[Feedback Friday #3] - October 31, 2014!

I smiled a bit when the snake 'deflated' while sleeping. Much better.

Now you'd want to work on telegraphing some things, because as the player it's not clear why the snakes are doing what they're doing.

1. The snakes should rear back momentarily and make a sound, before charging.
2. The snakes should shake their heads or something when they wake up.
   • Also, instead of Z's you could try little yellow birds... first three birds, then two birds, then one bird and then they wake up, shake their head, and return to normal.

Here's a thought. Make the rocks something you have to pick back up, and only give the player 3 rocks.

Jumping is still too weak, Maybe a 5-10% increase in jumping height would make a big difference. Every single jump feels like I *barely* make it and it's not adding to my enjoyment. I like the idea of throwing rocks at snakes and sneaking by them, not jumping on platforms I can *barely* get on. Also, jumping and barely moving 5 pixels left or right is starting to get on my nerves. I can only suspect this will continue to annoy more as time goes on.

 

I'm currently adding a boost system that will speed the player up while draining a meter. I think part of the player feeling too big may have to do with presenting a vertical shooter on a horizontal display. I tried rotating the viewport so it scrolled horizontally, but it felt very wrong.

 

The best is when your phone browser crashes when you've got three paragraphs written. :/

You could make it so that there's a button to toggle which current targeting mode you're using... a simple toggle between free-aim mouse mode and vertical shmup mode.

I greatly prefer the free-aim mouse mode, but in certain situations being able to strafe and shoot would be super useful.

 

Yes, very much so. Enemies are not just going to attack in a linear manner, but it's hard to get across with such limited content.

 

I'll look into this.

 

That's a great idea. I'm thinking right bumper on the controller/ X button on the keyboard.

 

The easiest way to think about it is to have a single Difficulty integer value...

When the previous wave of enemies is dead, use some pre-determined logic to build a wave out of prefabs and the likelihood of more difficult monsters spawning is randomized using Difficulty as a base. So badass monsters can only spawn if the randomized number is > a certain minimum value... which the higher difficulty is the more likely higher numbers will appear is.

Okay, I guess it is complicated. But it will save you $$$$$$ on building levels.

 

Interesting. I suppose you could also modify it to take the player's condition into account, something like Valve's AI director in Left 4 Dead.

 

Nice. I'll be sure to give it a go when I get back to my PC.

 

Jackpot. Then your levels are as long as you like. If you don't like the enemy composition, alter the algorithm... I was going to make this TDS (I probably am going to actually, now that I'm getting my S*** straight) and I was having issues with how to do the map, enemy layout, etc... in the past level design was a major reason why those games were so short. Most TDS can be completed < 1 hr.

If your guy is near death, wow crazy... that enemy dropped a life capsule. It's like the game knows what I need.

The difficulty keeps pulsating low, high, low, high and each peak is slightly higher than before.

 

@Misterselmo: I'm a firm believer in having enemy attacks and other dangerous events telegraphed. I implemented it for the venom spit and kind of for the charge by making them hiss right before they do it. Mainly I didn't do it because the next work the artist is doing is for the snake and I figured I'd wait til he gets the animations done. I had thought about making the snake's eyes pop a bit with maybe an exclamation point over his head, hissing then charging. The Zs was just my quick graphics. It would have taken hours for me to make stars or birds. And it is not worth it when the graphics will be redone anyway. I want to have them "seeing stars". The artist can probably knock that out in 15 minutes and it'll look great.

Ha ha and yep that is exactly why the rocks stay in the playfield. There will be a limited number and you will need to collect them Again a case of the 50 being a convenience for developmental play testing.

Putting the game here for feedback has been very valuable. I can see how much "inherit knowledge" there is in our works. In my head and on paper I have this large concept but in the WIP, of course, nobody can see the things that exist only in my head or on paper. Next time, I will include more info on the overall game. What you see here is just what I have implemented so far and is about 25% of the project.

 

Sounds cool. When you update it, I'll give it a try then too. It's interesting to see it evolving.

 

The new sleeping effect for the snakes sleeping works very well. Good job! I noticed that hitting a snake from behind in one instance caused it to scoot forward to the end of the ledge it was on instead of falling asleep. I think that exclamation points and stars are a great idea. They are well known symbols for exactly what you are trying to describe, if not universal (I wonder if the exclamation point was something else in the Japanese released of Metal Gear Solid). Being able to recollect your ammo makes a lot of sense. I believe in telegraphing enemy attacks too. Otherwise, it can feel cheap.

