heyvern

Based on the animation you posted you have made key frames. Those "red dots" or "boxes" in the time line are key frames. So called "normal" frames aren't and don't need to be "created" they just "are". The time line is the frames. If your animation is 300 frames long, you have 300 frames just waiting for you to put "keys" on them. (I won't go into "inserting blank frames" just now. ) A frame is just a frame. Keys exist "in" or "on" a frame. A "normal" frame is really just a "holder" for "keys". Keys do all the work. If you are sitting on the couch doing absolutely nothing, waiting for someone to knock on the door, the seconds and minutes that pass while you wait are "frames" in time. When the person knocks on your door and you get up to answer it, that would be a key frame. p.s. This one time I used linear interpolation while sitting on the couch and tripped over the table when someone knocked on my door. -vern

A frame is nothing... just a frame. A single point in time. If you have a 10 second animation at 24 fps the "frames" are all the frames of the entire animation. A key frame (or just "key") is a point of motion ON a frame that indicates a change in the object that was keyed (key frame). This key could represent a single channel (x translation) or it could represent several channels (x translation, y translation, rotation, etc). If you place a key frame on frame 10 that moves something up, AM will move the object up a "little bit" on each frame till it reaches the key frame on frame 10. It is the "little bit" on each frame that creates the motion. This tweened motion is called interpolation. To make one thing move at a different time than something else you would move a specific items key frame forward or back. This is why you need a key frame to make something "not move". That key is just as important as the key that indicates movement. If the ball in the example moved up gradually from frame 0 to frame 10 (the key frame) and you placed another key on frame 5 identical to the key on frame 0 then the ball does not move from frame 0 to 5. It starts to move on frame 6 and stops at the position on frame 10. Moving the key on frame 5 forward or back will change the speed and starting point of the motion. Some animators do not use key frame tweening or interpolation. They will place a key on every single frame (or every 2 frames, or every 3 frames). This is the way it was done in the "good old days". An animator would draw the important key frames and lowly less skilled artiists (just kidding but not really ) would draw the "inbetween" frames. Key frames are not specific to Animation Master, they are a standard for all animation programs and animation in general. -vern

Aha! The Psychic Software Syndrome strikes again... I think. It sounds like it to me. The problem you have is thinking of "moving" and "not moving" as two different things. They are the same. If you want to make something NOT move you need a key frame for it just like you need a key frame to make something move. The reason I call this the Psychic Software Syndrome is that new users unfamiliar with key frames and how the timeline works expect the software to "read their minds" and somehow know when they want something to start and stop. Just remember that EVERYTHING needs to be key framed. The software doesn't know the difference between a "moving" key frame and a "not moving" key frame. It knows nothing you don't tell it. If something is NOT moving from frame 0 to frame 5 and then moves from frame 5 to frame 10 you will still need a key on frame 5, the same key from frame 0, to hold it in place (not moving) and then the moving key frame on 10. You can help with this by changing the key filters... what gets keyed when you move things. If you set all the buttons to key everything then each time you move one item all the items get keyed at that spot. This can create more keys than you need though. I will stop there just in case this is not the problem you have. -vern

10 fps is probably too slow. You could cheat and make sure you speak and move really really slowly. At 10fps you might actually be moving a marker on the face so fast from one frame to the next there is nothing to "track". -vern

Dagnabbit! I was expecting some kind of "kung fu" animation technique. "Ooooh, cool! Action fighting! " Subject line needs a comma. -vern

The CD is required to launch AM but once it is running you can remove the disk. Of course if you haven't installed the libraries on the hard drive during the installation and you remove the CD those items won't be available (model and material libraries etc). Hash does offer a USB dongle that allows AM to launch without a CD. Contact Hash about purchasing that. -vern

You would call the "Transfer Department" on the 8th floor and ask for Jill. She has the "3D transfer forms" that need to be filled out and will put one in the mail for you. Fax the completed form back to the the Transfer Department (Make sure to include the "Transfer Allocation ID number" when you fill out the form.). Usually after 6 to 8 weeks they will get back to you with the proper documents and approvals from the background check for you to apply for a "3D Transfer License". It could take another 3 to 6 weeks before you get your official "3D transfer ID Card". Once you have your ID card, post here and we can tell you how to use it with AM to "transfer a picture to 3d". I'm just pulling your leg. there is no Jill on the 8th floor... it's Alice on the 6th floor. I'm kidding again! Seriously though, if you could give us a little more information about what you are trying to do we can help you better. AM can do a lot of things with a picture (decal it, warp it, make it move, rotoscope it, etc). If you narrow it down it may save time. And of course try to do the tutorials (if you don't do the tutorials Denise in Human Resources on the second floor will put mean post it notes on your desk). p.s. I should have been a bureaucrat. I could have designed really complex processes for getting information. p.s.s. There is no Denise in Human Resources. -vern

