Short Film: Minecraft meets Atom Zombie Smasher

Hello, everyone. It’s been some time since I’ve had an active WiP thread, but here is a new one. The concept for this short film is that it’s a sort of mashup of two games, “Minecraft” and “Atom Zombie Smasher”. Everything is seen from the POV of the main character, and it’s all done in one take. Because of these constraints, I’ve decided to make everything CG, aside from my hands which I’ll be shooting in front of a green screen.

Originally I’d hoped to have the film completed by the 15th of April, but with the amount of work to be done, I’m shooting for the end of the month.

I have already completed a fair bit of work and testing. The events in the film present issues I’ve not yet encountered yet- questions like “how do I control a crowd of characters, while still allowing a rigid body simulation to act on them individually?”. Actually, that’s the main question. There are some others, but that’s the big one.

I’m also continuing to try to find ways to get better results. A recent Blender Guru podcast (Episode 69) reminded me that it’s better to start with the big picture, and then work your way down to the details. Similarly, I think it will be best to create a storyboard or animatic for the film, to be sure that the actual story ideas are working, before getting lost in the technical details.

I hope this project may be of interest to you, and maybe you’ll even learn how to do something new!

Current Progress:


In “Atom Zombie Smasher”, there are giant mutant zombies that are larger than buildings. Luckily, there’s also a mutant zombie from a Minecraft mod, so I’ve decided to recreate him in Blender. He’s not rigged yet, or even properly laid out, but the basics are there.


I’ve been doing some tests with boids for the crowd simulations. I’m not sure yet if I’ll use boids or another method, as zombies need to be able to die when hit with an arrow, so that might require a more manual approach if it can’t be automated.


I’m using a Minecraft map someone has already made, because it saves a lot of time and looks better than doing it myself! This is the map: Greenfield
I’m still making slight changes to the map layout, as well as changing textures around. Textures are still WiP as well.


hey james, nice project, you play much minecraft?
recently I found some game of the same style (sandbox) but 2d call starbound:

you can visit diferent worlds, Im not sure if they are randomly generate, maybe could give you ideas too!

Hi James, the boids are a good idea. I’ve never tested them, though. Do they react to collisions? And can they be “killed” like other particles on collision? But most probably you want something more dramatic when one of those it hit by an arrow … :yes:

Jose - That game looks interesting, thank you for the link! I actually don’t play Minecraft very much- strangely enough, I get more enjoyment out of watching groups of people playing a game and interacting with each other. That’s probably partly due to not being able to play multiplayer with people.

minoribus - The boids do react to collision objects (for instance, going around the blocks in the above render, instead of through them). I don’t know if they can be enabled to avoid self collisions, but I don’t notice too many of those anyway. They can also be configured to follow or chase after an object. I may use a combination of boids and “follow curve” walk animations, depending on the shot.

For a few zombies, the main character will be using a normal bow to fire arrows. But quickly after, he will receive a special bow which is capable of firing hundreds of arrows per minute, similar to what is seen in this video (around the 7 minute mark):

[video]https://youtu.be/4fxjIYu95O8?t=7m1s[/video]

So clearly, some way of automating the process would be desirable, but it may just need to be done manually. I’d like the zombie bodies to go limp, like a ragdoll would, when hit with the arrows. I’d also like to have purple or green blood splatter out from the hits. Maybe I’ll be able to use dynamic paint to help with the process. Aside from that one action (shoot arrow, zombies go from walking animation to rigid body simulation and blood splatters), the rest of the video is all stuff I already know how to do.

Edit: Oh, and I want the arrows to stick into whatever they hit, and stay there. Walls, zombies, whatever they hit. I think that’s part of the fun of watching this. It’s funny, I take it for granted whenever stuff like this happens in video games. But when I have to recreate it in Blender, where I have TOTAL CONTROL OVER EVERYTHING, it seems so daunting. Maybe breaking it down into separate pieces will make it seem less scary.

I decided to start by focusing on getting arrows to fire and stick in whatever they hit. At first I stumbled around a bit with stickiness, damping and a few other physics settings. I could get the arrows to slow down a whole lot, but they still slid ever so slightly after striking a collision object. However, I found a much simpler solution; just click “die on impact” for the particles, but make sure to enable the checkbox to render dead particles. Now they simply stop moving as soon as they hit the collision object, but can still be seen. I haven’t tried it on a moving object yet, but hopefully this method will still work.


