[the main idea being that the player makes use of information they'd accumulate in the future.]

I am really excited about this. Have you seen Primer, OP?

OP I wanted to discuss designing for time travel; didn't want to hijack your thread & figured you probably have secrets for the game so I created this thread to discuss butterfly effect in time travel games: https://forums.tigsource.com/index.php?topic=57922.0

Unfortunately no, I have not seen Primer.  But that's a cool thread you started, I left my more general thoughts there!

One inspiration for Vision Soft Reset was also a time-travel movie: Edge of Tomorrow.  It was about a wimpy soldier that gradually learned more and more about how to take on an alien threat by living the same day over and over, trying new things each time.  The movie was also supposedly inspired by video games, where players could eventually get through a tough level playing it over and over, getting better each time.  I wanted to game-ify that in VSR, the main idea being that the player makes use of information they'd accumulate in the future.  This is where many of the puzzles and gameplay mechanics come from.  This is my first attempt at this and there aren't many sources to draw inspiration from; I sincerely hope what the game has in store will satisfy what players are imagining would happen.

Within the game's universe there actually is no "time travel" in the conventional sense.  There's the main protagonist Oracle, a clairvoyant, and that's about it.  Oracle has powers I'd imagine a clairvoyant could feasibly have, like seeing visions of attacks before they happen, but that doesn't include physically going back in time.  The time reverting in the Flashback mechanic is really just Oracle seeing what could happen from after the revert.

Having things make as much sense as possible is unfortunately the extent of my narrative skills at this time however.  I'm not really a writer, so even if I did attempt a huge narrative focus in VSR players would probably get annoyed.  Time travel is pretty hard to get right I'm sure.  Still there's conflict, agency, and twists as one would expect.  There's a story but it's mostly just there for context, and I expect players to be engaged by the gameplay rather than the story.

Sadly I have not seen Primer, but the way you describe it seems really interesting, as well as how butterfly effect kind of storytelling can be used in video games. 

This kind of time travel already does happen in games, sort of!  Like if you're playing an RPG where the story branches off into two paths based on a choice the player makes.  Well "the game" wants you to pick one, but it's not hard to just save the game before the choice, see how one option plays out, then reload and try the second option.  In a way, the player is travelling back in time to see how else events can turn out.

Games in the Zero Escape series, especially Virtue's Last Reward and Zero Time Dilemma, weave this concept into their narratives brilliantly.  And it wouldn't be right to not mention that Undertale does this a little bit as well, this post being made on its first anniversary and all.

But it'd be more interesting to talk about what time travel mechanics can bring to games that save scumming can't, right?  One thing I think would be cool is if a game started at the end of all this, and they need to back in time farther and farther making changes that prevent something bad from happening.  Stories that use this butterfly effect kind of thing like to surprise the reader by having changes result in unintended consequences.  But wouldn't it feel super satisfying to a player if changes they made actually did what they expected?  I can see that being a fun path to explore.

I'm sucked in by that bottom image alone.  The colors, the open doorway, the swiggly floor all work together so well.  It's unnerving, inviting, intriguing, and I want to see what it's all about.

This looks really nice.  Building your own engine and sprite editor is super impressive too!  If you want an extra gameplay "hook" you should definitely try to make something out of the sprite editor.  I'm not sure how it would fit or if it meshes well with your existing plans for the game, but unique ideas can come from anything right?  Good luck with Arctica.

This looks cool.
Will the Reset mechanic be used to backtrack and sequence break?

Yup!  While the Flashback mechanic reverts time by a few seconds, the Chamber Flashback mechanic lets you revert time all the way back to when a chamber was visited.  It's kinda like reloading an old save file, but you get to keep many of the upgrades you found.

Progression can also be split into "timelines".  For example, in timeline 1 you go left and find a locked door that requires a password.  In timeline 2, you go right and come across the password.  Then instead of walking all the way back to the left, you can just Chamber Flashback to timeline 1 where you reached the door, but now with knowledge of the password.  But if you already knew the password from the beginning (e.g. on a second playthrough), nothing stops you from just eschewing timeline 2 altogether and going straight for the door...

