So I'm trying to do some pixel art for a solo project I'm working on.

Can you guys give me some general feedback? Here is a recent sprite I made for an ally:



And here is a game screenshot:


So, any tips or general feedback? Some pointers on how to improve or a specific style to go for?

Many thanks!

I just posted the game in this thread: thread with Dream Catcher

But we should use this topic for the pixel art since I'm still curious about this part specifically.

Hi!

Alright, here are some things I noticed. Before I get started, I should warn you… I always tend to list more negative things than positives. Bad habit of mine, sorry for that... It annoyed many people before you. Please believe me… I only want to help you!
Oh, and I have only watched the youtube video.
Also, the things I mention are not all about the pixel art.

Alright! Art is not only about making things look nice, it is also one way to give the user feedback. I should clarify what I mean with feedback, since I’ve heard very different interpretations. For me feedback is about:

1. Making the user aware of the state of the game (at all time!)

• e.g. a ghost is materializing (harmless) or touching the ghost can kill you. So, how long will the ghost be not fully materialized? Does the player have to hurry to get out of the way, before the intersection counts as hit? (Death by a just spawned enemy is unfair and lets me never ever touch this game again.)
  
• Your depiction of the highscore is a good example for state-awareness. (Perhaps write “score:” in front of it or even better find an icon that you reuse in the highscore screen.)
  
  2. Notify about changes
  
  
  • Notify if the user has cracked the highscore! Use different colors e.g. for the background, so that the user is clearly aware of that he is now better than the old highscore. This can motivate and also stresses the nerves, making it all more challenging and exciting.
    
    3. Acknowledge user input
    
    
    • Make a “Click” sound if you click a button (that’s very useful on touch screens).
      
    • Change the appearance of a button if you’re hovering over it and it is clickable. This improves interaction speed and avoids mistakes (clicking next to the button or even worse on a wrong button).
      
      There are some things that need feedback and haven’t got any yet:
      
      
      • How close can a player get to an enemy before the hit counts? Simple solution: Warnings. Maybe tint the enemy red if you are in a certain radius. Use more red, the closer you get and let it blink, if it is very close. Blinking calls and demands for attention.
        
      • Award the player if he got away from a near death (textual is enough, giving points would change the core mechanics too much.)
        
      • The gray ghosts have a fixed trajectory. Why not showing that? It would help the player. Don’t be afraid you can make a game to easy, if you help the player a lot. You have other options to make it more challenging again. I don’t like it if a game is challenging because it lacks hints or can get you in unfair situations.
        
      • Depict the health of the end-boss (big ghost).
        
        Rewards and punishment. You should have more of both.
        Achievements! You can’t have enough of that, so reward the hell out of the player. (Double-kill, triple-kill ...)
        Perhaps add an item that makes you more agile.
        Punishment is important so that players can learn the rules of your game. But a game should be forgiving. One small touch and the player dies… What about giving a little recovery time? If you hit another ghost in that time the player dies for sure. Feedback would be very important here! The user has to know he is recovering. This would also be a good chance to add a smaller punishment. Perhaps make your player slower or more inert, cover parts of the screen or change the color scheme so it is a little bit more difficult to see enemies.
        
        
        You should have more goals. There are three types of goals:
        1. Immediate
        Dodge enemies and what else? Perhaps collect some more points that were dropped by enemies or just spawn somewhere? If the points vanish after a while, you make the player hurry-up. Perhaps, he’ll get more immersed by that.
        
        2. Short-Term
        A classical example would be: reaching the end of a level. So, let the player know, how far that end is! Short term goals help keeping the user playing and make it harder to put the game away again.
        
        3. Long-Term
        
        
        • Highscores (well, that’s not so interesting, unless you can share it.)
          
        • Classical example: Rescue a princess.
          
        • Perhaps something that helps you in future games? (Give it further replay value!)
          
          Notes on the high-score screen:
          
          
          • It took me a while to figure out, what all the blinking should tell me. A short notice that you unlock stuff here by spending points etc would be worth a small text.
            
          • Also the “+” button isn’t self-explanatory.
            
          • Group buttons! Don’t have some on the top and some at the bottom of the screen. Choose one place.
            
            General thoughts:
            
            
            • I would never have guessed that something that burns, has red eyes and looks so evil is an ally. Perhaps add some other sort of information that helps users to distinguish between allies and foes in general. (Once again, feedback.)
              
            • The shading of the sprites isn’t coherent. The player is lit from the left, the green one from the right.
              
            • Why is your player looking so sad? Is he afraid? Perhaps make a long-term goal out of that: Make the player happy again! The more points you gather over multiple rounds the more he smiles. That would perhaps add some replay value.
              
              Alright, I hope this helps you a little.
              Hm, looks like I mentioned one positive thing. Have I promised too much? I always tend to point out only things that can be improved.
              But, seriously: I like that game and it looks like some solid good work that has much potential. Also the feedback you're already providing is very good! (Much better than the stuff I usually get to see.)
              
              Bye!

First off, thanks. No, like, THANKS A LOT!

Seriously, your post is really helpful, you go in to specific parts and you make excellent points and suggestions!
I agree with almost everything you pointed out, but it has to be pointed out for me to realize, you understand?

This was the most helpful feedback I've ever gotten in my life O_o seriously.

