DrewMelton's Forum Posts

  • Okay, I'm working on the graphics for each floor. Here is the upper level, mid level, and lower level. I still need to do the sewer/basement, and underground mine or cavern. I'll do those soon.

    I've also added a few more props such as buckets, tools, sacks, etc. that I will spread throughout the castle to breathe a little life in it and make the rooms less empty.

    I've spent a lot of time refining my template with very careful naming conventions and organization so I can easily change materials on floors or whatnot and easily move my objects around.

    Here is the upper level. This level will have the best loot but also fully armored knights who won't be happy that you are there. This level will be more fancy with painted walls and whatnot. The lower levels will be more barbaric.

    Here is the mid level. It's not as fancy as the upper level, but not as dark as the lower levels. The loot here will be okay, and the guards will be moderate.

    Here is the lower level. It will be dark, and the loot won't be as great, but the guards won't be as ready to attack either (or as hard to beat).

    I'll be spending this week on the two lower levels beneath these and also working on some more props to spread throughout the castle. I will also be doing the floorplan for the castle so I can get started making these into actual rooms.

  • That's good to hear. I can always expand the visuals with some more animations. And not just characters, for example I could put in animated water, enhance the torch effect, or other bits here and there.

    My nvidia gtx 680 has 2gb of memory. I plan on upgrading it eventually. I do consider it to be high end even though it is a couple years old now. I expect people with even older or cheaper cards to have less memory.

    I am also keeping an eye on things because I'd like to use this template for future games either RPG or RTS.

  • — Thanks! Even though I'm mainly a 2d artist, I think 3d is more "fun" to work with.

    I guess when it comes to memory usage, I'm just cautious. This is my first game that will be completed and released. I don't want to find out after I do a ton of work that many people can't play it for some reason. I do remember Ashley saying in a post that going over 200mb will eliminate some users, but under 400 or 500 is probably still fine for most gamers these days. So, I have plenty of headroom to expand if I want to. Those were the last numbers I heard anyway, so maybe I can get away with even more.

    Actually, my main concern right now is just preview times. It gets harder to do tweaking or add in mechanics the more I add. I experimented once with making a version with no animations for faster previewing, but it just made it hard to keep two versions up to date (and I need the animations for much of the testing.)

    After I release a demo later on and see how it performs on other people's machines, I plan on seeing how far I can push it.

  • Okay, some updates!

    I have been playing with the level design a bit. I was thinking about how I am going create the different types of rooms and variations and what style would let me do the most. I need the level graphics to be something that can be easily arranged into new rooms since there will be probably over 100 rooms. (the graphics below are just for one or two floors. The others will be a bit different)

    Here is a new style. I raised the camera slightly so you can see the ground more and objects would not be as hidden. In my original screenshots on page 1, I did a sort of cut-away effect so you could see the ground better. With this new style, you can see plenty of ground without having to do that. I also mixed up the stone with some slabs and other materials to make it a bit more varied and make it stand out more.

    I also set up a highlight effect on the stairs, one for up and one for down. Here is the one you'll see if you want to go up.

    I was worried that with the angle adjustment ,I would have to re-render the characters. Which would not be hard, just time consuming. I think the change was minor enough that the characters still look good without changing them.

    Oh, and I set up a countdown when lock picking to prevent swiping stuff from a chest and running. If there is an enemy near the chest, I don't want people to just be able to grab the loot and make it out unharmed. I set up a red bar that gets smaller as you hold down the mouse over a chest so you could see your progress to unlocking. It only takes a couple seconds, but it alerts guards nearby right away too.

  • Try Construct 3

    Develop games in your browser. Powerful, performant & highly capable.

    Try Now Construct 3 users don't see these ads

  • At least I'm not the only one. I try to keep up with this forum every day, and these kinds of issues make it hard. I keep saying to myself "I've read this post!" "where's the new reply?" "where's the topic I saw mentioned on the main screen?" " he wasn't the last poster in this thread!"

  • Very Impressive on every single details! And the video's informative too!

    'Portrait with blood' is a cool idea stead of the usual health bar.

    Thanks! I swear I rehearsed that video so much that by the time I actually recorded it, my voice was coarse, lol.

    I actually got the idea for the portrait that filled with blood from Baldur's Gate II. The tricky part was inverting the percentage so that when you were at 80/100 health for example, the blood would only fill 20 percent. Took me several hours just to come up with the formula.

    I've had way too much tequila tonight to give any explanations, but I'll be back tomorrow if anyone has any questions or comments. Thanks again!

  • Ha I understand completely the time constraints of working alone (I think what you have is more than acceptable).

    You mentioned long waiting times to preview, out of curiosity, what is the memory use on the layout the demo is on?

    I have two rooms in the demo. The first room uses 146.33 IMU after export (and about 129.99 "before" export). It always seems to go up after export. The second room is nearly the same, but reuses the same assets.

