Advertisement

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

Entoforms: Blender grown creepies

35

Artist Dolf Veenvliet (of Durian fame and countless other Blender projects) is using Blender to 'grow' a series of creatures using some sort of DNA in Blender. He's printing them with Shapeways so he can turn them into a real entomology-style collection.

Dolf writes:

Hello everyone. Bart asked me to tell you a little something about this project I've been working on. It started with a strange idea about two months ago. I'm really not entirely sure where exactly the idea came from. It comes down to this. I am creating "Entoforms" in Blender. Entoforms are little creatures that I am "growing". And not randomly... they all have their own "dna string". The name is based on "Entomology" which is the study of insects. I am printing the Entoforms in 3D (with the help of http://www.shapeways.com), and presenting them as Entomologists keep their subjects.

I spent most of January writing Blender 2.5 addons that allow me to grow shapes. The cool thing about blender 2.5's python API is that we can now use python to do things that you'd normally do manually. In the old version we always modified the "data" found in Blender files.. but we couldn't manipulate meshes as if we were modelling. So I ended up with a script that could extrude/translate/scale/rotate multiple times all in one go. Which is very powerfull, you need to know much less about the data structure in Blender to make cool stuff happen. Thanks Campbell for making the API awesome! Even though it needs a bit more work.

About a month ago I was invited to visit Naturalis in Leiden. Most people in the Netherlands know it as a natural museum... but it's actually mostly a research centre. And I was given an incredibly in depth look into the work the biologists are doing there. It's very very rare for artists (or anyone who isn't doing serious biological research) to be invited there, so I am completely honoured to be allowed to have a look (I was seriously surprised they said yes). They showed me exactly how Entomologists "pin" insects, and why. There's lots of details to pay attention to. Without them I would have never known to use "wood free" paper to label my shapes with, or where exactly to pin the Entoforms. Getting details like this right is really crucial to what I'm trying to do. Somehow it's also really cool to know that they use roman numerals for the months when they note the date (because the day/month order is reversed in some parts of the world).

At this point... I've just received my first couple of prints. And well... I'm incredibly pleased. They turned out extremely nice. I struggled quite a bit to actually pin them. The plastic of the print is really hard! This afternoon I finally found drills small enough to finish pinning the first Entoform, and a pen fine enough to write the labels with.

The project is nowhere near finished yet... I have a whole bunch of improvements I want to make to the scripts. Currently I can just about grow a leg with a knee, but I really want to add mouth's, and eyes, and... well... lots of stuff!!! I just have to figure out how to do so in python ;)

Tomorrow I'm going to an art gallery in the Hague to show them the first Entoform. They have some prints of previous works of mine in their collection and I'm curious to hear what they think about this new crazy thing I've been spending all my time on.

Also... yes of course... I will release all the scripts at some point... but right now they're a bit too "unstable", and don't even work with a "current" version of Blender.

Follow Dolf's progress on his Entoforms blog.

About the Author

Avatar image for Bart Veldhuizen
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.

35 Comments

  1. Great work ! Very interesting concept and amazing results ! Congratuations. It would be nice to see a complete collection of these alien insects exposed in a museum !

  2. Suggestion : add also DNA information for colors that can also change and evolve (turing patterns for example)

    You are on your way to have a pretty nice insectoid collection.

    @Shapeways : wonder if you could print an insect so small that a magnifying glass would become mandatory to watch the specimen...

    @Shapeways (bis) : now that would be an interesting idea for a special event "build the smallest printable object"

  3. I work in the Information Systems Health Care field here in the US. I've read several articles about the possibilities of using 3d printing to grow organic structures for implantation where the donor cells are coming from the patient to regrow their own. Using the patient's own cells significantly decreases or eliminates the chances of rejection.

  4. Great. Wonderful. How about something useful :D

    Just joking.

    Partially.

