Quick Fixes

Today I’ve been fixing up little bits here and there, rather than concentrating on something bigger…my intention is to clean up some of the game’s little blips before moving onto anything excitingly new. This is following on from my previous to-do list, which as the days go on gets increasingly longer…

  • Apply more accurate collision masks to objects
  • This I have done today, although most of the problems with collisions were solved by reducing the collision mask around the player object.
    Previous mask:

    Based on head dimensions

    New mask:

    Based on lower body dimensions (much thinner!)

    This stopped the player from gradually sliding down corners and standing on thin air! This also means that collisions with obstacles are more accurate, as it is often the body that contacts that obstacle sprite rather than the head. I’ve increased the masks for the mushroom and swinging enemies to accommodate for this reduction, but only slightly.

  • Improve environmental tiles
  • I haven’t done much to improve the tileset, but I’ve cleaned up some of the edges by adding corner details and rotating some of my existing tiles to fit in more instances.

  • Make new sprites
  • There were just a couple of things from the end of the level that I’ve omitted to draw so far, so I figured it was probably time to put these in. Among the sprites still to create was the “end object”- the door to the next level. This stays closed until all the flowers are collected (although I haven’t coded for it opening yet as there is nowhere to go!) The idea for the new sprite came from this lonely little place, that sparked my imagination.



    The orange doormat represents the colours of the next level…

    As well as these items from my previous list, I started to work on some more minor details that I wanted to be ready in time to demonstrate to others. I’ve spent a lot of time working out a parallax scrolling system which affects the rate that the background imagery moves at, although this isn’t quite finished yet as it produced quite a jittery result. I’ve also added in a code to randomly generate the cloud sprites I previously drew. This works by producing a small amount of cloud sprites just outside of the player’s view, scrolling each cloud along at a set pace and deleting them once they are off the screen at the other side. The code then regenerates each sprite at the same coordinates back on the first side of the screen, resulting in an endless amount of clouds.


    This seems to work fine, until you enter a cave or building and exit back into the main level. Instead of randomly generating cloud sprites at this point, they seems to align along the Y axis and scroll across together. This is probably a fault with my coding, however I haven’t figured out how to solve the issue yet…


    So I’m kinda just creating more problems for myself right now! As a de-stress technique, I’ve been sketching details into my new level design, which looks a little like this right now:


    Personally, I think this level looks a lot more like its original kanji than the previous design, but it’s still pretty tenuous!

    Devlog Video 2

    This video pretty much sums up what I’ve been doing for the past few days, mainly with the background graphics and building interiors, and a few other graphical elements. From creating and playing back this video, I came up with an extensive to-do list based on bugs, flaws and things that don’t fit well or look right! My aim for this week is to fix everything that needs fixing, while away from the computer I’ve started to design the next level which now needs tracing.

    Things To Do:

    • Sort out problems with the lighting (solved)
    • Sort out problems with particles (solved)
    • Delete accidental “invisible platforms” (solved)
    • Reduce the amount of lights inside small buildings (solved)
    • Sort out collision with cave walls problem (solved)
    • Animate some existing assets, like flags and the Maneki Neko
    • Make health restoration item
    • Apply more accurate collision masks to objects
    • Improve environmental tiles
    • Make new sprites for rope bridges, statues, outdoor tables & chairs and Ikebana

    The biggest task here was the problem with the lighting system, which occurred when re-entering an already visited room. The code which draws the lights is called every time the room is entered, but because my rooms are persistent (changes to the room remain even if the room is left), the code was only being called when the room was entered for the first time. The result was that lights weren’t being drawn in revisited rooms, but would stay on from the previous room ie. the last place the code was called. I realised this by a long process of trial and error…

    The rooms had to be persistent (or so I thought!) to stop the player from being able to collect the same flower twice. In a non-persistent room, the flowers would respawn if the player left and re-entered. There are however a couple of ways to prevent collected items from respawning, without having to create a room which is persistent. I found this post from the Game Maker Community forum to be really helpful. One of the solutions suggested on this thread is to use a datastructure, in this case a list of all items in the level with their own unique ID.


