Posts about design & the design industry.

The Serpent Twins at Nantes Maker Faire!

We are very excited to have just returned from Nantes, France, where The Serpent Twins were one of the main attractions at this years Maker Faire in Nantes, France.

July 7th & 8th, 2016 Maker Faire Nantes was hosted at Les Machines de l'île, and the Serpent Twins from Oakland were one of the main attractions!


Les Machines de l’île, self explained as a "crossroads of Jules Verne’s "invented worlds", of the mechanical universe of Leonardo da Vinci, and of Nantes’ industrial history, on the exceptional site of the former shipyards".
We are so excited to have been part of this amazing group of artist fabricators and that a piece we worked on was among such amazing company.



Here is a great video from Saturday night at the Faire -- the amazing 'dance' of The Serpent Twins with Kumo, the spider....

And our front set up for The Serpent's resting spot, with signage and lanterns...


Revit to Radiance (via an expensive translator...)

So something I've done some work with over the years is a rendering engine called Radiance. Radiance is unique amoust rendering engines as it actually calculates the lighting, it doesn't 'fake' anything, so you can get some real design feedback from it. In other words, if your rendering looks bad, it means the lighting or daylighting is bad, and you've got design problems to fix. Most other rendering engines are focused on making a 'sexy' image ASAP (and rightfully so) and so you can't 'trust' the results for it's just faking everything.

However, Radiance isn't widely used because it's slower, harder, and requires you to know something about daylighting and lighting design to get good results out of it (that 'not faking it' comes back to bite you).  That and the fact that it's a very old school command-line Unix tool makes it's learning curve vertical, and if I didn't have a Unix god for a friend to pester with stupid questions I would have never learned it.

I've always felt that Radiance is kinda like the oil painting of Architectural Renderings: It's hard, and takes a long time to master, but will let you produce images no other medium can produce. Most rendering engines are more akin to watercolors, sketches, or airbrushed paintings: fast, loose, easy, and easy to make look great. Right tool for the job and all that, so Radiance isn't used much, 'cept by lighting designers, for most Architects just want a sexy image in the least amount of time possible.

Anyways, the right job came along, and I got to dust off my Radiance skills I haven't used in a while. We've got a project at my day job that's all Leeds and daylit and such. Had to get a good idea if it was really going to behave the way we designed it to, and Max alone couldn't get us there. However, seeing that we were doing the project in Revit, we had a nice, complete, clean model to work from. With a little bit of noodling I figured out how to go from Revit to Radiance with little pain (but a very expensive translator, sadly).

This is a 'falsecolor' image of the interior of the project, showing the lighting levels from the windows and skylight. The colors match that little legend in the corner, so that we now know how much natural 'free' light those tabletops are gonna get. Pretty cool, huh? (god I'm such a dork)

Here's what I do:
1. I export the 3D model from Revit as a polyface mesh to a DWG file, with elements turned on or off to filter out what I don't want in my final Radiance model.

2. I then file link that DWG into Max 8, which allows the Revit materials to remain 'intact'.

3. Revit likes to model glass as a solid plate. Radiance likes glass to be a flat plane. So in Max I select all the glass by going into the Material Library, setting it to only show the scene materials, picking the glass, and then using the very handy 'Pick Elements By Material' button on the Materials Palette. This then gives me all the glass. I then apply a 'Edit Poly' Modifier to the glass, and pick the 'Y' under 'Make Planer'. This 'flattens' all the glass, which as long as it's not horizontal (like in a skylight) fortunately for us has the y-axis pointing in the right direction to make this happen correctly.

4. I then export the model from Max as an .Obj file, with elements grouped by material, no normals, and faces set to quads. I've found this gives the cleanest mesh in Radiance so far, but I'm still tweaking with this to see if it can be better.

