You're blocking ads, which pay for BlenderNation. Read about other ways to support us.



An animation produced in four weeks by eight inexperienced but dedicated undergraduate students across the United States and Europe, collaborating over the Internet using Blender.

The making of:

Federico Burch writes:

A team of 8 dedicated students from the Department of Visualization at Texas A&M University collaborated on an animation over this past winter break with Blender as their software of choice. Their goal was to coordinate their artistic and technical efforts in a professional manner, while also working through a traditional production process.

They created concept sketches, animatics, matte paintings, color scripts, lighting keys, texture maps, character models, rigs, and particle simulation. The project served as a unique challenge in that each of them sat in different parts of the U.S. and, in one case, across the ocean in Italy, visiting family for the holidays.

Using their college student budget of exactly zero dollars, they employed every free Internet service they could summon to assist them in communicating information and sharing files. At one point, Federico Burch was forced to use Wi-fi from a closed airport shop in Italy in order to maintain an Internet connection to dropbox. At the same time, Wes Cardwell found himself sharing the tiny bandwidth of a 3G AT&T aircard with his dad to upload files, while Jose Contreras worked from his laptop on a sixteen-hour drive to Minnesota to visit family.

The animation was almost entirely created using Blender which they found to be a great fit for their project due to it's wide base of support on the internet and affordability. Challenges that professionals are familiar with tested them as novel concepts. Effects failed, rigs broke without warning, unexplainable artifacts reared their ugly heads, and all of it had to be fixed on the fly to meet their established goal of finishing before next semester.

Despite all of these obstacles, they persevered and submitted their animation to SIGGRAPH where it was chosen to be featured in the “SIGGRAPH Dailies!” along side other students and industry professionals from all over the nation.

Out of Breath Productions (OBP)

The people involved were
Federico Burch,
Wes Cardwell,
Jose Contreras,
Caleb Kicklighter,
Kevan Loney,
Anne Lynch,
Justin Sheehy,
James Storey, and
Thomas Storey

About Author

Bart Veldhuizen

I have a LONG history with Blender - I wrote some of the earliest Blender tutorials, worked for Not a Number and helped run the crowdfunding campaign that open sourced Blender (the first one on the internet!). I founded BlenderNation in 2006 and have been editing it every single day since then ;-) I also run the Blender Artists forum and I'm Head of Community at Sketchfab.


  1. Although that zero budget isn't true, you have computers, Internet, electricity, food, work hours... all that cost money ( or time, and time is money :P) thus should be include on the budget. If you don't have any money what so ever you can't do this :P

  2. @rogper - I'm sure they're talking 0 budgeted for this project. All you mentioned is day-to-day living expenses they would have had regardless of what they did project wise. A budget is for the project specifically. If they bought their computers FOR this project, that would be budget based. However, I'm assuming they already owned their systems.

    ON THE VIDEO: This looks great! Well done! Well done indeed!

  3. @Tadd:
    I understand your point of view because it was also mine on the past, but:
    If someone doesn''t have to care about life expenses, that someone is being sponsored :) and a sponsoring does count as budget for a short animation, I think :) Anyway I didn't meant it on a bad way, it was just a side note, sorry for that!

  4. I like it! The music sets the mood well, the timing is pretty good, the visuals and animation are well done.

    But man, another dead Blender butterfly! What kind of message are we sending as a community here?!
    (just kidding)

  5. Federico Burch on

    Thank you for the compliments, our team worked hard!
    Good point Gryphon, we were not trying to seem environmentally unfriendly.

    We are nearing the end of a second animation at the moment.
    Completely different animation but similar student backstory :}

  6. All this looks good...
    Especially the lighting, models and rigging.
    But, guys, you missed some things:
    - unrealistic jitter after hit (pls, don't rotate the cam on y-axis)
    - trails after "wheels" (the surface looks dusty)
    - particles for dust (really need some dust here!)
    - is the butterfly a robot too? :)

    p.s. the scenario looks quite... "short".

  7. Justin Sheehy on

    "Camera rotation on the y-axis": Do you mean local y (world z) or world y in Blender? I'm pretty sure I only played with f-curves for x, but there was a bug (in files created before 2.54 if I remember correctly) that caused noise modifiers on curves to persist even if the modifier itself was deleted, so it's completely possible it persisted and was overlooked. I'm very interested in the theory behind it though, whichever axis it was that we shook that we should not have =].

