Today we did the final touch ups for the game, got all of the documentation in order, and packaged it up. I redid the prison pipes textures last night, and have now implemented them in the game. The old pipes were far too realistic-looking, with rust, and old paint all over them, and they really didn’t fit. Since they are quite abundant in the level, I thought it would be worth while redoing them for vertical slice. The result is a nice, clean PBR texture, with a hand painted coat to make them more cartoony. With everything I’ve made implemented into the game, I focused on the GDD this morning. With a bit of spare time, I took some screenshots of the game, to give examples of the textures. Now that we have a pipeline for every aspect of the game, and now that we have all of the building blocks we need to create levels, I'm really exited about the next stage of development. I think things will be a lot quicker from this point on, and we can really focus on the level desings, which are really the most important thing. One thing I still need to work on is my documentation. I took a very relaxed approach to Hack n' Plan, and going forward I want to make sure that I update and log my work every day. Creating the textures for the game has been really fun. Texturing both the character and most of the environment , I've really learned a lot, and have even added some software to my skillset. Going forward, I want to develop my texturing skill set further, with an emphasis on environmental modelling and texturing.
0 Comments
This week was crunch time on the texturing. Most of the 3D props in the scene are completed, but the level itself was bare, or was using placeholder textures. I spent a bit of time researching cartoon PBR tileable textures, and I applied my new knowledge of texturing in substance designer to create textures in photoshop. I’ve found that creating PBR textures entirely by hand has lead to a far more cartoony look than using Substance Painters PBR materials. These are good for a base, but certainly require a final layer of hand. I’ve created a series of PBR handpainted tileable textures to be used throughout the Area 52 level in order to create a cohesive visual experience. Some materials suit an industrial cargo area, others work well for generic sci-fi walls and vents. All of them use the same palette of greys, and I’ve added a coloured wall panel to try and liven up a few areas of the game, as the level is quite desaturated and grey. We’ve used emissives and bright particles to brighten things up, and the main character is bright orange so as to stand out on the dark background. I went through the GDD today and wrote up most of the information relevant to art. The game is pretty much entirely textured now, the only texturing left to do now is going over objects with a hand painted layer. The textures involved a lot of trial and error. I've never really made tileable textures before, especially not cartoony ones, so most of them required quite a lot of tweaking. At this point, however, I am confident that texturing from this point on will be much quicker now that I have established a pipeline. I also made a decal to act as a sign within the game. It's a simple BARRACKS sign, to let the player know that enemy guards will be in the next area. We wanted to lighten the mood a bit and add a funny element, so I added some graffiti that the guards have scribbled on themselves. Below are screenshots of the textured applied to simple cubes in-engine, as well as a screenshot of the textures in-game. This week I’ve studied Jack and Daxter to figure out how they did their texture work. From what I can tell, they didn’t use normal maps in these games. The type of texturing that we want for our game would be a modern version of this cartoony style of texturing. Classic platformers didn’t have access to the modern software we have now, and I think this really informed the style. Hand painting all of the details in is really important, but I’ve also been making my own alphas so be used to create normal detail as well. I took a pattern from a Jak 2 wall texture, and decided to try and recreate it myself. Not being super confident in substance designer yet, I recreated it in photoshop to get a better understanding of the basic forms and layers of the alphas. I actually ended up using my alphas for the albedo as well. The lightest shades protrude the most in the normals, and the darkest recede, so it worked quite well. I textured a cube in substance, kept it really flat and cartoony, but just added a little touch of nose and dirt to break up the flatness a little bit. We’ll be texturing everything in a similar manner from now on. Using flat colours and hand painting as much as possible, just using substance painter to add a little extra texture and get the roughness/metallics just right. This past week was spent with a few team members at Armageddon. The week prior was spent getting a demo ready to take down with us. We ended up making a far simpler version of the first level. We cut out most of the narrative and made a gauntlet style lever where you progress through a series of obstacles. This way we could test each mechanic in an isolated manner, and ask people for feedback on each section as they went along. Armageddon was a really great experience. Scott, Juane and I drove down to wellington on thursday with all of the PCs, monitors and other gear, and we set up the booth. On Sunday we showed the game, and were really happy with the turnout. Children flocked to play the game, and there wasn’t a minute where either computer wasn’t occupied by a child. Children also give very good feedback, as they’re brutally honest. All in all, while our build was a bit last minute, it worked well, and we got a good feel for what works and what doesn’t. We know what we need to prioritise, which is level design (making use of the mechanics in the most meaningful way possible) and fine tuning. Kids liked the character, and most of them got really existed when playing the game. While we were up there it was up to the programmers to keep working on the game. Juane did manage to make a few small art assets on his laptop in his spare time, so we implemented those while we were there. I managed to get the basic textures down for the main character before we got down, so that players would get a decent idea of what the finished product would look like. Going forward, my aim is to texture a prop every day. BCT has a massive deadline next friday, so this way at least my work on the game will keep ticking