5. Then I use the Obj2Rad tool that comes with Radiance to convert the exported .Obj file to a Radiance file. I'm running Radiance under Cygwin using this fellow's wonderful pre-complied binaries. It's the easiest way to get it to work under Windows I've found. Someday I need to set up a 'heavy lifter' OS X or Linux rendering box I can SSH these jobs to instead, get them to run faster (Radiance runs at little better than half speed on Cygwin from what I understand)...

6. Now we are almost done. We need a .Mat Radiance material definition file to go with our .Rad file. what I do is use Awk, Grep, and Uniq to parse the .Rad file, pull out all the Material Names, sort them,  delete the duplicates, and then save them into a .Mat file. Once I get that process smooth I'll post it here, I'm still stumbling with it a little.

7. And now we're done! We just make a .Rif file, a sun definition file, and start rendering away. If we change the Revit model, we simply re-export to DWG, the change pulls into Max automatically next time we open that file, we re-export to .Obj, reconvert OBJ2RAD, and then we're ready to re-render.

Now. Here's a caveat. I'm cheap. so I'm trying to do this with the tools on hand. This company makes a version of Radiance you can use as a plug-in to Max, just as if it was vRay or some other rendering engine. I've used their demo, and it looks keen. but it's almost $2k worth of keen, so...

And someday I'd like to do this without needing Max at all. That would be swell. Gotta talk to some programmer friends of mine...

Maker Faire!

Wooo the Maker Faire is a ton of fun! Lots of great stuff here. We're also right next to Mr. Jalopy, our favorite blogger and a really funny guy. We made him this sign so that he wouldn't hate us from all the noise we're making with the Shopbot.
    So the PRTalpha benchtop Shopbot was kind enough to send out here for us to use at the show rocks. It's faster than the one we've got at home, and for someone that just wanted to make small stuff, like guitar bodies or signs or something it would be perfect. It's working quite well (once a fuse that worked loose in the shipping was found and set back into place!) and it's quite a draw. Several folks seem to be having the same little ephiany that we did when we first learned about them. You can see the gears turning in their heads, thinking of all the things they could make with one. Gotta say, Shopbot is really good at making accessible and affordable CNC gear for everyday folks.
    The difference between the one we have at home and this one we've got at the show is that this one is an 'PRTalpha' while ours is just a 'PRT'. What this means is that the one at home is always running 'open loop' while the one here at the show can switch from 'open' to 'closed' loop on the fly when it needs to. 'Open loop' is where the computer tells the motors to move the bit a certain distance, but doesn't check that it actually moved that distance. It just trusts that it did. This means when you're doing something that's really complex, or need something to be really accurate, or cutting something really thick, you need to slow it down a bit, for otherwise it can 'loose steps', i.e. not really move as far as it thinks it did and get off register. With the Alpha, it can also run in 'closed loop' where not only does the brain tell the bit to move, it then checks to make certain it did move the right amount when it's done. This means that you don't have to worry about loosing steps when doing stuff. However, it's slower to run closed loop all the time, so the alpha switches between the two automatically depending on what you're doing. So in the end, it runs fast, yet is unlikely to loose steps. Pretty cool. Best thing is that the PRT and the PRTalpha are only different in the stepper motors that drive it and the brain, so at some point in the future when we've got the money we can 'step up' and not lose steps anymore by just swapping out those items.

Design -> Revit -> Blender -> CNC -> real life!

Here's a sneek peak at a big job we're in the middle of. We're doing a lobby for a local game design company, and we're very excited about it. The overall theme of this lobby is based upon an orginal creative work (a fun PSP game and a neat comic) of this company called 'Death Jr'. It's theme is that you are Death's kid and you've got to save your dad (who's in trouble because of something you did). We loved this image of Death's desk in his study from the comic:

So we proposed to base the reception desk design off of it. We were very excited that the client liked the idea, and we designed and modeled the desk in Revit:

Which formed the majority of the desk. However, we wanted to do some very custom parts for the desk. We mocked up these wings for the front in Revit, but we wanted them to look more carved instead of flat. So I exported the wing from Revit into Blender, and then remodeled it until we liked the look:

And then we exported that model from Blender to the CNC toolpathing software, and then carved it out from a solid chunk of walnut we got from Pals down the street (certified sustainably harvested, Pals rocks). Those extra bits in the model are to save some of the wood, so we're not carving it all away:

Then we sanded and finished the wing. We're very happy in how they came out:

They were a nice touch to the overall desk.