But anyways, I've updated the game once, and I've given up on it, actually, which probably wouldn't not happen if you had posted this sooner xD
The game was really unsuccessful thanks to very low organic growth in the Android Market (nobody searches for "Dream Catcher" looking for this type of game) and zero advertising. So I decided it was a lost cause... But your feedback has inspired to improve it!

However I'm now caught up on a new game, which I'm developing with my cousin (having two people makes the process much better), but I'm keeping the code reusable so some stuff could be plugged directly in to Dream Catcher and further improvement could be made thanks to your feedback!

I may decided to completely change the theme though, to make it more "mainstream" and have more organic growth.

Anyways, I would love if you would be able to look at some future games of mine in the future and give me more feedback, really!

Thanks again!

Hi again,

We all have done tons of unsuccessful games / prototypes.
The key is to figure out why it has been unsuccessful (otherwise it would have been a waste of time). Therefore you have to ask yourself the right questions and you need to observe players.

Do players keep forgetting how the functionality is used and where they find stuff in the menus? -> Then your game doesn’t help the player to build up a Mental Model. If you have “learned” a game then your brain subconsciously looks up the adequate action from the mental model. You can do many things to improve and speed-up the creation of mental models, e.g. by employing metaphors or make use of genre-specific habits.

Are the reactions on actions of the user not transparent? Are they surprised by the things that occur? Then you haven’t taught them the meta-game (the rules). As an example look at your fire-ally: Is it obvious that you can defeat enemies with it? You could make it more obvious if your enemies flee from you if you have it. Likewise your game could lack feedback. For instance if you forget to tell the player that he has now a new power-up, which allows him to defeat enemies (This one you have done great!).

Do they have no idea what to do? Then you have probably forgotten to give them goals.

Do they have tons of goals but can’t remember them? Also keep in mind that players could pause, do something else and then continue. They most definitely have forgotten the “state” of the game and probably the goals. So always help them to remember things. E.g. in Witcher 2 is a small notice on the screen of your next goal. It hasn’t been there in Witcher 1.

Do you have enough achievements? (Usually the answer is no, since throwing in achievements is time consuming.)

Do users find your UI pleasing and helpful? Then you have followed the Gestalt principles.

Do users understand the feedback you are giving them? Besides, you can improve the processing time of information if you know about perception and cognition. You should also know how much information you can throw at the user at most (it depends on the way you do it). One point in the topic of perception is the attention. How do you guide the locus of attention? (The place the user looks at.) One obvious way is blinking. As an example: sometimes the user has to search for a thing on the viewport. E.g. when he dies, he spawns randomly at a new position. How can you help the user finding this position faster? For instance draw a small arrow at his last focus of attention which points to the new position.

There is a small mantra when it comes to UI design.
Who are your users? (What abilities / experiences do they have?)
What are their tasks? (Find the place where the player spawns after death etc.) -> So how can we help the user?
What is the context of usage? (Do they play on the bus or at home? Adapt color schemes and your sensitivity of user input accordingly.)

As I side note: Usually you should engage users at every stage of your game development process. Ask for there opions, find out their tasks etc. But you can save a lot of time and locate the major UI related problems if you "invent" some imaginary users. (For the fine-tuning you always need real users, though.) Give those imaginary users some properties (experiences etc) and then write down a ten-minute scenario. What does the user think, when he first sees this or that item. What does he try to do with it, etc. Those imaginary users (you shouldn't give them names by the way) are called "avatars". They are one very helpful aspect in scenario-based software engineering.

You not only have to learn from own mistakes. Why not analyzing commercial games?
What goals do they have? What kind of rewards and punishment? What are the states that have to be communicated? Which state changes are there? What kind of feedback have they used? (Visual feedback can be overlooked; sounds are perceivable regardless of your current locus of attention.) How do they steer the attention of the user? (This doesn’t always have to be on the UI. It can be very subtle. E.g. look at Assassins Creed. I remember there has been at one point a labyrinth and you had to pass it in a certain time limit. If you come to a crossing you simply had to take the turn where you could see a dim light. Players always tend to go to lit places. Level designers use this fact very often. (Since I knew that I didn’t even take one wrong turn…) That’s why often mods are confusing, since there lights are sometimes placed randomly, so players can’t find exits or small items they should collect in a room.)

Perhaps this post seems a little off topic at first. My intention here is to help you to learn from your project.
In conclusion: I think it doesn’t really matter if you move on to the next project or pick up the old one. What matters is that you have gained experience and have learned from your mistakes.

Bye!

I think the things you pointed out are important to make the whole game experience more compelling, but I don't think they are the reason the game didn't succeed in my eyes. I believe the problem is lack of content. I mean, the only objective you really have is getting a high score or maybe defeating the big ghost, while there should've been more stages and in-between objectives, so they could spend more time with the the game.

Maybe that content could be, like you said, more achievements, and maybe that would work just fine, especially with some pleasing feedback as soon as you achieve it.

There is also zero incentive to get premium power ups(power ups that you need to complete advertising tasks in order to get), and that was the only way the game could succeed commercially.

I think I did learn a lot with Dream Catcher, especially on programming. But I don't think I'll really be sure of improvement in the "game design" section until I have actually made a successful game commercially too.