Layers panel, click on the 3D layer to activate it, if it's not already active, and then take a look at the 3D panel. You'll see everything in the scene listed here, including the environment, the scene, the model and its materials, lights, and cameras. Click on any of these, and then you can adjust the item's various attributes in the Properties panel. For example, if you want to work without shadows, select the scene in the 3D panel, and then in the Properties panel, change the surface style to Unlit Texture. In the 3D panel, select one of the materials, and take a look at its attributes in the Properties panel. Next to where it says Diffuse, click on the little file icon, and then in the menu that pops up, click Edit Texture. The texture image will open as a PSB. (Alternatively, you can also double-click sublayers in the layers panel to open texture PSB's, just like in CS4, but the Properties panel is far more convenient.) From here on it, it's just like it was in CS4. In the PSB, activate a layer to work on, and start painting. You can paint in 2D in the PSB, or paint in 3D directly on the model in the PSD. Changes will reflect in both. That should get you on your way. Any additional questions, come on back and shout 'em out. Happy panting. ETA: I'm not sure what you mean by "blend from back to front". Could you explain? I want to make sure I'm addressing the right issue.

Ah, I missed the part where he siad it looks good on the ground. Been one of those days.

Deezy Abruzzo wrote: Cool, what sites have free mesh? While I'm usually the last person to say, "Google is your friend," in response to a forum question, in this case, I have to say, seriously, Google is your friend. Type in "free 3d models" and you'll get over 76 million results. Take a look at some of them. Do be aware that not everything you find will be suitabe for use in a realtime environment like SL. The majority probably won't be, in fact. Even the ones you do find that are game ready will probably still require some tweaking. I would suggest you consider them to be concept art, not end product. Use them for inspiration, to see how different 3D artists do different things, and then do similar things of your own.

Too many polgons per material, perhaps? What's the poly count on that thing?

I agree with Qie. Because it happens that naga do not have legs, you can use the avatar's leg bones to animate the tail. With creative use of joint offsets, there's actually quite a lot of capacity for animating a really convincing serpentine tail. There are ten joints you can use, when you include both legs. Take a look at the two figures below. The one on the left is the avatar skeleton, in its default pose. By moving the leg bones, you can create the figure on the right. That will give ten sections for your tail, which is plenty. Create the right animations for it, and it should look great. For a creature that has both legs and a tail, you'd have to sacrifice control of a limb, in order to control the tail. You luck out with the naga, since it has no legs. A friend of mine did something similar, a while back. He made an angel, with really good flapping wings, by using the arm bones to control the wings. The model's apparent arms were fixed in a praying position, leaving the actual arms free to be the wings.