I haven't yet shown off this mechanic in detail but I intend to bring attention to it in the next update.

You can probably expand your enemies' repertoires of reactions (including reacting to walls and even reacting to complicated moving elements like moving platforms) without having to run a forward-simulation.

So long as you restrict goombas to only interacting with deterministic elements outside of the player's influence (so as many moving elements as you want, just not ones the player can choose when they move or make them disappear), you can just initialize each room with a second, "ghost" copy of every moving game object, already in-place as "one second ahead", and step it forward with your normal physics calculations on each update.  You don't have to predict one from the other or have them interact with each other, except for things like deleting a "ghost" goomba when its "real" goomba dies.

This is a great approach for goomba-like enemies, but it would tricky to set up for different kinds of behavior.  Enemies that need to react to the player somehow (e.g. not attacking until they see the player, aiming at the player) wouldn't be synced up with the ghost copy.  This also applies when enemies die, as you mentioned at the end.  This could be solved by "resetting" the ghost version when a dynamic change in behavior happens, but at this point is it easier than predicting movement and making visions on the fly?  It's situational and worth me investigating more.

You couldn't get them bouncing against each other with this calculation; you'd have to have it so that enemies walk through each other.  Goombas *themselves* aren't purely deterministic -- because they can disappear due to player agency -- so by the restriction above they can't be allowed to interact.  If the player were to kill a goomba a half-second before it was set to bounce off another goomba, it would "de-sync" the two simulations; the future goomba (the surviving one) would already be walking away in the wrong direction.

Yup, any time an enemy interacts with anything it complicates things, so I keep that to a minimum.  For enemies that move back and forth there are predetermined paths that they can't leave.  Deterministic enemies are generally more boring to fight though, which makes it challenging to make them fun.

Gameplay video 2!  This one focuses on the significant improvements to the presentation since the last gameplay video.





New features and levels are constantly being worked on, but none are shown in this video.

Thanks for the support!  The vision mechanic is indeed challenging to implement because of the restrictions it puts on enemy behavior.  The game has to "know" where the enemy will be in the next second, so their movements have to be somewhat planned out.  But if they're too planned out then their behavior becomes predictable, making the game boring and the vision mechanic useless.  Hitting a balance is tricky and forced me to think differently about how enemies work.

My idea was to cheat by making the "vision" the actual enemy and making what appears to be the actual enemy a version of this that lags behind a few frames or something. So instead of predicting the movement of the enemy and their actions I would actually act out their actions but draw a sprite of it that is delayed (and make it so the actual attacks don't trigger until after the delay passes too).

This is a really clever idea, wish I thought of it!  I might have to try this out...

[Sealime]

I feel that combat can very easily make a game feel repetitive and stale, which is strange because combat is usually added with the intention of giving players more to do.  To circumvent this problem in Vision Soft Reset, I want to introduce a large variety of enemies that all feel distinct from each other.  This includes differences in movement patterns, appearance, and even their cries upon death.  By mixing different enemies together for each encounter each battle can feel like a fresh new challenge.

And here are four such enemies below, wonderfully animated by Elias Frost, VSR’s second sprite artist.

Sealime:

A green seal-like creature that is usually content with just flopping around.  But some Sealimes are more dangerous than others, and will fling a barrage of projectiles if they feel threatened.  The Vision mechanic can be used to reveal which one will strike before being in the line of fire.

Ciurivy:

A squirrel-like creature that can lob explosive acorns.  They never seem to run out, but luckily they’re not very hard to avoid.

Magoom:

A mole-like creature that moves quickly while aimlessly slashing its claws.  Magooms will occasionally burrow into the ground and surface in a different location.

Smosey:

A small mosquito-like creature.  Not very threatening on its own, but in groups Smoseys become much more formidable.

This looks spectacular.  The look, the characters, the setting are all intriguing and makes me want to play.  In-game screenshots of levels with actual design instead of being completely flat surfaces is a good sign that it's mechanically sound too!  I'm definitely keeping an eye on this.  Good luck!

