[go: up one dir, main page]

Indie game storeFree gamesFun gamesHorror games
Game developmentAssetsComics
SalesBundles
Jobs
Tags

Robert Yang

122
Posts
5
Topics
4,777
Followers
143
Following
A member registered Apr 28, 2014 · View creator page →

Creator of

Recent community posts

https://radiatoryang.itch.io/rinseandrepeat

Enjoying this but I feel like the weed pulling (at least in the early game) demands a lot of attention for an idle game. I could never quite leave it alone / it was some unexpected low key anxiety

reminds me of the legendary "fruit mystery" flash game

well, I'm a hardhatbeetle in this version, which is kinda cool tbh

I probably don't need to point out what's broken, so instead I'll stick to the big picture:

- you made a lot of great progress as a whole, doing group debug is a really great way to learn development... honestly I'm a little happy you refactored because now you're all collaborating hardcore, despite how stressful it is for you

- but uh obviously in the future, beware of doing a project-wide refactor as you approach a project deadline... most game studios would never refactor everything as late as you've done... or, maybe there could've been a different way to approach the refactor? or maybe the refactor could've been more modular and slower? stuff to think about re: your development approach for the next 2-3 years of projects here at game center

I got to 999m!!! really great progress, it feels like it's coming together

- add sound!! very important for the game feel here!!

- make sure death camera doesn't keep scrolling horizontally, and don't let death camera drop too low (or else it destroys the illusion of your buildings)

- imo you should be more generous with the jump timing, there's lots of instances where I jumped off too late but I felt like I should've been able to jump anyway? in platformer design we call this "coyote time" where we let the player jump for a little while even after they leave the ledge... let me know if you need help brainstorming how to code this

great progress! love how big the map is, and a lot of the core features are working, and the variety of enemy behavior is impressive

I can't figure out how to get money though? everything I killed dropped either nothing (99% of the time) or a heart (1% of the time)... is that how it was in the original version?

the knockback from enemies feels kinda intense and laggy, if possible you might want to tint the player sprite (or use a coroutine to make the player sprite flash red) to explain the stun frames / knockback cooldown before we can move again... also maybe tune down the knock back a little? idk

in general the enemy collision feels too big and ungenerous to the player, like if I just barely graze the enemy I get knocked back 20 feet? if possible, make the enemy collision smaller so that we have to get closer to an enemy before we get knocked back

(1 edit)

love the catfish crash zoom mechanic, kinda wish the fish would've reacted to being looked at? great surreal feel to it tho... and wikipedia says this type of catfish can breathe air and walk a little?? intense!! didn't know this game was so IRL

controls seem a bit laggy sometimes, and sometime link's facing direction doesn't match the movement direction / input?

controls are actually reversed in game,,, for me, X is sword, Z is shield?

cool to see combat working! sword attack doesn't feel super reliable though, sometimes it feels like I should be hitting enemies (and visually the sword touches the enemy) but then the enemy takes no damage? this is mostly a problem with the little bouncey blob monsters... so make the damage raycasts more generous? you can make the raycast longer, or use something like circlecast (which is like a thick raycast)

add in some basic sound for various actions and player inputs will make the player controller feel more solid, so maybe it's time to add a bit of that in

implement pits / damage soon, it's hard to test the combat when we can move wherever and enemies don't hurt us

you might need to scope down in other areas to focus on the core combat more

good progress, and art is good!... no more sprite tasks! focus on the other aspects now lol

movement feels a little heavy and laggy... how are you doing input?

yeah the knockback is uhhh intense lol... some ways to improve physics stability:
- make sure you're using rigidbody velocity or force in the knockback, don't use transform.translate and don't edit position
- weaker knockback force?
- set player's rigidbody to continuous
- if all else fails, you can also raycast behind player to find the nearest wall, and then apply a weaker force based on the distance to the wall

the enemy's fireball shoot rate seems a little weird? is it like that in the original?

good progress though! make sure you leave a little time for juice and game feel... like you need some basic sounds (especially for combat!) and also change the font to be a pixel-looking font

other notes:
- parallax offsets should always be slower vs. the further back you go into the background... i.e. it's weird that the back parallax is scrolling faster than the front parallax, tune the parallax more
- falling animation and rolling animation feel solid enough to me, but I think running animation needs another pass, leg movement is a little awkward, is there a misplaced frame?... also maybe adjust the framerate on the animations, could be a bit faster or more fluid?
- core gameplay is almost there, now push it more and make sure the juice and tuning and sound is there!
- it's ok to replace the credits with your own names in the final version, as long as you make clear this was a exercise for school and you're not actually trying to steal Canabalt lol

https://imgur.com/a/aX96e2F ... bug report: I got stuck on the edge of the building and now I can't fall down :( ... you might need a special collider / trigger on the sides of buildings, OR use PlatformEffector2D to disable side friction, you'll have to see what works best for your scene setup

(2 edits)

(windows, standalone desktop build) took me a couple tries to really get the controls, but once I did, it felt mostly solid and I was able to navigate around between the rooms! good progress

