Rodney

I love the mood you are creating with the lighting and textures. Simon already hit on the way your compositions are directing our attention. Very nice.

I apologize for burying this link here in this topic but it does relate to 'The Good Dinosaur' in that it's director, Peter Sohn, is a featured panelist along with Steve Martino (The Peanuts Movie), Roger Allers (The Prophet), Pete Doctor (Inside Out), Charlie Kaufman (Anomalisa) and Richard Starzak (Shaun the Sheep). The video is a one hour long 'Animation Roundtable' conducted by The Hollywood Reporter's editors Gregg Kilday and Carolyn Giardina. LINK One of my favorite take aways from the interview comes from Pete Doctor who, agreeing with the panelists (and referring to lack of dialogue in efforts like 'Shaun the Sheep') suggested that one pass of reviewing a film in production at PIXAR was to view the film without any dialogue to make sure it still works. Of course, there are many more gems of value to be found in the video and accompanying article.

I think he may mean 'access' as in 'turn on' rather than open a file that is already created. If that is the case: In the Project Workspace listing under Post Effects>New>Post Effect (Shortcut: Double Clicking on the Post Effect label will get you to this stage) The above creates an empty Post Effect1 container. Now Right Click on Post Effect 1>Change Type to>Hash Inc> (Select the type of Post Effect) These Post Effects can be used either on a Camera (drag and onto the Camera in the Chor) or via A:M's Composite feature (In the PWS listing: Right Click on Image>New Composite. The Post Effects you create can now be dragged/dropped onto into the Composite where you can add effects to any previously rendered imagery. Please note that a few Post Effects have more powerful equivalents in more recent GPU Post Effects which can be turned on in the Render Panel. Blur is an expecially useful improvement as the standard Blur has very little effect while the GPU Blur is what one would create blurring as expected. The rule of thumb of which Post Effect type to use would be that if the Post Effect exists as a GPU option (via the render tab) that should be used instead of the older standard Post Effect which is manually added to a Camera. I've added an image that shows the basic differences for accessing PWS and GPU Post Effects. The PWS Post Effects on shown on top/GPU Post Process on bottom.

It looks a bit like there is a shelf on the wall but... no wall. That may be an element that is causing some confusion. In other words, there are flying things (drones) that aren't flying and then there are stationary things (shelf and object on shelf) that are flying? That may be part of the riddle of the image but if intentional then perhaps additional riddles can complement that idea. If unintentional perhaps something could be done to further orient the viewer. higher camera angle to see that the objects are on the desk), hint of the wall, etc. My eye wants to move to a specific location (I think it wants to go to Santa's face) but it can't quite get there because it's trying to ascertain other stuff. With the theme of 'rein drones are falling on my head' I'd love to see a down shot where Santa is looking up and/or reaching up to a flying drone. Something falling/flying... something 'on/over my head', etc. Speaking of over my head... I'm in over my head here so I'll say, "Keep it up!" (Nancy, this is a first draft of a post that I was refining but browsed away from. There are several inaccuracies on my part which I wanted to fix such as referring to your post title as the theme etc. I'm posting this 'as is' for now. Sorry for any confusion this may cause)

I note that one of the guys that was heavily involved in the Bluth forum. weekly seminars and attended the masterclass was asked if he wanted to move to Phoenix to help with the project. In today's terms the book doesn't seem to dig quite as far as one might hope but back when it was first released it was a very welcome guide as there were very few books available on animation. While it has a lot of good visuals it's strength (as I recall) is mostly in word; practice and theory of hand drawn animation. For instance, a page or so devoted to more obscure topics as inspiration such as listening for 'the blue note' and courting the muses. Speaking of muses... music is covered very well in the book too and up until the time of Don's book music hadn't been emphasized much. While Don is certainly a master of animation my personal interest in Don's work tends to gravitate toward Layout. Don is an amazing layout artist. Once upon a time when I asked for a quick sketch, other folks wanted drawings of Banjo, Mrs Brisby etc. but I asked for a simple 'layout construction' sketch which Don happily assented too do. In a few seconds he'd captured a tiny sketch that captured more about layout than I recall every seeing in an animation book; just a few quickly placed lines of a thumbnail drawing in a tiny little sketchbook. As for working on the project, I'd say that if it does get going (the film.... not just the pitch) then a whole lot of folks will be vying for the opportunity. I admit to being jealous of Lavalle's opportunity to work with Don (assuming he chooses to do so) and I hope that if the opportunity arises there will be work there for you and many others too. One of the reasons I went to Don's master class was that I felt that if I didn't go I would never again have the opportunity.