Combat feels good and rhythmic.  It's nice how you never really stop moving when doing attacks.

The roof mechanic wasn't doing it for me.  Getting on and off feels clunky.  My sword and arrow can hit enemies on the roof while I'm on the ground and vise-versa, which doesn't make much sense.  Currently it feels less like a roof and more like a separate bounded area that isn't elevated off the ground.

My session ended with a game-breaking bug.  So I was on the roof and tried to jump off near the door.  The room the door leads to was visible for a second then faded away, I think?  And then I got stuck here:



Maybe I hit a trigger for loading the next area during the jump off animation and it messed something up.

The gameplay is solid, basically all the problems I had involved the roof somehow.

Thanks that's great to hear!

Hi everyone.  It's been a while since the last update, and much of that time was spent reinventing the visual presentation of Vision ~ SR.  My own art that I had been making was not meeting the quality standards the game deserves.  So here I welcome Arron Johnson, a skilled sprite artist that has been working on improving the in-game assets.  It took some discussion, but eventually we decided on a great new look for the game, a taste of which is below.

Pretty much everything has changed; from the UI elements to the ship to Oracle herself. The way Oracle is rendered was even modified.  Instead of an ordinary sprite sheet, there are now two separate sprite sheets used to animate Oracle: one for the upper body (e.g. aiming) and one for the lower body (e.g. running).  It's a simpler workflow on both the artistic and technical sides, allowing me to easily change the stance of Oracle's legs based on the slope of the platfrom she's standing on.  It's a nice feature you don't see in many sprite-based games.

Other characters have gotten face-lifts as well, while still retaining my original vision for them.  We also finally got profile images for the text area, handily provided by concept artist Molly (who you might remember from the previous update).

The island tilesets and backgrounds have received improvements as well.  Levels are being reworked to account for the new look of these assets, and they are not quite ready to be shown off yet.  But the slightly-obscured screenshot below shows a bit of what it'll look like.

The .gifs that explain the gameplay mechanics on the first post have also been modified.  Here's a new .gif of the Flashback mechanic:

And below is the new .gif showing the Vision mechanic.  The shader rendering the silhouettes also changed; now they appear as a solid gray color rather than a grayscale version of the sprite.  This was done to reduce noise and more clearly differentiate the difference between attacks and their intangible visions.

Quite the improvement, wouldn't you say?  I'm excited to think it'll only get better from here.

Oracle Concept Revamp

After showing off some gameplay last month I realized I wasn't satisfied with how the game looked.  Despite my best efforts the sprites I made looked too plain and simple, and it would take a disproportionate amount of effort for me to get Vision ~ SR to shine on my own.  Over the month of May I've been chatting with some skilled freelancers, and a few have signed on to work on this project.  By living cheaply I'm able to offer fair compensation from what I make at my day job.  One could look at my game's development as a "really expensive hobby", with "expensive" referring to money and free time.  But since I'm an adult now I can do what I want.

On this post I highlight the efforts of Molly Heady-Carroll (http://mollyheadycarroll.com),  who will be creating  concept art for VSR's prominent characters and bosses.  So far she has been great to work with and you can expect more spectacular creatures to be shown off soon.  But for now I want to focus on the evolution of the design of Oracle, the main protagonist.

Oracle

Oracle is a cybrog, the design of which was inspired by BR-W9 from the 2009 game Contra Rebirth.  The decision to have VSR star a cyborg that can see the future was made for gameplay reasons; I needed a character that had most of their abilities from the start, but would be unable to use them until the player progressed through the game (I’ll go into why in a later post).  With this reference I initially just dove into making the sprites from there without drawing it out first.

As it turns out, getting a cyborg to look appealing while not falling into the uncanny valley is quite a challenge.  The first design had a stoic, creepily serious expression that never sat well with me.  I later replaced the face with a singular “eye” to make it look more appealing.  It wasn’t until I decided on getting a revamp did I consider updating the look further.

