Visualising Mechanical Heritage in Motion

The matchlock's mechanism is simple, but firing is over in an instant, so it is hard to explain with pictures and words. That makes it a good fit for letting people control the speed themselves. Also, Kansai University has worked with Sakai City on matchlock research for a long time, so we already had scan data and earlier CG and AR work to build on.

In 2021, we used a 3D scanner on a matchlock owned by Sakai City, and we also took photos for the textures. The scan data was too heavy and its surface was not smooth, so it was not good for real-time use. So in 2022, we made the model again by hand, using the scan data as a guide. In the new model, every part is separate, so the parts can move and come apart.

The controller tells us how far the trigger is pulled, as a number from 0 to 1. For the trigger, the sear and the serpentine, we set two angles: one at rest and one at firing. Then we use this number to find the angle in between, with quaternion interpolation. So the parts move exactly as much as your finger moves. It is not a video made before.

Many visitors were wearing a headset for the first time. With MR, they can see the real room around them, so they feel safer and less nervous. Also, they can see the printed stock in their hands together with the CG gun, so the real size is easier to feel.

There are three reasons. First, the real gun is about 123 cm long, so it is hard to hold it with one hand for several minutes. Second, if the real object is long, people can easily see that it does not match the CG gun exactly. Third, a stock is light and easy to carry: many of them fit in one suitcase. It also makes it harder to hit other people by mistake.

During the exhibition, our staff stayed with each visitor and showed them how to use each function, for example, pulling the trigger slowly to watch the parts move, taking the gun apart, and showing the part names. So most people did notice the mechanism, and many told us that they understood it. In the questionnaire, 98.5% gave 4 or 5 points for watching the firing structure, and 33 people wrote in their comments that they understood the structure better.

Of course, left-handed people can play too. We only meant that for right-handed people, the action is the same everywhere: with a real gun, in computer shooting games, and in our system, they use the same index finger. So it feels a little smoother for them.

This is a question we take very seriously. Our university's ethics committee approved this study, and because the matchlock is a gun, they asked us to follow stricter rules than usual. In the game, the targets are old objects from history, not living things, and when you hit one, there is no injury or blood, only a particle effect. However, the game has not been given an age rating yet. In the future, when we show it to younger children or in other countries, we would like to put a short notice at the entrance.

You are right. We did not give a knowledge test, so we can only say that people felt they understood the matchlock better. We cannot say that they really learned. Next, we would like to give a short test before and after the experience and compare the results. We are also thinking of a quiz inside the game, where you shoot the target with the right answer.

This study was done at an open campus event, and our main goal was to see whether visitors would accept this kind of experience. So we did not have a comparison group. Because of this, we cannot tell whether the good results came from MR itself or from the content. In the future, we would like to use the same model and content, and compare MR with 3D on a tablet or with a video.

Next time, we would like to write the questions as statements, such as "The controls were easy," and ask people how much they agree. We will also make 5 points always mean the best answer, use a standard questionnaire such as SUS, and add short interviews.

The exhibition was very busy, and most visitors were high school students. We wanted a short questionnaire that was easy to understand, so we made nine short questions ourselves. The weak point is that it is hard to compare our results with other studies, so next time we would like to use a standard questionnaire.

Yes, that is possible. However, the questionnaire had no names, and it had nothing to do with entering the university. In the comments, some people also wrote what they did not like, such as labels that were hard to read, or that they wanted the gun to shake. Also, on Q1, easy to operate, and Q8, understanding, teenagers and visitors aged 40 and over gave almost the same scores.

That is possible. We think there are two reasons. First, the labels were hard to read: when the parts were moving, the words were hard to see, and some people could not read the kanji. Second, pressing the stick is a move that young people have used since they were children, but many people aged 40 and over are not used to it. Both point to the same answer: make the labels easier to read, and use a button that is easier to find for showing the part names.

We think three points can be used again. First, use a kind of game that visitors already know as the entrance, for example, a rhythm game for a machine for making cloth, or a puzzle game for a lock. Second, make only the part that people hold, and show the rest in CG. Third, let the visitor's own hand move the mechanism step by step. This works well for machines that move too fast to see. But so far, we have only tried this with the matchlock.

We have already shown it many times in and around Osaka, and also at an academic conference abroad, and more exhibitions are still going on. Also, one elementary school wanted to borrow a matchlock model for a class, but it was hard to lend one. Our system is light and easy to carry, so it could also be brought into schools.