Just a note to say that with 13 days remaining this project has less than $3000 to go. Looks like Don and Gary are about to have to get to serious (animation) work, making the pitch for the movie, etc. Update: The goal of $250K has been reached with notice stretch goals soon to be released. .

Nice test. So far so good! I like the idea thus far.

Yes, perhaps we should bring that back. I can see why it disappeared... users would have to remove that chor (or create a new chor) prior to being able to build something of their own but for new users (and upon renewal) it was a nice touch to welcome folks in such a way as that. Perhaps if it was just added to the Library folks could open it from there.

I vaguely recall someone modeling Playmobil sets/characters but I don't think they shared any of those with the community. A quick search didn't turn up anything.

For those that would like to use Renderman with A:M and those that have been making inroads to use A:M with Blender this should be of interest... The (initial) Renderman for Blender plugin was released back in Jun/Jul and allows use of the non-commercial license of Renderman. I don't know much more on the subject but wanted to update this topic. There are quite a few issues with the plugin and some documented pitfalls can be avoided by following the development of the plugin on github. For the latest Renderman release (currently v20) see: http://renderman.pixar.com For some getting started info: https://community.renderman.pixar.com/article/906/introduction-to-prman-for-blender.html

This is a bit of a pitch for University of San Francisco but provides some additional behind the scenes as well. The video highlights the contribution of USF alumni and PIXAR storyboard artist Rosana Sullivan on 'The Good Dinosaur'. xhttps://vimeo.com/148139574 And here's a link to an exclusive fandango video that covers a bit of background on the team real world research and a couple seconds on volumetric clouds in the film. http://fandan.co/1IHV8ck

Yes, that'd be a great way to go (for long shots) and it'd reduce your patch count from it's current number to somewhere around... 20 patches. That number of patches will save really fast. Of course you'd want to keep (at least sections) of the more detailed building or some facades for close up shots such as for use with someone standing in or looking out of a window, walking through the front doors, etc. Not to complicate the approach that Robert is suggesting but as desired using a transparency map the (black) of the windows could be made partially (or completely) transparent as well. I'm not entirely sure how well that wil always work with displacement maps also in the setup but as long as the displacements aren't too crazy it should work well. Added: Another thing the approach does is speed up the entire process for a production. If we don't spend a lot of time initially on models that require less detail we'll have more time for additional detail later where we deem it necessary.

Congratulations Dan! Aside: I find it interesting that other parties are already selling your book (at a slight markup) via Amazon's link to 'new and used'.

Thanks Matt, apparently my inability to follow the link and get to the resource was a permission thing. Admin's that can't see what everyone else can... what is the world coming to!!!** I figured something was up when I only got two responses for a search on 'pinetree' one of which was this topic. I see it now. Thanks! I like your latest but wonder if it wouldn't benefit from a little post processing (I'm thinking blur for the shots with depth primarily). But as always...as long as the client is happy! **I can see it now but don't have access to download. For that I'll probably have to log in as non-admin. This seems all very familiar to me.

I like the styling of that. Its within that range where you cant help but wonder if parts are CG, parts are stopmotion and if all the various elements have been composited together.

