pouët.net

1) Artist(r)
2) Comp-U-Com
3) You

...get the joke?

http://www.pouet.net/prod.php?which=10053

i think they're all 3 painted by a monkey on acid

who cares, painting is a dead form of art.

I guessed the middle was done by a human, because the brush strokes look very natural (as if by a person scribbling as fast as they can), so it's pretty impressive that it was done by an algorithm.

I guessed the bottom one was by a computer because it is the most detailed, and a computer can add an arbitrary amount of detail by being left to run for an arbitrary amount of time. But I guess most of the details happened by "controlled chance" from splashes or smudges of the ink, so it didn't take that long.

As for the top one, while I was still reading the text and I saw it only out of peripheral vision, I thought it was a black and white photo of a jet airplane. I was surprised when I finally looked directly at it and the details I thought were there were suddenly gone. I guess the painting is designed to resemble a literal object (of some sort, not necessarily a jet) in the lower frequency domain, but it resembles simple brush strokes in the high frequency domain. So I didn't think a computer would figure out how to do something like that.

In retrospect, it seems the main difference in the computer generated one is that there are no "trends" throughout the image. That is, in the top one, all of the brush strokes are collected together to make a shape. In the bottom one, the picture gets less dense and less smudged as it goes to the left, and is composed of clearer, bigger blotches as it goes to the left. In the computer generated one, it doesn't look like there is anything cohesive or form-building about where it chooses to do the brush strokes. ie. the computer doesn't notice overall patterns in the image and try to build on them, the way an artist might.

The computer-generated one has some typical paint-software doodles (eg. look at the continuous scribble in black at the top-left, and the white scribble overtop the black area right of center). It's easy to do in software, but something I would be unlikely to do with a paintbrush (and even if I tried, it would be difficult to reveal that it's real paint through varying thickness of the paint). The human ones lack this kind of scribble. It's a cheap way to pick out the computer-generated one though.. because it's through a difference in the medium. They ought to give a robotic contraption the same kind of tools and canvas to make it fair.

The top one seems the most organized to me and seems to follow some composition guidelines. Also reminded me of a hangman contraption or something like a cross, whereas the other two are too messy for me to bother picking out any interpretation.

Damn! The bottom one was yours? i thought that to be Kline's because it looks more informel than the other ones. You are artsier than art!

Nice drawings texel (these on your site) :D

I was immediately pretty sure that the 2nd was computer generated since it presents a mild level of complexity and because of the sharpness/regularity of colors and shapes.

Then I was undecided between the 1st and the latter. I anyway supposed that since often modern artists tend to use minimalism it could be Franz Kline. For sure the last one was at least 10x the complexity of the others.

Said that I think that by fine-tuning drawing algorithms (=much time) one could reach any of these "levels" in a generative/parametric way.

I've heard in a movie someone saying: "there's no point in painting realistic things now that you have photos"

I'd say that depends on your intentions.
I'm pretty sure it could be a useful experience, just as covering other musicians' tunes can be.
Having said that, I am able to admire and appreciate the talent of those artists that are able to mimic reality - because they can do something that most of us can't. And that's pretty much my definition of art.

xernobyl: depends how you do it... sure, there's not much point painting something infront of you as realistic as possible (except perhaps as practice, or to show your skills.. much like a boris valejo copy :) if you can photograph it. But you can paint something realistic that doesn't actually exist.. how about that?

Also, a good artist can capture a bit more than an average person with a camera.. something of the feel of what is there, or the personality/soul of a person.

That doesn't mean there's any point in this, either:

Abstract horror! That's a strange mixture of good and ugly to me.

Contemporary and realism paintings are not excludent, and there is not one better than the other... these are just different things. If I ask you "what is better, opera or cinema?", for sure you will choose easily the kind of art you prefer or you are more used to or you better understand, but it doesn't stop both being art, both have sense as arts... Not understanding a form of art doesn't mean the art has no sense. By other way, in every form of art there are more pieces of crap than pieces of good work...

In my case, sometimes I paint with near random ink spots and other times I do gigerian pigs.

You can call paint tossed randomly over a canvas "art", but that only detracts from the value of the concept of art, it doesn't add anything to the painting. Of course you can delude yourself into appreciating it anyway, but that tends to amount to judging a painting by the person who painted it, not by what it actually looks like.

Let see it this way: don't call it art and artworks but activities and productions. Every activity have a set of rules for the creation of the productions. The set of rules also helps to rate a production by other people; as best as a production fits to the rules, the best the production is. Since some (or all) the rules may be subjetive, then the ratings may be also subjective.

One example: the set of rules "representing a scene of reality in the most similar possible as what eye captures using painting over canvas". This activity would be called, for example, "realistic painting".

Other example: the set of rules "representing a scene of reality not using visual correct perspective, but different perspectives for every object and even several perspectives for every object. Also using the colors the scene makes feel instead of the visual correct colors. All done on canvas with paint". This actividy would be called "cubism".

Some people, in the human tendency of classifying things, would say "realistic painting and cubism are similar activities since both try to represent a scene of reality and both are done on canvas with paint, let's create a super-category for this... lets call it 'painting'.". So now, we have a supercategory called "painting" wich contains two activities and two sets of rules.

Now, get a cubism production and rate it with the realistic painting rules, what happen? Since some of the rules doesn't fit at all, the rating would be always poor, no matter how good would be rated if it were being rated using the cubism rules.

A person knowing only the realistic painting rules and thinking cubism productions have the same rules of realistic productions, would rate every cubism production poorly, and will finally think all cubism productions are not good.

Now an example in demoscene. If you see a production from the demoscene supercategory, it is a "256bytes", but you don't know it, you think it is a "demo of unlimited size activity", your rating would not be coherent with the context.

Other example more. You learn the rules of cooking activity. Then you are given a painting from Velazquez. You think it is a cooking production. Since it doesn't have a good flavor (one of the rules of the cooking activity), you will rate it as not good.

Well, finally I have to say that all activities, when rated, have more bad productions that good productions. But, to rate you need to learn and understand the rules. And also, do not confuse supercategories with activities.

texel, you got a point there.

battle droid, so you would say that jackson pollock wasn't an artist?

nosfe, no good will come out of this discussion ..

íf you picture dont have sports car, dragons, tits, swords or spaceship, aliens, etc. it's not art. it's called "degenerate wannabe art which takes away opportunity from real artists like franz frazetta". only way measure picture is how impressive it depicts subject or how firm woman breasts are in picture! if there is something that i dont understand it's not picture at all. it's garbage. like garbagebag! look at art of ddr or cccp or nazi germany. this is art what i understand! all other art is degenerative:(

disclaimer: above post was not written seriously.

If you feel oppressed, escape to the USA ! Look at Fischinger, he did it too... :D

Oh and btw, throwing references to nazis and CCCP into the discussion doesn't make it easier to debunk, uns3en_.

Of course, but still : always nazis references... Godwin point and such. I'm getting tired, even when it's meant as a joke. :)

unssi: you got a point, though :D

My guess (before looking through the thread) was: Texel, HAL, Franz.
Iow, I confused you and Franz. Bost are post-modernism crap. (No offence)
But more importantly I recognized the computer simple because it drew white on black. Was I really yhr only one who saw that? :o