PlaytestingDuring playtesting I was hoping to get feedback about the TRUST in the game, however players didn't even make it to a point in the game where it was introduced. The puzzle mechanics were broken, the camera was extremely non-friendly, the character movement was game breakingly buggy and the game as a whole was appropriately called 'Asset Soup'. Things wrong:
I put on my leader hat and assigned rolls to the team according to importance and time needed to fix it. After all the system work was delegated it was my job to redesign the whole world... again. I wanted to focus on the tutorial area and insure players wanted to progress through the game. In the playtest most people quit in the middle of the tutorial due to the unfriendly nature of the experience. I also needed to make it feel 'not like Journey'. A whole new worldI feel that Journey has somewhat ruined any games that are set in a desert. Sure i'm deriving inspiration from the game, but the fact that even having sand is a "Journey Ripoff" really annoys me. Also since we don't have animators the only road we could go down was that of the cloth people, I admit, this does relate to Journey's character design. So. How to make a game feel like journey but not look like it? I went with a suggestion I got, and it worked. Night Time. But surely just by making it nigh time wont fix my problem? That's why I made sure right from the beginning of the game the players are presented with Aztec style objects to make them know that this is no 'Journey'. TeachingThis was a big problem since every player needed to be told how to do things by the devs instead of having the game teach them. I imagined these little followers of the gods telling you controls, so I made a little 'teaching totem' that would tell you things based on your proximity. Mechanics OverviewThe mechanics in the game had to be based off of a prompt, I chose 'Praising'. They also had to be related to trust in someway. The praising mechanic comes from the praising of the shrines which keep you safe from the fog.
Mechanics:
A & O's trustThe trust in the game comes from the praising mechanic. Basically there is going to be 4 types of shrines to 4 different gods. Only one of them is guaranteed to keep you safe, one is always going to fail you and the other 2 are going to be hit and miss. By experiencing the praising enough it is assumed the players will come to trust the shrine that is keeping them safe the most (all shrines have a different look to them). I feel that this is trust, however a weak form of it. I won't to make it stronger, but with the current game it has pretty much reached its capacity of SCOPE. Plus having awesome ideas is cool and all, but actually knowing how to carry them out is the problematic thing. More time is spent researching then actually creating, and this leads to all nighters the day before the project is due and a general lack of quality with the finished project- due to the gallery I don't want this to happen. I want the game to be FUN.There is nothing worse than seeing people playing a game you have worked tirelessly on and them just having a blank expression that imposes the funnest part of the game for them will be putting the controller down and walking away. Not many people get to experience that feeling, I envy them; It is by far one of the worst feelings- shame. This happened with my last game and I will do everything in my power to ensure it doesn't happen again. The one thing I felt like I needed to focus on was the sense of FUN the players will feel while playing. For the prototype this is going to be very minimal due to the time constraints. Fun things are generally aesthetic based, sure they will have a mechanic behind them, but I feel the fun comes from the visuals and the feedback. To Puzzle? or Not to Puzzle?I constantly feel that my games lack a certain depth and that they are just an aesthetic journey with no real purpose. Something I have gotten into the habit of is putting puzzle mechanics in my game to counteract this shallowness. I don't like this at all- it's a terrible compromise. I want to remove the puzzles in A & O, but they are already done and if I do I would still be left with the shallowness problem. I need to think of a way to solve this differently.
Now that all design relating to the mechanics was set and stone and good to go I turned my focused to creating all the environment assets for the game. The question was not what had to be done, but how to do it? Thus began my super exciting weekend of taking my knowledge of 3DS MAX to a whole new level! Rocks ROCK!First I began with a cube and a spherify modifier, the turbosmooth then created the extra geometry that would give it detail later in the process. If I felt like it needed more detail, I could just come back to this modifier and change the iterations to a higher number. After this I added two displace modifiers, the first would give it its shape with a cellular map, and the other would give it its detail with a noise map. By playing around with these I was able to get 3 different rocks in a short amount time. I kept the poly count at a medium level as I knew the computers we would be using for our game were not the best performance wise. These 3 rocks later became the 3 big mountains and rock piles you'll see in the final game. This picture shows how changing the turbosmooth iterations produces much higher quality rocks (at the cost of A LOT of polys). I then imported into Unity to see how the cell shader that Tim wrote would look on the rocks. Since the camera is mostly going to be pointing at the ground (constant angle of declination), I just payed attention to the areas close to the base of the rocks. It looked good, thus I started on improving the sand aesthetics. Making believable sand-While keeping in mind I couldn't do a high poly model for the sand due to performance costs, it was very hard to find the right balance between simplicity and detail. This took some time to get right. I began with the editable patch tool in 3DS max, this method produced some pretty awesome dunes, the only problem is that it took a considerable amount of time and it was a very detailed model. I took a step away from this method and began experimenting with the method I used to create the rocks above. By using the displace modifier and a mixture of different maps I ended up being able to make a 'Sand Ocean', but not 'Sand Dunes'. Then I wondered upon the holy grail of heightmaps. By using this heightmap alongside the displace modifier and cellular maps I was able to create a very low poly map that I was rather happy with, as was the team. From this process of a trial-error experimentation, I was able to cut the tri count from a whopping 51,200 to a mere 3200. Shrines of the gods-Philip Jackson does some awesome sculptures that managed to pop up on our FB page. I instantly fell in love with these and think they can fit extremely well with the feel of our game. These are now a big part of the inspiration for the design of the shrines.
|
What is this BLOG about?This is basically my thoughts and whatnot as I improve as a Developer. Archives
June 2017
Categories |