Wednesday, September 16, 2009

dElitist Software

Screen shot 2009-09-16 at 10.59.47 PM
Hypershot 2010 in fullscreen mode with some quick SmoothTeddy models

Today, one of our students was talking about the future in Thesis class. He was contemplating a body of work that embraces the rapid changes in the world; but was unsure about taking on an optimistic point of view because he really didn't believe it. Well, in at least one respect, I am looking forward to the future. Not a future of increasingly sophisticated and complicated systems, where only those of us who are willing and able to memorize and control complicated workflows can create. I am looking forward to the generations of software and hardware that we can understand by playing with them, where there are results between technical success and failure. Naked edges in NURBS, incompatible filetypes, out of gamut colors, drivers that won't install...I think this stuff is dying. Designers are increasingly capable of giving us studio tools that barely even need to be taught as technical subjects in college courses. What's my evidence? For now, Hypershot.

When I started 3D modeling, less than 10 years ago, the modeling process was complicated enough - generating anything approaching photo-realistic renderings was pretty much rocket science to me. 3D Studio Max has about 100 dialog boxes with sub boxes within sub boxes - it's super powerful and too difficult to learn and has way more features than a designer would EVER use. Flamingo was awesome and fairly straight forward - but was still more difficult to setup than a photography studio to take pictures of physical work. When an auto designer at GM introduced me to Hypershot my first words were "this is how rendering should work." I took existing data that I had created with significant labor, and used Hypershot's virtual camera to take pictures of it in views and materials that were realistic and compelling. This software DSLR is faster, has more flexibility, and can shoot in dynamic points of view that my Olympus Evolt 500 could get into only if I fitted it with a robotic tripod arm and a lens with infinite depth of field. And I was realistically rendering within minutes of getting the model into rough shape. I could confidently 'digitally prototype' my design well before the piece was "watertight;" making decisions about whether or not something was right before I even finished the form, before I hit "print" on any of our 3D printers, before I went through the 20 hours of finishing one of my castings.

The new render queue in Hypershot '10, visual, simple, it persists even if you have to close it out, crashes etc.

The key advantages, to me, of Hypershot include: it is always rendering in real time. As soon as you move your viewport, it starts casting shadows, reflections, etc. There are NO LIGHTS. You get dynamic light from an environment map which not only illuminates, but provides the reflections to accentuate your form (and you switch between them at will to change the mood of the image). There is no visible horizon to worry about accidentally seeing in the frame. If you are looking, like me, for some kind of neutral infinity background, it uses any .jpg or .tiff image as a 'backplate' behind your rendering to cover the environment/ground behind the model. And there are few complicated choices to make in the selection of materials and output etc. You drag and drop, slide qualities up and down, choose between few standard formats for the image.

Recently I started beta testing the '10 version of Hypershot. And its ease of use, professional power, visual simplicity, and available customization is an example I wish more companies would follow. It has gotten more sophisticated, there are more options. But they were designed in such a way that they are fairly intuitive, they don't obscure the process of image making - they clearly facilitate it. When I demo this software for my students tomorrow, it will be with an awareness that my traditional role as vessel of technical knowledge is fast becoming obsolete. I am looking forward to this future - when I can concentrate on helping them with their bigger problems; like becoming effective designers - not mechanics. My role as chief know it all is all but over, right when I was about to run out of steam staying two steps ahead of them on the latest version of software...

Hypershot '10 is the '64 Mustang, Chester Carlson's photocopier, the first Macintosh. One of those things you didn't know you needed or wanted until someone put this beautiful, simple, game changing widget in front of you and said "hello," I love you. Now, if we could just work on getting the price down, you could all see what I mean.

My experiments with Hypershot (2)010 so far can be found here. Many more images from my students and I using Hypershot 1.5-1.9 can be found here.

Thursday, August 20, 2009


When I started exploring fluid modeling a lot of things excited me about it; some of which I've been following up on, and many more which will take me years of work to understand. It was wet, shapely, ephemeral, controllable and chaotic, and deeper than any set of technologies I'd yet plunged into.

I have been playing with the "fill object" emitter within Realflow since I began learning the software - giving my scenes bodies of fluid into which I can drip objects or other emitters in order to generate dynamic "spatter." But it can be used on an imported element as well; in my case one of the very fluid orphans I've been creating over the past few months. That filled object can then be under the influence of various forces and other objects, and can change over time - degenerating new primary and secondary elements. I can take an otherwise and relatively smooth object and give it bubos, force those nodes to break off or homogenise into a puddle. This is my first thought (flickr images)...

(some thumbnails from flickr)

I'm not really ready to go full Terminator 2 on it; but it does seem like a reasonable way to start layering all these data sets together into more "maniacal" forms. Whether or not they'll have legs...who knows?

Prototypes being rendered on a Dimension 1200SST 3D printer

Monday, August 17, 2009


P1013749 1

I've been spending a lot of time with the thousands of fluid meshes I've orphaned over the past few months. And I've finally gotten around to one of the first uses for which they were created; composing and constructing objects from them without having predesigned the object in advance.

I am going to be a part of a show called "Re/Thinking Design for Consumption" at the Scarab Club in Detroit next month - and I wanted to be able to show some new jewelry beyond my spatter pins.

Over the past few days I turned a number of the elements that have been rendered in red and pink ABS by our FDM machine into a new necklace called Beam. The cord is black silicone. No clasp.

More images of this piece can be found here.

It can be seen in person (along with other recent work) Sept 2 - October 18th at
Re/Thinking Design for Consumption
curated by Christine Bossler and Erica Bartels
The Scarab Club
217 Farnsworth St
Detroit, MI 48202-4018
(313) 831-1250

Thursday, August 13, 2009

lower case

On a couple of occasions I have been fortunate to have been included in some of the marketing literature for one of the creative products I use.

A year or so ago Hypershot featured my Contralat brooch in some of their mailings about the Rhino plugin they use; you can see it here.

My flickr images about this work is here.

Today I got an email that my Wake hairbush process video is being featured on the website for the NextEngine 3D scanner I use.
See it here.
The page about my work is here.

My flickr gallery about this work is here.

Thanks Dan and crew at NextEngine!

New: Matt Sederberg just posted my Flume flatware on the website. Thanks!

Saturday, August 8, 2009

Playing with Fire

I am often feeling a little 'out of step' at a barbecue. I don't drink, I am not super into classic rock, even when wonderfully played by talented musicians, I don't smoke, and I am a vegan. I am the guy lighting straws on fire or scribbling on the tablecloth waiting for something interesting to happen. Maybe reading a book waiting for a concert to start. Could even be trying to start conversations about the intricate details of someone's vocation when they just want to bullshit.

I recently gave a short lecture on my work at SIGGRAPH - and someone in my small audience asked me if I thought my current work would be satisfying for very long. Maybe it will, maybe it won't, but there have to be other things in the mix right?

It rained like hell today, and we were pretty sure this mega party in and at the Mecca of backyard BBQs (Clarkston, MI) would be a bust. I swung some kids around by their arms, got tired and dizzy, and the dog was locked up and unable to come out and I figured I would start the research for my next piece in a small bonfire that started about 9 o'clock. This is research draft 1 (set to the hauntingly beautiful music of Paolo Nutini - "Candy")

Thank you so much Shelly!