The last night was by far the best night technically we had no problems with the drum at all, and the sensor never broke or gave false readings.

As for attendance I’ll admit I was seriously worried. Most people go out on Friday nights and I kind of thought that we wouldn’t find anyone at Frightling all night. I was wrong.

Attendance wasn’t amazing, but it was definately much better than expected. One of my friends Joey Bernardo, said that the exhibit was pretty cool and liked to see how fast he could get the whole storm to show blue skies (he was by far the best at this task). Cal Ribbens (my professor for another class) came by and told Greg that he was glad he brought his whole family.

At the end of the night attendance slowed and then my roommates Sean O’Mara and Brent Boggs played a visit. Sean said that had never had to build anything so big in Electrical Engineering and was very interested in how it worked.

The night was overall pretty awesome, at the end someone from the Ghosts of Nature group shut our door which locked everyone out of Frightling… luckily with only 20 minutes remaining this proved to be a minor problem and I was still very happy with our project and our result.

Tonight we continued to have the vibration sensor snap in half a couple of times. We came up with the solution of placing padding between the drum head and the vibration sensor. The padding acted as a low-tech shock absorbing system that actually worked throughout the rest of the night.

An older man in a suit visited the exhibit and started to hit the drum very calmly. After drumming for a couple minutes he started to hit it harder and harder until he began to resemble a young child playing we his new toy.

As for tonight’s attendance: less than expected. Approximately half to one fourth the number of people that attended last night’s showing attended tonight’s showing. I am worried about tomorrow’s attendance being dismal — we’ll see.

Set Up

Tonight’s set up went without a hitch — well, almost. The amount of AMPs allotted to 216 Kent Square proved to be a problem. With both RevoOVER and Frightling on at the same time the circuit breaker kept getting switched. The solution was to unfortunately turn off RevoOVER during the exhibit.

We’re Live

The storm live on the wall (Credit: Sabrina Killian).

Frightling proved to be a very popular exhibit — especially with young children. This one little girl, maybe six years of age, continually came back to the exhibit to play the on the frightling exhibit.

Truman Capone, the director for Virginia Tech’s School of Visual Arts seemed to enjoy pounding the drum to see how fast he could calm the storm.

After a while, Frightling became almost a game through the night as people tried to see how fast they could calm the storm.

Vibration Sensor Problems

The vibration sensor continually fell from the drum as the public would hit the drum too hard. In testing, we seemed not to have this problem however you never know what the public is going to do. After an hour into the exhibit, disaster, some one broke the sensor in half. We kept the storm going, but without the essential drum prop. Greg resoddered the sensor together, which fixed the problem temporarily for the rest of the night.

Tomorrow we are going to either duck tape the sensor to the drum, or utilize the motion sensor instead for data.

When we set up the storm we noticed a problem: The lights in the room appeared after the lightning on the screen. This resulted from a pseudo bug in the system: The audio for the thunder always happens in the real world after lightning appears in the sky, so that’s how we had set up the animation. Unfortunately, the audio for the storm needed to actually appear before the strike so that the audio could go through the system which reads the audio data and transfers that data into the amount of AMPS given to the lights.

I added a dealy variable to the text file and a delay script attached to each lightning strike. Now each strike will delay for X seconds, and this value X can be changed on the fly by Bobby’s software.

http://www.chrisritter.org/lightning7.html

Today we found out that we had some serious problems to work out.

Problem 1: The lights flash way after the strike appears

This just looks silly.

Problem 2: The projector cannot broadcast from the ceiling

This was a problem that at least had an easy, dirty solution. We broadcasted the image backwards on the screen and set windows to display the image backward. 

Problem 3: Not enough lightning strikes

We only have five lightning strikes thus far, which isn’t going to cut it for the actual piece. Greg and I have to put together at least double that.

Problem 4: Backwards projector = backwards controls

The projector being backwards on the screen meant that we no longer had access to the controls. I came up with the solution of just using a screw driver underneath as a switch over the on button — as that’s the only control we really needed.

Today was our first serious look at the room that we have for set up (we have seen the room once before, but we never knew we would even be able to use this room). We set up the projector to run our early version of the lightning storm and it actually worked pretty well. I think after one more day of work we may have something pretty awesome.

One problem that we will have will be sectioning off the room: A table, computer, and various RevoOVER supplies are in the room. We will have to use a curtain (as Simone suggested) or some kind of wall to seperate the room.

Over the last couple of days Greg and I have been hard at work drawing and animating the storm to appear more realistic.

The Rain

With the rain I rotated the angle 180 degrees, because Greg noticed the rain I had drawn/animated didn’t go the same direction as the clouds.

The Lightning

Greg redrew his lightning strike using YouTube videos as reference. I applied the shaking effect and now we have some pretty realistic lightning to work with.

Link: http://collegiatelabs.com/chris/lightning6.html

Here’s a big milestone. 

Lightning

Greg drew out each from of the lightning strike in photoshop (to get the most realistic strike possible). The strike is then imported and animated frame by frame in Flash and I have purposely shaken the frames a bit so that the strike shakes a little… This is defianetly not something we will use in our final project — but proves that we can integrate the lightning successfully. I told Greg that actually don’t like the strike we have now, so told him to use real strikes from you tube and examing them fram by frame to generate a more realistic strike.

Rain

The rain had some big changes over night. I integrated the rain program into the main flash movie. Greg commented to me that something didn’t look right. He suggested angeling the rain. I added a rotation to the rain and now we have a much more realistic rain product. I also changed the size of the rain to be much smaller (and realistic).

Clouds

Greg commented that I had the clouds backwards, the big clouds were supposed to be at the top, and the small clouds at the bottom: I fixed this as well in this animation.

I have just written my first attempt at creating rain. The script works by using 65 rain drop dots that are each assigned a random scale value, random alpha value, and random position on the screen 30 times per second. 

Link: http://collegiatelabs.com/chris/rainprogram.html

The current version could use some work. The rain appears to be a little longer than I had imagined, not to mention I feel as though we may need some more drops in the final project. In any event I feel as though it’s a good first attempt at creating some good rain for our project.

Over the past week Greg and I have been talking about how we were going to draw/animate the clouds properly. As discussed earlier, we have problems with the cropping artifacts showing up in the animation. Greg has taken every cloud (we have 36) and blurred the edges and added synthetic components to make the clouds look more real — kind of ironic that we had to make them more fake to look more real.

Over the last week I’ve written a good deal of ActionScript to dynamically control the cloud animation. The clouds are split into three levels the front level moves the fastest, the middle clouds moves at the middle speed, and of course the back clouds move the slowest — this is to make the sky seem more real. As the clouds go off the screen they are queued so that we have an unlimited supply of clouds. You could what this animation for days and it would constantly generate new formations because of the queuing system combined with the variable speeds of the clouds.

As for I/O, Bobby’s program that simply takes vibration data and writes a text file must integrate with my flash program: the “spped” text is actually the data from a text file on web server. At some point, Bobby’s program will write that text file with the necessary speed data.

Link: http://collegiatelabs.com/chris/lightning2.html