    The current time to run a preview is about 40 seconds. This is the time from clicking the run button to being able to play. I recently deleted some old junk and did some cleanups, so it's down a bit from the previous 45 second times I used to get.

    Most of my memory is used on the hero and enemy characters. I tested it once by deleting them and running the preview, and it only took about 5 to 10 seconds. I guess after I get the rest of my characters done and add in the level graphics for other rooms, I'm going to be waiting even longer. Oh well, I've got most of the features working already.

    Of course, once I export it to nodewebkit, it only takes about 5 seconds to run.

  • GenkiGenga Thanks! I'm happy with the way they have turned out.

    I'd love to be able to spend more time on the animations of course, but I have so many things to do working solo, I can only spend so much time on each part.

    Plus, there's the limitations of image memory. I also need need certain triggers to be on specific frames consistently (such as the attack box for damage spawning during frame 15, or spawning the arrow, or whatnot) in order to keep combat fair for enemies and the player.

    Modo has been pretty good for animations. I don't think it's as good as Maya, but it's half the price, and their update policy is much more forgiving. Once you come to grips with it, it can make very nice animations.

    This is the rig I built after following of couple of very lengthy tutorials. (I adjust the fingers with a channel haul that is not pictured). I turn off textures/materials and smoothing to make it animate more fluidly.

    Soon, I plan to add a few more enemies, but I think for the next step I should create the floor plan for the castle and start building some more rooms. I can get started making my props as well (rubble, various castle objects, columns, anything to place in the rooms to make them less empty feeling).

  • I always seem to use one big event sheet that is well organized into groups. I've never heard of any difference between the performance of one event sheet versus many. The most important thing seems to be how the groups are organized so you can use the debugger profiler to see what resources they are using or to enable/disable them as needed.

  • tunepunk Thanks! Yeah, I like older RPGs. I still play Baldur's Gate II enhanced edition on Steam. It's amazing how well the graphics of some of these 15 year old games still hold up. In fact, the only complaint I hear about BG II EE is the somewhat low rez character models, but even they are still decent looking.

    bclikesyou Thanks! Always good to hear that it's a game people want to play. The more interest I get from people, the more motivation I have to keep going.

  • ahr Ech Thanks! For the first few months of my project, I was going with more of a hand-drawn style. It was either going to be pixel art or more of a stylized hand-drawn look. As I got further into development, I wanted an easier way to create realistic characters and environments. After all, my main inspiration for isometric games came from things like Baldur's Gate II, which was fairly realistic in its overall look. Plus, when I decided to go with fully isometric (as opposed to multi-angle like Golden Axe on the Genesis) I needed an easier way to make things.

    I then made the change to 3d. I now quite enjoy 3d. I feel like I'm actually building things. Plus, I like digital sculpting (like in Zbrush). Animating in 3d is fun too. I'd probably choose to be an animator if I had to pick one job. Of course, I grew up doing drawings and paintings, so part of me also likes to do fantasy art (like the thief portrait). I will be doing a full illustration later on for the title screen and promotional art.

  • Great work! But I think that in the future it will use a lot of memory. Isometric perspective need a lot of sprites and animations for mobs and Character... More mobs = more animations = more memory usage...

    Thanks!

    Yeah, I'm keeping the rooms limited to maybe one enemy. I haven't decided yet. The game will be hard enough just with that. Enemies will not be a pushover in this game, well except for the scrub enemies that are going to be intentionally easy to beat.

    The main character and that enemy have most of their animations done already. Any more at this point would be optional.

    What you see in memory use in those screenshots will be pretty close to what the final game uses. I'm trying to keep it under 200 IMU, but I believe Ashley said in an old post that even 400 IMU is good for PC games, so I have some head room left if I want to make a room with more enemies in it.

  • C-7 Thanks! I just started learning 3d last summer, though I did have a bit of experience with 2d animations before that. Most of my art was sort of like illustrations, so I had to know how to draw the figure in action poses. I think that helped with learning 3d animation

    Since image memory usage is a huge concern for 2d games, I have to be conservative and mindful to get the most out of it. I spent A LOT of time designing a template in Modo that basically centered the character perfectly so I could use the auto-crop feature in C2. This made importing isometric animations a breeze and allowed me to be efficient with the memory. I set up 8 cameras in Modo, one for each direction. So, when rendering animations, I just go Camera A - render animation, and put it all in one folder, then Camera B, and so on, and then import them go to C2, and I can import all the frames for each animation at once, shift-crop the whole thing at once, and save myself a huge headache over my old ways of doing it.

    The number of frames varies. The biggest one is the attack animations. I set them to 40 frames just to slow things down a bit. I didn't want combat to be so fast that you would not have time to observe and think. I also thought that combat should be fun to look at, so I gave it priority over the other animations.

    Idle animations use less frames. For the bow's idle animation, I even use ping-pong to make it last a little longer while not needing more frames. The dagger idle is a bit fancier, so no ping-pong. They both use 30 frames.