Amusingly one person told me that Oracle’s sprite resembled a “purple fox furry”, and someone else drew it like that.  Clearly this was not the intention.  But hey I got my first fan art of the game, so that’s a milestone right there.

Anyway I had Molly design some concepts, and we eventually reached on a final design shown below.  The alterations to Oracle’s “hair” was a great change; making her look more sleek and futuristic without being creepy.  Basically the concept artist made an illustration based on the pixel art sprites… which seems backward in retrospect.

Larva Oracle

Larva Oracle is Oracle’s true, suitless form.  While in the suit Oracle is a dangerous force to be reckoned with, and I thought it’d be funny if it was actually being controlled by a cute bug the entire time.  I still wanted a hint of mysticism in the design, as she’s still powerful with her ability to see the future.  In fact in larva form Oracle’s clairvoyance is a bit stronger, but she lacks any combat ability.  The four spots on her back are receptors she uses to commandeer the suit.

Currently Larva Oracle is only briefly shown in the game on the title screen and during her death animation.  The form will be more significant as the game continues.


Thanks for reading, I’ll be sure to share more designs and other fun insights soon.

I really want to try this out for myself.  Keeping an eye on DANDARA to see how it turns out

Implementing the Vision Mechanic

"Vision" describes the passive ability of seeing silhouettes of attacks from enemies a shortly before they happen.  It empowers the player by giving them an extra bit of time to react to danger.  Vision is a unique mechanic, and to my knowledge there hasn't been a game before that offers a feature like this.

This idea came to me when considering how gameplay would work with the Flashback mechanic, where every time the player made a mistake, they could rewind time and correct it.  This lets the player learn what the enemy is going to do next and play accordingly, which is a neat concept, but the execution didn't seem much fun.  Rewinding every time the player gets hit seems like it would get boring after a while.  Having visions of the future was my proposed solution of preserving the concept, but removing the tedium from rewinding time.  The name was inspired from a similar concept featured in Xenoblade Chronicles, a game I knew about from the Super Smash Bros. series but never actually played.  Later I found out that the vision mechanic in that game works much differently than my idea for Vision ~ SR.  So...it's still unique!  Still my idea!

Visions of enemies (and projectiles the visions spawn) are rendered as cloned versions of the original enemy with a simple grayscale shader.  The visions make no noise and have slightly transparent sprites, which makes it clear they're intangible.  But they're not so faint that the player would miss them.  I also try to have enemies attack in such a way that movement is involved, resulting in a vision that moves around and stands out.

Functional Restrictions to Enemy Design

But that's not the only way this mechanic influences enemy behavior.  Obviously for Vision to work we need to know exactly where an enemy will be and what it'll be doing in the next second.  So for the span of time from when a vision appears to when the attack actually happens, the behavior of the enemy needs to be deterministic and predictable.  This is surprisingly difficult to implement, as it turns out, and forces interesting limitations on enemy behavior.

Consider a Goomba from Super Mario Bros.  It's pretty hard to think of an enemy that's more simple than something that just walks into you.  If we wanted a Goomba-like enemy in Vision ~ SR to create visions of itself, we would need to be able to predict exactly where it would be in the next second.  How hard is that?

If a Goomba walks in a straight line without accelerating, then the position it'll be at is a simple function of speed and time.  But what if it walks off a ledge?  Considering where this ledge is or how above the next platform it is, it can be tricky to tell where it's going to land.  This might be manageable if we didn't also consider that Goombas reverse direction when hitting a wall.  What if, in the span of a second, a Goomba falls to a lower ledge, walks a bit, turns around after hitting a wall, and walks a little more?  Where will he be then?  Short of simulating a second's worth of physics and movement in a single frame, there's no reasonable way to tell!  And that's not even considering the possibility of bouncing off of other Goombas!