    This is a script called gameInit, which is called once when the game starts. Gabriel Verdon uses a similar script for The Archer, to set global variables and position control objects etc. You can see that this script determines the keys used which correlate to the user controls, and generates the player object, HUD objects and the effects control object- all of which are persistent and remain throughout the game once called. At the top of the script, I added a generate list command, which will hold the information of flowers collected throughout the level.


    This code is from the create event of the Sakura blossom object. The item is placed onto the list when it is deleted recorded by a unique ID, and this code commands a respawned item to be deleted again if it is on the list.

    So now my lights stay on in the rooms they’re supposed to be on in! Most of the other fixes on the list were relatively little, some involved reducing collision masks and the rest involved clearing up some of my human errors. What I really want to see now is finished visuals.

    Continue

    I’ve looked over the timeplan for this week and partly be mistake, my objectives are mostly the same as they were for the previous week, to achieve a “good representation of the final game.”

    So, there’s still a lot of visual work to be done! There are still a whole range of tiles to be added to each tileset-nothing is finished. Over the weekend I came up with a silhouette background to fill out some of the background, which has made a huge difference to the way the game looks.

    I got the idea for this style of background from some screenshots I took of the game Nevermore 3, which uses layers of distant silhouettes in its backgrounds.



    In my repeated background image, I’ve created silhouette buildings, trees, statues, shrines and pagodas inspired by Japanese landscapes:


    The idea is to use a darker coloured background on top of this to add a better sense of distance and break up the monochrome.

    Today I advanced on some of the graphics decisions I made yesterday, starting with the background trees.


    This was the initial structure that I had from yesterday’s rough test, however after looking at examples of real and drawn cherry trees I felt that it could do with some improvement. For a start, this sort of tree tends to have a lot more branches- the lack of leaves really emphasises its branches more than other trees. I painted this new design based on additions to the old one:


    Before, I’d represented the tree’s blossoms as blocks of colour, but realised this wasn’t really appropriate based on the nature of the tree. The blossoms create very different shapes to leaves, and I wanted to get the texture of the tree right.

    To get this right in the digital version, I took the 16×16 pixel Sakura Blossom sprite and reduced the size by half, which ultimately distorts to shape and makes it seem more abstract while keeping a lot of the colours intact. I then copied and pasted it into the design, to create a repeating pattern which looked a lot more natural and detailed than the look I was going for before.


    This tree is made up of three layers: a top layer containing blossoms, a middle layer with the tree trunk and a bottom layer which contains a darkened set of blossoms to create depth within the tree.

    I repeated this process with a new tree design, which I sketched out but didn’t get round to painting unfortunately. The idea is to have two contrasting designs so I can hopefully repeat these two throughout the level design without having to draw any more VERY time-consuming trees…


    Some screenshots from the level:




    They’re surprisingly flexible! However, not much good on their own. I’ve had to adapt the design for the Bonsai style tree to match this one, or it looked seriously out of place. I used a similar method, however I converted the colours to the Sakura Blossom sprite to look more like a leaf than a flower. Here’s the result:





    My main objection to the design is that everything is currently still very murky, despite the odd light here and there. I’m thinking of increasing the span of some of the outside lights to make everything a little brighter, however for now I’ve seen a great opportunity to hang lanterns from my trees. Everything so far is still in test phases, and this is no exception! This test has resulted in light overkill, which I’ll sort out if I decide to keep this in.



    Auspicious Everything

    Today I started to crack down on Hanami’s background imagery, not just the background tiles which in a way make up the foreground…But the actual far-distant background that doesn’t affect gameplay in any way. So far, my background has consisted of a grey colour which has been made a murky pink colour by the ambient overlay from the lighting system.

