The project was done in 2004, first implementation in shockwave, later flash. I redid the whole thing in js+canvas a few weeks ago, finally with a much smoother zooming algorithm than before.
Feel free to shoot me any questions you might have!
We worked with single steps of 1024x768 pixels size with a center part of 50% left out. An illustrator would either work inwards or outwards of the existing portion. Outwards that would give a photoshop document of 1024x768 with the previous step locked on a centered layer of 512x384. Inwards with the upscaled previous image of 2048x1536 as the frame, and the respective centered paintable area of 1024x768.
The illustrators then could easily blend their work into the other peoples artwork. The whole thing evolved very un-planned. There was no guideline as to what should be depicted. Illustrators just had to pick up what others have left for them. Best part of the process for me.
When you watch the whole thing and look at some other page afterwards, it's moving in the opposite direction.
I found the artistic aspect and the amount of work that must have gone into this project more interesting than the technological aspect. There are 15 participating artists though so at least they could split the work.
This fantastic interactive artwork makes me seriously want to watch the show being promoted (Tin Man, a TV miniseries). Many of the pieces in it are stunningly evocative (like the Asian city) and bring to mind some of the best adventure games of yesteryear.
Bigger, longer, and more unhinged than the original. This brings me back to the collaborative online art projects and people getting together on forums to do cool things for the heck of it.
An endless (not really), silly pastiche of meme-y images, courtesy of the goons of SA.
- There's a tradition of "collaborative building" projects on the SomethingAwful Forums, which tend to be full of in-jokes and pop culture/videogame references. Some examples:
The art is compelling, but I don't understand the post title. Is this, in your opinion, some amazing feat of the underlying technology? Doesn't look particularly computationally intensive or anything like that. Yes, canvas lets you draw 2D graphics inside a browser.
I think it's largely because of the title, which suggests that this (and JS + HTML canvas) is technically innovative in a way that it isn't.
As long as they're informed criticisms and not too aggressive, that's a good thing. If people are potentially going to criticize your project for not bringing anything new, point out where you've oversold it, etc. then the positive feedback means a lot more. Plus you (and others reading) might learn something.
It's a piece of art, enabled by web technologies that are far more accessible now than in the past. I love web art stuff. I'm both an artist and a scientist/programmer, so things like zoomquilt are a perfect marriage for me. Gives both sides of my brain stimulus.
It appears to be similar to previous version: http://zoomquilt2.madmindworx.com/zoomquilt2.swf. Only this time it's a slightly different script language, different rich client and a different vector-based rendering engine. The obvious difference is that now I can't save it easily to be enjoyed later.
Other differences: the art work is 100% different and you have to wait for the flash version to load.
Luckily my browser has the ability to "bookmark" urls, so while I am not going to save it for "offline" viewing, I can go back to it whenever I am connected to the internet.
Isn't that also true of the JS version? It took several minutes to load for me. There's a loading bar in the middle of the screen, which slowly fills up until the demo is ready to run. That could just be due to the server currently being hammered, though.
"MHTML, short for MIME HTML, is a web page archive format used to combine resources that are typically represented by external links (such as images, Flash animations, Java applets, audio files) with HTML code into a single file."
I haven't tested it, but it seemed like a neat idea.
"Microsoft Internet Explorer, as of version 5.0, was the first browser to support reading and saving web pages and external resources to a single MHTML file"
Actually pretty amazing, artwork and technology! So how exactly does it work? The art couldn't have been done in a JS canvas, could it? I imagine it's just a really high resolution image that gets zoomed into and looped back? That's actually hard to understand too...
It's a series of images at different zoom levels. Since there's no freedom of movement, you "just" need to leave a section in the centre empty, have someone draw a large image for that section, copy in a scaled down version of that image.
When displaying it, once you've zoomed in a certain amount, you simply start copying in the next image in the series suitably scaled down, and keep zooming until that image fills the screen, and fill in the next level in, and so on. Roughly.
Repeat for as long as you have images, and optionally make it look nicely back to the beginning as they've done here.
It's nicely done, but technically very simple - "infinite zooms" like this dates back to the early days of the demo scene at least, so at least early 80's (and it wouldn't surprise me if people have done it with "manual" animations before that).
I dunno about everyone else, but on this 24" Dell IPS I can see the next image being loaded and then scaled, does it every few seconds. The details starts getting a bit blurry them pop, new asset super imposed and we start scaling.
It's certainly not a single image, you can see how the content gets blurred at the edges. It's similar to the kind of image you get when you point a camera at a TV and display the picture on the same TV. It looks like infinity, but actually you just have a rectangle in every picture through which you can "look" at the same picture, which in turn has another rectangle etc. In zoomquilt, it's not rectangles, but more complex (vector-drawn?) shapes which let you see through each image to next. Similar to a card stack where every card has a hole, and the last card is also the one before the first.
That's exactly what I was thinking when I saw it. I know I have seen it in swf/gif form in the past. While I don't dispute HTML5's canvas(+js) is very cool I was hoping for something straining the limits or doing something new.
I was going to say something along the lines of "...rather than seeing an old idea/concept rehashed in canvas+js" but then I realized that could include people doing things like getting old games (Quake/Doom) running in the browser with canvas+js. The counterpoint would be that this idea could be implemented in gif form (While it would lose the ability to toggle your direction/speed) whereas I don't think anyone would enjoy gifQuake/gifDoom :).
TL;DR - It would appear I'm jaded when it comes to things like this. And that's probably not a good thing.
after seeing it loop for one time, I came back and started reading the comments (on my iPad).
I've had the feeling, while looking at the comment section that it was zooming out (away) from me as if it was to compensate for all that zooming in...
The project was done in 2004, first implementation in shockwave, later flash. I redid the whole thing in js+canvas a few weeks ago, finally with a much smoother zooming algorithm than before.
Feel free to shoot me any questions you might have!