    I would love to see some l-system in Blender for trees, bushes, foliage... As cool as entoforms look, are as usefull as wings in boat...

  5. Very cool!

    I was wondering Dolf - you say they are not being grown randomly but each have their own "DNA string"

    (1.) Does this mean you change certain parameters within the script for each one, or that you change pieces of a separate DNA string that the script consequently interprets? Both would function the same way I guess.

    (2.) Are the DNA strings generated in a random manner (or likewise by an algorithm) or do you experiment and custom design each one, essentially modeling all aspects of each bug via python?

    (3.) Would each DNA string produce exactly the same bug every time?

    Anyhow the bugs look great.

  6. I'd love some more info on the script behind this, are you at any point (if you are reading this) planning to release this to the public? I have the feeling that some pretty cool stuff could originate from this project besides the cool stuff that it's already delivering.

  7. Hi everyone! Thank you for all your encouraging words. I've been getting a lot of responses to this project, also outside the Blender community.

    In reply to some of what you ask/mention here...

    aws357: yes I'm thinking about colours, but sadly right now you can only print colour in sandstone... and sandstone doesn't allow for much detail... and to be honest... not enough detail for me.

    samiboy: I am not coding this as a feature in blender, but as a collection of addons (so no build needed just a bunch of python scripts)

    twitcher: There used to be an "l-system" python script for blender 2.49... Google is your friend here.

    snowglobe: Well, some turn out to be quite odd... and a few just show me bugs, but all in all... I'm pretty chuffed with the results.

    Robbie Losee: 1. Currently when creating multiple entoforms, I change 1 letter in the dna string every time I generate one. 2. I could "model" the strings, but then there's little point in using them at all... then they'd just be "settings"... so yeah I put in specific texts. 3. Yes... there is no randomness at all, it is completely predictable

    Heke: Sorry nope... no randomness at all. Random is really quite crap... because well... it's pretty impossible to find and fix bugs properly when you use random generation, and... computers don't do proper random anyway.

    Seryan: Yes I will certainly release the scripts at some point. Though I doubt the actual "creature" script will find much use... I have a bunch of addons that I think people will find useful though.

    Filip: Thanks, I'll keep it in mind in case I ruin my current set ;)

    Thanks everyone... I'm making some progress... I've almost got the script generating 'eyes' now! ;)

  8. Wow, great work! After reading the title of this article and seeing the first picture I could only think of one thing: "Why didn't I think of that?! That's frickin' brilliant!". Of course I didn't know about the DNA-based growing then yet, which only adds extra awesome-points.
    Amazing work Dolf, both from an artistic and programmer's point of view! =)

    -Patrick

  9. This looks neat. I'm interested in possible l-systems as well. I've found the scripts often create overly complicated geometry for believability, or on lower settings (for background perhaps) they really do just look like crap, i wich there was a 'fake it' mode for the low res versions.

    I was interested at the possible 'DNA' being talked about. I'm assuming it's a bit like flipping a switch on a panel of switches and ANDing all the results together. Is the generation of the model 'instant' from viewer perspective (generates new model wholly out of nothing) or does it start with a sphere/cube/dodecahedron and evolve over time, morphing and what not?

    The only thing i can say is, i don't find the shapes particularly pleasing, aesthetically or functionally. I kind of jump the boat and imagine live moving things, but these have more the shape of a flower or virus, stationary objects for the most part. I see all the sensors,antennae,flagella whatever those tentacle protrusions are supposed to be, and it seems like overkill. Overspecialized to the point where it's bred itself out of certain environments. Just difficult to imagine a moving entoform with those unwieldy appendages.