It would seem we have different philosophies on these things, Crysantha. Where you see a "gotcha", I see a perfectly reasonalbe, very standard, and very much expected, business practice. If you aren't comfortable with the concept of subscribing, that's fine. Go ahead and buy the products outright, just like before. Nobody's stopping you from doing that, and Adobe will be more than happy to keep you as a purchase customer. Nothing has changed for anyone, in that regard. That said, there's an interesting discussion to be had here, I think. Let me try to respond to your concerns, since I do feel pretty strongly that the subscription is a better option for most users. The items you appear to take issue with are the annual commitment, and the things that happen if you cancel early. Let's talk about each. 1. The Annual Commitment By referring to this as a "gotcha", you seem to be implying a couple of things. One is that it's something the company tries to keep hidden, and the other is that it's somehow bad for us users. In my own experience, neither of these is the case. The fact that the subscription pricing is based on a one-year commitment is clearly marked in all the literature, and was said aloud by the Adobe phone rep I spoke with, within literally the first few seconds of the conversation. So, it hardly came a surprise to me. I'll take your word for it that your own phone conversation was lengthy (mine was, too), but I can't help but doubt that a delay in getting to the subject of the annual commitment was what necessitated the length. Correct me if I'm wrong, but in all likelihood, you probably discussed many other things, as well as that, throughout the whole of the conversation. As for the implication that a term commitment is inherently a bad thing, it's not. Subscriptions, by definition, are intended for people who use products or services regularly, after all. Adobe's is no different from any other, in that respect. It makes perfect sense from a practical standpoint that any company would ask for a term commitment from a customer, in exchange for offering thousands of dollars worth of products and services, for just a hanful of dollars each month. I'd have been truly shocked if Adobe had not reqiired that sort of commitment from me. The only part of it that I found surprising was that it was only a year. I would have expected two or three years, at that price point. Surely you didn't expect they'd set it up so that a casual user could just pay $20, whenever he or she feels like doing something with Photoshop this month, and then skip it next month when he or she feels like doing something else, did you? That would be an unsustainable business practice. Quite frankly, those types of stick-and-move people aren't the kind of customers Adobe should want. On a side note, I actually wanted to pay the whole year in advance. There was no way to do that directly, though, which seems awfully silly to me. Adobe's billing system for Creative Cloud apparently is only set up for monthly payments, no annual option. The phone rep suggested I throw the annual cost onto a prepaid moneycard, and just use that as the card to bill. I thought that was a great idea. I plan on picking one up at Walmart as soon as I get a chance to get over there. I'll plug the card number into my Adobe account, and then I'll cut up the card, so I can't spend the money on anthing else. 2. Cancellation This a total non-issue to me, and to everyone else I know who's made the decision to subscribe. We use the products literally every single day, and nothing short of worldwide devestation would cause us to stop using them. That said, it's still worth talking about, of course. Your issues appear to be the fee, and the fact that you'd lose access to the programs and any data you've stored on Adobe's servers. Let's go point by point on these. The fee is less than I was expecting. Asking for half the remining term cost is hardly unreasonable, especially when you consider that at most it's only 10% of the purchase price: $100 for a single title, or $250 for the suite. (If someone cancels in the first month, there's no cost, per Adobe's standard 30-day trial policy. To be charged the maximum cancellation fee, the customer would have to cancel in the second month.) While of course, every customer's first instinct is to want all fees to be zero all the time, I think we all also understand that there's no practical way for that to happen. Like it or not, cancellation fees on subscriptions are a necessary fact of life. It's worth keeping in mind that Adobe is taking a not unsubstantial risk with the one-year term. As we discussed earlier, it takes at least four years of subscription fees to equal the purchase price. If someone cancels any sooner than that, the company loses money. In the unlikely event that I were to cancel, I would do it it the end of the year, BEFORE I commit to another year. In the even more unlikely event that I just wanted to get the heck out, and be done with the whole thing, then I'd just suck it up and pay the fee. Either way, I'd still end up spending less than I would have had I bought the suite, like I always did before. Losing access to cloud storage upon cancellation is absolutely to be expected. The same thing would happen if you cancelled any other remote storage service. That's why I NEVER archive anything in the cloud. Cloud storage is good for sharing and collaboration, but it should not be used for archival purposes, ever. If you want to make sure you can keep somthing forever, store it locally, always. For what it's worth, I don't plan on using Adobe's cloud storage. Between Dropbox, Google Drive, and other services my various clients insist on using, I've already got way more stuff in way more places than I'd prefer. That just leaves the part about losing access to the products and services. Of everything you mentioned, this is the only one I can relate to. I definitely understand that the transition away from the purchase model we're all so used to is uncomfortable. I felt that same discomfort initially, and to a certain extent, I stil do. Old paradigms die hard. But when I ask myself to think rationally aboyt WHY it's uncomfortable, I cannot come up with a good answer. I subscribe for lots of other things, and it doesn't bother me, so why should this be any different? If I cancel my Norton subscription, my security software won't work anymore. When I cancelled my WoW subscription, I could no longer play the game. If I stop paying my cable bill, none of my internet-dependent software will function. If I don't pay my electric bill, nothing in my house will work at all. If I don't regularly buy gasoline, my car won't work. Etc., etc., etc. My Adobe subscription is just one more item on an already very long list, so again, why should it feel like a whole new thing? The only anwer that my brain can conjure up is, "It just does." I can't base my business decisions on such a flimsy answer. I have to work wth facts, logic, and the reality on the ground. In this case, the reality is I'm spending less, and getting more, in comparison with my previous history. Logically, this makes the subsciption the only sensible way to go, from here on out. The discomfort over the paradigm shift will eventually go away. The money savings and the other benefits will remain. Here's the bottom line. If you personally don't intend to use Adobe's products on a regular basis, then subscribing would not be in your interest. Neither would buying, for that matter (unless you've got unlimted money, in which case it doesn't really matter which way you go). The very casual user is better off with less expensive alternatives. For those of us who DO intend to keep using the products regularly, the subscription makes an awful lot of practical sense. By the way, at the risk of going off topic, I'd be curious to hear what you meant by "sounds good but isn't anymore", in regard to Sprint. In the last several years, the only service they've stopped providing has been annual phone upgrades. (Now it's an upgrade every two years, just like with every other carrier.) Nothing else has gone away or changed, at least not for me. Has your experiece been different? Sprint remains the only carrier in the US to offer truly unlimted everything. All others either cap or throttle your data after a certain limit is reached, which is extremely not cool, in my book. That's why I've remained a Sprint customer for so many years now, myself. I've had no reason to feel like the plan they offered me merely sounded cool, but didn't stay that way. They've delivered everthing they promised me. Would you mind explaining what you meant?