Unless there is something significant happening (such as free stuff being given out) I will probably post updates to this project only on my blog. That way I won't spam the forum with other folks desire for you to give them all your money. The occasional exception may be made such as when/if the project is funded or if some reward is added that is significantly relates to Animation:Master. It is worth noting the new project is now just 50K away from it's goal of $250K with 39 days remaining. Quite a difference from the slower fundraising over at Kickstarter. As such I can't foresee any reason the project will not be funded. Unless I missed something there does appear to be one casualty of the move over to Indiegogo and that is the digital course notes I mention above, 'Digital Animation: One Way to Approach It'. That seems to have fallen off the list of rewards. Oh disregard. Gah. I see they've added that to the $5 tier reward. I'd say that the info contained in that 39 page handout will be well worth five bucks. I see at the $15 tier they are giving out digital copies of the now out of print books 'The Art of Storyboarding' and 'The Art of Animation'. Both of these books should be in every animator and cartoonists library.

Objects don't tend to move for no reason so we'd just have to figure out what that reason is and address it. I'd say one of the most likely reasons is that something is accidentally animated. There are several ways to get a group of objects to move as one object. The first would be to make them the same object (i.e. combine them). Another approach would be to establish a Relationship that treats them as the same object (or as Parent/Child). This is usually the approach of using Bones. Where the mesh on one or more objects is assigned to a single bone those objects will move together as if it were one. Where two or more bones are involved that aren't in a Parent/Child relationship objects are constrained (manually linked) together in such a way as to dictate their location and orientation. It should be no surprise here that the method used here would be to use Constraints of which there are many types; Translate, Aim At, Orient Like, Path Group, etc. Quite often a Bone that doesn't have any mesh attached to it will be used as the target of a constraint. A special bone called a Null is often used for this purpose. An example of this would be having two eyes (each having their own single bone) following a Null via an 'Aim At' constraint. In this way wherever the Null is located the eyes will follow. The most likely reason objects move (when no movement is desired) is that we have created that movement on the timeline. This can usually be avoided by making sure we are on frame 0 of the animation (as opposed to some other number where a keyframe will be generated resulting in the unintended movement. Another way we can prevent keyframed movement from occurring is to turn off the option to create keyframes (via the Animate button - the A button on the top menu). Just be sure to turn it back on or the problem will reverse and the question then become, "Why isn't anything I animate moving?"

Here's a behind the scenes look via Wired (covered by fxguide's Mike Seymour): xhttps://www.youtube.com/watch?v=VZ9UGFsWexM

There's an article at Huffington Post that hits on a lot of elements of what the old story would have been versus the film that launced into cinemas. For those interested in the world of 'The Good Dinosaur' it's well worth reading. It details why there were so few dinosaurs in the film and outlines the changes that made the two primary characters younger. http://www.huffingtonpost.com/jim-hill/by-thinning-the-good-dino_b_8738842.html Now, here's the thing... Technically, they could still make (most of) the originally planned story as a sequel to 'A Good Dinosaur' although a few things (character plot points) would obviously have to change. I'm actually glad the additional cast members were set aside because the entire film would have taken on more of a 'Land Before Time' feel. I do wish they'd hinted at a little more of the outside world (specifically the characters that inhabit it) as that would have sparked many a creative imagination. Of course that may be very intentional on the part of PIXAR as the lack of detail gives them considerable time so that at their leisure (that is to say... based on the economics of creating divergent properties related to the film) they can develop the world as they see fit. .

I've sent my photos Mark! Thanks and my apologies for the delay. Roger, I'd take your photo for you but.... you may still be somewhere in the Midwest that isn't near to me.

Okay Matt, you've got me intrigued BUT... your link doesn't go anywhere for me and I don't see any other recent topic related to trees posted.

Yes, there are multiple ways to approach this the simplest way being just to rotate the instance 30 degrees and be done. For more automation (of turning of many trees) I'm sure a random expression could make sure every tree in a scene is turned slightly different that the others. A breakthrough in Modeling for me occurred when I saw other people Modeling in a Choreography. I was like.... whoaaaa... we can do that?!?! To keep file sizes low we would then just save the entire Project (or Chor... some people don't save Project files... only Choreographies.... but this is a tale for another time). If the desire were to be to have everything as one mesh (and not modular parts of a whole) then all we have to do is export to Model from the Chor.