As it turns out, even a basic Goomba would be too hard for me to use the Vision mechanic before, as its movements are based off of reactionary behavior.  Enemies with Vision-friendly attacks would need that span of a second to be completely planned out.  In other words, position and state would all have to be determined from a pre-specified function of time.  For example, enemies that move back and forth in Vision ~ SR do not react to the presence of walls.  Instead, they move within static predetermined invisible bounds that make movement predictable.  Given their speed, starting position and time since they first spawned, we can create a formula that gives us exactly where they'll be at any other time.  These predictable, deterministic functions are essential in creating enemies for Vision ~ SR.

Making Vision Fun

And here comes the really hard part: making sure these predictable movement patterns are actually fun for the player to fight against.  On its own this predictability is boring and defeats the purpose of having the protagonist see the future in the first place.

Introducing randomness is a simple way to make enemies more interesting, and while we can't add it to the spans of time when a vision is formed, we can instead use it at any other time.  The movement behavior of an enemy when it isn't attacking can be wild and varied.  The time it takes for an enemy to prepare an attack can change as well, as long as it's longer than the duration of the vision.  We can create enemies that feel spastic and unpredictable when they actually aren't for a specified block of time.  This creates a more engaging encounter that makes the Vision mechanic useful for the player.

Another way is to make enemy attacks absurdly powerful and over-the-top.  Vision gives the player plenty of opportunity to dodge, so while it's easy to not take damage (and even easier when combined with the Flashback mechanic), the player can still feel accomplished for outmaneuvering devastating blows.  Basically the player is OP, so in response I also made the enemies OP and that's called game balance.

The player will also learn to value the mechanic, since Vision can be temporarily lost if the player uses Flashback too much.  Predicting enemy movements without the aid of Vision is much harder, and will force the player to take a different approach, making the possibility of losing Vision a very important inclusion to the game.  It also separates visions from "tells" of enemies in most other action games.  If visions were essentially just tells it would be a pretty worthless mechanic, so I wanted to come up with distinctions between the two.  To satisfy this Visions will occasionally be formed for non-attacking actions as well, such as teleporting or using healing techniques.

And here's one last observation about the mechanic: visions cannot be made in response to anything the player does.  For example: it would be impossible to create a vision of an enemy successfully landing a hit, taking damage from a bullet or dying.  I want to devise enemies that take advantage of this, although at the time of this post there are no enemies in the game that do.  Counterattacks and "combos" from a hit confirm are examples of attacks an enemy can perform without creating visions while still remaining consistent with the Vision mechanic.  As these moves would likely be confusing for new players that are still learning how the game works, these will be restricted to late-game enemies and bosses.

Future of Development

At the time of this post Vision ~ SR is very early in development, so barely anybody knows about this mechanic.  I have no idea if this idea will be received as fun, obstructive, or forgettable.  It's a risk that's going to require a lot of playtesting and tweaking, and I hope people enjoy it!

EDIT: Position is now closed.  Thanks for the interest everyone.


Hello.  I'd like a sprite artist to help me out with a game I'm working on.  As a solo dev I've been giving it my best shot but I'm unsatisfied with how it looks so far, and I'm interested in commissioning a more experienced artist to improve the presentation.

The game is an action-adventure platformer about a cyborg that can see the future.  Along with running and shooting the player can also rewind time and see visions of enemy attacks, among other unique features.

Webpage for more info: https://visionsrgame.com/

GIFS of the main mechanics in action:

Seeing enemy attacks before they happen:


Rewinding time to undo taking damage:


Some other screenshots:

Climbing the ship exterior:


Base landing in a snowy area


Ambushed


Cave area


The foreground tilesets and enemy design are what need to be improved the most, so those are the fields I'm interested in.  I'd also prefer someone that has genuine interest and prior experience with the game design.  It's one thing to just make some sprites, but it's another to really consider how a design might work with or affect the rest of the game.

This is a paid opportunity with flexible compensation.  To keep it simple and fair, you'll be paid per asset (such as tileset, enemy animation), or per hour if that's preferred.  PM me for more info regarding rate and my budget.

PM me with your portfolio if you're interested.  Thanks very much for your consideration!

I really get a kick out of hand-drawn level plans on graph paper.  It's pretty neat comparing your recent 2-1 screenshot to what you planned out on paper to see how it evolved.

