Cool, thanks for sharing.
reading through it id like to bring up some points I find interesting.
The chatgpt spent 45% less time studying than those doing it traditionally and scored 11% lower on their retention test (57% vs 68%)
In one section they discuss that the biggest disparity happens on complex subjects, it seems a structured course is better. This makes the next part I mention below an interesting choice to me.
From what I read the content for studying was "AI/ML topics spanning foundations, methods, applications, and ethics". This seems to be a pretty complicated subject to me. The students chosen for this test are Business Administration students. Im guessing its a pretty large learning curve from that to machine learning.
They do mention that there is still disparity on other topics but it looked like it was about half as bad.
In my personal opinion I do think it is a crutch, a stiffer crutch than google even, however I think they may have painted the picture a little worse than it is.
There is a lot of things in this world to know about or understand, sometimes knowing where to find an answer is our best option.
This discussion does remind me of math class at least when I was in school, simply putting the correct answer on the test did not give you full points, you had to write the process you used to get to the answer, demonstrating you actually understand the process. If you truly want to learn something effort, time, involvement, execution all greatly help in retention. For me putting it into real world practice is most beneficial.
Lordricker
Creator of
Recent community posts
I have tried it again.
-The ai racers does help engagement a little but i dont think they can make it around the track, at least in my attempts they didnt.
-I tried multiple ships and could barely tell a difference
-i used the w for throttle and a and d for turning control scheme, it felt pretty good to play
-I like the bigger tracks, the smaller tracks were pretty hard to drive on, although i wouldnt mind smaller sections on a big track.
may i ask why you have it so that it is zoomed out to see the whole track? I would think having a fixed camera zoom that follows the player around the track would be better, but that could be just me.
Hello!
I am excited to share what I have been working on.
It is a pixel art fighter game were you defend a castle.
For this post I am just going to go through it chronologically.
Before the initial commit I had been working on the first character of my game, at the time simply known as Red Knight. I hadnt done pixel art before so it took me a bit to figure that out.
March 5th- First commit, at this point I was wanting to start the game dev part but as it turs out that wouldnt happen for another month. It was kinda addictive polishing frames of the Red Knight animations. I had started with a basic thrusting stabby animation and that went well, i also did a slashing animation and a walking animation and idle animation, surprisingly the walking animation was the most difficult.
April 8th- I watched a few tutorials (thanks brackeys) and finally opened up Godot to start putting the animated knight to work. Using a sprite sheet i added the frames and had him running around. I had found an asset bundle on itch called tiny swords that really fit the vibe I was going for so I use a lot of sprites from that pack to fill in stuff that would just take me way to long to make all myself. Added a black knight so i can wack him and test hit boxes and stuff.
April 10th- I was having an issue that stumped me for a while where when a character changes facing direction the hitboxes did not flip over to the other side. I think this was an error in the setup of the character scene where I used a node 2d as the root and put the animationsprite as a child, I am not quite sure, anyways the solution I used was to put a Pivot child and put anything i wanted to flip over the y axis as a child of that and then when changing directions the pivot gets flipped.
April 13th- Up to this point i didnt have any hp yet, i had added some knockback so i could see if a hit landed but at this time I decided to implement the "hook" that i had been planning. I drew up some containers and implemented filling them, one was for HP and the other one i call the Flow Bar. Basically when you press an attack button it doesnt immediately attack, it fills the flow bar and you press the button again in an attempt to hit the small green window inside the flow bar. A successful green window hit will execute the attack immediately on the second press. If you miss then you deal 80% the full damage and you have to wait for the bar to fill the rest of the way before the attack executes. I did this to try and simulate weapon proficiency, requiring skill to use a weapon efficiently.
April 15th- For the tests before I had simply placed an enemy in the scene and he would do his thing, walk toward the castle and try to kill it. So on this day I added enemy spawning mechanics. Its pretty simple, there is a spawn point on the right and left side of the map, spawner randomly selects a side and spawns an enemy every so many seconds, they have a path they follow and oscillate along. For the spawner i added a curve graph for spawn rate and max enemy count where you can adjust how fast or how many spawns there are at a given elapsed time during a run.
I also added player spawn and castle death at this time.