It's very interesting how feedback differs when it's blind vs. with explanation before hand. I guess that developing the ability to parse it before our projects go out into the wild is a great reason to post here.

 

Beautiful. I'm imagining a saw wave, where the peaks are a function of the player state vs. difficulty, and the valleys are the inverse, with a touch of old school random encounter generation (and now I'm really missing the old D&D books I got rid of years ago). I think the hard part will be not just determining the difficulty of an enemy based on its health, speed, flight pattern and offensive capabilities, but also when grouped into formations, and the combinations of different formations so that they can be placed algorithmically. I'm really going to have to think about how ranged weapons are distributed as well, possibly redo the entire system.

 

It's probably not very noticeable due to the size and poor collision registering but the snakes only get knocked out if struck in the head. When struck below their head and facing you they are enraged and rush forward. When struck below head from behind they are startled and rush away (forward). My idea was the player could use their behavior to kind of lead them and scare them away to make things easier.

 

I noticed something with the hit box, but not the other behavior. They only get knocked out and the front, higher up.

 

Will it be possible to aim the rocks more, or would that break up the flow?

 

Yep that is how it is programmed. However, I battled an issue with the Unity polygon colliders not being very accurate. The polygon colliders Unity creates are mapped around the images with a gap ranging from a few to many pixels. It was much worse but in a discussion in the 2D forum someone did some testing and discovered it is based on image size period. So the images for the snakes are actually double the width and height of of their in-game size. This causes the colliders to be generated closer to the image. I scale them down to 25% of that size so they appear correct size in game and the colliders scale too keeping the closer edge mappings. I may need to quadruple the images before importing into Unity so it generates proper colliders. That just seems so wasteful though.

Things like this annoy the hell out of me about Unity. They put things in that are very good but the execution of them is so poor that we have to do all of these work arounds to actually use them. I may work on building my own polygon collider generator. I'd rather just focus on the game dev though instead of always fixing issues with the game engine! Certainly could make progress a lot quicker that way! lol

Just to illustrate... here is a polygon collider Unity created for the snake at normal size:


I think it is obvious how loose the collider is which results in very poor collision checking.

Doubling the width and height of the image and importing into Unity, we get this collider:


Much better overall but you can see there is now a bit of the back of the snake that is not included in the collider and still we have that big gap on the front.

Increasing the size by 4 times the width and 4 times the height, Unity creates this collider with the error message shown to the right:


As you can see, overall the collider traced the image better but also looks corrupted on the back of the head and failed to include the end of the tail at all.

Anyway, so it is things like this that make Unity game development (at least 2D) so much more difficult than they should be. I guess I just do not understand why even bother to include these things if they, in reality, cannot even be used? It makes no sense and just wastes my time. If they had not had a polygon collider 2D then I would have set out to make my own detailed collision system from the beginning as I have always done in the past!

Okay, end of rant (for now! LOL)

 

Worms style rock throwing... get some precise trick shots out of that

 

Whoa... I wouldn't use 2D anything in Unity. I would use 2D graphics and primitive colliders for everything. I would also be raycasting to check for rock collisions each fixed update. It's unity3D. If you want a snappy 2D toolkit without the frills, you should check out game-maker. It can publish to mobile. You're having a lot of frustration from unity, due to having to work around features you don't need.

 

I bet you could get away with 3 bounding boxes.

 

@Misterselmo and @BeefSupreme, I updated that post with some screenshots.

Oh yeah, believe me I have had a ton of frustration with this stuff. It has been 2 steps forward 1 step backward the entire time.

Like I said, if there had not been a polygon collider 2D I would have focused on using multiple bounding boxes or simply made my own collision system. Because there was a Polygon Collider 2D I expected it would actually work!

It's fine though. I gave it my best shot to get them to work. They sound good on paper but as implemented are simply not usable in practice. So, I will come up with my own collision system. It will be good to have it anyway so going forward with other projects I do not need to wrestle with this issue.

For the rocks, they actually are using ray casting. Not Unity ray casting, but my own implementation. Unity ray casting would not help in this case I think because the issue is with the Polygon2DCollider that is being generated. So, I can either make my own polygon collider generation system or make a system to generate multiple box colliders for an image. The latter is how I have implemented this in the past.

 

Ouch. Unity done lost its mind. Indeed frustrating when a tool produces results like that.

 

I would file a bug report, this is not right. They need to know about this.

 

They should know. Several of us have been talking about these issues for months now.

This issue was documented in a discussion already here.