- start thinking about the level design and tutorial now... you need to train players to hold down left-click to aim, etc. and I didn't even know about the right-click thing
- the input feel could be streamlined a little bit... it's a bit strange that clicking = pick up object, but then holding down = throw object? it doesn't map neatly onto a hand metaphor exactly... or show a UI for charging up our throw? also maybe show a reticule for where our mouse is pointing, or always point the character toward the mouse cursor? etc.
- the preview arrow, when we're holding left-click, needs to be a bit more obvious (and maybe inverted, to preview our future trajectory instead of our throw trajectory)
- the clickable area on each object needs to be x3 bigger, right now it feels very stingy and ungenerous... make picking up objects feel juicy and nice (hover state, sound, particles, etc) to encourage players to use them!
- in one of the red dot rooms, I entered a weird soft-fail state where I was very slow and stuck in a corner with no objects near me!... do you have a plan for catching this soft fail state? make the walls bouncey? make all nearby objects automatically gravitate toward the player? (it seemed like maybe you were doing that? if so, the gravity felt a bit weak) etc.
- I mean, there's a scenario where you make this into a hardcore hard-sci-fi simulation where it's brutal and stuff... but the cute character and art style make me expect a smoother generous experience that doesn't punish me too hard, etc.

webgl on chrome

- the player vehicle physics are pretty solid, I was able to drive around pretty easily
- but I had a lot of trouble trying to pick up the capsules...  I think the capsules might be too big, or the claw is too small? I kept hitting the capsule around, until the physics spring got broken and it blasted the capsule away and it fell through the gap
- do more level design, think about what the easiest capsule placement would be... I think picking up a capsule on the ground seems a bit hard for a first challenge, is there an easier situation you could design?

webgl on chrome

kinda brutal, I just kept dying on the first spike pit, was never able to really feel like I'm skating... probably need to tune the player physics to be more generous? maybe A/D can add slight impulse forces left and right? idk

could never really get SPACE working, or if it was working, it didn't feel like I was climbing the rope

also try designing an easy skateboard momentum level, what if level 2 was just a nice half pipe / ramp that lets us do a big jump? instead it's a very short drop with a  wide spike pit, which is pretty hard

and even in your test level prototypes, make sure you clearly differentiate between platforms we can click on vs platforms we cannot click on... give them different colors

Windows, Client+Server (singleplayer?) mode

I was able to pick up a stick and throw it a very short distance? but if I didn't know that was an interaction already, I'm not sure I would've done it, it didn't look very important or clickable

main thing here seems to be playing with the dog... but it's very hard to know the dog's state, or what the dog might be thinking? in real-life we look at the dog's animations, face, behavior, etc. to guess... in this game, you might need to do different thought bubbles or emoji, or draw a line between the dog's position and dog's desire, etc. try to visualize the dog's mind so the player can understand the simulation better?

(1 edit)

- in first screen, the clickable tutorial book should be 4 times bigger, make it very obvious
- the first puzzle should be easier? we don't even understand the goal of the game yet, and we need to rotate the chair? this should be level 3


imo you need a lot more UI feedback...
- preview to player when they are dragging object into an invalid position? it is annoying when the game snaps the object back into inventory with no warning
- show to player when the object is in correct place and giving progress toward next level... see zipzap

good progress though, the game concept is pretty clear and I was able to complete it

I enjoy the contrast between all three spaces. The outdoor area is like a container / hub area that helps us understand the two houses as part of a neighborhood to interpret. Then the peaceful house is comforting and organized, while the surreal house takes a very different chaotic approach and the unexpected drop down into the water is a great moment.

Now, "Disaster Walk" is a cool title but I must confess I still don't really get the theme of "disaster" here really? I suppose the bedroom in the surreal house is a bit disaster themed, with the broken walls and garbage and food everywhere... but it made me think more of a hoarder or hermit more than a disaster area. In that sense, the original title of "Head Spaces" might make more sense.

But if you wanted to push more of that "disaster" theme then, maybe one way is to emphasize the similarity / contrast between the two houses. Right now they feel very different and almost unrelated. However, what if they resembled each other more, with similar floorplan and hallways, but then very different rooms and decoration? Then we would compare them more readily, and understand one house as "pre-disaster" and the other as "post-disaster" more clearly. Maybe that's what you were going for? If so, it would've been cool to see an uncanny similarity pushed more.

Altogether, interesting work and structure, just wish there was more of it!

needs reuploaded windows build (with _data folder) + mac build!

needs new windows build!

needs mac build!

interesting use of physics and the projectile example! I wish there was just a little bit more level design though...

really interesting take on platforming, I like how it's kinda tactical / turn-based? I do wish the level was a little bit more forgiving / easier in the beginning though...

nice use of level design! I found the rick astley!!

wow, so much visual art! the comic format is super interesting, and the relationship between the monster(s) is already compelling

I like the interactive detective elements, inspecting the background scenes is a nice touch, and the quick intro is also an effective hook, well done!

I think the game maybe got stuck between talking to rebub and pranking wendelin... but I enjoyed the simulator aspects, really wanted to destroy wendelin by the end! great characters

... but the visuals felt very consistent and fitting with the theme, and the dialogue / characters felt very plausible, almost as if drawn from real-life experiences,

bug report: code crash when I selected an option... I forget what it was... but it was a bit after the shooting star convo

great dialogue and characters, exciting cliffhanger ending, wish I got to see what happens! love the sense of world and themes going on here

the puzzle design here is intense, the bbq sauce reminds me of sierra adventure games with instant game over! brutal world but love the interactions and choices here

very funny, great characters, love the focus on personal expression and customization... I got the good ending I think, because I'm good at games

gorgeous art, great use of sequences and animation, very evocative... I got the good end because I'm a pro gamer