April 16th- This is the day Posted the game to Itch
Also I added a healing monk and a blacksmith guy in the castle, blacksmith sells upgrades to help your run. on the 17th i added coins and stuff so the upgrade system could function
April 19th- ARCHER HAS ENTERED THE FIGHT. I added a playable archer character
April 20th- now that i have two characters i decided id like to add multiplayer, I just had to find out why everyone says dont do this...
needless to say it took multiple days to get it working, still its not perfect. I went with a peer to peer setup so a player starts a run and they are the Host and someone can join it as a client, most of the game runs on the host and the client runs his own characters animations and flow bar and sends it to the host. It seems playable enough to me.
April 26th- I added a main menu and stuff
May 3rd- I added some new enemies, Archer and Spearman
And that is about where I am at!
Right now im working on a logo thumbnail thing to hopefully draw some more attention. I also plan to do sound design and more enemies, those are the big things to add.
If you made it this far thanks so much for reading. I feel like this game is going much faster than my last attempt, I am very grateful for anyone who tries them out, I hope you have a wonderful day.
Hello
i tried out the tutorial and the painter
My favorite part was the cube boss, that was engaging.
The art itself looks good, it was a little odd looking for the character to just change vertical lanes as if he was running in the air. you might try a higher view point looking down more instead of the straight side view.
Definitely feels like a rhythm game, story seems like it will be interesting as well.
Thank you so much for your video playtest. Watching someone on their first time is always invaluable.
Definitely took a lot of notes. I have some ideas on how to help convey what to do on the very first attempt to help avoid turning away new players.
To address some of your points of confusion:
-The game is multiplayer, when you are starting a run it gives you the session id that others can join and allows you to make it a private run.
-The flow bar idea is to simulate your proficiency with handling the weapon, figuring out the timing (or cadence, or rhythm) will basically double your dps. maybe its too much realism but someone good with a bow would be able to shoot much faster than someone just starting out.
-On the archer i added a combo mechanic, i have not added the description of it yet tho so sorry about that (not your fault). Basically the if you hit consecutive shots the arrow trail goes from white, to green (1.3x dmg), to purple (1.6x dmg). The green window does not affect the combo, (i figured that would make it too hard). If you are getting purple arrows and the green window then the archer is pretty devastating.
-Your thoughts on the balancing of the early game are really what i needed since i literally cannot test that myself (i already know how to play)
i do have spawn rate curves and stuff already implemented so its a super easy change, after watching you play i do see it gets overwhelming too early.
I do have one question for you, could you see yourself playing it more if it was more fleshed out?
Again thank you so much for your time.
Thank you so much for your feedback!
shop refresh- I think I agree, it is a little annoying to have to wait for the one you want when you have a lot of coins. Also would fix spamming attack only (need to fix that)
Attack Range- question on this, did the range feel too short? i assume you mean on archer, his j attack fires with less power than his k attack.
I have been brainstorming a way to aim the archer since yes it is clunky to aim (literally just moving up and down)
Melee interrupt- is it actually stopping your attack? (that isnt supposed to happen) or is it just annoying they knock you back.
enemy indicators- yes I think that is needed now that you mention it. I think ill add lookout scouts to the towers maybe that indicate threat level in their direction.
castle hp- adding to list
thanks again for playing, 7 minutes is a pretty good time
alright i started a playthrough yesterday, i was skeptical at first but i think i can see the draw.
I see the missed events and it does have an impact on me.
There seems to be an issue with missions, i started 4 of them yesterday and they all just say returning today. I dont see any button to finish them
In the reports section for the events that have happened it gives flavor text but i dont know what the actual effect is, did kicking out the ration thief help morale? did it help our rations situation?
This kind of reminds me of clash of clans, in that you do a few things log off and wait. Unfortunately i think the big draw of that is that it is multiplayer, im not sure if that is in your scope but it would definitely help engagement.
hope this helps,
-just a guy who plays games for fun
Hello, I did try your game.
I like the art style quite a bit.
Unit path finding works great from what i played.
I couldnt quite get into it though, i played the first few levels and my brain just really wanted it to be an RTS type game instead of a tower defense.
one issue i had is once you start the wave whatever units are on the field that is all you get, in a tower defense game usually a tower is permanent, in this game units are not towers and they die but dont come back? all my units died and there was only one enemy archer left, the king did not attack the archer.
My recommendation is to play some popular tower defense or unit management games and pick it apart to see what about them makes them popular.
The original incursion game was a lot of fun back in the day, it had plots of land to put a building and each building had a type of unit in a set of 3 (archer, warrior) and you could upgrade each unit individually, for example the archer building could have two long bow archers and one bomb archer. if one dies they have a timed respawn.
https://armorgames.com/play/14138/incursion
Again i think the art looks great, just needs some meat to the mechanics.
