Friday, February 20, 2009

Long Day's Zooming into Night

Hey, look at me! I've got texture!


Whoa.  That Algorithmic Worlds thing of Sam's is just the sort of thing I was talking about.  If it wasn't for guys like Sam I'd say Ultra Fractal was the biggest software rip-off of all time.

As for texture; Even Mondrian's plain colored squares with black outlines have it.  But I think that's because Mondrian didn't have Ultra Fractal -- or Photoshop.  The texture was a by-product of the materials he used: canvas and oil paint.  What he probably wanted was something more like glass or plastic, which wasn't available at the time.

Mondrian's famous square collages are a good example of "art".  I saw one once in a gallery somewhere (I think he made a lot of them) and you know what I did when I saw it?  I moved in close and took a look at the surface of the picture -- just the sort of thing that your binocular images allow viewers to do with your artwork despite the fact it's viewed on a monitor.

I remember seeing a painting by uh, Klee, Paul Klee, I think.  It was full of squiggly little creatures against a surreal and colorfully painted background.  I'd seen it before in a book.  The kind of photograph of art that you've mentioned, and is lacking in detail and texture.  When I saw the original hanging on a gallery wall (and I took a good close look at it -- no hanging rope barrier or plexi-glass case in the way) I hated it.  It looked like something I'd painted in art class.  In the book it looked more professional, but in high-resolution on the wall (more like "full-resolution") I could see the brush strokes and it looked like Klee has just slapped the thing together in a few minutes.  That shouldn't matter, really, but the point is the high-res version -- with lots of texture and detail -- looked worse, not better.  To me, anyway.

Getting back to the Zoomify things, I think they're pretty gimmicky.  Sam's "...atl2" is interesting, but the other one doesn't benefit from the zooming.  Janet's is not a good one for zooming either; probably because it's too fractal, actually, and the recursive pattern is repetitive rather than revealing of something deeper or more subtle, which is what zooming into an image is usually done for -- to show you the artwork's underlying architecture.  In defense of both of them though, I'd say that the Zoomify feature is relatively new and they're just experimenting with it at this point.  These examples are just the beginning.  (Although Jock Cooper's Zoomables are better done and with much simpler technology, too.)

To compare my "Sterling-Worlds" with the Zoomify flash applet, the first thing I'd say is that Zoomify doesn't go deep enough to mimic fractal zooming but it's not a good magnifier either because it goes too deep for that and presents you with details that are not even noticeable in the top-level view.  What it's good for are maps and diagrams where the relationship of the detailed view to the top-level view is abstract and doesn't have to be related to what you've currently zoomed into (a street intersection or the connections between several atoms on a molecule, for instance).  What I'd prefer to see is a parameter file that you could load into Ultra Fractal and explore like you can the Sterlingware parameter file I posted.  I'm not even sure that's possible with Ultra Fractal.  The program is so "refined" that it's abandoned its fractal origins which the single layer programs have maintained and developed.  Ultra Fractal is more a graphics program now that does creative layering employing fractal themes in a trivial, decorative way.

I find Fractal Art (art with fractals) appears to be simple to define on the surface, but when you start to consider it's algorithmic nature (deterministic, mechanical) and particularly it's abstract, non-representative characteristics (it doesn't really look like anything), evaluating it according to established principles of what is good art and what is bad art, is something I still find to be elusive.

The other thing is: I consider the computer monitor to be a adequate "canvas".  I guess that's why I consider the parameter file, generated world, to be the only real zoom or exploration that counts.  If someone was to print out my images and see something more, or less, than is visible on the computer screen, I'd say stop looking at prints and stick to what you see on your monitor -- that's the real thing.  Whatever extra shows up in a print is just artifacts, by-products -- as artificial as Mondrian or Klee's brushstrokes.  Is that crazy?


Blogger Jenn Dixon said...

You're right. The few days that I've been playing with Sterling2 have reminded me why I fell in love with fractal art in the first place. Though I do use UF, I cut my teeth of programs like Tierazon and Fractal Explorer. Both are great for zooming.

2/20/2009 3:35 PM

Blogger Sam said...

Thanks for the advertisment.

I agree with the first part of your mail. I had the same feeling in front of famous works of art as well, I liked the overall composition, but the details were just ugly. Picasso is maybe the artist which disappointed me most from this point of view. On the other hand, Dali and some old masters are amazing at high resolution. I don't know if you're aware of it, but there are gigapixel images of famous works at the Prado.
Check the second picture from Hieronymus Bosch. I would even say it's impossible to fully enjoy his work without a virtual zoom. You can't get so close for such a long time in a museum.

As well, it is impossible to fully enjoy my works on a 700x700 image, simply because only a tiny part of the pattern is visible. I do believe both images benefit a lot of zooming, even if 20061012 is not among my most detailed works. And the image itself is not so big, only 0.1 gigapixel. As I said, it is mainly a test, with an image that I had rendered for other purposes.

About texture, I don't see why a work of art should necessarily have a "painted" texture. If one is looking for such a painted look, then the best thing to do is use paint and not a computer. Using filters on a fractal image to deliberately lose resolution doesn't make any sense imho. In my works the distinctions between texture and pattern depends on the scale you're looking at, because the texture is produced by smaller versions of the pattern. This unity makes much of their interest, imho. That's exactly what you can't see well on a small web image, and the reason why I started using Zoomify.

I thought about letting the computer of the visitor compute the image so that it would allow him to zoom and move freely into the image (and it would spare a lot of disk space, too). I even asked Frederick, who said that it could be possible with ActiveX. But that's way over my programming skills. And it would allow anybody to retrieve the parameters and the algorithms used, something I do not want. So zoomify is the best option for now.

And about these continuous and pointless attacks against Ultra Fractal, maybe you should just start displaying some fractal images that obviously cannot be realized in Ultra Fractal. I haven't seen any on your blog so far. The ease with which it is possible to implement ideas into algorithms and then works (especially with the new object oriented programming) makes it for me without any doubt the best tool available for algorithmic art. Anyway, a constructive approach would be less boring for your readers.

By the way, people interested in updates of gigapixel images can subscribe to a RSS feed on this blog :



2/21/2009 12:58 AM

Blogger Tim said...

That Prado link is a great one. It's funny how zooming into some paintings reveals exquisite detail and into others, which may be equally great works of art, reveals crude brushstrokes that are almost childish. I'll bet many of Van Gogh's works look rather ugly on close inspection, and yet at regular, "stay behind the rope please" view, they're very impressive.

This is part of the disagreement between Terry and me. I see texture and fine details as a quality of some artwork but not a necessity for all good artwork. In particular, digital work, like fractals, has neat clean details unlike anything a human hand could make.

Ultra Fractal's greatest advertisement and defense is the artwork made with it. That's the case with any program. Maybe if the "Good Ultra Fractalists" like yourself were more open about how you made your work and gave out more parameter files then there'd be much more good work like yours made and less of the "Revlon Fractal" layer-itis stuff that is unfortunately much more common and easier to make.

Of course, you have every right to hold onto such information, just as Frederick has every right to make a closed-source, proprietary program, but it would benefit fractal art more if it were different and more open.

But, for the record, you make great work and it could only have been done in Ultra Fractal.

2/21/2009 10:17 AM


Post a Comment

<< Home