If you don’t have 3D glasses, you can click the ‘In 2D’ button above most of the 3D videos. If you would like to get 3D glasses, there are places on the web that will send them to you for free, if you send them a self-addressed stamped envelope. Find them here. You should ask for Red/Cyan glasses.

“Post mortem” might be an overly gruesome term to use for this retrospective evaluation of the 2010 bird house. After all, unlike 2007, no babies died this year. Still, not everything went as well as I had hoped.

On the plus side, I got tenants, a nest, and babies. I got color video with audio, and the viewing angle was good – for some things. The ability to reposition the cameras was critical for capturing both the birds’ explorations of the house before building the nest, and the babies’ activities later.

There were lots of negatives, though:

• • I got the same old boring wrens as last time.
• • The 3D quality wasn’t what I had hoped for.
• • My infrared LEDs burned out (I had suspected I was putting too much current through them – I guess I was right) but even when they were working, they did a lousy job (probably the camera’s fault more than the LEDs’).
• • The videos had a low frame rate, making for jumpy images (this was partly due to the low light levels inside the bird house, and partly due to the underpowered computer).
• • The ability to reposition the camera was good, but not good enough. I would’ve liked top views for the egg laying and incubating, and side views from the other side when the parent(s) got in the habit of standing in front of the cameras.
• • Once the eggs were laid, I was scared to approach the bird house to reposition the cameras, lest I cause the mother to abandon the nest.

I have elaborate and ambitious ideas for addressing all these problems in the future:

• • Making a platform for robins instead of a house for wrens would help me get different birds.
• • An open platform would allow me to put the cameras an arbitrary distance away, improving the 3D effect. (Unfortunately, they can’t be an arbitrary distance away from the computer.)
• • Instead of infrared LEDs, I should have another camera with true night vision. This might be expensive and hard to find.
• • A platform would have a lot more light on it than the bird house had in it. This might solve the video frame rate problem. However, I might also need a faster computer. That would definitely be expensive (relative to the value of this endeavor).
• • A camera pointing at a platform would most likely be mounted on a telescoping pole. This would give me a lot of flexibility in positioning it.
• • And here’s the elaborate and ambitious part: I could have a remote controlled motorized set-up for moving the camera. That way, I could move it without scaring the birds off (at least not permanently).

I make no promises as to whether or when I’ll do this. Expect it when you see it.

They all left today. I came home and the bird house was empty (well, except for a nest of course). I have video of them departing, but if you’re hoping to watch them march out in an orderly way, you’ll be greatly disappointed. It may be unseemly for me to criticize my own videos, but here goes:

First, the birds have gotten a lot of dirt on the clear plastic wall that protects the camera from them. The video suffers accordingly. Second, the birds are scrambling in and out of the doorway, so the light is obstructed most of the time. The video is dark accordingly.

I pieced together a few scraps of typical video. It’s dark, there’s lots of movement, and lots of noise. But at the end, there’s no one left.

Here’s something a little puzzling. In the following video, the parent arrives (let’s assume it’s the mother), she stays in the doorway a moment, as indicated by the intense darkness in the birdhouse, and gives the babies some food. Then she comes into the house and gives the babies some more food. Then -- and here’s the puzzling part -- she reaches to the doorway, gets some more food, and gives that to the babies.

So where did that food come from? Is the other parent sitting outside, handing her food, or has she parked some food in the doorway, using it as a shelf?

I tend to think the latter. If there were another parent there, he’d be blocking the light. Also, if he were there, why wouldn’t he just feed the babies directly? So probably the mother just came to the birdhouse with a lot of food, put most of it aside, and the fed the babies a little at a time.

By the way, I’ve given up on 3D, at least for now. Running two cameras is just too much for the PC I’m using. It works, but the video freezes up when there’s a lot of motion. Also, the cameras are too close to the action to get good 3D. However, just as this birdhouse was more ambitious than the last one, I have ideas for the next one that improve on things.

Now, here’s the video:

As I foretold, the birds are bigger and the videos are better. This video shows three different feedings within five minutes (I’ve edited out some of the waiting around).

I’m back from my brief vacation, and I was happy to see we now have baby chicks. There’s lots of videos; a parent enters, feeds the babies, and leaves every 3 to 5 minutes. That’s a lot of feeding!

I present for you here a single video with two feedings. It leaves a little to be desired: the chicks are barely visible at the bottom of the screen (this should be less of a problem as they get bigger), the parent blocks our view of them, and the action is so close to the camera that the 3D effect is not working well. So we have 2D only today.