    Obviously, one hugely important feature of the background is trees. I would consider myself a failure if I made a game about cherry blossoms and didn’t include the trees that they came from. This has always been a part of my level design thoughts, and although I’ve never mentioned it, the trees were among the first elements I started to think about when developing the art style.


    As a I knew I was going to be using tiles but wasn’t sure what they would look like at the time, I wanted to make sure the trees didn’t conflict with the rest of the style too much. I tried to think of ways to make trees from tiles just in case making stand-alone tree objects didn’t work. From playing around with rock features etc. I’m pretty confident in saying that wavy tree objects aren’t going to be an issue against the square shapes of the buildings and other elements.

    I discovered this photo a while ago and knew I had to make trees that looked like this! They fit in very well with the style and look naturally Japanese, which helps with the consistency of the game’s themes.


    I reduced the size of the image and essentially traced the shapes in Photoshop to create these tree objects.


    I imported these into the Game Maker project to see what they looked like with the rest of the tiles I’d already created, and here is the result:


    They fit into the landscape better than I expected, however I felt that they needed rescaling to act as background elements. The background is better off made up of fewer large objects than lots of small objects cluttering everything up. I had a similar idea for a Sakura tree this time traced from a branch as opposed to an entire tree, which never really got finished.

    Auspicious Clouds


    Another element that’s been a part of my designs for a while now has been this swirly cloud pattern, which I’m hoping to use for cloud platforms towards the end of the level. It’s a very Eastern cloud drawing technique, started as the auspicious cloud pattern in China and adapted by the Japanese. Anime enthusiasts may recognise a similar pattern used as the Atatsuki symbol in Naruto.
    It’s the spirals that really set this design apart from the standard cloud-like patterns you tend to see in Western productions, so I started arranging drawn spirals to get these cloud shapes:


    This is actually an early page from my sketchbook which I cleaned up and recycled, which is why Hana has appeared in the bottom left corner!

    Because I want everything to look Japanese, including things like clouds which tend to look the same all over the world, I started working on implementing this sort of design into the background. I used the previous screenshot to draw straight onto to try out a rough version.


    From this I made a cleaner version, which I attempted to shade and make consistent with the rest of the game assets. I kept the scale roughly the same, as I thought the cover from one cloud was adequate. However, in the game I felt they may be slightly too big compared with other elements such as buildings, although I won’t change anything now until I’ve added in more background. These will definitely go in though, they don’t look as though the conflict or compete with the foreground and represent a part of Japanese culture.


    This kind of gives you a rough impression of how they look in the game, although for now I’ve simply repeated the designs infinitely along the X and Y axis and seem to have created a cloud overload. In this example, I’ve added a slight blur and changed the opacity of both clouds to make them seem more cloud-like.

    Auspicious Trees?
    You can see from the sketches I did above that while I was drawing the clouds I thought to myself “I could use this pattern for trees also…” Although this decision has been a battle in my mind all day. Yes, it works well as a tree design in theory but how well does it work in practise? I tried both designs for the bonsai tree and the sakura tree out with this auspicious leaf pattern on the mock-up screenshot:


    While both designs are nice on their own, I just don’t think its appropriate for implementation into the game, as this really does create clutter and mess. As competition for the design, I created a more realistic looking Sakura tree, inspired by this lovely painting. I wasn’t sure on the scale of the tree, so I made this second design slightly smaller and compared it with the “auspicious” design in the Game Maker project. This is the comparision (minus clouds!)


    Unfortunately, I felt that neither of these designs were suitable. While the more realistic design works better in union with the game’s tiles, the “auspicious” design is much better in terms of scale. So, to put any tree designs into the game I have to increase the scale. I think the height that seems to be working is about 150 pixels.

    Auspicious Bushes?
    I had a real lightbulb moment this morning when I drew a spiral pattern at the base of the tree and realised that I could add bushes to my level design. These won’t be spirally, but expect bushes. There is a definite lack of plant-life in my game world so far.

