Monday, May 30, 2005

The supernova has a name

Hey remember the supernova post I started his blog with, well it now has an official designation: 2005ca - recent SN discoveries are posted here.

The Visualization of Astrophysical Data

Wednesday-Friday of last week we hosted a Visualization of Astrophysical Data workshop at KICP and Adler. I had hoped to blog each night, but I was spent after the long days. Anyway I've rested up this weekend and can start filling you in. The workshop went really well, all the feedback I've gotten has been glowing. We had an interesting mix of people: scientists, graphics people, programmers, museum people and educators (with most people fitting in at least 3 of those catagories).

Some of the highlights:
Andrew Hamilton from the University of Colorado showed off his Black Hole Flight Simulator - really cool. Everyone always asks what goes on inside a black hole, and now you can find out. When you fly through a charged black hole (we don't expect black holes to be charged, but charged black holes behave very much like rotating black holes - and we do expect black holes to be rotating) then after you pass through the event horizon you pass through a wormhole out a white hole and into another universe -wild stuff. As you fall into the black hole you pass through a point where light has piled up, and you can view the entire history of the universe in one blinding flash (Andrew put a cool shimmering visual effert here). Creating the simulator presented some serious viz challenges. OpenGL, DirextX, or whatever, all assume that we are in Euclidian space. To create the BHFS, Andrew has to do the general relativistic ray tracing at each time step. When he was at the Univerity of Chicago six months ago, the simulator kept crashing his laptop-- now he had this giant AlienWare laptop and it ran really smoothly.

Mark Whittle presented his work on Big Bang Acoustics. All the structure in the universe was started by sound waves in the aftermath of the Big Bang. Mark has transcribed those sounds up 50 octaves so that they are in the range of human hearing. Check out his site to hear them.

Curtis Wong from Microsoft gave an insightfull talk on the role of context and narrative. He has always been interested in doing a project on Astronomy, so hopefully the workshop gave him some ideas. One thing he showed which impressed everybody was the Leonardo da Vinci CDRom he worked on at Corbis. It is currently out of print, copies are going for more than $100 on eBay.

Frank Summers talked about Scientific Visualization using Hollywood tools. He told us how he used Maya and Shake to create an IMAX short for the Space Telescope Science Institute.

John Dubinski was inspired by our GeoWall demonstration and rendered out a version of his Milky Way Andromeda collision simulation in stereo. At the end of the day Thursday we hung out in Adler's 3D theater and watched it. John seemed really convinced about the value of watching the animations stereoscopically.

Chris Mihos showed some of his viz work on galaxy collisions. One thing that I really like, and we talked about adapting for Adler, is his Galaxy Crash applet. He also shared an anecdote where the viz led to new scienific understanding (the offset of gas and stars in interacting galaxies). We all decided that collecting these anecdotes is important since much of the scientific community still is not convinced of the value of visualization.

Well there were many more highlights, but I'll stop here for now. I'm helping with putting together the website for the workshop, and one thing I plan to include is a videocast with some of the clips that have been donated. I'll blog again when it's up.

Sunday, May 22, 2005

Panoramas on the PSP

Playing aroung with the PSP, I noticed that the image panning feature works pretty well. If the image is zoomed larger than the screen size then you can move the analog stick to pan across the image. While you are panning it shows a little graphic showing the whole image and a box denotng the part you're looking at(even a nice little showdow around the finder graphic to pop it out). Once you stop panning this graphic disappears - it really well done. For me the panning feature works best when you're constrained in one dimension, otherwise it can be a bit disorienting. So I put together some Astronomy images that fit the bill and added them to the Cosmus PSP site. First a set of all 10 ull color 360 panoramas from the MER rovers, and then another set of 10 from the Multiwavelength MilkyWay poster. When you load the images up the PSP scales them to fit the screen, you need to zoom up to 100% and then pan with the analog stick. It works pretty well as a psuedo-qtvr, the only drawback is that the quality level is quite low while panning.

Friday, May 20, 2005

A new way to do 3D

Most of my visualization work the last few years has been for passive stereo displays: the Adler 3D theater, the U of C geowall, SciTech's VR exhibit. All of these work with dual projectors, a silver screen and polarized filters and glasses. We use linearly polarized glasses which can cause a problem it you tilt your head, the EVL folks prefer circular for that reason, but circular only really works at one wavelength, and there is some leakage at the surrounding colors. Anyway one of the guys coming to our Visualization of Astrophysical Data workshop, Toshi Takahei from the Riken Institute in Tokyo is bringing a new solution - Infitec glasses and filters. These work kind of like the red and blue glasses but with much more precision. The red, green and blue parts of the spectrum are each cut in two, half of each going to the left eye and the other half to the right. The big advantage is that you no longer need a special screen. Toshi is interested in projecting this on a planetarium dome to create the first stereoscopic planetarium. Anyway I'm looking forward to seeing how well they work, the only drawback is the high price of the glasses ~$500.

Thursday, May 19, 2005

A rocket on the street

Yesterday on my way to the Machine Learning Summer School I came across a rocket parked outside - weird. Then on a break between lectures I stumbled into their press conference. It was the the X-prize team "Canadian Arrow". They are starting a company to offer sub orbital flights that will splash down into the great lakes. They were in town because they are partnerning with a Chicago businessman/polititan Chirnjeev Kathuria. So they came down to International House at UofC with one of their rockets in tow.
Afterwards I had a short conversation with PlanetSpace's president Geoff Sheerin - interesting guy. He got excited when I heard I was an astronomer. He suggested we think about what kinds of observations we could make with an instrument in the nose cone (4 minutes in space). He also was interested in speaking at Adler sometime. Finally he pitched a vision for me where Project Space had a barge just offshore downtown Chicago (or Meigs field perhaps) where they would launch the rockets carrying passengers and Adler designed scientific instruments and then splash down in Lake Michigan. It's a brave new world.
Anyway, it's not everyday you see a rocket parallel parked on 59th st. (although come to think of it I have seen some strange things there: mideval sword battles, Korean drumming, pow wows...)

Monday, May 16, 2005


A few weeks ago I starting playing around with creating science content for the PSP (playstation portable). Its really a fantastic device, absolutely beautifully designed. Almost an ideal portable media player, with two main drawbacks: the flash memory is really too small to store much in the way in video or sound, and infuratingly Sony purposly limited the video playback off the memory card to half resolution. Anyway I put together something for the cosmus site. Right now there are just some photo folders and our "Mapping the Universe" video. The PSP is really great for browsing images (using the L and R buttons). I'm thinking about replacing the photo essays with something more like magazine articles. Once we get a full suite of content I'd love to mount a couple of these in CyberSpace.

Friday, May 13, 2005

I discovered a new supernova today

One of the things I do each morning is to check for new supernova canidates from the SDSS spectra. As the galaxy spectra come in we have automated code which checks to see if there is evidence of a supernova on top of the normal galaxy spectrum. Most of these canidates turn out to be false alarms, but a few ike the one this morning turns out to be real. I sent an email to the Central Bureau of Astronomical Telegram informig them of the discovery. Once the SN receives an official desigation I'll post it in this blog. This one was a type Ia observed about 8 days after maximum light. You can follow along with the supernova search here.