    The walk animations are set to 29 frames (well, 30 if you count frame zero). My main concern was making sure it looped smoothly during the walk cycle.

    A couple of the shorter animations only use 22 or so frames, like the death animation (not shown in the video).

    Anyway, as far as easing new players into it, I am going to use the main text area in the to give some tips. When you mouse over an icon or whatnot, it will give some info on it. Also, the "tips" icon will bring up additional info that players may miss. And as any good game should, I will make sure the first few rooms the players enter will help them get to grips with the game with easy enemies and whatnot.

  • Okay, so I've been putting off posting an actual thread about my game, or even showing it off, for a very long time. I think I now have enough done that I can start posting about it.

    This is an isometric RPG that I've been working on for about 1 and a half years. Of course, a lot of that time was also spent learning 3d, learning C2, and learning about game design. I also took a couple short breaks when I reached sticking points to try some quick prototypes. I still have a long way to go. I need to build the levels and create some more enemies, but most of the gameplay is in place. Since the preview time is already about 45 seconds, I decided to hold off building any more and just focused on gameplay features for a while. Soon, I'll be building out the rest of the game.

    The target platform is PC. I hope to get it on Steam eventually.

    Anyway, here is what I have to show.

    This is just a test area. I plan on building a full castle with at least 3 or 4 levels plus 1 or 2 underground areas. Each room will be small, but I plan on mixing it up with different sizes or layouts for the rooms, different enemy placement, and whatever else when I get to that stage. It'll be sort of an "open world" castle. You can go between rooms as you please or seek alternate paths.

    You can spend skill points at any time. They can also be saved for later. You will need a good amount of skill points to beat some of the enemies in the game. Skill points either improve your abilities with weapons or your overall strength and health. As I get more done, some of this may be modified for game balance purposes, but so far it works pretty well.

    Combat relies on hit rolls. Your attacks are compared to a 1d10 die which rolls 1 through 10. Your skill with a weapon can be from 1 to 10. If your skill is equal or higher the the computer roll, you successfully hit the enemy. I have it started at 3 which hits 30% of the time. It can be increased to 10 to hit 100% of the time. Enemy attacks roll from 1-15 or 1-12 depending on the enemy and roll against your agility. If your agility is equal or higher, they miss.

    Loot will be random. I use functions to determine what loot you get from the containers or whatnot. Whether the enemies will drop something has yet to be decided. Everything you do gives skill points so you can level up. Objects that you can interact with will be highlighted in yellow. This was partially inspired by Deus Ex. I also wanted some randomness since this game was originally meant to be a full rogue-like, but deviated slightly. I am still probably going to feature some rogue-like aspects though such as permadeath.

    I have done a lot of work to make a nice HUD for the game. I think I have redesigned it about a dozen times by now, not to mention small tweaks or experiments. I'm finally happy with it, though of course, I may or may not revise or improve it as time as goes on.

    The goal of the game right now is to collect loot. Since you are a thief character, it seemed like a good goal. I may let the player choose the goal amount or whatnot. It just depends.

    I chose a thief because it had the most varied skills. You can attack at close range, hide in the shadows, set traps, use a bow, pick locks, and whatever else I add before it's done. I didn't want the game to be about straight combat. The enemies do not attack right away. If you get close or do something suspicious, however, they will come after you. I made it this way so you could decide if it's worth attacking a difficult enemy or bypassing him.

    The game is pre-rendered 3d. I use Modo and Zbrush. I also use Photoshop for the HUD. I render everything on different layers. For example the castle floor, back wall, and containers are all on one layer since they are rendered together. The front wall is on a higher layer so it always appears in front of the characters. The characters themselves and any miscellaneous objects will use z-order sorting. Characters are rendered in 8 directions. So, each animation has 8 variations. I explain more about this in the video below.

    I built most everything from scratch. I am using my own character models, my own animation rig, my own levels, icons, portraits, and eventually sound. This has been a solo project.

    Here is a video I made. It lasts about 18 minutes. I explain quite a bit about the game. Please do note that this is a very "casual" recording. So, forgive the occasional "um" or "you know" that sneaked in. I was a bit tired when I recorded it... <img src="{SMILIES_PATH}/icon_cool.gif" alt="8-)" title="Cool"> (oh and try to watch it at 60fps in Chrome if you can).

    Subscribe to Construct videos now

    Anyway, I most likely forgot to mention something, but the game still has plenty of time left before it's done, so I'm sure there will be plenty of updates.

    I'm fairly new to game design, but it seemed like it would be something I would be good at. Technically, I'm an artist. That's why I chose C2. This game is the first one that I am taking to completion. I've been reading books and studying everything I can to make sure I get started off on the right path in game design.

    Update 8/22/15

    New screens. (see page 3 for details)

    Update: 9/3/2015 New video!

    Subscribe to Construct videos now

  • You do not have permission to view this post