    Auspicious Everything Else
    So I now know that I will have at least two types of tree in my background, as well as possibly some bushes if these don’t get into the foreground. I’ve mentioned before that the background will probably mainly consist of mountain shapes, to look like far distant mountains. Japan is full of mountains. It’s very much a Japanese thing.

    I went back to the Photoshop screenshot mock-up again and drew in some really rough mountain shapes. These fill a great amount of negative space, and work so well in creating depth.


    I’ve yet to develop these any further, but getting the background down is one of my top priorities now as I’ve realised how much difference it makes to the feel of the level. I will probably create a seamless string of mountains that will be repeated across the X axis at the top of the level like this.

    Hazard Evasion

    Hazard progress is well underway, and has led me to discovery a whole range of previously untouched Game Maker functions. Here’s a run down of what’s gone on in the process…

    The Deceitful Blossom

    Not quite so evil-looking at 16 pixels high… This has been the easiest obstacle to create. I’ve created an obstacle parent mask that means that I only need one collision command when the player runs into any obstacle. When this mask object is contacted, the global.health setting (which controls player HP) is reduced by one, as this is a constant for all obstacles! I’ve added a character flash-to-alpha effect when an obstacle is touched, which temporarily disables health from being reduced for a set amount of time while the player has a chance to get away. Those familiar with 2D platformers will be familiar with this, and would definitely expect something similar. I’ve compiled this video montage so you can see what I mean…

    In the case of the Deceitful Blossom, the obstacle is destroyed on contact. This prevents the player from frustratingly walking into the same blossom more than once.

    Hello Mushroom

    yummy!


    With the Hello Mushroom, I’ve looked into particle functions in order for the object to release periodic spore clouds which cause damage to the player. I’ve used particles before in motion graphics, and Game Maker’s handling of particles is very similar. You can completely customise particle aesthetics and behaviour without having to create a sprite first. For my small spore cloud, I’ve gone for a short release of small particles which shoot straight up into the air. They have a slow speed and a short life-span to make them look light and airy.



    I’ve placed the Hello Mushroom in this cave here, and you can see a part of the cloud dissipating. I’ve essentially used a timer to create a period burst, and custom collision masks so that the spores cause damage as well as the mushroom itself. I used this little tutorial for help with a custom collision mask, as Game Maker doesn’t allow mask changes for objects that don’t change shapes themselves, and won’t allow collisions with particles.

    The Hanging Adversary
    The green thing that looks like an upside-down cannabis leaf in the screenshot above is my Hanging Adversary sprite, which currently swings like a pendulum and takes advantage of Game Maker’s abilities to draw connecting lines. This is another example where the darker green line isn’t part of a sprite which I have pre-made, but is constantly being drawn, refreshed and redrawn by the software. The clever part is, the swinging process isn’t based on a set movement, but is defined by settings such as mass and velocity…


    The drawn elements originally created looked a little like this. This was adapted from a code I found on the Game Maker Community forums, which creates randomly swinging circles on these very long lines! The grey lines demonstrate the boundaries of the swing, so it was really simple to get from this to a smaller, non-random swing motion.




    I think the sprite may need changing for this one… I made all of these sprites really quickly to get them into the game, so expect changes.

    Natural Hazards…


    This week I’m thinking about all the features I want to have in the game before handing it to others for feedback! I think in my original time-plan I wanted to base the product of this week on feedback from participants, but I’ve gone into some of the graphics in a lot more detail than I was expecting to and as a result have a few other things that need rounding off/actually making… So my goal for this week is to create a working prototype ready for testing either at the end of this week or the beginning of the next.

    One of the major things which I have omitted until now is, to summarise, how to loose whilst playing Hanami. I’ve implemented a really basic health system so far, which can currently only go down, and instigate an immediate game-over is it reaches 0 (which it can’t, because I haven’t put enough hazards in yet!) This is one of the things that needs a lot of improvement this week- it especially needs something to build it back up.

