Leaderboard

This is a long time personal project that I've had on my mind to write, script and animate. I finally put it together and It's a little awkward. I've used A:M to make low-fi "retro cgi" art, over the last few years. However, I think the software is very capable of making any variety of style or appeal. I've never done a video essay and it did not come naturally to collect my thoughts and lay them out like this. I'm glad i did anyway! The video essay on YouTube

Wow, that was amazing. What a great essay. Those 'y2k CGI aliens' aren't dead, I'm working on a full length movie they feature in. Still using AM v13s, seven pc render farm, windows 7, no internet connection, no plugins, no subscriptions and most definitely NO AI. At all, none, not even upscaling, I'm rendering the whole thing 4k and most of it 16pass multipass at a few hours a frame. I'm only 30 mins in of 75min after six years of production with likely another 5-7 years to go. Also all orchestral music is done on 16 vintage roland synths and even using hardware reverbs and effects. Pretty much the most slow and tedious way of doing it but it's completely future proof so I can make movies until I die lol Don't get me started on people moving to Maya and other software, most of my team back in the 2000s constantly went on about moving to Maya or Soft Image or whatever. The funny thing is, my mate is also making his feature film but using Maya and Houdini and modern pipelines. He's constantly having licensing issues, can't open old models without renewing subscriptions and can only run one copy on one machine cause its too expensive to have more. And something even more funny, he was over the other day seeing my workflow on my old PC and I pressed play in the AM timeline; two shaded characters and a wireframe set, and an embedded reference video, started playing at the same time at atleast 30fps. his mind was blown "I can't do that" he said, I can only watch it realtime once I've rendered it. (That said, he's a pretty good artist and his film will be awesome.) The trick with AM is to lean into what it's good at, curvy lines with razor sharp edges, reflections (lots of reflections) and keeping things simple. I can't stand the look of modern CGI movies, some of them are completely unwatchable with ridiculous amounts of colours in every shot. And the lighting and surfacing is so real all you see are pimples and eyebrow hairs, it's distracting and unnecessary. I also hate all the animation as animators think they are actors, shoot video of themselves and act like they think animated characters move. The result is it's all the same, it's all shit, over arcing over animated mess. Ok, not all films but most of the slop that comes out now. Oops, sorry for the vent. Your video is very inspiring and makes me feel I'm on the right course and all the work is worth it. Thankyou! Dylan

That is 51 frames of ASCII images

Some other attempts. Viewing an illustration of Lovecraft I modelled a sculpture for it. Then I was inspired by the ShadowRiders from In the Ban ofthe Ring. gif won't load. discord Finally I illustrated a screensaver from Win98 to Quake.

Also, I had the pleasure of attending multiple Hash Bashes that Martin put on back in the day and even got to direct (badly) a couple of scenes on Tin Woodsman of Oz. The old Hash crew were great to hang out with and Martin is a legend and a great host. It's a pity what went down with his team but he soldiers on to this day. Robcat is a godsend, his level of knowledge on all aspects of AM is unparalleled and his videos and advice have been a key part in me making my movie. I'm also working with Aldrich who's here on the forums and makes any new models and characters I need. Another Hash legend I don't post here much but I sandbag and watch whats going on. This is an amazing community of extraordinary people and I'm proud to be a part of it. Go AM! Dylan

Wow @Yopachi, I didn't realize how much I needed to see that. Outstanding essay that captures so much or the Animation:Master experience (and early 3DCG history) in such as short video.

Here's that same GPT generated airplane with some minor tweaks (mostly to control point positions):

So, after many hours, over many months I finally got the Sample plugin to compile. First let me state that I am new to C++ programming. I started programming back in the mid to late 70's on an Apple I (Integer/Applesoft BASIC) in high school and a SWTP 6800 with an IQ terminal (Hex op code) at home. A few more languages (Cobol, RPG, Assembler, other Basic flavors) and moved on to other endeavors. Tried learning C, or maybe it was C++ (Borland), but I never got into it. I started again, quite a few months ago, with the assistance of a few others who had in recent years started their journey, to learn C++. One reason was to dabble in and experiment with the creation of plugins for AM. I now have a list of cleanup tasks on this project, then compare it to the C++ Sample Plugin project file, available to download and the existing info/tutorials on it to perhaps bring it UpToDate with the available Microsoft Visual Studio (2026). Which is what I used to compile it. I also have to get it working for 32bit systems. Hopefully that will be as simple (😁) as changing the configuration of the project in VS, but we will see.

I'm still plugging away, mostly with backgrounds sets. This latest was more of a reflection exercise as I usually avoid blurs and reflections. I was fighting this shot for quite awhile trying to get just the right camera angle... just the right reflections... etc. etc. The moon is one of my (pat. pending) 5 point patch circles. (and it'd be good to keep the moon in place for parallax purposes) I made no attempt to update the proxy trees. Project file should have everything embedded. LakeFrontPark.mp4 LakeFrontPark.prj

I'm on the third part of the animation. I made sequence of a Thunderbird on Forrest Stubborn. One of a teaparty, walking scene of Parker & Brains. At last I tried a scene with Lady Penelope. I.m not such a great character modeller. My attempts ended in a granny like figure I'm a bit ashamed for. Fortunately it declines the time period I'm in. I have the idea the camera is a bit trembling, or the image files are rendered in a misligned way. https://mad-red.com/animationmaster/

