Hi. I'm David Rosenthal, a senior at high school and you have found my supercomputing page on the web. This is a showcase for a bunch of images that I have generated with various software that I have written for a supercomputing class at the Thomas Jefferson High School for Science and Technology.

Daveray is a recursive raytracer that I wrote in a couple of weeks towards the end of the semester. It supports multiple colored lights, planes, spheres all with several different surface attributes. The code is all object oriented and should be easy enough for someone who knows the basics to understand. The code is here. It runs only with stdout and should be very multiple platform happy. It writes to a .TGA file which you can view with whatever. Have fun.

Very "realistic" image.
It shows off the procedural reflection
and roughness functionality in the surfprop struct. (ts.jpg - 145K (1280*1024))

This is one of my favorites so far.
It shows off soft shadows nicly. (rainbow.jpg - 64K)

Same type of scene as above.
Nice show off of blurry reflections (in the spheres). (twospheres.jpg - 54K)

These are just a few of the images I have made with a program that I wrote to visualize the Mandelbrot set.

Note the automatic color scaling.
Standard Mandelbrot set (-2,-2) -> (2,2) @ 50 iterations. (mandel.jpg - 28K)

A zoom to 200 iterations.
Notice seven-fold symmetry. (dmandel.jpg - 243K)

These images were generated with a program which calculates a projectile's motion by an iterated simulation. It uses a gravitational attractor that can both pull, like normal gravity, and also push.

A note about reading the graphs: On the y-axis, launch angle varies from 0 at the bottom to 90 at the top. On the x-axis, launch velocity varies from 0 at the left to fairly hard :^) at the right. The projectile starts at (0,0) and the target is at (1,0). Color indicates how close the projectile came to the target. Colors range from white (very close) to red, orange, etc, to black (not very close) (the gravitational attractor is at (0.5,0.5).)

Another note: All of the garbage that you see in the images (especially the ones with the gravity) is from using a time step that was too large. However, I needed to do that because else these images would not be on this web page. They would still be generating themselves on some poor computer somewhere.

This is an image with one gravitational
source that pulls in the projectile. (simupull.jpg - 51K)

Just thought that I should throw in one
of the plain-vanilia simulation results image. (simupush.jpg - 29K)

Same deal as the top one, but with the "gravity"
pushing instead of pulling. (simupush.jpg - 44K)


One last thing. I was bored one day and decided to play with the scanner our lab has. I scanned two images, one from a HP Deskjet 660CE and one from a friend's Alps MD-2300. The Deskjet costs about $300, the Alps was $600 discounted a little bit at the local store. The Deskjet print is on normal paper while the Alps print is on their special paper in photo-output mode with photo inks. On the Alps, an 8 by 10 inch printout costs about $1 for the paper plus $2-$3 for the ink. It's expensive, but I think you'll agree that it's worth it if you need good quality.

Here are the images: (both were scanned on a HP Scanjet IIc @ 1600dpi)

The Deskjet

The Alps

Any questions about stuff? Graphics in general?
Write me at davidro@ibm.net