From the screenshot it looks like some areas are walled off.  I assume that's part of the background, but in-game it might be confusing since their tilesets are so similar to the tilesets of the tangible platforms.

Man this game looks cool.  How far in advance do you plan your map?  Do you have an idea of what will go into all of the gray and dark blue areas? 

[how the design of the game came to be]

Figured I'd start this dev log from the beginning.  So before I forget it, here's how the design of the game came to be.

Initial Concept

Among many other sources of inspiration, the core idea behind Vision ~ SR came from the 2014 movie Edge of Tomorrow, where Tom Cruise’s character learns more and more about how to stop an alien invasion by repeating the same day over and over.  This movie was supposedly inspired by hard levels in video games, where the player fails a level so much that they eventually beat it just from remembering where all the enemies are and what they’re going to do.  This always seemed like cheating to me though, as the protagonist within the universe of these games would have no way of knowing all this information about the level ahead of time.  But what if there was a game where this was the entire point?  Would it work?  Would it be fun?

Contra and Contra: The Movie

In the summer of 2015 I came up with the following concept for a game.  The player is given a short amount of time (e.g. 15 minutes) to collect four keys in some sort of labyrinth.  The majority of the game would be preparation; using the short 15 minute periods to enhance the player’s own abilities and determine the location of the keys.  The paths taken would resemble a tree starting from one root location.  One path would have the player heading right, one heading left, one heading left and then down, and so on.  Ideally each path would serve a purpose, like determining the location of a key or unlocking a new ability.  Once the player has learned everything they need to know about the labyrinth, they must use their knowledge to plan a route that would allow them to snatch all four keys and unlock the final door in the short 15 minute time limit.

This structure interestingly resembles basic concepts used in graph traversal.  The initial runs are like depth first searches of a maze, and the final speedrun-like segment resembles a solution to a “travelling-salesman” problem, a classic “hard” problem in the Computer Science field.  For a while the game was called “Nondeterminism” for this reason, but the name was dropped for being a mouthful to say.

Vision ~ Proto

With the general structure of the game planned, that left the question of what the minute-to-minute gameplay would be like.  It could be anything really.  I chose a 2D platformer with shooting combat just because I like those kinds of games and had experience developing them.

I also wanted unique features that fit the “seeing the future” theme.  This is where the titular Vision mechanic comes from, which allows the player to see silhouettes of enemy attacks a second before they actually happen.  Going backwards through time, denoted Flashback, was also thought of at the beginning but initially I wanted it to be an unlockable ability the player would get later on, and wasn’t even sure the mechanic would be fun.  But I figured if there was even a small chance of including this mechanic, it would have to be integrated into the game’s code from the very beginning, which is why it was developed first.

To test out the minute-to-minute gameplay, I created a prototype version of the game called Vision ~ Proto.  Proto was very short, and merely consisted of waves of enemies that the player had to defeat.  It showed off the gameplay, but little else.  Proto was a two-month project with a hard deadline of November 20, 2015, the date of that year’s Rensselaer Game Showcase (RGS).  RGS was an annual event at Rensselaer Polytechnic Institute, the university I was attending, that allowed students to show off their game projects to the student body.  Naturally it was a great opportunity to collect feedback from players.

Vision ~ Proto at RGS

Vision ~ Proto Screenshot

People enjoyed it, I think.  Nobody hated it.  That was good enough for me, and I made the push to expand the game into Vision ~ SR.  I knowingly added features I knew would be removed for the real game, like a scoring system and a simplified menu, which resulted in some wasted assets.

Pushing the game to the next level was a long and arduous challenge that required me to work on it at least a little bit each day.  Menus, upgrades, dialogue, a tutorial, multi-screen rooms, multiple rooms, a map, more enemies, more environments, and even the initial concept had all yet to be implemented.  It took 4-5 months to evolve Proto into the game’s current state.  Now that many core features are now implemented (with the notable exception of music), interesting updates should hopefully be coming out more frequently.