Everything is generic (modified cubes) in this test but I wanted to get the patch count up higher for testing purposes. If I calculate correctly this Chor has the equivalent of 77,000** plus patches which are instanced from 1 original file (consisting of 6 buildings apiece that contains approx 11,000 patches). The Project file (with everything embedded in it) saves immediately. While I'd have to check, I believe the Project file also contains the individual story of one building that is duplicated three times in height (to get 4 sections/levels per building). If doing this again I'd probably make the windows of a type that could appear more detailed as well as give the tops of the buildings some type of detail. All of the buildings where turned 45 degrees via an action before being exported out into groups of six buildings. In the image you can see top center a building that is larger than the others. This was a quick test to resize an instance (not the original) via a Chor Action. To get differentiation in similar objects that have varying details... that's where Poses (pose sliders), Actions and such come in... but none of which are in this test. I know your setup is more detailed but am just going through the paces of testing to see what might be causing the specific issues you are dealing with. If you have tons of patches that will definitely take more time for A:M to process. If you have copy/pasted tons of patches that aren't optimized that will take additional time to process. In the image you can see top center a building that is larger than the others. This was a quick test to resize an instance (not the original) via a Chor Action. It can help to think of Modeling in similar way to Animation. We don't blast every bit of information into one frame but put what is needed where it is needed throughout. One of the benefits of computer animation is reusage where we don't have to recreate everything... every time... from scratch. We can just point to the resource and say, "(As a starting place) reference that." Edit: Nice example Nancy! I wish I'd thought of that. *Generic project is attached **According to A:M''s Info setting (RIght Click and select Info) the Chor contains just over 86,000 patches (remember though that only 11K of those are original patches). 81K plus.prj

I think we need to introduce you to the art of instancing. It's far easier (from the programs vantage point) to save a link to a resource than the resource itself copies many times. A question that might be asked is, out of those 800 or more decorative roof braces how many are exactly like each other? If the answer is 'They are all exactly like each other just put into different places' then you can drop that patch count tremendously just by referring each one back to the original. The the file only needs to save the original plus the 800 textual links (not sets of patches). I'm in the middle of a test (creating a cityscape of hundreds of buildings with hundreds of windows) to see how much it slows down my computer. At present A:M has ground to a halt while it's attempting to save the model. In a little while I'll try the same thing with referenced instances of original models. I guess what it boils down to is determining how many unique shapes you have in your setup and then optimizing your project from there. You mentioned 240 windows. I assume that is 240 of the same window duplicated over and over again. ...and of course, once you have the model rendering perfectly that'd be a good time to capture that render for use as a decal so you can place the resulting image back on a model with minimal patch density. In that way a 300,000 patch model can be whittled down to (easily?) less than 100... give or take areas where you may need close up detail.

I know what you mean. I think we all like to see our efforts rendered. I trying to cure myself of that! There are some ways we can get a good look at what the render may spit out for us. Render Lock Mode (Shift Q) is the main one I use (although I'd like to try to cure myself of using it). It's the one that we need to hit the Escape key on to tell A:M we don't want to see that preview. If there is a specific part of the screen/render we need to see we can hit the Q key by itself and then drag our mouse (while pressing down the left mouse button) and that will preview render that selected area of the screen. This won't render to file, it's just for us to get a quick look at what the final render will look like. There are some other options as well but I don't use them often. I try my best to stick with the realtime view and just get a sense of what the final render will look like based on that. The more complicated the scene (lights, particles, etc.) the more the final render will diverge from that previs. One thing you can to also is just render one frame of a sequence before turning the renderer loose on the full sequence. \ Like I said, something is very wrong there. It'd be good to figure out what that is. I must assume here that by 'save' you don't mean 'render' as those are two very different things.