This is pretty cool!
The tutorial worked great, I understood pretty much everything I saw very quickly.
It reminds me a lot of a board game type game.
I did play a few rounds and then i attempted to stop the character on one of those crumbly blocks but i stopped one short and fell to my death, definitely liked the puzzle aspect of this.
One bug I did have was right after the tutorial it congratulates you for completing the tutorial and that message didnt go away, couldnt see the middle card in the vending screen.
Art looks good!
No need to apologize, you have constructive criticism and thats what I need! I just wanted to figure more of your view point.
I just tried it again and I do see the issue where it looks like you hit it but it didnt hit, I was apparently just ignoring it. I have an idea to fix it.
A couple tips on the knight, he has longer reach than black knight, the k thrust attack gives invulnerability, the L spin attack has really strong knock back.
You mentioning it feeling stiff has crossed my mind, I believe ill try starting the flow bar as soon as you press the button instead of waiting for the animation to pause, should speed things up.
I wanted to like this, since the visuals look good and i really like sailing games.
-no camera control, stuck in one view, need a way to look at the nice visuals
-nothing happened when i ran over the chest except the character got slower.
-tried boarding the ship, player doesnt move with the ship, its basically just land. also water and ships have no interaction.
-fell off the boat and sank to the bottom, could have the player bob in the water instead.
Thanks so much for playing it!
My intention was for it to be rather difficult like certain old arcade games to give it a higher skill ceiling.
sounds like you played the archer, Im still working on his L attack animation so thats why it was missing, probably some sort of kick.
Archer J attack shoot a slower shot and does not pierce, archer K attack shoots a piercing arrow at a much higher speed and deals more damage at the cost of a slower charge and smaller window.
On the green perfect shot part did it fire immediately when you got it? or did it turn grey and keep filling to the top then fire? the main benefit to hitting the green is that it basically allows you to shoot twice as fast since it executes it immediately.
Ive been brainstorming a way to aim the bow since i agree that would be nice. For some reason in my mind im avoiding using the mouse for controls, not sure exactly why but i think there is a reason.
I tried it out, just on keyboard.
game ran smoothly so that is good
i tried the three different control schemes here are my thoughts.
-mouse only is too easy, if you have ever driven go karts at a place where you just hold the pedal down the whole time thats the feeling i got from that, no skill ceiling
-A+D and Space was not too bad to get the hang of
-A+D only, this was pretty difficult, however im not sure that is a bad thing, you might even consider doing a mash up of the control schemes where certain ships have benefits and draw backs, eg: ship goes really fast but has only two inputs (left and right thrust)
i was a little confused on the stats of ships, i tried the grey ship which had low top speed and high turning stats first, then i tried the blue ship that had high top speed and low turning but the blue ship turned way faster than the grey ship, it may have gone a little slower but that wasnt very noticeable, really the speed of any ship i tried felt about the same.
cool concept!
if you could please try out my game, the one im working on currently is this one https://lordricker.itch.io/cadence-blade
I am looking for feed back on this project I am working on. It is an arcade style fighter where you defend a castle.
Current features:
-fully plays in browser https://lordricker.itch.io/cadence-blade
-Tested on Windows and Mobile browser
-Playable Knight and Archer
-multiple purchasable upgrades (lasts only during that run)
-2 enemy types
-TWO PLAYER ONLINE (im pretty sure this works, hard to test by myself)
-roughly a 5 minute play loop
Current road map:
-add more enemy types and behavior
-add a playable rogue character
-spruce up some ui and artwork
Questions for you!
Is it fun to play?
What was your best time?
If anything was possible what could be added to this to make you want to keep playing the game?
I think I will have to try that, starting the game with a premade tank, the issue in my mind was that the tank data is saved to the players AppData folder so at first i didn't think that would work but now that I think about it that can probably get forced on first load up of the game.