    Bravo for fabulous work, nonetheless, looking forward to reading over the script and making my own DNA controlled lifeforms. And this way i won't have to kill something just for a specimen box :)

  10. The Masked Lurker on

    Fascinating! The idea of DNA in this case isn't anything like ElectricSheep is it? that would be an awesome project! you could use a similar setup with a database of traits that get mixed with other traits (mating) and display the results over a network of users with a client that allows for the offspring to be killed or allowed to live and bread it's traits with others. Kind of like selective breeding. I'm a little too tired to get the idea out effectively but look at Electric Sheep ( http://electricsheep.org/ ) for what I'm talking about, it basically does the same thing with fractals and displays them as a screen saver, great stuff! The thing I like about the idea of applying it to your concept is that we could actually have community guided meshes that evolve. We could save them and trade them. Don't think there would be much practical use for it but it could become something like trading baseball cards or something.

  11. Very beautiful and interesting work, Dolf. Congratulations!!

    Of course, we are now looking forward to the scripts: we are a biology lab and do protein work,
    I suspect that something of the kind of entoforms could be used for cellular settings: the way oligosaccharide (complex sugars) chains are added to proteins and lipids, the growth of cellular appendices (in neurons and dendritic cells, for example.)

    Also, pushing on the biology side: DNA in live things is present in two copies, almost identical. In the almost sits the 'randomness' of real life forms: all mosquitoes are almost identical, but in the details they are all different (true for humans as well).
    With a double DNA it would be possible to create an evolving system, in which creatures are generated with a very little amount of random input, and selected according to some feature that the environment (in this case you) decide.

    Cool Cool cool
    can't wait to have the scripts!
    Monica

  12. Impressive!

    I can't help but make several links:

    1) D'Arcy Wentworth Thompson's most famous book "On Growth and Form", notably its chapter XVII, "The Comparison of Related Forms," in which he explored the degree to which differences in the forms of related animals could be described by means of relatively simple mathematical transformations (http://goo.gl/wbJRZ Look for its illustrations on the web).

    2) A stop motion animation I’ve seen in the early ’90s on TV, in which the artist successively photographed/filmed each bone of a rabbit’s spine (individual vertebra) in their anatomic order (according to their level in the vertebral column), starting from the top of the neck (atlas) to the end of the “tailbone” (coccyx). Each frame would display one single vertebra, which would be replaced by its closest neighbour in the next frame, oriented exactly in he same position as the previous one relative to the camera. The speed of the animation would prevent the spectator to detect the substitution. As a result, the spectator would think it was the very same bone that was gradually changing shape (as smoothly as in a morphing). One would then realize that the “design” of all vertebrae obeyed the same biological equation, and that, despite their different sizes and shapes, they were only slight variations on the same theme.

    3) Self-organization in nature and cellular automaton which can explain the color pattern of some seashells (such as “conus textile” a.k.a. “Cloth of gold cone”), the leopards’ spots, the zebras’ stripes, etc. in terms of diffusion of pigments and their concentration: each shell or skin cell secretes pigments according to the activating and inhibiting activity of its neighbour pigment cells, resulting in spots, stripes, etc. (cf. aws357’s comment)

    Hope this sparks even more ideas in relation with this great project.
    I'm pretty sure biologists could benefit from this artistic work!

  13. GentlemanGerman on

    This is quite interesting. I would like to say that from an artists perspective I wish there was more detail to these critters. Perhaps purely using a script is not the way to go. I can understand the necessity and the ease of modeling the basic form with a script, but I think once you have the basic shape of the Entoform that sculpting detail into it (segmented body lines, exoskeleton pores, wing details, eyes, etc.) would accel these to proper alien like life forms. Don't forget that the shapeways prints can always be painted as well. I would love to see one of these pinned up with full sculpted detail and a convincing paint job, as a collection of realistic looking non-existant critters.

    Great job all the way around so far though. This is the thing I love about Blender and it's community is the ability to create in new ways and the creativity that comes from fellow artists all around.

  14. Any plans to include armatures and motion in these critters? It would be awesome to see if you can evolve things that are able to move in different ways - swim, climb, jump, pounce.

    I really love these things. Keep them coming.

Leave A Reply

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

Advertisement

×