    "[Lack of] trails after wheels": *double face palm* =p. That is the one thing the vast majority of people notice. By the time someone (outside the group) brought that up though we were already far past the modeling and texturing phases, so to keep with our project goals we couldn't afford to go back. Keyframing the texture and bump mapping would have been too time consuming for something we had never done before, and the unwrapped mesh wasn't really dense enough to add realistic blend shapes either (which would still need to be textured).

    Particles for dust were deemed too expensive, as the particles for grass and smoke were already terribly taxing. Non of us could really have been considered any more than beginners when we decided to start this, so while we had solutions for many problems, a lot of them (like particles) were not terribly optimized or fully understood =].

    And as far as it being "short", we had a very "short" time budget as well (roughly 4 weeks, not including rendering) XD.

    @rogper: I suppose our parents sponsored us then =]. After all, we were all just college sophomores.

    Thanks everyone for the views and the compliments, we appreciate the encouragement!

  8. @Justin:
    If speak simply - just hit "R" and rotate your cam in the Camera view. It's like that. I didn't completed my earthquake animation (unfortunatelly) and there I was stuck on the same. But after some time I did understood that it must be the very center point on the cam's stand and some special "angular" vibrations to be looking like this. So.. only "the jumping" is good (if no special conditions).

    about "trails": hmm... I thought that you used procedural textures there. then maybe it should be easier to handle. You know... I could make this in such a way:
    camera projection for matte, matte painting (just good mixes and blends of your texture using any painting program) for the eroded ground place (probably this can be made with two or even one projection) then just another projection for trails (overlaying the texture)... then deforming of this mesh will be looking "smooth" and nobody will get that something's tricky is here :)
    then you will not need to "keyframe the texture" (sounds strange for me here) Commonly it should be done with keyframing the blend texture (that will be revealing the texture for your trails)

    hmfff) I need to make a little project on this theme, I think! My explanations aren't best.

    wow, four weeks! it's fast then! :) BTW - did you used Renderweb for rendering? I found it's very fast (and free).

  9. Justin Sheehy on

    Thinking about it, that makes a lot of sense that it all depends on how the camera was set up and the conditions for the shake. It probably is out of place, just with how stable the camera is in all the other shots.

    Ah, good point on the trails. By "keyframing the texture", I meant pretty much the method you described, (keyframing the alpha blend to reveal the texture little by little). That might look believable enough by itself. Yeah, for continuous tracks that probably would work. I think @Federico brought that method up actually when we noticed it, but there just wasn't time left to do it. If i remember correctly, that was near the end of week 3 or the start of week 4 that someone pointed it out to us haha.

    I think your explanations are fine, they at least make sense to me =].

    Nah, we used just whatever computers we could get locally. Our school had has a small lab with (21 i think) Macs in it that we set up a Distributed Render in (very inefficient by the way, but it was all we had), and so we just used those computers in their down time. That was before we even knew we were going to submit to the Dailies, we were just having fun and trying to learn at that point! I've heard of, never of Renderweb though. I will never go back to Distributed Rendering if I don't have to, Network Renders are so incredibly much more efficient and less trouble to set up. I'll give Renderweb a look!

  10. @Justin:
    yeah, cool :) I'm glad that my methods have sense and my explanations are readable.
    21 PCs for a small render farm is more than enough!

  11. Justin Sheehy on

    @Moolah: 21 sure seems like a good amount, but there wasn't an actual renderfarm set up on them, and we couldn't install new software on them either. So we had to manually set up distributed renders and move the blend file around to every computer whenever the smallest thing changed, recalculate the frames, estimate the time to complete again, over and over. It was over 200MB of data, so it got really tedious really fast, and we had to be careful to balance our efforts so the lab of computers didn't go from being an advantage to a burden (which it was at times to be honest). Fortunately though, our group is playing a small part in helping to get that lab turned into an actual renderfarm, and hopefully before we want to render our current project!

  12. Well done, guys. I hope Blender Foundation has some advanced courses to train talented students like you who need just a "little refinement" with your work.

  13. @Justin:

    If you're not worrying to much about your copyright then using RenderWeb and all your computers will be the ideal scheme! There you can set up of how much CPU to use, choose any project you want to render first... Just one thing (and it can be stopping from rendering difficult scenes) - you can't render smoke and water sims (till RW guys find a way to store baked data somewhere without need to upload it from their server to every slave-PC).

Leave A Reply

To add a profile picture to your message, register your email address with To protect your email address, create an account on BlenderNation and log in when posting a message.