Friday, February 4, 2011

Invader Zim embroidery design - GIR

2 friends of mine is really into Invader Zim, so I wanted to embroider GIR onto something for them. Suprisingly, there are not that many designs out there. I found 2 at Needlework, but the site design worries me. So I decided to make my own design from scratch in Embird.

I had been playing with fonts in Embird, so I decided to try the digitizing software. It is nothing like I expected. I was expecting the digitizing software to import a design, and get 80% of the way to done automatically. I can even think of how the algorithm would work in my head, so I didn't even think that there was another option on how the software worked.

The software doesn't work anything like that. The steps to digitize a design are (at a high level) these:
1. Import a picture or design as a background image.
2. Draw objects on top of the background image.
3. Switch colors at strategic points.
4. Draw objects in next color
5. Draw the outline
6. Test the design using the sewing emulator
7. Save as whatever format your machine takes
8. Embroider on your machine.

There is absolutely no help in actually digitizing the design to stitches in the software, you have to draw it manually. There are different types of objects (circles, columns, fill patterns, etc) that help, so you are not controlling each stitch manually (though you can), but it is not fast to do. Again, the actual software itself is extremely esoteric and slows you down substantially.

It took roughly 8 hours to do the first design that I tried. It worked out pretty well and I learned a ton. The fact it took so darn long kind of explains why designs are so expensive. If this was a business and it takes 8 hours to do a design, then I need to sell at least 40 of the design itself at $5 to make up for the time spent making the design.



One of the big learnings is that the pull of the thread is substantial in the direction of the stitch. Looks to be roughly 0.5mm or so in the fabric I was using. This means the fill areas must be expanded to make up for the difference so you don't get unfilled areas. That can be compensated for in the design as long as you think ahead.

The other big learning (that I already knew really) is that the software doesn't do nearly as much as I would expect, especially for $450. I'm glad I am using the demo version. The Brother-PE software isn't much better, though it will do 1 color automatic digitizing. There is a ton of room for new players in the embroidery software space, mayhaps I will make my own open-source version that is much more user friendly.

No comments:

Post a Comment