The Zoom Quilt

November 30, 2004 3:28 PM

The Zoom Quilt (uses flash)
posted by criticalbill (19 comments total)

One million times I've seen this! I demand my money back!!!


But seriously. Double post.
posted by jonson at 3:34 PM on November 30, 2004

It's news to me. All of the illustrations are gorgeous. Does anyone know how they managed to stitch those together? It's not one big file, and on zoom it looks like each frame is its own rendered file. But that would be a huge filesize...
posted by still at 3:42 PM on November 30, 2004

Original here, though the url be broked.
posted by eyeballkid at 3:45 PM on November 30, 2004

Well I searched "Zoom Quilt" and nothing came up. And the url didnt get a match. Sorry.
posted by criticalbill at 3:50 PM on November 30, 2004

I was wondering how they did it too, but I imagine that its just a rectangular pattern mapped to a cylindrical surface with perspective.
posted by Photar at 3:59 PM on November 30, 2004

I'm not sure you can map like that in Flash.. I could be wrong.
posted by still at 4:02 PM on November 30, 2004

criticalbill, the other one was "zoomquilt" and the search engine didn't pick up on it. I do believe that Matt's working on it though.
posted by fenriq at 4:05 PM on November 30, 2004

despite the fact that it's a double post, I've never seen this before. so thanks, b/c it's amazing.
posted by mai at 4:08 PM on November 30, 2004

How it works, I believe: Imagine the field of view is divided into a 3x3 grid. The artist draws the image in the outside eight squares, leaving the middle one blank. That's [Image 1]. And centre blank square is the basis of [Image 2], which is it's self a 3x3 grid with the middle square blanked out. [Image 2] is drawn so it's outside edge merges into [Image 1]'s inside edge, so it looks continuous. The blank space left in the middle of [Image 2] is the basis of [Image 3], etc.

The animation, then, displays [Image 1], with [Image 2] scaled down to 33% in the middle of it, [Image 3] scaled down a further 33% in the middle, [Image 4] scaled down a further 33%, etc. etc. - You probably only need maybe 10 scaled down images before the smallest "middle image" is pixel sized, and no further iterations need to be drawn.

Zooming, then, is just a case of making each sub-image larger.
posted by Jimbob at 4:41 PM on November 30, 2004

Very cool, despite it double-postedness. As to how it works, I think the hard work was done outside Flash. What we are probably doing is navigating a series of still frames in a loop.
posted by pmbuko at 4:42 PM on November 30, 2004

If you click the Xs at the bottom of this page, you'll see the graphics used. Click where it says HTML/JPG (1,5 MB) link to see the different sizes of each image that were used.
posted by iconomy at 4:43 PM on November 30, 2004

Like that quilt was used in the description.
posted by thomcatspike at 4:47 PM on November 30, 2004

Very cool. Thanks for the link
posted by Outlawyr at 5:04 PM on November 30, 2004

This seems to be more thematic and coherent than SITOs Gridcosm project, though roughly the same idea.

I posted it here. I think mine might have been a repost as well.
posted by loquacious at 5:11 PM on November 30, 2004

Cool link even though it is a double (ya fecal freak).
posted by dobbs at 5:34 PM on November 30, 2004

Thanks!! this is mind-numbingly cool...
posted by Balisong at 8:01 PM on November 30, 2004

That was cool, even if it was a double.
posted by dazed_one at 9:48 PM on November 30, 2004

was this a double post?
posted by Satapher at 1:58 AM on December 1, 2004

I don't know. If only someone would tell us!

But this is really neat. I like the part with the killer trees. KILL TREES, KILL.
posted by jennyb at 4:10 PM on December 2, 2004