I've been having a blast and exploring a lot. Here, have a pic: Okay. What exactly is this? This is just a quick illustration of a utility I threw together that allows for extracting parts of models based on Named Groups, saving them out as separate models. the shape on the top is the original model and the 3 shapes on the bottom are those three sections extracted from the model and drag/dropped back into the Chor. A simple example but particularly useful for complex models where we might have parts that lack detail and we want to extract that part to plus it up and drop it back in separately. The simple interface: For the moment I'm keeping all groups inside the extracted models to ensure that groups that effect the extracted splines and patches still get the effects of those assignments. This does tend to cause an error in A:M when Groups don't have any assigned CPs and I'll likely add a pass that removes those Groups in the extracted models. As a more complex example, let's say we have this town square with a number of proxy buildings, pave stones, fountain: If barely seen in a shot the proxy will likely be fine. But for those objects we need to plus up we can extract those and plus them up. If we extract the Barber Shop for instance: We can steadily improve upon it until it meets the needs of our production. This update is still in the 'proxy' arena but its starting to look a bit more like a barber shop.

There is a trend of youtube traffic going strait to discord as a sort of after party. lots of gen z and alpha are already used to this pattern and might not ever step foot into a forum otherwise. could be a way of netting more animation master traffic for younger audience. and I'm gonna push the traffic this way if i can. but my suspicion is that the kids wont come over here unless they are serious about learning. so there are two functions and they can feed into each other. rodney is right about the real time nature there is already some quick back and forth and learning going on

Do we need an alternate to the forum? Is there something the forum is not doing?

TWINSDANCE.prj

I was having GPT model shapes made out of tubes with varying degrees of success. For instance, this cartoon dogs head (here modified to connect the tubes and eyes added manually by me) The drawing of 'tubes' by me in an effort to have GPT shape and place tubes in 3D space. One thing GPT can do short of actually creating the models is help plan the model by creating the plan for modeling via an image. This assuming you don't want to draw it yourself. Then you can just model the thing yourself.

Okay, what's going on here? In trying to model some curved splines and patches I had GPT create a treasure chest (not particularly successful). A treasure chest needs gold coins though right? So, had GPT made stacks of gold coins. Successful but time consuming to tweak and rerun with variations. So, what to do? Answer: Have GPT create a python program to create stacks of gold coins with easily adjusted settings. Many of the variables were informed by my failures to create good looking stacks of gold coins. For instance, if too close on top of each other the stacks look too much like long tall objects. Even though the coins have random shades of orange and yellow. So, need some distance between coins vertically. Perfect stacks horizontally don't look good either. So how about a 1 in 10 chance the coin will go in the same direction as the last coin placed? Etc., Etc. I started with GPT creating clusers of stacks with 1000 coins total. Not the best starting demo and we want the user to set all those numbers as well (min and max for stacking etc. too) Here's Take 1 results out of the python program that replicated the basic process GPT was using. Not bad. Save out a file with the settings for that (in case we want to recreate the same or similar set of coins (seed value allows us to get same results with random numbers) Try 1000 coins (the programs current max count) as these models are being generated immediately and... As each coin/object has it's own group we can grab any coin we want and adjust. Don't want to stack coins? Point to a different model, such as a sheet of paper. These processes are pretty good at plussing up the Duplication WIzard. Which reminds me. I didn't add an option for rotating each object as it is placed.

After having GPT extend the buildings, storefronts and street with cars another 10x wide I set up a few cameras and ran through the scene. The 'helicopter shot': storefront_helicoptershot.mp4

I'm wondering why a children's book needs to look Noir. If i had to get a dozen images for a children's book... I'd call Rodney! Rodney is very clever doodler. He makes great-looking characters. If it needed to be hi-contrast I'm sure he could do that too. (I hope it's OK that I showed these )

I'm certainly cheating a bit by generating models with flat planar surfaces but... again... method to madness. Had GPT add books to the bookcase and while it did it the first time the books looked more like boxes because of how wide they were. Second attempt: This is the point where I've introduced GPT to the concept of Group Folders. Before we were getting a lot of benefit out of Named Groups but organizing Groups into Folders will be even more useful. Specifically, the set of groups that is used just for generative modeling won't be very useful to the users so those groups can be placed in their own Folder. It is there where temporary data can be stored as well. Consider for instance where the group might store instructions and once a task is successfully completed that task/group is moved from one Group Folder to another. Groups specifically used for targeting to apply bones, decals, materials, etc. Some rules likely need to be devised in order to determine what groups can be removed without effecting the look of the model and which are essential.

Something I think of importance: Note how GPT has named all of the various surfaces of the model as Groups. This is useful on many levels but primarily for identification. If we were to tell GPT to connect SeatTop to the SeatFrontFace it might not actually do it correctly but at least it has named those parts so it knows what they are. Thus giving it a fighting chance of success. From a users perspective it saves a lot of time performing manual identification.

Short answer, every hair is like three or more CPs whose position needs to be recalculated on every frame, with motion and collisions and stiffness and weight, etc taken into account. A head of hair will easily be more points to track than most character models typically have.

GPT is struggling with the understanding of Hooks (although making some progress). It would be good to feed some information about Hooks from the A:M SDK. Of note: Animation:Master does a very good job of repairing models where it can. Where I am currently with Hook testing is getting GPT to generate the model, A:M identifying the error upon loading, and A:M fixing the error to best of its ability. In some cases this requires manual fixing to take the model to its intended state. (although not in my latest test as A:M fully repaired the problem) I then feed back the repaired model to GPT (along with any error messages) so it can compare the good model with the bad and make appropriate adjustments. Here's an example of a generated mesh that A:M repaired upon opening

Two static entities for Quake I made with the Animation Master! craneA.mp4 craneB.mp4