We'll be posting a lot more photos of this when we're done, but here's a sneak peak of what it looks like now:

The bookcase returns...

So it's been a while, but I was working on a design for some bookcases for our upstairs here in the Barn, and I finally got around to modeling out one of my sketches. Here it is:

So those ribs would be cut via the Shopbot, and jointed together with some rods, bolts, or something along those lines. Maybe a keyed joint, like a puzzle bit too. This is just a design model to get some ideas, thought I'd post it here, for it's my favorite so far. However, I've got to make at least the bottom part and stand on it to be certain that it's joints can take the weight of all the books...

Big news about the MAKE Show...

So some big fun news to share! We entered into the upcoming "Maker's faire" that Make Magazine is putting on, for we love Make Magazine and figured since it's local, why not try to see if we could get into the show? We're going anyways, we might as well go and show some people some stuff too. So, amazingly, we got picked! We're going to have a table there, and show off stuff we've made, how we make it, and talk all about CNC tools for everyday folks like us.

...but it gets better! Jillian somehow got the Shopbot folks to actually send us a mini-Frank. Apparently Shopbot loves Make Magazine too, and just bought some ad space in the next issue, but wasn't going to be able to send someone to the show. But when Jillian talked to them, and told them we were going to talk about their CNC routers, they decided to ship us out a "tabletop" model Shopbot for the show. We'll put it together there, and then make stuff for folks right on the spot! How cool is that? It will be tons of fun showing people how great these robots are, meeting lots of other folks into making things, making things right there for people, and maybe even selling a lapdesk or bookshelf or something (or two).

This "tabletop" model is about a third the size of ours, and instead of the whole top moving back and forth like ours, this one moves the table back and forth as one of the axises. When Jillian asked Shopbot if this one is any, well, quieter than ours, their response was a "ha ha ha... No."

Now, while we'll be in the 'workshop' part of the show, where other people will also be making things and showing tools off and such, still... between the dust and the noise, let's just say that we're not going to be making any friends within our immediate area... But if you want to drop by, I'll be posting more info as it approaches and we've got a better idea as to where we'll be. Tickets are $12 too, so you should go, for it looks like it will be a ton of fun...

Big news... That it seems everyone knows already...

I was going to wait on this news, but it seems the cat's out of the bag, and news travels faster than 'blogs:

So I've up and decided to join Gensler as a full time employee. Next week is my first day, and I'm wrapping up all my consulting gigs.


Well, simply, because Gensler rocks. I mean, how much more do I have to say, it's like going to work for Apple or Google or Pixar. They are a great company that does great things no one else can, so why not go work for them?

Also every firm I've had the pleasure of working for or with I've always had the same worry: that their understanding of basic business concepts was poor to nonexistent. After working for a reseller and a software company, and reading a lot on my own about the subject, I came to realize just how bad most design firms are when it comes to business. And how much their lack of business skills stands in the way of what those design firms are trying to do. Gensler is one of the few firms out there that seems to really understand how to be both a good business and a good design firm at the same time, and I'd like to learn more about that.

Also this is a very interesting transitional period in the design industry. New technology is enabling new ways of working, but many design firms are unable or unwilling to take advantage of those things, or can only bite of a small chunk of it. Gensler seems committed to putting together the best team it can to really try to leverage BIM in real-world projects in the best and biggest way possible, and in ways no one else can, and that's something I want to be a part of.

Also everyone I've met there is very smart and nice. The Culture they have there went a long ways towards my decision to join up.