over. I really need to divert my energy to my other assignment, which is really frustrating because I’m really excited about the game now and really just want to work on that. The texturing is going to have an overhaul later on, so for now i’ll be trying to keep it simple and flat. We’re going to go for a hand-painted look, and so far the textures are looking a little too realistic. The good thing with textures is you can always revisit them, but I don’t want to leave too much texturing work to be done later on. Week four was a very productive week for the team. With Wellington only a week and a half away, we still had a lot to do. My main goal was to start texturing the environment and characters so that they’d we could start seeing the atmosphere and mood of the game. After a lot of research, I decided to teach myself Substance Designer and Substance Painter, so that we could produce stylised PBR materials for our world. Learning this software has certainly slowed things down initially, but after completing the main enemy for the first level in painter, I feel that my skills have improved massively. Learning Substance Designer was a little trickier, as its node based, and logic isn’t exactly my strong suit. My first attempts at a tileable metal floor tile for the prison have turned out better than I thought they would, but may need revisiting later down the track to try and get them to look a bit cartoonier. The textures for the guard have turned out really nice. Again, I may need to revisit some models later to add a layer of hand-painted texture, but for now i’m really happy. The design tenants that i’m trying to stick to texture-wise is to keep details low, keep colours saturated and dial down normal detail and really any noise that I can in any of my textures. By keeping things subtle, flat and brightly coloured, I think that after a bit of practice, i’ll be able to get a good pipeline down. The guards need a weapon, which isn’t going to be in the game by wellington. I have a few rough ideas for it, which I will be developing further, in about 2-3 weeks time. We want to keep our game violence free (well, relatively), so the guards won’t kill the player. Instead, they’ll zap the player, stunning them. For this, they’ll have a cool, old-school looking lazer gun, that has an electric eel strapped to it for ammo. Other version of the weapon could be slippers rubbing on carpet, two balloons rubbing together, etc. There will be one gun model, with different ammunition modelled separately that will snap on. Perhaps different ammo types will indicate the strength of the gun/enemy. For now, Scott will be animating the guards with a rough greybox of the weapon until the model is replaced later on. Ideally, by next Wednesday, the first level should be done and textured. We have most of the puzzle pieces floating around in different unity scenes, I think that Monday should ideally be trying to get them all in together. That way we can see how the game feels, see how all the systems interact, and go from there. This week, we decided to sort out the overworld. We had a vague idea of what we wanted it to be, and we only need a rough version working for when we implement the other levels of the game. The Vertical Slice will end with the overworld, and the other levels greyed out. We decided that a 3D map of the ‘world’ - essentially a little island where the game takes place - would be the most effective. It’s essentially an interactive menu, in that instead of scrolling and clicking with a mouse or controller to select levels on a map, you instead drive the protagonists van around, selecting levels that way. We took inspiration from Overcooked, but we took it a step further. We implemented one button driving and drifting. Our game will be targeted at children, so we thought that an more interactive menu that also adds context for the overall experience would be richer for this demographic. Our van would hoon around little city blocks, and off into the desert and other areas, banging into stuff, skidding around and making a bit of a mess. We are keeping in mind however that it shouldn’t take too long for a player to drive to a level, so we’ll be figuring out what the longest time should be when traversing the map between level, and never go over that. The map will be easy to make, with very low poly modular assets, and objects lying about the place that the player can send flying when bumped into. We currently have a flat textured van model which we will be updating later.
I spent most of the week optimising my security guard model, and getting it ready for texturing. I talked to Scott, who has taken on the roll of rigging and animating, about what would be best mesh-wise for his rig, so we’re confident that the animations should look really clean and juicy. I had to re-do a few parts of the model because as I was handing it off to scott to rig, I noticed that the shoulder topology was really weird. It seemed okay, but I could see that the edge loops were all wrong. I spent half a day looking at reference and asking other students for help before I got it to a level that I thought was sufficient. While our guard character would most likely not be posing very extremely, I was worried that later down the road if we decided he’d do an overhead attack or reach up to grab something, that the mesh deformation would be really noticeable. I decided that the extra time spent now would be worth preventing possible problems in the future. I spent the tail end of the week in Substance Painter. This was my first real foray into the package. I had previously used it for the prototype to texture characters, so I knew my way around, but I had only previously used basic paint layers. This time round I had to learn how to use generators, masks and a few texture maps i’d never used before. After a lot of frustration with generators, my research led me to curvature maps. I had previously not used these before, and in all honesty had no idea what they were used for. I tried to generate a few of these, but had horrible results. After again consulting my classmates and researching further, I did end up with a reasonably decent curvature map. This week ends with me having finished the characters mesh, and having a substance painter session set up for texturing early next week. I spent a lot of time this week in new ground, with a lot of trial and error and perhaps wasted time. I am heading comfortably into next week however, knowing how the pipeline works for future models. I’m hoping that the time spent this week will save time exponentially as we head further into the project. A bit of a theme this week ;) Vertical Slice #2