    I’ve mentioned possible “hazards” or “enemies” before, and I’ve sketched out a few ideas in some of my level designs. The main feature of all enemies/hazards is that they cannot be “defeated” because there is no combat in the game. They are a part of the environment, and will not actively attack but will stand as a hindrance to players. As the collectable items are based on flowers, I’ve also based my enemies on plants, creating a good/evil balance throughout the natural world! Each enemy is also based on a unique movement type, to keep them varied and keep the player actively working out how to evade them.


    The first enemy type is one that I’ve been using as a health system test, and is based on the Sakura blossom object. The idea is that it lurks in shadows and looks similar enough to the real Sakura object to lure players towards it, only to hurt them if they make contact. I’ve called this one the deceitful blossom, which is currently a working title name but may stick! Its movement type is nothing, it’s the easiest enemy to avoid as it simply sits in once place.

    This enemy type has a few influences from existing games, not so much in terms of visual qualities but in attack style! I’ve looked at items and enemies that disguise themselves and attack at the last second. I thought of Vileplume from Pokemon which disguises itself as a flower, and the mimic from Braid which hides under the soil with a flower under its back. In a way it reminded me of the Mario “know you mushrooms” design seen on bags & T-shirts etc. Many Mario mushrooms look similar, but have very different effects, good and bad if acquired…

    The second enemy happens to be a mushroom, but nothing like a Mario mushroom unfortunately. Unlike the other enemy types, a name didn’t pop into my head straight away with this one, so it is currently called Hello Mushroom…for a number of irrelevant reasons… This enemy doesn’t move itself, but it sprays a vertical line of deadly fumes into the air at random times through one of its many sphincters, which will deduct health points if touched. Most of my house mates have a serious aversion to mushrooms and try hard to stop themselves from vomiting when I cook them, so I’ve made this one super gross to fit their opinion of them. I think mushrooms are really yummy personally.

    To get the motion of spore-release, I’ve been playing around with the particle functions in Game Maker today. I found a great guideline to all the available functions in a downloadable PDF here, which literally misses nothing! But so far I really have only been messing, so I’ll write up about my proper particle experiments later!
    This guy’s kind of inspired by the many monster mushrooms in video games, like Funguar from Final Fantasy VIII, the Fume Shroom from Plants vs Zombies, and of course the deadly mist emitting Black Fungus from Kingdom Hearts.

    The final enemy is one that moves horizontally by swinging from ledges and cave roofs etc. I’ve called this the Hanging Adversary, mainly because it was the first enemy I came up with and I wanted to differentiate it from any other potential creations! The hazard here is really sharp leaf-type structures- I said I didn’t want to feature any cliched spike-pits, so this is my original equivalent. I’ve fashioned it after a venus fly-trap to some extent, simply because the venus fly-trap has those naturally evil-looking teeth which make for a great game enemy. I’m sure they’ve inspired many monster creators to make plants that bite.


    I had to be careful that this guy didn’t end up looking too much like anything else from the gaming world, although influences can natural be seen to the Mario Piranha Plant, and similarly the Venus Flytrap from Braid which was probably based on the Mario enemy! My favourite of the carnivorous plant monsters from games has to be the Deku Baba from the Legend of Zelda series, which looks so spiky and evil even with the lowest of poly-counts!


    I’ll hopefully get all of this into the game tomorrow, and adapt the health system accordingly!

    Weekend Catch Up!

    Tying up some loose ends…

    New Characters
    Here are rough designs for three more characters, all with specific roles (should they make it into the game!) I’ve tried to create one unique character for each of the game’s levels, as well as recurring characters who will appear throughout restoring health and saving the game.