Hey guys? Isn't December 1st like... on Saturday? Not pushing or anything just curious. -vern

At first I also thought too many splines in the nose but on second thought you may NEED that density for creating facial expressions. I would consider keeping patch count lower on the rest of the model. -vern

Ramon, you could use that same thing to "blur" your models. Much easier than key framing it by hand. -vern

I always use custom settings for "previews". Multipass set to 1 is much faster... no antialiasing. With MP off you still get that antialiasing step, setting it to 1 turns of antialiasing completely. I also turn off EVERYTHING, shadows, reflections, decals etc etc... I think the "low" setting does this by default... but you can turn off images there. Of course this depends on what you need to see in the preview. This is fine for previewing the "action" or motion but if you need to see lighting and shading... you need to turn some of that back on.... although single frame renders might be enough for judging that type of thing. -vern

The way the leg and foot are splined depends on how much detail you need for the toes or "bumps" on the elephants foot. Extrusion or lathing for something like this would seem to be the logical solution. However if you need more sections to create those toes then you end up with a ton of extra splines in the whole leg just for the foot. You could try having a "simple" leg (6 point extrusion) with fewer cross sections then use hooks to add in a denser mesh just for the foot area. Making sure the hooks would be in a place on the leg that doesn't bend. A 12 point extrusion circle would work better for creating toes on the foot. You might need more than 12 for the foot though depending on the number of toes needed. Think of each "point" in the extrusion as a possible "toe" that can be pulled out. Assuming a 12 point extrusion you could close the bottom of the foot treating it like a "grid". You would end up with 4, three point patches at the "corners" of the closed area, but the flat part of the foot would be "smooth". The 3 pointers could even be adjusted so they would be in the flat area and not even show that much. Just connect all the points straight across on with the last extrusion. You could "pull out" toes from the outer ring of the mesh of the foot. Then as I described above connect the foot to the leg with hooks. If the leg only has 6 points in the extrusion then the 12 points of the foot would hook perfectly, with alternating hooks between each spline. This explanation stinks without pictures to back it up. The "dangling" points on the outside represent where the foot would "hook" or connect to the last extrusion of the leg. You can see how the shape of the splines for the foot and the bottom of the foot are arranged for "toes". The bottom of the foot would be pretty "flat". The 3 pointers might cause some bumps. You could increase the spline extrusion count to create more toes or have more patch detail for defining the toes, the only trick then would be how to connect to the leg. Maybe increase the leg section point density to 8 or 12. -vern

That's it John! Yes this solution is fantastic. I especially like the "sagging" loose chain effect. I can't remember exactly what I used it for, but I recall "reverse engineering" it for another project I had at the time. It might have been early early... early... in the TWO project when I was working with chains and gears for the workshop set. -vern

That's it! -vern

Somewhere on the forum is the perfect bicycle chain tutorial with an included project. It is perfect. I remember playing around with it because it looked so fantastic. I hope it isn't "lost". I suppose stuff does get archived and this would have been quite a long time ago. -vern

I can't remember who it was but someone a while back was working on a project and had the most gorgeous soft puffy fuzzy clouds using sub frame "jitter". They looked fantastic. I remember there was a sort of stylized "Knight" character. The whole project had a beautiful global illumination look to it. -vern

You could "blur" the texture using the same sub frame technique or blur the texture in photoshop first. I think blurring one texture in Photoshop might be easier. For sub frame decal blurring you could animate texture size at random amounts in the sub frames (is this possible in AM?) or apply the texture multiple times in different "random" positions and animate the opacity of each decal in the sub frames (I know this is possible). Or you could create a set of image sequences with the texture using photoshop. Randomly move around like 3 or 4 copies of the same image in Photoshop then in AM use those images as a sequence for the decal and randomly assign the frames in the sub frame render. Just enough frames to create a pseudo blur. The trick in AM is the "random" movement. You really need it to be random in all directions including maybe a slight random rotation so you don't see "sharp" versions of the decal. This could be done in an action so you can easily "tweak" it. I can see how physics would be an issue. In that case you would need a duplicate copy of the model NOT jittered, like a proxy, that doesn't render but is used only for the physics. The "jittered" version would render but not be effected by physics. Both could use the same action for movement, but the jittered version could use a short repeating action for blurring. -vern

Try rendering to an image sequence if you aren't already. Rendering to a "movie" type file means that whole thing has to live in memory until it is done. At least that is how it use to be. This way once each frame is done it just "starts over".... sort of. Frees up more resources. If I have this wrong someone tell me. -vern

