Making the Voria Logo

Thursday, December 22, 2005 by darco
Posted in , , ,

voria_logoAs many of you may know, sometime around 2002 I began work on starting my own company. In January of 2004, that dream was realized as Voria Studios. That company has come and gone, but the logo that I made for that company lives on in Synfig. So where did this logo come from? Well, it's an interesting story. And like most stories, the best place to start is...at the beginning.

When I was a little kid I was horrible at Trivial Pursuit, as most little kids are. One part of that game did make an impression on me though: those little pie-piece things. You know what I'm talking about, these things:

trivial-persuit-pie

You know you want to eat one

I thought they looked neat. It seemed like the whole point of the game was to bling out your pie with these cool colored pieces. For some reason that just seemed irresistible, even though I lacked enough breadth of knowledge at the time to be any good at it.

Fast forward a few years. One day I was using hotbot (This was before I found google), and I noticed a tiny Inktomi logo at the bottom of one of the pages. For some reason, it reminded me of the above mentioned Trivial Pursuit Pie thing, so I was immediately taken by it. At this time I was starting to get good at using Adobe Photoshop, and making logos for non-existant companies was a hobby of mine. I'm sure this influenced some of my earlier designs.

Initial Inspiration

Stained Glass

The "Stained Glass" Composition

Fast forward a few more years. At this point, I was working on what is now an ancestor of Synfig. I was making all sorts of pretty, fractal-like pictures using the software I was writing. One of these pictures was called "Stained Glass", which seemed to be rather popular with my friends.

stained_logoOne particular element of this picture stood out to me, because it reminded me of the aforementioned random stuff. A part of the picture was shaped like an upside-down shield with a very pretty texture. So, I started going about isolating this element into it's own composition.

After hours of hand-coding and tweaking the composition file, this (rather familiar looking) image was the result:

testlogo

I was amazed at the sense of depth and detail, but it took forever to render–I think I let that particular image render for over a day. Not that long render times are a problem.

For a while I thought this would be the perfect logo for something, but I just wasn't sure what. It slowly became the unofficial logo for the software that made it.

However, the design did have some serious drawbacks:

  • Its beauty was in the details. Scaled down it looks fairly uninteresting.
  • It looks even more uninteresting in black-and-white. Even worse when printed.
  • Full color is expensive to print on t-shirts, signs, business cards, etc.

In other words, the ideal logo would look good in black-and-white AND full-color. So when it came time to come up with a logo for my company, I knew that I would be doing myself a huge favor by simplifying the logo.

One thing that made the original logo so appealing to me was that it was entirely algorithmically generated. I did not paint any of the details, they are just a natural part of the math I used to create it. There is a bit of beauty in that, and I would be taking that beauty away by refining it further. Nonetheless, it was a necessary action.

I created a mockup of the final logo in the Gimp, because Synfig was not quite ready at the time. The mockup shape was more similar to the original logo, but the design was now solid blue with black strokes to demonstrate the detail of the original.

oldlogo

The idea was that for black-and-white printing, the blue color would be changed to white, leaving only the lines of the logo—which were easily recognizable.

One thing that I always intended was for the final logo to be created entirely using Synfig Studio, so once the software was at a point where it could be used to create the logo, I did so.

logo-1

At this point, you can tell that the shape has changed slightly. The base is a bit tighter, and the logo a bit thinner. This was a stylistic change. I just thought it looked better that way.

I also added the stripes in this version because the background just seemed too bland as a straight blue gradient. At one point, the logo looked like this:

voria_imperial

But I feared it would remind people of the Japanese Imperial Flag flown in World War II:

imperial_japan

So, I decided to scratch that version, even though I liked it better.

Getting the Trademark

After the logo was complete, I decided it was time to get a federal trademark from the USPTO. This process took forever, but we finally received notice that our Trademark had been registered a few weeks before we closed up shop in December of 2004. Go figure.

Making it better

I was never totally satisfied with the logo. Even with the gradient, it looked rather flat, and the stripes never looked right when printed in color. I really value simplicity so I was a bit unwilling to try to add more "bling" to the logo to fix it up. I had just kind of accepted it.

I started to notice the whole "candy" look (pioneered by Apple's marketing team) was really becoming popular.

apple-logo

Do not eat

It was never really my intention to rip off this trend, but I did want to see if I could achieve this effect with Synfig Studio. So one day, I sat down and started playing around with the Voria logo. Before I knew it, I was looking at something that looked much better than the logo we previously had.

logo

Do not eat

I loved it. I showed it to my employees. They loved it. So I made it official. BEHOLD! This was the new logo. Sound the trumpets. We are gonna be famous, and this is the symbol that the world will recognize!

Of course, the whole famous part never quite worked out like I hoped, but I still love the logo. I even made a few wallpapers featuring it:

eroded_metal

Eroded Metal Wallpaper

So that's the story of origin of the logo—from the idle fascination of the trivial pursuit pie pieces to algorithmic rendering models.