No, not that sort of technical difficulties – everything’s working fine. I haven’t posted recently because there’s still nothing interesting going on. Clearly, the next big news will be when babies hatch. I don’t know if the eggs have been laid yet; clearly there’s an advantage to having a roof-mounted camera as I did three years ago. For the best of both worlds, maybe I need multiple cameras in multiple positions.

But during this hiatus in the action, I thought I’d share some behind the scene struggles with the technology (thus the title, “Technical Difficulties”).

I’m using Microsoft’s LifeCam Cinema web cams. They’re good: they have remote-controlled focus and zoom, and very high resolution (for a web cam). The recommended PC hardware for them is a 3GHz dual-core system with 2GB RAM. The computer I’m using has half that power, and I’m running two cameras on it. Miraculously, it mostly works.

But I do get crashes every few hours. I’ll be out of town for four days in June, and crashes like that would be unacceptable. Fortunately, I’m a software engineer. It was a simple task to write a program that checks every minute for a crash dialog and gets rid of it. Then, it checks if the web cam program is running and, if not, starts it up.

I accumulate about 70-100GB of video a day. In just two days, the disk on the system I’m using would fill up. Fortunately, I had a 500GB external drive lying around. It wasn’t doing anything important – just backing up my vital data. But as those of you who follow my Galactic Studios site know, I have a brand new PC (in a very unusual case that I built myself) with brand new disks. My data is probably safe for at least a month. So the backup drive is being used for videos.

Night vision in the bird house has been a nightmare. I have infrared lights in the bird house. But the web cams that are so good in daylight do something very bad with IR light. They show a large, bright white spot in the middle of the image. Even in daylight, it ruins the picture. It’s weird, since the IR lights are not, themselves, anywhere near that spot, or pointing towards it. It’s clearly an artifact of the camera. I don’t think it’s caused by the fact that IR light focuses differently than visible light, because changing the camera’s focus doesn’t eliminate the spot. I think it’s a problem with the sensor.

Anyway, I set up a timer to turn the IR lights on at night and off in the morning, so the daytime image isn’t spoiled. But I don’t have high hopes for seeing anything clearly at night either. Three years ago, each egg got laid around midnight, and I had hoped to capture that this year. I don’t think I will.

One bit of good news: I got a free wireless mouse from Microsoft. The web cam settings aren’t being saved correctly, and Microsoft technical support couldn’t figure out why. So they sent me a free mouse to curry my favor and win my forgiveness. Who would’ve thought Steve Ballmer could be so humble?

Yes, birds really do feather their nests, and here’s the proof. This video shows the wren going in and out of the bird house with feathers in her mouth, and depositing them in the central depression. It’s a little strange that she leaves the nest with a feather about as often as she comes in with a feather. Wouldn’t the feathers all be flying in one direction, into the nest?

Through my own clumsiness (and a certain program’s lack of error checking) I lost the video from the right camera, so we have 2D only today.

I feel a little guilty about posting new videos, but honestly, there’s nothing special going on. The wren now spends nights in the bird house, but not in the central depression. She sits on the side. She wakes up around 5 AM and spends much of the day looking out the door. She’ll occasionally sit in the center, but it’s not her favorite place.

The cameras are positioned a little above the side of the nest. They can’t go any higher – a shortcoming I’ll remedy if I ever do this again. Sadly, I don’t think we’ll be able to see the eggs. We will, however, get great videos of the chicks being fed.

If you want to see eggs, check out the videos from the my previous bird house. You can see the summaries of all my postings from that bird house here. The egg shots are clearly labeled (click on the article titles to see the whole article, including video).

This is just a quick note to let people know I’m still here and checking the bird house every day. There’s just nothing new happening. The bird still comes, dives down, and doesn’t do much of anything.

At least that’s what I hope is going on. I hope it’s not that my view is blocked by the side of the nest. However, three years ago, the first egg wasn’t laid until May 31^st. So I’m probably OK.

Well, the time lapse strategy didn’t work so well. Rather than showing progressively more sticks, as the first time lapse did, it just showed grasses being shifted back and forth. (I guess it’s interesting that grasses have been added to the nest, but visually, it was pretty dull.)

So I’ve switched back to motion-triggered recording. And I have zillions of videos showing the next stage of nest building. The wren comes in, dives down under the surface, then starts shifting things around. I suppose she’s weaving the grasses and sticks together to tightly form a cohesive nest with an egg laying area in the middle. Occasionally, she’ll poke her head up, then dive back down. It reminds me of a submarine.

All the videos look pretty much the same, so here’s just one sample: