Design How to handle being "retro" and being "fun"...

Nocturne

Friendly Tyrant
Forum Staff
Admin
Moderator
So, my current game is a retro-styled platform game, based loosely on the Manic Miner / jet Set Willy games. here are a couple of videos to give you an idea:


I've obviously modernised the controls and a lot of the core gameplay elements, but one thing I've kept from the "good old days" is the lives system. The player has 5 lives to start with and there are extra lives hidden throughout the game... HOWEVER, that means that if you lose all your lives the game is over and you'll have to start again. That's pretty damn hard-core, but then, that's also the way it was with the games that inspired this creation and yet people still played them and completed them!

Thing is, I don't think the modern gaming marketplace really wants that kind of super-hard challenge, or at least the majority of gamers probably don't? But, on the other hand, the whole game revolves around the challenge of recovering all the "dark matter" as quickly as possible and without dying, so not having this lives system feels like it might make things too easy.

So what's the solution? I can think of a few things to have:
  1. Checkpoints. You lose all your lives then you go back to a previously saved game state/checkpoint.
  2. Continues. You lose all your lives you have an option to end the game or continue.
  3. Game Modes: Have a "hard-core" game mode and a "casual" game mode, where hardcore is "as is", and casual has more lives and/or checkpoints and/or continues.
  4. Options. You can set the number of lives and/or continues before you start the game and so set the level of challenge.
Which of these options is the best? I can think of good and bad reasons for all of them... :( Also, in general I think giving the player a choice is the best idea, so really the third possibility of adding "options" would be included as part of the solution to the main problem of making the difficulty "scaleable". However, I'm not sure what options to give, and I obviously don't want to make the game TOO easy. Or do I? I mean, there are a lot of people that play games to have a good time and don't WANT it to be too challenging... I AM SO CONFUSED!

Also note that the game has an online leaderboard, so how do I deal with that if I impliment any of the options above? At the moment rankings are based on % of the game explored, time taken, and amount of dark matter recovered... Should online rankings only be permitted if the user plays in the default "hard-core" mode? Or should I have an optional field for "continues used" or even a separate leaderboard for people that require the use of checkpoints or more lives or whatever?

Keep in mind, that I kinda like the idea of people speed-running the game, or trying to 100% it and stuff, so there HAS to be a hard-core challenge, but at the same time I don't want to alienate casual players.

So, what does the GMC think about this? Any and all suggestions are welcome and appreciated, especially if you can give examples or reasons for them!!!
 
Last edited:

Alice

Toolmaker of Bucuresti
Forum Staff
Moderator
I guess it sort of depends on motivation for the players.

Some folks might prefer to play the game for challenge, and possibly to rack up as high score as possible, and for these I'd think the usual lives system works pretty well. You can even add score based on lives remaining when the game is completed. The scoring would only apply until someone runs out of lives - any subsequent gameplay is mostly for completing the game and possibly also getting a better grasp of it to tackle challenges in the next run.

Others might want to play the game for exploration/completion and prefer not to lose too much progress all at once. For these players, I think some kind of continues system - even indefinite continues - should work well enough. To prevent the game becoming too easy, you can have the player respawn not where they died, but rather around some checkpoint. The frequency of checkpoints is something to be found out through playtesting and such.

In summary:
1. For challengers, who want to get a high score and/or complete the entire game on a single run, the basic run-out-of-lives-and-die system should be appropriate.
Any run after using continue option wouldn't qualify for leaderboards, and the "Continue?" popup should make it clear.

2. For those who just want to complete the game, they could use the continue option to start another run with full lives at some checkpoint. The player can use continues indefinitely, but there's still a challenge in reaching another checkpoint. With well placed checkpoints even indefinite continues shouldn't make the game too easy.
(it's somewhat similar to Rayman 2 - most level would be divided into section, and when you run out of health, you go back to the start of section rather than replaying the whole level)

At least, that's how I would approach this kind of problem.

It's worth noting that "Continue?" popup delays the decision until the player actually runs out of lives, as opposed to game modes. It's not like the player will get a run of their lifetime in casual mode without being able to record its score - they can go as far as they can on a single set of lives and record their score, then keep going with continues to see the rest.
 

GMWolf

aka fel666
you go back to a previously saved game state/checkpoint.
This is nice as it requires you to develop enough skill to beat a section of the game with a set number of lives. You get to control that aspect of the difficulty based on how close the checkpoints are.
Where this isn't so nice, is that if you want a higher difficulty and move the checkpoints farther apart, people may get frustrated at having to replay large sections of the game.

Continues. You lose all your lives you have an option to end the game or continue.
This is nice as it give the player a choice in the moment to give it another go or go home.
Hardcore players will do "no continue" runs.
Where it doesn't work as well is that most players will simply choose continue and lives have much less meaning then.

Game Modes: Have a "hard-core" game mode and a "casual" game mode, where hardcore is "as is", and casual has more lives and/or checkpoints and/or continues.
This is nice for the above reasons.
It will get more people to play with no continues as the game mode suggests it to them. However lives may still not have that much meaning in casual mode.

Options. You can set the number of lives and/or continues before you start the game and so set the level of challenge.
I think this is too free form and people won't know what to set it to.
When beating a game they would probably have the feeling that they cheated by changing the initial rules.
It is nice for speedruners who want to create their own categories though.


Where game modes don't work as well IMOnis most players will choose the default game mode in order.to experience the game as it was designed to be played.


If you have a score system, a good solution IMO is to have continues go back to the start of the level, but reset the score.
Those like me who like to progress through games and don't care about score so much, will simply choose to continue and try again.
Others that like to get high scores and generally try hard will restart the game in order to achieve that high score.

In your case you could lose all your dark mater for example, but keep the time and % the same.

Another option is to have a severe time penalty for continues. Or as you said, have separate leaderboards.

I would avoid separate leaderboards and game modes as it makes casual players feel like they are not experiencing the game "correctly".
I think having the game mechanics nudge the more hard core players (like with a score reset) is a more elegant way.
(Disclaimer: I'm not a game designer)
 

Nocturne

Friendly Tyrant
Forum Staff
Admin
Moderator
Thanks for the replies! So... just to clarify something about the scoring system in the game:

The MAIN scoring metric is time. How fast you get the required amount of dark matter. Then the secondary scoring metric is the percentage of rooms explored and then the third scoring metric is the dark matter. You should be able to complete the game without actually collecting all the dark matter or encountering all the rooms (about 90% dark matter and about 95% of the rooms). This is why time is the major score metric... The idea being that skilled players can complete the game faster, with the added challenges of 100%ing the exploration adn 100%ing the dark matter.

I think this is too free form and people won't know what to set it to.
I agree... although I do know that some players like to have options! But yes, I'm coming around to the idea that this would over-complicate things needlessly.

In your case you could lose all your dark mater for example, but keep the time and % the same.
Couldn't really do that, as it would then require users to revist all the rooms they've passed to recollect it since it is essential for completing the game. The basic principle is that you need to collect a critical amount of dark matter to convert all the people in the laboratory back to human again. :)

1. For challengers, who want to get a high score and/or complete the entire game on a single run, the basic run-out-of-lives-and-die system should be appropriate.
Any run after using continue option wouldn't qualify for leaderboards, and the "Continue?" popup should make it clear.
I really like this idea and it would be the simplest to program! I'm more inclined to do something like this rather than have checkpoints as the game is kind-of-non-linear and so the placing of the checkpoints would be difficult to decide... Although it could be done, maybe at the transition between biomes?
 

GMWolf

aka fel666
Couldn't really do that, as it would then require users to revist all the rooms they've passed to recollect it since it is essential for completing the game. The basic principle is that you need to collect a critical amount of dark matter to convert all the people in the laboratory back to human again.
I meant for the scoring system.
When computing the score, ignore any dark matter collected before the last continue, or something to that effect.
 

Yal

🐧 *penguin noises*
GMC Elder
If the main scoring metric is time, you could have a time penalty for dying (and a much larger one for using a continue) so players that go past the default 5 lives won't get good leaderboard times to begin with? (Could even be justified in-game as your body needing to be rebuilt by a cloning machine, which is a slow process)

A lot of modern arcade games also add 1 point per continue used (with the smallest "real" score increment being 100 points, and the continue 'bonus' ending at 99 even if you use more continues) so a score that used continues will be instantly recognizable without metadata. You *technically* get more points by using continues this way, but there's usually "no miss" bonuses (clear level without dying once) or other life-based bonuses that are more valuable so it's impossible to cheat a good run by continuing 99 times.
 
If we look historically, the main reason games had timers for score and were so difficult is that we would keep feeding the game with quarters at the arcade.
Most retro games fell in the "Easy to learn, hard to master" category. This must have been a crazy time for devs... nowadays, if someone buys your game and plays it once, you still get 50$...back then you'd get 0.25$ and that's it!!!
That makes it so that we can now afford to make games that are a little bit harder to learn, longer, and more involved generally. We also don't HAVE to make the player die every 3 minutes to afford some butter on your bread anymore, and we can afford to let him save the game without fearing he will beat it and get bored with it.
But unfortunately, there's no one-size-fit-all recipe for a video game, some will love the "hardcore"/"retro" play style, some will play more casually or go for speed or 100% completion.
I think the key to please as many people as possible, as you mentioned, is to give them options/mods. And that, we can do without limits, these days, even the cheapest cellphone would handle that.
But as far as the base difficulty goes, I've heard many pro devs adopt the saying "Make the game easy for you, and then make it easier", and I also try to vouch by that.
 