You have revealed to me a key instruction I missed in the editor, it says you have to purchase the ai files first but it should also say you have to load them. I didnt even realize the menu popped up before loading, instead it should show a message saying to load a file or something, completely missed that since i normally just click load.
anyways it is functional you just have to load the file you want to edit first then the menu works properly.
were you able to figure out and follow why the tanks were making the decisions they were making?
was a charming little platformer. didnt really notice any bugs. music was nice.
the spring pads are little easy to miss, also i might have missed it but the wall jumping mechanic i dont remember being mentioned. took about 10 minutes for the demo.
all i can really say is add some content, i would have been happy to continue.
i made it to the top of the huge ramp, not sure but i think that was the end. has a trackmania vibe to it.
the car looks great, and it handles well at low speeds, but at high speeds it gets too bouncy. That being said this looks like an old muscle car so that might actually be realistic. I also think it would be less of a problem (the bouncing) if you smoothed out the track. something i think that would really help is level select and a timer that saves your fastest time for the track. I did not have much issue keeping myself entertained even with that tough curve in the last level (a little more grip would have been nice right about then.) Sound effects, some track textures, they are not really necessary to make the gameplay fun but would help immersion.
slight bug when resetting it should remove velocity, sometimes it would throw me off the track immediately.
Hello, I tried it out. It works but there is not a lot to it.
you might think of putting some upgrades in that you can purchase after a run or something like vertical move speed or horizontal movement? maybe have some moving spikeballs to add some challenge or things to try and collect to get extra points.
the parallax was cool, however it sped up a lot and the spikeballs didnt speed up that much
5. As far as i know it operates the same way its just to the right instead of down. lets say you have ifenemy and it has two connecting lines from its output with ifrange>100 on top and a ifrange<50 under that, the ai will check the top branch first (>100) and if that is false it will check the (<50) and if that is false it will then backtrack to the next available branch up the tree behind ifenemy. setting up the logic had my head spinning for a while, so i understand lol.
7. that is fair.
8. i will implement this, i think ill just make it like fire 0 by default and the description will tell you that you can input a number that will lead the target before firing, i think at one point i thought being able to fire without aiming would be useful but cant think of an application now...
9. oh, i could do that.
10. i believe there is slight acceleration functionality but i may increase it to help smooth out the jerky driving. I do know that tank weights affect speeds and stuff, along with the engine stats, but it may not be noticeable enough, hard to tell unless you have two tanks driving along side each other. You can blame the game, i can take it. all for the effort of making a better product.
I am actively working on some stuff to alleviate the "cant find each other" problem. one solution especially for early levels is to use mapcenter instead of wander for enemy navigation. Also having more than one tank should help fill out the field and provide more vision on the map so less lost puppies. Somewhere at the midpoint the tanks start using COMS node which allows them to see their allies vision and make actions based on that (ie. tank 1 sees enemy, tank 2 vision cant see the enemy but with coms it will chase, or fire at the enemy based on tank 1's vision) penalty for coms currently is slower driving and slower ai node processing (how fast it makes decisions)
Thank you so much for the detailed feedback, and actually trying the AI editor, since I feel it is a pretty big part of the game and I was really needing some eyeballs that are not mine to look at it. Originally I wanted the editor to look like unreal node editor but I couldn't even get a straight line to do what i wanted much less a spline.
1.2.3 Your description of your experience definitely gives me some ideas to make it better, the overload of info right at the start makes sense (as in don't that)
5. the guide does mention that if false it will backtrack, I haven't used the unity behavior designer but the screen shots seem to show similar tree but it goes down instead of to the right. I guess my question is if the backtrack part mentioned in the guide was missed? if you did that line (maybe you at the time you wrote point 11.) and were still confused ill have to revisit, maybe green output bubble and retry an explicit description.
7. A visual indicator for tanks vision cone did cross my mind, i dont think it will hurt anything to add that. Did you see the debugs in the top left of screen when you follow a tank? It shows what ai paths are currently in use.
8. I struggled quite a bit on how to implement leading a target, trying to keep it as modular as possible and not make the text in the node excessive. I think what you are saying is give fire an output and connect lead target to that? is there a reason for fire to ever not use lead target, instead just make them one node.
9. 14. As you play through the campaign the shop populates with new AI files to purchase, these files are what the enemies used, nodes dont actually unlock but they do get shown in the later files that enemies use.
10. 18. taking note on QOL features that would be nice, currently the SubAI node can be used to copy another ai tree directly into the open file, allowing modular design.
12. im planning to rework the campaign to start with more tanks, probably 4 as just 1v1 is rather stale.
13. tank selecting targets: i did try to implement a tag system, i need to test it some more.
15. ill have to check on that, in my tests id notice they would start turning to flee but it takes a while. I did not notice them driving the wrong direction so maybe i broke something lol.
16. The turret and body operate independently, the turret can point backwards (and shoot) while the body drives forward, the body cannot drive in reverse currently, only rotates and drives forward, i ran into some issues on backwards driving but might revisit if it makes sense.
