Pixel Inventor Goes Back To the Drawing Board 304
lawpoop writes "Russell Kirsch, inventor of the square pixel, goes back to the drawing board. In the 1950s, he was part of a team that developed the square pixel. '"Squares was the logical thing to do," Kirsch says. "Of course, the logical thing was not the only possibility but we used squares. It was something very foolish that everyone in the world has been suffering from ever since.' Now retired and living in Portland, Oregon, Kirsch recently set out to make amends. Inspired by the mosaic builders of antiquity who constructed scenes of stunning detail with bits of tile, Kirsch has written a program that turns the chunky, clunky squares of a digital image into a smoother picture made of variably shaped pixels.'"
Re:Suffering ? (Score:3, Insightful)
Re:Suffering ? (Score:3, Insightful)
Why are we suffering from it since so ?
I did not read the article, so I don't know if it's answered there.
+1 blatant
Re:Huh? (Score:5, Insightful)
In a word? Jaggies [wikipedia.org]
Wait... (Score:3, Insightful)
Re:Huh? (Score:5, Insightful)
Exactly. Besides, you have to have some kind of regular pixel on a physical display, so it has to be some geometric shape that meshes well with itself: squares, rectangles, triangles, or hexagons. Squares are the easiest. To overcome the blockiness, you just have to decrease the pixel size enough, and increase its density enough, so that the human eye can't perceive the individual pixels. Modern displays have pretty much achieved this.
It sounds like this guy's trying to invent variable-size pixels, but that doesn't make sense. Sure, you could come up with algorithms for dealing with them efficiently, but making a physical display that shows variable-size pixels is anything but trivial, and pointless since we can already make square pixels so small.
Re:Huh? (Score:5, Insightful)
But those are pretty easy to solve. The most complete solution is simply to increase display resolution past what the eye can perceive. Have small enough pixels, no jaggies can be seen. We are working towards that bandwidth of the interconnects and cost being the only hurdles, and those are going away slowly. As a quite effective stopgap, anti-aliasing can be applied. It is very easy to do on modern GPUs for little cost.
Now, take a variable size, variable geometry pixel grid. Tell me how your process that, how you store it in memory, how you rasterize images to it. Sound like some complex problems? They are, very complex. So solve all that, and in such a way computers can process it in realtime with cheap hardware (if it is even possible). Then you get to tackle the REAL hard part: Building a physical display that can display said pixels.
So, you can do all this, which I am unconvinced is possible, OR, we can simply work on making displays with more pixels. Get displays up in the 300-400PPI region and none of this is a problem anymore. While that will take more bandwidth than our current interconnects provide, engineering higher bandwidth interconnects is a well understood problem and there are a number of solutions (such as simply running more channels in parallel). It will also require working on ways to bring the cost of high density displays down but again, we've had a great deal of success with that. LCDs went from VGAish resolutions that were quite expensive and small to massive HD displays in about a decade.
To me, it seems like we have the solution to the problem. This new solution sounds far, far more complex and likely impossible.
Re:Huh? (Score:5, Insightful)
Re:Huh? (Score:1, Insightful)
Agreed that variable pixel sizes sounds like a silly pursuit; the pixel's boundary would require too much information to be useful.
An alternative regular tessellation does make sense, though. Hexagons can be packed with a greater density than squares (e.g. animal kingdom's genetic algorithm settled on it in many places, including our eyes and honeybee combs). The neighboring scheme for adjacent hexagons is in some ways more sensible. Spatial FFTs on hexagonal images might be nice
The downside: addressing schemes for hexagonal images would probably be quite tough compared to rectangular, which matches nicely to Cartesian coordinates.
Re:Huh? (Score:4, Insightful)
Anti-aliassing is essentially a form of blurring.
You eliminate jaggies at the cost of sharpness.
Using non-square pixels is an interresting (although perhaps not practical) way of tackling this issue.
Re:Favorite graphic designer story (Score:3, Insightful)
Re:Invented the pixel? (Score:3, Insightful)
Yeah, but that was hand tinting [wikipedia.org], which was (mostly) more like painting.
The Russian guy actually did it with light filters and three separate images. Way ahead of his time.
Re:Favorite graphic designer story (Score:3, Insightful)
Are you honestly suggesting that, by having more pixels on the screen, the picture won't be blocky when they zoom to the pixel level?
No, I'm suggesting that by increasing the resolution the designer will be able to draw smaller shapes. This solves the designer's problem of "I need to be able to get a smaller shape in here but it's all too blocky."
Re:Invented the pixel? (Score:3, Insightful)
I'm surprised you've never seen "The Color of War" on History Channel or PBS. Anyway youtube has lots of footage. And oh yeah, they didn't use video. It was film
Re:Favorite graphic designer story (Score:3, Insightful)
Favorite Jerkass story (Score:5, Insightful)
Okay. Minus five points from the "graphic artist" for not knowing how to resample the image. Plus one point for trying to improve her knowledge instead of suffering in silence.
Minus one point from you for not knowing something that's not directly in your field.
Minus ten points from you for not even trying to help.
Minus fifteen more points from you for being a jerkass about it on Slashdot.
So she's down four, and you're down twenty-six.
Re:Exceedingly silly (Score:4, Insightful)
I completely agree with you about the fact that everything in the paper seems to be pulled out of thin air... But I do see two reasons why his compression algorithm might be better than JPEG or other lossy codecs in some situations:
1) the decompression performs no arithmetic on the pixels, hence you can perform gamma correction or color change losslessly (like in a square-pixel image)
2) aside from the choice of mask, the compression is entirely deterministic, which is a plus in scientific imaging: when you have a "triangular pixel" with value 200, you know that the average of that zone was exactly 200 (with JPEG, you can't know anything for sure as the compressor could add artefacts or remove detail as it sees fit)
Why are you maximizing contrast instead of minimizing error like any sane person would do, WHY?
In fact they are equivalent, assuming that the masks are equal-area:
square of RMS error
= Variance(residual)
= Variance(maskedimage1) - average1^2 + Variance(maskedimage2) - average2^2
Since Variance(maskedimage1) + Variance(maskedimage2) remains constant (we just shuffle pixels between both masked images when we change the mask), minimising the error is equivalent to maximising
average1^2 + average2^2
= 1/2 * ( (average1-average2)^2 + (average1+average2)^2 )
Since average1+average2 also remains constant, this is equivalent to maximising
(average1-average2)^2
which gives us the maximum-contrast method.