Think of difficulty as "frustration". Whenever the player feels something is frustrating or "unfair", there's a chance he'll stop playing.

As Slow Fingers mentions, lives were tied to quarters and there's no real need for it now besides following a canon.

My approach would be to not limit lives at all. Restart to last checkpoint when dying but keep whatever was collected.

Extra : Collected stuff should have at least one way of using them in a meaningful way, typically as currency for a power-up or whatever. If they are useless besides scoring, it won't develop a strong attachment with the player.

Add a time trial with harsher rules as extra game mode. At least they'll have time to leisurely explore the game before going head on into that challenge.
 
Totally agree that the fact that having an in-game timer, and in-game stats makes it much more prone to be adopted by speedrunners.
To me, Goldeneye 64 is still one of the best example of cool rewards for speedrunning and grinding levels (unlocking cheats/mods).
If you make the reward appealing enough, people will replay.
For me, I love unlocking stuff, and wouldn't replay games such as Mario Kart or any fight game if there were no players or stuff to unlock
 

Roldy

Member
While this is a large topic I'll keep it short and suggest that you investigate using checkpoints. In general there are plenty of examples of successful challenging games, both niche and mainstream, that as they increase difficulty they reduce time between checkpoints.

Terry Cavanagh designs challenging games:

Super Hexagon the difficulty is high, but you only have to last 60 seconds.​
VVVVVV with its checkpoint system can be seen as a serious a extremely difficult minigames that last about 5 seconds each.​
Consider the end to the first Uncharted game:

Uncharted used checkpoints. Leading up to the end of the game the time between checkpoint decreases. Eventually the final sequence, heading towards the helicopter, the checkpoints are reduced to 3-5 seconds apart.​
Consider a game like Lode Runner that is room based:

If you die you have to repeat the room. But once you complete the room you progress to the next. There is no lives or death limit. This is basically room based checkpoints. Combined with time being score it works well.​
Well placed checkpoints can help mitigate frustration as well as provide quick attainable goals. The fact that the checkpoints are visible to the player in VVVVVV provided a clear and short goal to get to.

IMO the key to really successful checkpoints (and games in general) are INSTANT restarts. Reduce delay in restarting to zero and players will pull the slot machine lever again, and again.


 
Last edited:

samspade

Member
Personally, I don't hate lives, but I don't prefer them. But if you really want them, I would probably just have the ability to turn them off.

There's a tiny, but very cool game called Overwhelm. You job is to descend into this hole in the ground and kill all the bosses. You can fight any boss in any order, but each boss you kill adds an ability to enemies on the map, so the order you choose to fight them in effects things. The trick is that you only get to die 3 times. If you die more than 3 times you start the entire game over from scratch. This is great for the game, as it's the only real way to really force you to learn how to fight the bosses and experience the way enemies evolve and change based upon the order that you fight them in, but it is also incredibly punishing. Fortunately, there's an assist mode where you get unlimited lives. Eventually I turned it on and beat the game.

Also, you're probably familiar with these but I think that these two Mark Brown videos (Game Maker's Tool Kit) are fairly deep discussions on the topic. The first one particularly offering about a dozen different possibilities (some but not all have been suggested already) and the second discussing why things like assist mode and similar things are good ideas.


 

curato

Member
I agree with check point for your lives. A lot of people these days don't do the large play sessions that they used to. They really don't want to spend a bunch of extra time replaying to get back to where they were.. I would add achievements or something to reward people being able to make it through with the initial set of lives if you want to have a play through like that.
 

Nocturne

Friendly Tyrant
Forum Staff
Admin
Moderator
Thanks for all the replies everyone! I think what I'm going to do is simply keep everything as it is, but add checkpoints every X number of rooms. So the gameplay would be:
  • You start with 5 lives.
  • You lose a life, you simply restart the room.
  • You lose all five lives you get to restart from the last checkpoint, only losing the progress since that point.
Since the main game metric is time itself, I would also consider having a time penalty for using a checkpoint... maybe 15secs or so added to the time? I don't want to overly punish people for using a checkpoint, but I also don't want to punish people that can complete the game without them by letting checkpoint users potentially get faster times. ALTHOUGH, having to go back to a checkpoint and redo a couple of rooms would itself be a time penalty...