And finally, I never really wanted to be a consultant. It was work I could do to pay the bills while I tried to move along my own business with Jillian of making cool things and designing cool things. But it quickly swallowed up all my time, so much so that going to work for a steady gig like Gensler, even though it will be busy, will allow me more free time to focus wholly on making Because We Can a great little design and production firm, and allow it to be a creative and experimental as possible (and remove the whole consulting part from it, making it much more straight forward).

I'm very excited about this, and I think it's gonna be great.

The Pencil.

I talk to a lot of people about new technology. A lot of designers, that is. And there's one common thing that happens, over and over, and for some reason it's been happening a lot of late:
It's the Pencil Thing.
It's that whenever a designer is challenged by a new way of working, or feeling threatened by technology in general, and are unable for some reason to grasp that change, they hold up a pencil weakly as a shield against these threatening forces and proclaim 'this is the best tool for Architecture yet'.
I hate The Pencil Thing.
Every time someone does it, my heart sinks. It's like I can feel a vast wind-blown casum opening up, a giant hollow sucking space, and it just makes me want to leave, for it's obvious that no one's really trying to grow at that point, and instead wants to stay and rot and die. No one's thinking about how to do something great, they are just wanting things to not change. They want things their way, totally ignoring whether their way leads to a better Building or more value for the client, and they don't really want any changes, thank you very much.
It's even sadder when someone my age does it.
If I was a snarky man, I'd ask them if I could have that pencil, for obviously there must be something different about that pencil, it's a super-pencil, must have belong to Corbu or Frank or the Other Frank, damn that pencil could make me rich! Best tool ever! Builds whole buildings with one fell stroke!
If I was a geeky(er) man, I'd hold up my Wacom stylus, and tell them that it's my pencil, it's just that it writes on my laptop instead of paper.
If I was trying to prove a point, I'd say that I thought that our minds and hands and eyes were the best tools yet for Architecture, and everything else is secondary.
Sometimes they have a point.
Like when a senior partner at a firm I used to work for talked about how they were more profitable when working on paper, prior to AutoCAD, so why would this new Revit thing be any better? That's a focus on Value, and that I can respect, for AutoCAD already made them less profitable, so they want proof that Revit is better than paper. I can see that, for that's a wise assessment of the situation.
But once the pencil is in the air, they aren't listening anymore, for they are looking at your finger when you're trying to point at something, instead of looking at what you're really talking about in the distance.
And once the pencil is in the air, they are deaf to the sound of that huge train of progress that's about to render them obsolete by flattening them.
Once the pencil is in the air, they are no longer thinking about Architecture. No longer thinking about Building. No longer thinking about adding value to the whole Process. No longer trying to make great things.
That's why I hate the Pencil Thing.

Upstairs Bookshelf 'round two... FIGHT!

So a second, more 'organic' approach to the upstairs bookshelf. I'm diggin' the idea of making it via these CNC-cut 'ribs'. While I think this one is interesting, there's a third version that I like a lot more (that I'll post as soon as I have time to generate a model of it).
Second Bookshelf Try...
Last week at the full-time gig, so hopefully I'll soon have time to get the website looking better, get the Shopbot put together, yadda yadda yadda... All for now, too busy to post more!

Idea for the upstairs bookshelf...

Working on a design for a bookshelf we want to make for our place. The plan is for it to be a big ol' U shape, for then the low middle part can act like a bench and the two 'towers' will frame what we hope to be in the future a screen (or the wall) for a video projector to aim at. This is being optimistic that someday we'll be able to afford said video projector, but hey, it's a cool shape to work with anyways and it would be good to have a better bookshelf up there.
Wavy Bookshelf
I don't like this one, but it's a start, and now back to 'real work'. I like the idea of the flipping shapes cut on the CNC table to make up the body, but it's too rigid and too 'woody'. I think I need to make it more organic...


Subscribe to RSS - Design