Hey gang;

Well, I’m slipping in my last-minute entry. I’m really not pleased at all how it turned out but in the spirit of the competition, I thought it would only be fair to post it.

I’ve been following the FXWars forum for several years now, always being tempted to enter something. However, due to the workload of the full-time job, that hasn’t happened until now.

This theme sounded especially cool so I really tried to set aside some time to work on this. Instead, I manged to have a total of about 6 full days to get this done out of the very generous 2 months Roberto provided.

My main interest in entering the FXWars competition was to try different techniques for improving my realistic rendering and effects. As a result, I used a lot of camera mapping and projection methods in this sequence, along with a lot of camera optical effects. Mainly to try and take some of that digital edge off.

Anyway, it was still a good learning experience and I’ll definitely be trying more in the futre. Kudos to Roberto for having done such a stand-up job with this forum, and to all the wicked submissions that made it. Below I’ll post all my work-in-progress for this sequence.



I found the original models on TurboSquid, which is one of the first places I generally look. Since my main interest in this competition is lighting, rendering and compositing, I wasn’t interested in modeling any assets. Add to the fact that I suck at modeling and TurboSquid immediately sounds even more appealing.

Like many free models, these had a lot of problems. They were way too dense, didn’t meshsmooth properly, and had numerous flipped normals. Some time was needed to fix these up to a semi-decent state.

Here’s the models I ended up with after some modeling and texture touch-ups. Much more was done after this render as well.



The animatic, of course, is one of the most important aspects of any project. I spent most of my time working on the timing and animation of the shots. The camera work is very important as well but it didn’t turn out as good as I’d hoped. The “Pearl Harbour” movie is a great reference. There’s definitely much more room for improvement but given my short schedule, it’s ok.

The intention was also to have a fourth shot in which the plane crashes to the ground in an epic explosion. Also, the first shot was actually a lot longer than what you see here but I edited it down because I felt it dragged on too long and therefore killed some of the energy.

->FXWars Animatic<-



The rendering was all done on my home computer, a Q6600 without many bells and whistles. I didn’t use my company’s renderfarm because I also wanted to manage my rendertimes and keep them reasonable. The longest rendertimes averaged about an hour at 1280x720 rez, but most of the rest averaged 15 minutes / frame.

Everything was rendered with VRay using global illumination and HDRI lighting with raytraced motion blur.

Generally, I render many, many layers for lots of control in my comps. However, I tried to keep it to around 4-5 for each element in this project to save time. Here’s a sample of some layer breakdowns:

Airplane Diffuse Pass

Airplane Ambient Occlusion Pass

Airplane Muzzle Flash Pass

Airplane Bullets Pass

Background Pass (with projection mapping)



The compositing stage is the most interesting stage for me and the main reason why I entered the competition. Whereas as I used to always rely on 70% rendered 3D footage and 30% comp for my final result, it’s been the other way around now for a long time. I find it’s much more efficient and controlling to add smaller elements into the comp from pre-existing photographs than to model, texture, animate, light and render them. Quite often, I may not know exactly where or how I want the element to sit in the comp so a 2D element gives me full control over placement, scale, rotation, etc.

A good example of this is for the third scene in which the planes are whipping past the background. The foreground has really fast passing boulders in front of the planesthat are all done in comp - mainly for the timing reason: so that I could control how fast, how big and how dark they would be.

Integrating the digital elements into live actions plates (planes into bg’s) is always the toughest part. It’s critical to first and foremost match the whites, the blacks and the gamma. Then you can move on to color balancing and grading.

The other huge part of compositing is what I call the optics pass. These are all the optical effects and phenomena that are inherent in live-action film such as film grain, color grading, diffusion, vignetting, lens flares, etc. A good optics pass can do wonders in taking some of the CG edge off renders.

The color grading stage is super important and, if done correctly, can make your sequence feel cinematically epic.

I do my compositing in After Effects because I like the software. The raw renders in the previous post have clearly undergone a ton of color treatment.

Here’s a few final frames from the sequence:




One thing that I’ve noticed about this sequence is that it’s slightly overexposed. This is not the first time it’s happened to me and I really gotta keep an eye out for it next time.

Here’s the links to the final sequence:




Like I said in my first post, I’m not pleased how this project turned out. It wasn’t a matter of not having enough time, but rather not managing my spare time efficiently so as to devote enough effort to this sequence. 6 days is definitely not enough for this and pissing away the 2 months Roberto provided is just sad. Here’s a few things that bother me at this time:

- Most shots (especially the first) lack plane integration with the bg. They need atmospheric and volumetric passes. Currently, you have no sense that those planes are disturbing these environments in any way.

- The planes are a bit flat and require a separate highlight pass. I always do this but I thought I could get away with it this time by trying to render it all in the diffuse layer. What I originally had in mind was to have sunlight glints catch the wings as they rotate in the air.

- The muzzle flashes and bullet passes are pretty weak. But they serve they’re purpose in telling the story I suppose.

- I wish I had more Afterburn experience as I would have liked to create a better explosion for the last shot. Instead, I relied entirely on stock footage that was warped, retimed and treated in the comp.

- I created the sound design in about an hour and since it has the power to enhance the visuals so much, I should have devoted more time to it.

It was a good experience and I’m looking forward to participating in more of them in the future - preferably with better time management! :wink:



Great Stuff mate I really liked it, short but sweet ;0)

Nice one.


Hey Steve;

Thanks, dude! I’ve been carefully following yours and it turned out great - the volume of work you managed to pull through is outstanding! Ironically, this is exactly the biggest problem with me and this project - I need to manage my time a lot better next time.

You going to give the next challenge (Nuke) a shot?



Hey Rich thanks for following mate. I had a few months off work so I guess that was my advantage on this challenge.

Nuke? hmm ,I might be tempted to join in lets see on the deadline & when it starts. =0)


This is really great work, I am always happy to see projection mapping pulled off properly!


that was awesome, nice projection though.


Hey guys;

Thanks for the comments! ParamountCell, nice entry - I was following your thread attentively like the others!

Tavo; thanks. Really, it’s the first shot that’s a good example of projection mapping. Everything else is just standard mapping with additional elements comped together in After Effects.

I’m looking forward to the next challenge…



I just came across an artist spotlight on the Luxology site that features a lot of cool WW2 fighter shots: http://www.luxology.com/community/profiles/wiek_luijken/-Rich


This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.