Week two, and the first thing on the agenda was to build a world for our game to live in, and a character that would make sense in this world. After a lengthy discussion over a couple of days, and a few character concepts, we decided on a setting and a character. The setting for the game would now be Earth, in an alternate future (or present). The protagonist would play as a dispossessed alien, imprisoned deep within Area 52 for scientific experimentation. The player would be a small, fat alien, very simple in design, reminiscent of characters like crash bandicoot, ratchet or Daxter. An engineer by trade (perhaps from a planet of engineering aliens), with no physical aptitude at all, the character built themselves a marvelous robotic arm that could be used to perform a myriad of tasks, originally intended for engineering purposes. Abducted by earth scientists, and their arm taken for testing, they are trapped within the confines of the prison block of area 52, and the first level of the game takes place here. The player must escape their cell, traverse the level, which consists of the prison area, a scientific lab, and a parking garage, retrieve their arm, and leave area 52. The arm is a new addition to the game, and is our way of contextualising the crazy arm physics we had in our original prototype. Originally the character had two crazy arms, but we realised that one would serve the same purpose. Our character (who is currently unnamed) IS the arm. They are nothing without the arm, just a feeble alien, and we want to make the game revolve around the use of the arm. The arm will have a lot of utility within the levels, and part of the fun for players will be figuring out how they can use and apply the arm to progress further. We also talked about gadgets. Gadgets have been on the table for a while now, and at this point, we aren’t going to have them in vertical slice. The arm can be used to achieve what most of our gadgets were going to do, but with a lot less work. If we can justify gadgets later on, or the game feels like it’s lacking something, then we’ll revisit them. We definitely like the idea of gadgets, and we’re going for a stealth/combat-less version of old school platformers that we played as kids, so they would definitely fit aesthetically/thematically. Our character is going to have a reasonable intense rig. We want to have crazy face deformations, and extreme facial features. We’re going to use blend shapes for pre-made facial expressions to animate the characters face efficiently. One of the most difficult things that the programmers are going to have to figure out, is how the arm is going to work. We’re going to need it to stretch, and this poses a lot of technical problems. Juane is our technical artists, and he has a good background in programming, so he’ll be working on the rig, and will be working on getting the arm working alongside the programmers. I spent a day or two whipping up concepts for our little alien character, the environment, and the game’s first enemy, the area 52 security guard. Juane then took my alien concepts, and made the 3d character. He unwrapped and rigged the character, and i’ll be texturing it later on. We had a few rules for the character, that he had to be short, to make hiding easier to facilitate, he has to be really simple, so that the crazy arm takes the spotlight, and he has to be bipedal, so that kids can project themselves onto the character. I made the base mesh of the security guard off my concepts. In my concepts, I had angry human faces on the guards, but we decided that having a helmet/mask would make the guards feel less human, and a bit scarier. Also, our world does have aliens in it, so we didn't want to make the guards blatantly human and bring up some kind of political aliens vs humans agenda. We’ve made a pinterest board for the environment, so once the world is greyboxed, we’ll be doing proper concepts. We decided to go with codename ‘Grand Theft Office’ going into vertical slice. Initially, we decided on ‘Vertical Slice’, out samurai robot fighting game, but after to talking to tutors and a lot of other people, we decided that the issues we were going to face would be too difficult for us to overcome in the short period of time we had. Over the holidays, I worked the entire time. I didn’t really have a lot of spare time to spend on uni work, but I did manage to steal away for a good day’s work at uni. I talked to Scott and Josh, who had been in the previous day, and they had written up a bunch of ideas for the game. We solidified some design rules for the game, and thought about where we could take it further. Once uni kicked back into full wing, the whole team got together, and we spent the first two weeks reconceptualising the game. We wanted to keep the core of the game in tact - stealth/stealing/humour - but we wanted to take it much further. Our initial prototype was a simple room, and the gameplay loop was very short, and abilities and points were arbitrary. Going ahead, we decided that the game would have an overworld, and that the player would have a series of levels to choose from, unlocking harder levels as they went further. We began to discuss how we could redesign the character, as, while the current character was pretty cool, they weren't that interesting, and we didn’t really think that they could grow much along the way. We wanted a character that could grow as the story progressed. We realised during this discussion that we needed to further develop the setting of the game, as the context would define the character. We needed the context of the world to fit with our core principles. We showed our game at Chromacon before the holidays, and the reception the game had was really exiting. We had a bunch of kids lining up to play the game, and they seemed to really enjoy it, as simple as it was. One comment that really solidified the game in our minds was something a mum said while her son was playing the game. When asked what she thought about the game, and the stealing aspect (which we were slightly worried parents wouldn't like) she said 'I don't care if he's stealing, i'm just glad that he isn't killing anything for once'. By the end of this first week, we have a good idea of what needs to change and be developed further. We have a good grounding, and we have rules set in place so that the game doesn’t stray too far from it’s roots. |
AuthorContrary Scholars ArchivesCategories
All
|