Hash added "jittering" to hair because in past versions someone came up with the idea to use sub frame jitter to "smooth" hair so it doesn't need to be so dense. Sub frame motion is cool. ---- I love these types of threads. It starts out all negative with no easy solutions and then very quickly the problem is solved. -vern

jpappas you are a genius! Ramón, this should do EXACTLY what you want. DANG! I should have thought of that! I did something very similar a while back using sub frame motion blur to create "vector" lines from moving 2D shapes. I can't believe it... this is such a simple solution and can be rendered all in AM. -vern

the problem with a "post effect" is that in AM a post effect is applied AFTER the render, thus the term "post effect". It would be difficult to apply a blur to an object that is in front of other objects. The object needs to be blurred, then some kind of "3D" mask to determine how the blurred object fits into the 3D space. Using a material to simulate a blur could work but it is going to be difficult to create this effect on say a character model with textures and surface properties. You won't be able to get a uniform blur on model details like face features or shapes. You would need to have two copies of the model if you wanted a "non blurred" version. Also you wouldn't be able to transition between the two smoothly. Don't be afraid of compositing. Try it with something really really simple. You may find it isn't that hard... and it might save you a ton of time. You could try another thing that would simplify the compositing. Let's say you have a moving character or model in a choreography that you want to "blur". Render out the blurred model in the scene by itself making sure you have an alpha channel. Use DOF to make the render blurred or if there is a blur post effect (is there one? Not sure) use that. Import this rendered animation as a camera rotoscope and set the properties to be on top or above the scene. Now turn off the model in the cho and render. This saves the trouble of actually doing any "complex" compositing. You are just using a camera rotoscope. ----- Side bar: A while back I tried a crazy idea based on camera rotoscopes over a chor. I used the OUTPUT rendered image sequence for several layered camera rotoscoped animations with an offset and transparency. When each frame rendered, it used those rendered frames as the rotoscope creating a bizarre... motion trail effect. I had to put that aside until Hash fixed an issue with image sequences. It was fixed recently so I may have to try fiddling with it again some day. -vern

That would work for very simple objects but wouldn't actually "blur" a complete model. I think Ramon is trying to achieve a "real" blur effect on an otherwise "normal" model. The only way to do that would be to use DOF or post processing with compositing. I think this is a very specific and unusual effect. It isn't something that would come up very often. It would be 10 times easier to create this effect in post then to create a feature in AM to do it. The return on investment of a feature like this would be small since it wouldn't be used very much. -vern

No way to know with out looking at the actual files. It could literally be any number of things. It could be not enough ram to load all of the models or it could be a problem with one or more of the models, it could be too many lights etc etc. Some things take up more ram in a model than others, like big gigantic textures, raytraced lights. Materials use a lot more processor power when rendering, which slows things down as well. So if you have a bunch of models in one choreography with big images and materials, a ton of lights and raytraced shadows... etc etc... and the machine can't handle it then it causes trouble. Do each of them render okay by themselves? Render speed is mostly based on the computer's processor. Having more ram means you can have "more stuff" in the project but as far as I know it won't speed up render times. I have a PC and a Mac. The PC has a much faster processor than the Mac but they both have the same ram. The PC will render projects much faster than the Mac but they both can open huge files without much trouble. I had a project once with several models that all used HAIR (lots of hair). Each model rendered fine by themselves but all together in one cho with lighting etc... the machine choked. -vern

Have not seen that. This effect would have to be done with some type of compositing. There is no way to do that to a specific model in AM. If you want this effect you will have to render your project in "layers" and blur them in post. For example: Render the object you want to blur all by itself with an alpha channel for compositing. You could render with the shadows or you could render the shadows as a separate "layer" for compositing. Turn off this object and render again. Now you can import those 2 or 3 layers back into AM or into another video editor and blur the object independently of the rest of the scene. This would be a good reason to have the objects shadow on a separate layer so it can be blurred more or less as needed. If using AM to composite you would use depth of field for blurring but this would be a "fake". I don't have the latest greatest version of AM so there might be "compositing" tools built in now that allow for "blurring" layers. I'm just guessing. I know the latest version has a built in NLE compositor. -vern

Keep in mind AM is a "3D" application. It simulates the "real world". If you want something "fuzzy" like a "furry ball" then you could use hair. If you want a fuzzy ball of light or energy you could use glow, or a volumetric light or a glowing spherical combiner with transparency fall off. The type of blur that exists in the "real" world is depth of field, or focus. AM doesn't have "painter" type effects like "blurring". You could render different "layers" of objects and import them back in to AM and composite them on layers and use DOF to "blur" one or the other. This would be a way to "fake" object blurring. Could you tell us exactly what kind of effect you are trying to achieve? A glowing ball? A fuzzy, hairy object? An out of focus object? Without know what you want to do it is hard to offer more specific suggestions. This stereo image used render as lines and glow for the "cube". I used a material with "blurred" edges for the glowing stripes coming out of the screen. -vern