Good point, Ciaran. In case anyone's wondering, Adobe's educational discount pricing can be found here. Creative Cloud subscription is offered at 40% off. Purchases are 60-70% off, which may or may not be better deal, depending on your needs. It's worth noting that Adobe appears to have changed their license terms for students. It used to be that if you use the products for commercial work, you'd no longer be eligible for the student discount. The logic was that your commercial work deos not magically bcome less commercial, just because you happen to be in school. However, according to Adobe's FAQ page; they've relaxed that riule, and you can now do commercial work with a student license, no problem. Here's the relevant quote: Adobe wrote: Can I use my Adobe Student and Teacher Edition software for commercial use? Yes. You may purchase a Student and Teacher Edition for personal as well as commercial use.

Nya Jules wrote: I smooth bind the pants to the skeleton, then select the waist area of the pants and set it in weight painting mode at hips left and hips right to 0. This way it looks definitely better but the upper back of the pants is somewhat distorted when standing still already. Do I need to paint something else? In nearly all cases, it's best to work additively, rather than subtractively, when painting weights. In other words, instead of painting less influence from the hips, paint more influence from the pelvis. You'll find that the weighting process works best when you start from the extremities, and work your way inward. For pants, here's the basic procedure: 1. On one of the legs, start with the cuff area, and weight it 100% to the ankle joint. This will make the cuff move when the foot flexes. Right now, the cuff is going to move way too much, but that's fine. We'll correct that in the next step. The important thing right now is to make sure the ankle, and only the ankle, is controlling the cuff area. 2, Now go up to the lower leg, and with a somewhat soft brush (meaning give the brush a decent amount of falloff), additively paint it to be 100% weighted to the knee joint. Cover the whole lower leg, including the entire kneee, all the way down to about the edge of the cuff area. Since the brush is soft, some of the white paint will bleed over the cuff, causing the cuff to behave more naturally, as it will now be pulled on by both the ankle and the knee. Test the movement, and if it's still too much, simply add more knee paint, until it behaves the way you want. 3. Now go up to the thigh area, and in similar fashion, weight it 100% to the hip joint. Cover the whole upper leg, including the hip area and the buttock, all the way down the edge of the knee area. Just as before, some of the hip paint will bleed over the knee area, allowing it to behave more naturally than if it were influenced solely by the knee joint. Test the movement, and if the thigh still has too much influence from the knee, simply add more from the hip. 4. Repeat the process for the other leg. 5. Now go up to the waist, and paint it to 100% weighted to the pelvis. Once again, some of the paint will bleed over the already painted areas, including the upper buttocks. Again, test the movement, and if the butt isn't behaving right, add more pelvis weight until it does. Ditto for the crotch. Now you're done. The first time through, it might have taken you a while, but once you get the hang of it, it's only a few minutes worth of work. You'll find that by working this way, from the outside in, you'll always get good results, fairly quickly. In any of the steps, if you mess up an outer area by painting too much influence from the next inner one, it's super easy to fix. Just reapaint the area, additively, from the outer joint. Then go back to the inner joint, and again paint additively, to fix any problems created from the re-painting you just did. Eventually, you'll see how to arrive at the right happy medium for the area. As you've already discovered, if you try to work subtractively, you lose a lot of control. You can end up spending all day playing whack-a-mole with stray vertices that won't cooperate. As soon as you squash one subtractively, another pops up to misbehave somewhere else. It can be very frustrating, and time consuming. So, don't do that. Additive is just about always the way to go. By working from the outside in, entirely additively, you'll never encounter the kind of trouble that subtraction causes. A rig that might have taken you a whole day or more to do subtractively can be done additively in an hour or two, or in many cases, just a few minutes. To put it in terms of hierarchy, it's always more effective add your way up from the bottom of the chain, than to try to subtract your way down from the top of the chain. If you were doing a full character, you'd start with the toes, and work your way to the feet, then the ankles, then the knees, then the hips, and then the pelvis. Then, for the arms, you'd start with the hands, then the wrists, then the elbows, then the shoulders, then the spine. After that, start with the head, then the neck, then each spine joint, all the way up the chain to the pelvis. Make sense? Nya Jules wrote: Edit: Oh and I've got another question: does someone know what to do with the "Avatar Weights Source" that comes as a layer in the standard sizing package? Edit: Oh dang I think I got it, I copied the source to my pants and it seems to work now without manually painting even! Yes, the source, as the name would suggest, is for copying weights to the target model. If the topology and/or shaping is similar enough, you might not need to do any tweaks, depending on which copy options you use. A lot of times, you will have to adjust some things by hand, though. In those cases, follow the same basics outlined above. Start with the extremities, and work your way inward, to the pelvis. You'll might be painting smaller areas this time, instead of whole sections, but the same principles still apply. When an area behaves problematically, it will almost always be because it needs more weight from the next higher joint in the hierarchy. In those relatively rare cases in which you truly do need to zero out the influence of a bone on an area of a surface, that's generally best done in the component editor, rather than with the paint brushes.