    The idea for the purpose of Most of the characters is that they will give Hana a flower petal, which can be combined with other petals to create one flower. Five characters in each level will give Hana a petal each, which will then make a whole. I really hope I can get this to work, otherwise they will simple have to give her a whole flower that they have found (to make it easier on my lack of coding knowledge).


    Left: a guy who wears a face mask to protect himself from germs, or possibly to protect others from himself. Apparently, the lovely Japanese people wear a mask when they are ill to protect others from catching the bug.

    Middle: a resident of the “pink” village- the game’s first level. He is based on Tsukimi from 51 Japanese Characters, and prefers watching the moon (“tsuki”) to the flowers. He sits high up in the mountains at the end of the level.

    Right: I think this guy will appear in the next level. He is found in a hot spring, hence the lack of clothing!

    Old Characters
    While I’ve been working mainly on level design, I’ve also been doing character pixel-art for the characters I created a while ago.

    Priesty


    Monky


    Panda


    Sleepy


    I’ll give each of these characters walking and fidgety animations so that they can move. They’re all a very similar size and shape so that I can apply similar animations to all of them.

    Level Developments
    I’ve started jotting down ideas for the next level I’m going to create, which revolves around Japanese forest rather than the current emphasis on mountains. The colours scheme is orange, so I have a new Kanji to work with! There isn’t so much a structure as individual features at the moment.

    Draw, Animate, Code & Play etc…

    Drawing
    Over the past couple of days I’ve tried to prioritise sorting out my environmental tiles, but still haven’t really come to a conclusion. I tried to test out my new idea for rocky tiles with real parts of the level design, but so far can’t seem to make them work well as they do on paper! To stop myself from ending up in a rut, I’ve discarded ALL rocky texture tiles for now and replaced them with a plain purple colour, which can easily be swapped for a textured tile when I decide what it will look like!


    I made a little progress with the rock pattern around cave entrances however. I considered using straight edges around rock outcrops so that straight edges around other level features wouldn’t stand out so much, but didn’t feel this worked as well as the more natural, uneven design which is closer to my previous trace design.


    I tried this design here with my previous rock texture, but decided that the rocks looked better against a more plain background. Even if I bring a heavily textured tileset back for platforms, I may stick with a plainer tile for cave walls.

    Whilst trying to ignore all the complications of seamless tiles etc, I’ve diverted my attention to creating more Japanese-looking objects for the level. I’ve started by taking objects I’ve already made and adapting them to suit various instances, which is great for level continuity.


    Amongst these smaller features I’ve been working on some large objects inspired by specific Japanese things, like this shrine gate:


    and this bridge:


    Although, the reason for the bridge’s funny shape is because it follows the shapes of my platform mask tiles. I may round this off later to make it look nicer and have the character follow the tiles rather than the shape of the bridge.

    And I finally got round it pixelising the small food stand that I designed about a week ago. You probably can’t tell even if you’re Japanese, but I attempted to write ramen “らめん” in hiragana on the sign!

    Without the ground texture tiles, the game definitely doesn’t look as “nice” as it did before, but the more empty spaces of the level are starting to fill up. (I haven’t built the lights in this screenshot either, which is why it looks so dark!)

    Animating
    In time for a proper working level prototype, I wanted to get many more character animations in. Before now, I’ve simply had one continuous running animation! I’ve only created the basics so far: running, stopping, jumping and climbing ladders.

    To make the character stop you could use a single frame of the character just stood still, but I like to keep the character moving to ensure the player that the game is still running! Before now, I’ve used a breathing motion by making the character move slightly up and down, but where Hana is made of so few pixels, any rearrangement of pixels results in something far from “subtle”. I think a breathing animation is out of the question, unless I find a clever way to do this. For now, I’ve created a couple of frames that make her look fidgety when stood still, and a blink frame that flashes up irregularly.




    For the jumping animation I’m currently only using one frame, although I would later like to add another to differentiate the character jumping up and coming back down.