17. the ai file can be programmed to mitigate the collisions using wait commands and ifrange, i need to put that into the base files.
you mention smooth feeling tank behavior, im not exactly sure what you mean? like physically the parts of the tank move very rigidly, maybe the body needs a slower acceleration to mimic reality a little more. or do you mean inside the ai editor the constraints on design are too rigid?
Anyways thank you so much for your feedback, I know it takes time but without it we would head down wrong paths and waste our time anyways.
Alright I played through the demo, it was really well done. It is quite difficult but i enjoyed the challenge. Its not necessarily what i think of when i think cozy, but my definition may be off. I say it was difficult but i guess i mean difficult to get perfect score, after the first two levels i did fill all the ships. I did save scum a few times (the fuel wipe of darkness can really put you in a bad spot if it comes at a bad time) also I went around preparing everything before lighting the lighthouse, not sure if that was intended.
The one thing I didn't fully grasp was the main engine fueling the light, the first time I died I wasn't really sure what happened, i then realized i kept letting the light go out and it does show the warning symbol. However once the screen starts greying out and the purple smoke comes in (very atmospheric, well done) and i start filling the fuel back up it doesn't make the smoke go away till it is at 100 %, so do you keep dying unless its at 100%? then in the last demo level it wipes your fuel tank so if you were trying to keep it at 100% you now got punished for it.
That was the only part that wasn't quite clear, still had a lot of fun playing it, before i died the first time i was like "man what kind of feed back am i going to put, its all done real well..."
art is cohesive, story is good, oh wait another small feedback note, the text did not match the talking speed very well.
Yeah good stuff here! Good luck.
Ive been looking at some game Jams and it seems pretty common to have a zero tolerance ai policy. Id love to join but my current game would not be eligible.
I am interested in where the line is drawn since I believe most software includes auto complete or predictive features. When people say no AI do they just mean prompting a chat bot? (in reference to the coding side, I think the art side the line is pretty clear)
Hello, I am looking to improve the onboarding experience to my game. I have had a little bit of feedback and figuring out the tank assembly and starting a match is catching people up more than i had hoped.
Id also like to know if editing the tanks Ai files for the tank is achievable by an outsider as opposed to me who created it.
Thank you for trying it!
I will consider starting the game off with a tank already configured.
The way it is set up currently is a little harder to understand than I had hoped. You need to first purchase one of each of the 5 categories from the shop and then select a tank slot and then equip one of each category to that slot.
This was a nice little rpg! its a pretty cute flip on you being the slime getting attacked by the humans.
The tutorial was pretty thorough however it was just a lot thrown at you all at once. im kinda just being picky though. I had a little confusion about equipment, pretty much everything is consumables and you just chuck them at the opponents i guess and each of your characters has an affinity for specific projectiles.
It did feel a little easy, you get so much stuff it would be pretty hard for any competent person to fail, but its a light hearted game and that does seem pretty fitting. Lots of options to expand as far as story direction and difficulty later on so not really a complaint, just feed back.
played in browser with no issues, i left it open since i didnt finish, they just keep going. I liked the story enough to read it so that is good.
The ui art looks decent, however the map art like houses and stuff are quite basic, but hey you have to start somewhere and this is a good start!
It took a few minutes to figure out how to play, when you start the actual round you immediately get attacked, so i died cause my bot didnt have a gun lol. I could see day night cycles but the enemies didnt seem to care if it was day or night, they just pop out when you are near by.
This led to an issue later where i went searching for more crystals, a lot of enemies start popping out and it basically lagged my game out, it didnt crash but it was unplayable. Having lots of enemies come out is fine but this game doesnt really strike me as resource hungry so im not sure why im having trouble running it (to clarify I am on a laptop so its nothing beefy but it isnt an old laptop either, and i did download the windows version as the web version froze pretty quickly for me).
a few QOL things that id like.
- hold click to fire continuously, maybe make it not super fast so you can panic rapid click or something but having to click both mouse buttons on a double laser bot for every shot is kinda awkward.
-delete and rename options for the custom bot builds (torlai and i had the same idea apparently with the double drill guy.)
Last thing, the capsule and name of this game give me no indication as to what type of game it was, like i have no idea what a root warden is, do they protect trees? I actually played it cause torlai commented on my game and i saw he commented on this one too. The objects in the capsule are too obscure, a triforce and a button looking thing, i could not tell there was a bot on there from just a glance.
The potential is there and I could see the fun, wish you the best!