It turns out animated meshes were a problem- arrows just stuck in the air wherever they hit; not altogether surprising. But at least it will work for static objects (walls, ground, etc).

I’ve now shifted focus to creating rigid body zombies. Here’s a short clip from working earlier today.

[video]https://youtu.be/gNq6mromwxk[/video]

I’ve decided to change the format of the video substantially. The basic plot will stay the same, but other aspects are now different. I’ve decided to shoot the video live action, with myself as the main character, and to film it in a more conventional style rather than purely POV in a single take.

My reason for changing the format is because I realized I was becoming apprehensive about the work, and found myself beginning to procrastinate instead of working. I think it was due to how big the video was becoming, and how long it would take to render. Speaking to Jose earlier in the week, we were talking about the same thing, and it really stuck in my head- if I want to become really good at making videos, I need to learn how to work quickly and complete projects efficiently.

Along similar lines, Brandon Laatsch (formerly the other half of the Freddie W channel) was commenting on an older video, and he said something that I found noteworthy. He said, “We didn’t build up to the number 5 channel on YouTube by making tons of work for ourselves. We did it by figuring out ways to not make work for ourselves, and figure out how to get things done in a way that was effective and efficient.”

And this is something that I think I need to begin to keep in mind more often now. If I want to reach 1,000,000 subscribers in the next 3 or 4 years, I need to learn to find a balance between my “ideal” version of the film and what is practical and efficient.

Besides that, I think it will also help the video if it features myself as a character, instead of just a faceless, nameless body in a fully CG POV video. Indeed, I think including myself in most of our future videos (whether a small role or a big one) will help me, as it will give a consistent face to the channel. I think people are more likely to subscribe, and more likely to watch our videos, if they recognize a familiar face. I think it’s just a psychological thing.

A bit of a ramble there, but more actual WiP stuff coming soon. I’m feeling much better about the new changes to the video.

Ha, ha, ha, there is a “last man standing” amongst your zombies :slight_smile: Nicely done. But too bad to hear about the problems with the arrows.

Since you wrote about your plans with the channel something came to my mind. Aren’t all the big YT channels built around one central topic in one or the other way? Would you agree with that? You know more about this stuff than I do, for sure. So I’d be interested to hear your opinion :slight_smile:

Yes, I think that’s true, minoribus. Having a central theme or topic that your videos focus on helps entice people to subscribe, and indeed to stay subscribed. I’ve seen channels with hundreds of videos, but with so many different topics that don’t really fall into any particular categories. It becomes very hard to filter out what you’re interested in from what you’re not, and indeed there may only be one or two videos on that channel that even appeal to you. I think, generally, the more specific your content is, the higher your chance of becoming known for that type of content.

There is a channel where every video is simply a man smoking a pipe (I think each pipe is different). It sounds like the most boring type of content, and indeed if you’ve watched one video, you’ve watched them all. But for the people who are interested in that sort of thing, he’s known by them all!

Here is his channel: https://www.youtube.com/user/SMOKERSOFCIGARSPIPES/videos
Don’t ask me where he gets all those pipes, I have no idea! Maybe he’s in the business.

Back to the topic, though; I try to keep my content focused primarily on VFX. Moving forward (as I mentioned in my recent update video), I want to really start focusing intently on short VFX films based on computer games, instead of just VFX in general.

I think the most important factor in choosing a central theme or topic for your channel is picking something you’re truly passionate about, and probably will be for years to come. Even if it’s something obscure or niche, it’s best to go with what you yourself are interested in. I think people can tell when someone is not really passionate about their work, and may just be doing it to try to appeal to as large an audience as possible.

Another interesting bit of trivia: in the early years of YouTube, for 31 days, the most subscribed-to channel was that of Peter Oakley, AKA geriatric1927, an OAP who shared stories from his life to his audience. In November of 2006, he had 30,000 subscribers (which was was huge number at the time). That’s over 4 times as many subscribers as I currently have. I think that’s proof enough that if your content is genuine and sincere, you will find a dedicated audience (eventually at least).