    My climbing animation is currently universal, used for climbing up and down ladders. This particular animation works best for climbing down ladders, so I may later add another one which looks more like climbing up a ladder.

    Coding
    The main bulk of the coding I’ve done over the past couple of days has been in the character step function, which controls the way the character moves. With all the new sprite sequences, I’ve had to customise things like image-scrolling speeds for each individual animation and can no longer rely on one over-ruling command.

    I’ve split character movement into two separate “modes”-one fore running, jumping etc. and one for climbing. If the player is not on a ladder or in mid-air, the player is on_ground (this is a ready-made variable that comes with the Grandma Engine.) If the player is on_ground, the rules of horizontal movement apply, including sprite sequences and player input. If the player crosses a ladder but does not climb, ie. does not press up or down, then the rules of on_ground still apply.

    if on_ground && place_meeting(x,y,obj_ladder) && !keyboard_check( key_up ) && !keyboard_check( key_down ) {
    can_c = false; }

    But, if the player crosses a ladder and does press up or down, can_c (short for can climb) becomes true, and the rules of ladder climbing apply. This code is pretty similar to the code I wrote before for vertical movement, but I’ve added sprite sequences and image-scrolling speeds. As you can see, I’ve applied the climbing animation twice, once for each vertical direction, so if I wanted to I could use two animations that would represent each direction.

    else if (place_meeting(x,y,obj_ladder)) {

    vspd = 0;
    can_c = true; {

    if (keyboard_check( key_up ))
    vspd = -S_MAX_V /4;
    sprite_index=spr_hana_climb;
    image_speed=.1;

    if (keyboard_check( key_down ))
    vspd = S_MAX_V /4;
    sprite_index=spr_hana_climb;
    image_speed=.1;
    }

    I haven’t written much other than this, but I’ve added a few lines of code for more warp objects throughout the level. This has helped me create some clarity where caves are involved, as I have created two new rooms for caves that are joined to the main level by these warp points. Now you can see a definite distinction between the outside and inside of the cave parts!

    Outside:

    Inside:

    Playing

    Adding a lot of visual features doesn’t occur easily whilst play-testing, due to the fact that in Game Maker the platforms are made of these red-block objects that cover the background imagery. They need to do this so that I can see where I’ve put platforms! But at the same time, I can’t see if I’ve made a mistake with the imagery below. So when I’m editing tiles and want to see how they look in-game, I have to add these red blocks temporarily and delete them again afterwards. One major criticism of Game Maker is that it doesn’t allow bulk actions to be applied to all instances, so each block must be added and deleted individually, which is a looooong and tedious process. My main focus is still on visuals, but I’ll build the complete level for a play-test from volunteers next week.

    This concludes my lengthy summary or the past two days!

    More Changes to Environmental Tiles

    I’m still not 100% about my mountain/rock tileset, which seems to be expanding increasinly to accomodate for varying edges and crossovers. I had a rethink and came up with a more simpler, more flexible solution. I’m going to try this out for now and not expand on it too much, as I don’t really want the graphics holding me back at this stage…

    Here’s some of my most recent screenshots from the level test in Game Maker:



    I’ve taken on some of the advice I’ve received from feedback sessions, such as using the blue brick tiles as a background layer to show depth within the mountain, however I’m still concerned about the scenery looking overcrowded, messy and repetitive. This is before I’ve even filled the level with many of the items/objects which will be added later. So I’ve gone back to my traced level design and doodled this new tile design, based on a much simpler and organised idea.


    There are parts of my original design which clearly show just a few neatly aligned rocks jutting out, and I feel this suits the building graphics better than the more clustered tiles I was previously using. I created this test mock-up in Photoshop to check their effectiveness:


    You can see that I’ve also changed the colour of the grass to match some of the colours used in the building, so the level shouldn’t seem so disjointed any more. I’m currently working on the rocks around the cave entrances, and wondering if I should create a non-tiled pattern specifically for all of these instances, although all of this still needs a little work.