Hmmmmmmmmmmmmmmmmmmmmmmmmmm

I guess I should impliment checkpoints FIRST then see how that works before deciding on adding a time penalty or not! :)
 
Thanks for all the replies everyone! I think what I'm going to do is simply keep everything as it is, but add checkpoints every X number of rooms. So the gameplay would be:
  • You start with 5 lives.
  • You lose a life, you simply restart the room.
  • You lose all five lives you get to restart from the last checkpoint, only losing the progress since that point.
Since the main game metric is time itself, I would also consider having a time penalty for using a checkpoint... maybe 15secs or so added to the time? I don't want to overly punish people for using a checkpoint, but I also don't want to punish people that can complete the game without them by letting checkpoint users potentially get faster times. ALTHOUGH, having to go back to a checkpoint and redo a couple of rooms would itself be a time penalty...

Hmmmmmmmmmmmmmmmmmmmmmmmmmm

I guess I should impliment checkpoints FIRST then see how that works before deciding on adding a time penalty or not! :)
It could even just be a little "perk" if you don't use checkpoint, like a little star , or something, in the highscores. Doesn't require much tweaking, and for some reason, people LOVE getting stars in videogames 😂
For example:
1:48 -> Time using checkpoint
1:48 🌟 -> Time without using a checkpoint
 
Retro gaming is one nice sector and fun to and also easiest to program. I personally think that sounds and graphics is what can make an old school game look good. In case of 8 bit style games, simply using different more modern colors can make the game more attractive without making it loose it's retro style. For example, instead of bright green, go for a more pastel style with a 1 or 2 gradients instead of making it simply single color on black background.

As for the sounds, the 8 bit style sounds are, to my opinion annoying to hear. For example, that constant fish splashing in the water in your first video, without changing the sound, to stay true to the 8 bit style, positional sound would be good. When your character is below the body of water, you can make the fish not sound anymore until you get near it again. Same thing for your laser beams, I think a simple visual indicator on the side to prepare the player is good enough once you are past the casting point or when the laser casting machine is out of sight. Having the sound depend on where you are adds depth to your game and makes you feel that you are evolving or entering somewhere even if things are visible on the screen, if we hear them just when near, it makes us feel that moment with a bit more intensity.

What makes a game easy to sell at first is it's image and sound. That's the ads and what people get attracted to. Then, in videos, they can see and feel the environment before playing. So I think that good and well thought graphics and good music and sounds will give a good first impression. After that, you need a sense of evolution in your levels. Something that tells the player that he is advancing. Something as simple as a change of platform color is enough to feel the sense of achievements. For example, a Fire boss can change from orange to green to show it is now poison. You did not really change anything aside from the color. You can throw 1 extra projectile being poison instead of fire. Simple things in retro games can prove to be enough to be big for the player. When Super Mario could throw fireballs, they did not change the sprite, just the colors, that was enough to feel powerful. It made Super Mario boring and we felt we needed that power.

Huge foreground shadows and small background shades of the same sprites give depth to the game. You can use a brick block, turn it completely black and make it's size much bigger and set it in foreground. They use that technique a lot in shadow games where foreground grass in entirely black. It gives a sense of depth while not hindering the gameplay. And in the background, it gives some texture when you do the same thing but with different shades of the same color. For example, bright yellow becomes dark yellow and tiny in the back giving some depth to the game on top of a normal background. In your case, it's a single color dark shade, add a few tiny little platform elements here and there, no need for a completely fill space, just 2 or 3 smaller and darker platforms in there and it's good to go. The space will not seem cluttered and still have this dark background but it adds a bit of decoration. You can even use this technique to show game progression where the more you move forward, the more small elements appear in the background. It's simple but effective depending on your game style.

Retro is fun because you do not need to focus on HD graphic details, just on color and size. As for sounds, you can also play on volume to represent getting closer or further. This also gives a sense of accomplishement of evolution in the level. You hear your beams from afar but do not know what it is until you hear them loud and clear when beside, then, the further you move on, the quieter it gets. You have a sense of accomplishment and it's just sound volume. Like HEY! I'm getting near something, then HEY! I've finally passed it and moving on.

We are not in the retro era where we are limited anymore so cheating a bit on computer resources and technology makes retro games fun, feel modern while still keeping the retro feeling.
 

vdweller

Member
@Nocturne

A very retro mechanic is passwords for each level. Does this fit well with your game?

Also, maybe a hybrid approach like classic Hydorah where you have eg the ability to save your progress only 2 times and you can use it at any point before a new stage.

Lastly, you can differentiate between an easy and classic mode with limited/unlimited retries respectively. I mean, you know this already, just confirming it sounds an OK option.
 
Top