Well stated, as always, Code. I don't necessarily disagree with any of your points. To be clear, though, subscribers do have the option not to keep up to date if they don't want to. The subscription comes with an application manager, which tells you when updates are avaialble, and what's in them. If you don't want it, simply don't click the Install button. The only potential catch is I'm not sure if there's an official way to roll back, if you decide you don't like a new change. For those who are very concerned about that, I guess setting a system restore point prior to each installation would be a good idea. As for the involuntary beta testing thing, I do know what you mean, but I've got no reason (so far) to see it as an issue with Adobe in particular. I was asked to join the closed beta for one of their programs about a year ago. Even though the chatter in that group is entirely on point, the sheer amount of it has been so extensive that I've actually had to tune most of it out, or I literally would not have time for anything else. From what I've seen, Adobe's commitment to not releasing anyhting until it's ready is amazingly thorough. This has opened my eyes to why sometimes newer versions of their programs have missing features that later reappear. If an older feature doesn't work perfectly with the newer version, they won't include it, until it's been fixed. Of course, that has made me angry in the past, when I've bought an upgrade, only to find that some of the features I use aren't there. (Auditon CS5, anyone?) But now I at leat understand why it happens. With the company's income being largely dependent on the CS release schedule, they have to keep that schedule more or less set in stone, and sometimes full development of any particular thing simply cannot fit within it. I'm guessing this is yet another reason why they decided to do the subscription service. They still get to offer the suite as a suite, but they have a lot more freedom to let each program develop at its own pace.

Helium Loon wrote: What abou those of us who tend to keep and use software without paying for unneeded (for us) upgrades every few years? What about you, indeed. Quite obviously, this thread doesn't apply to you. For those who are neither going to buy nor subscribe, the entire topic is irrelevant. Remember the part where I said, "if high cost is your only reason..."? Well, for you, it's not the only reason. As you stated throughout your post, you simply don't want the new features, which is fine, for YOU. I thought it rather self-evident that my message was aimed at people who DO want that stuff.