Hi all, I am about to start teaching in a master university course, with game engines in general and UE4 in particular as the main topics. I plan to give a theoretical introduction to game engines, how they usually work, what parts/modules they are often made of, etc. Then I would like to switch to UE4 soon. Since it’s only one semester, I dont have much time, so I will only be able to touch some basic things like the editor in general, maps and levels, assets, actors/pawns/characters, the Blueprint system, shaders and materials, some AI, particles, animations, some C++. I know there is much more to say about UE4, but, yeah, time unfortunately is quite limited …
Now I am thinking of what a good way of performance assessment could be. Theoretical test(s)? A small project to create on their own? Anything else? The decision on how this assessment will look like is mine, but obviously it has to be fair and comprehensible. For example, if I ask the students to create their own small poject over the course of the module, what could be a the indicators for me to be able to compare those different projects in a fair manner? Or should I go with a theoretical test, which on the other hand might not be the best choice for such a topic? Or maybe a combination of both?
Your feedback and thoughts in general are very much appreciated!
+1
Exams are not productive in any ways.I remember,when i was doing a maya course years ago,they decided to make a written test and it was a fail.Why?Because you understand more about it if you do it in the program and you visually see your mistakes.(this is not math).
-Make one small level and make sure everyone copyes what you do.Introduce third and first person perspective so that they feel more motivated.You will notice that some people will do it in a different way.That is the start of creativeness being awaken!Let them do so.
-Than with the same assets tell everyone to make a small personal game level(make sure you keep restrains on them and that the level is small.Two- three rooms at most or they are lost!)
-And at the end,try to pair them in groups of x2 so that they can make a small,more complex game together.This is the most difficult part as one of the two students will try to rise and force his way on doing things on to the other more ¨¨timid¨¨ person.DONT let that happen or they will not learn nothing.Be the bridge between them.Even when you guys are having a small break,go and drink something in the coffee shop with them and make sure they chat between them about the work.
For the sake of all good that is left in this world, drop all thoughts about doing written test of any kind. Assess your students in form, that is alike to game jam or something like that.
I would echo what most people are saying and strongly advice against a written test.
If was a taking such a course I would want to end the class with a fully functional prototype that addresses all major areas of game development in the engine.
One suggestion would be to work on a single prototype during the semester. Start with a simple template, like the thirdperson template, and as students learn about the different topics they add new functionality to their game. So, once they learn about animations, their assignment is to add X number of animations to their characters. Once they learn about materials, their assignment is to make X number of materials with specific properties, etc
At the end, their prototype will have custom meshes, animations, materials, BP and C++ interactions, and even cutscenes!
You can grade it on the efficiency of code / BPs and creativity. Maybe you ask them to create a specific system and see how they solve the problem?
Talk to each student. Have them explain what they’re doing, what their challenges are, and what approach they are taking.
The good ones will know the surrounding lay of the land of their challenges, and may even be able to talk about multiple options and pros and cons.
The bad ones will “just mash it” or “try things until it works” or maybe “I know it when it feels right.”
Also, make a small project. Look at the solution. How does it perform during edge cases? What if you order -1 beers at the bar? What if you jump and hit your head on the underside of the moving platform? What if you set the end goal zone on fire? And so forth.
This is actually a lot of work for the teacher, which is why written assessments are so popular – the seem impartial, and they require minimal effort to prepare, give, and rate, and they lead to no discussion of partiality or unfairness. It’s just that they aren’t particularly good at assessing real skill.
Thanks for all your feedback, guys. I definitely don’t want to have written tests, that would be my plan B or even less. I would like to have small projects instead. My only concern is that it might be hard to compare all the solutions of the students, if they are working on different projects. And that at least a few of them would complain about their marks. With a written exam it is easy to mark in a comprehensible way …
I understand your challenges. There’s a lot of things that can be done using game engines.
Yet, if your focus is on game dev., i assume a set list of features that have to be implemented in a final project is a good start to make project evaluation relevant.
I read this branch and I understand how backward my university was!
Teacher asking on the forum how best to teach his students)))
In my college teachers have been taught by the old training assignment assistance for years without changing anything in their training programs(((
Well, you know, teaching is not my real job, I am a game developer, doing that teaching stuff just for fun.
And I told myself, if I really do this, then in such a way that it makes sense for me any students …