Those are my thoughts on the subject of channel content!

Thank you for your detailed reply, James. I knew you have more experience in that area :wink: And I think that the subject you have chosen is able to raise a big number of subscribers, indeed. I’m not much in to games but I do see how much interest there is and how huge the number of fans is.

This is looking good, i’ll look forward to updates.

i would suggest trying the demo version of minecraft, downloading mineways (free program), build your world in minecraft and convert it over to blender with mineways. My opinion, but you get better textures and far quicker results using mine ways. Guessing this is a blender only project though? Looking good :slight_smile:

minoribus - Funny enough, I actually don’t play games very often (haven’t really got the money for it, or the time). But I do enjoy the different worlds and ideas presented in many games, so I think that’s where I’m able to get a lot of enjoyment out of thinking of how to build a short film out of them. I hope that building an audience with videos based on games will help me to one day be successful in creating my own original videos.

MonkMonk - Thank you, I’m glad you’re enjoying it so far!

sheepHD - I actually used Mineways to convert one of the maps I downloaded, it’s the city seen in the first couple of posts (I used some different textures in places). Since I decided to shoot everything live action, I’m now just using character rigs from Rymdnisse: https://rymdnisse.net/downloads/minecraft-blender-rig.html
plus the giant mutant zombie I’m creating, and a few other things. I do have more Minecraft videos I’d like to create in the future, though, so I’ll probably be using Mineways for those. It’s a great program, very easy to use!


I’ve been busy with other work most of the week, but after finishing it tonight, I spent a little more time on the chopper. This is the helicopter which airlifts the survivors out of the city. The idea is that it’s made out of Minecraft blocks. What are your thoughts; how could it be improved?


A nice addition, James. The helicopter looks much like what I know from Minecraft, although I’m not an expert. I like it. The rotor blades could perhaps be more of the style of Minecraft’s diamond swords?

That’s a funny idea, minoribus! It made me laugh. I’ll try that out tonight!

I went ahead and added swords to the chopper. It looked a little funny, so I trimmed down the handles a bit. I might still adjust them a little, but already I think it looks better than the realistic blades did. I also added the name of the chopper from Atom Zombie Smasher- it’s called the Pleasant Pheasant. I’m not sure what else I should add to it to complete the design. I suppose it should have a door of some kind to enter it. Any suggestions?


I’ve not yet decided on the final colors. I think part of that will depend on what the actual footage looks like, so I may let that go for now.


And finally, been trying out the motion blur to see more of the full effect. I think once sound effects are added it will come to life a little more.


I’ve started working on the walk cycles of the zombies and testificates. It’s my first time creating a walk cycle. It’s probably a good thing I’m starting with the zombies- it’s looking a bit awkward so far!

I favor the yellow version. Just curious as to why there is such lens distortion on the buildings in the last render in #16

The new blades are cool. And yes, the helicopter needs a door to let people enter it. But in addition to that this type normally has a tailgate. And I would also favor the yellow version. White looks a bit like UN and the brown one is more like a military one. The light blue color is just something I would not expect on such a workhorse.

I’m still leaning toward the yellow version as well. To save time, I think I’ll just have the civilians “phase” into the helicopter, so just fade opacity to zero as they get close to it. I might still add a door and some details to fill out the look of the helicopter.

I didn’t notice much distortion on the buildings, but maybe that’s because I use that HDR map for testing so often that I’m used to it. It is just the raw environment map, not a proper backplate, so maybe that’s what the distortion is from.

I’ve designed a work schedule to try to improve my rate of uploads, and if I stick to it, the film should be completed in two weeks. I’m filming on Sunday, and the rest of this week is devoted to final testing and preparing all production assets (props, recording gear, etc).

I’ve designed a work schedule to try to improve my rate of uploads

I remember that you also tried several tools to speed up the workflow of asset creation, like substance painter and so on. If your schedule allows - and really only in that case - I’d be interested to hear something about your experiences. I myself am using 3D Coat more and more for texture painting and in the upcoming version 4.5 they will have PBR materials integrated. The feature previews on YT look really promising.

Just a short note because you might find that interesting. Finally that whole PBR story is coming from the game industry and if at some time you take up a next generation game as your subject you eventually must copy their PBR style.