I think it is basically like @Misterselmo said. It is a 3D engine and for 3D it is probably damn good but the 2D stuff seems to be filled with glitches. I guess many people never run into the glitches though. It probably depends on what game you are building, the shapes of your images and so forth. Personally, I have run into glitch after glitch since the beginning which is why I currently am using my own physics system, my own animation system, my own ray casting etc just like normal. lol I really just need to be able to display images, play music and sound effects and write code to implement the game-play. If I can do that I am more than happy to write the code to handle all of the supporting stuff like collision checking, physics and so forth. I just figured if they already had polygon colliders why bother wasting time reinventing the wheel writing my own? But... yeah the screenshots show why I will be doing it.

 

I would have just done a 2.5d system, unity2d is new where the old 3d stuff is crazy tested. I hate these types of issues. I will build a collider out of smaller box colliders before I touch some shiny "super poly colider" or something. If I can't build it, if I don't understand how it works, it ain't reliable

I made puppets consisting out of more than 20 pieces using the animator... and threw a sphere collider on each limb. They were fully animated and could detect collision on any limb. Yet the display was 100% ortho 2d.

So are you running loops, doing 2d rectangular collision on interpolated points, is that your own "raycasting"?

 

That's a great route to take if you want to get into creating 3D models and such. I just wanted to make a straightforward 2D game. It certainly should be possible. I mean really all that is needed is to make a quad and set the texture to an image. Presto... sprite (or tile). I am sure (or would hope at least) that is what the new 2D Sprites are doing.

I agree absolutely. That is what I always end up doing too. Every time I try to use the "canned" stuff (except I must say Blitz3D, DirectX and XNA I had very few to no issues with their stuff) I run into things I need to work around. And even if it does work I still like to roll my own for the exact reason you mentioned... if I write it I know exactly what is going on and if there are bugs I can fix them. If it is not my code and I cannot access it I cannot fix it other than trying to spend time working around it.

I just figured good grief, how hard can it be to make something scan an image and generate a polygon? I am not sure how they did it but I would just scan it, box it (building boxes much like you'd do for box colliders) then connect the boxes. Simply outlining the outer box lines would do it. Alternately they could connect the corners and add a new point where necessary.

 

Game-maker actually uses per pixel collision detection, it's made for 2d games.

 

GameMaker is definitely not my cup of tea. I downloaded GameMaker Studio. Opened it looked around about 10 minutes and closed it never to be opened again. My ideal dev environment is straight code. That is what I am used to. But I am getting used to Unity. I do like it, I just find things like this very frustrating because in my head if you are going to do something do it right. If you are not going to do it good (usable) then don't do it at all. That's all. A lot of what is in Unity is excellent. I am not knocking the product as a whole just certain things like this make development much harder than it should be.

Alright enough griping. Let's get back on target!! Someone throw out a game for us to play test!

 

If you drag a script onto a game object in game maker, you can write it all yourself. All that weird gui stuff I never even used. I thinj if I was making a game like yours, I'd favor simplicity of a tool like that, but once you need to do anything other than 2d sprites its over.

 

I read once that you can only ask your friends to test your game once. Your mileage may vary
Gigi

 

Lol these guys have been great. I'll throw it out again after the artist completes the new snake animations. Probably be a couple weeks. By that time I should have a programmer art spider and possibly bat in place.

 

So... my 45 minute lunch break was spent drawing a spider.

And this is a mock-up of it in-game:


Alright, you can all stop laughing now. This is my programmer graphics at its best! =)

 

Poor little sprite dude looks like he's about to be lunch.

 

I can't help but imagine how horrible that creature would look if it were real.

 

I am currently working on a small game, nothing like an MMORPG. It is my first game and I am new to Unity and scripting and game development as a whole, so I am sort of learning as I go. You can play the my game here. View any updates in the thread here.

I would like to know if the controls feel... natural. What I mean by this is do the controls feel right or does it make it harder to play? Also, would you like different modes of gameplay (you could see the ideas I have for modes in the thread)? Most of all... do the graphics look good and does the music fit with the graphics and the overall mood of the game? Is this good progress for my first game?

Thanks,

- Chris

 

@Blue Bean Apps - The game is very cute. Has a catchy tune, and brightly colored art. I also like the simplicity of the interface and the controls. I think the game could be even stronger with a few minor tweaks to the design:

• Since up/down is overly rigid without a joystick, consider adding a spacebar to give the fish a vertical 'boost'. With a few tweaks, your game becomes a flappy type of game, similar to 'Retry' by Rovio, Flapy Bird, Whale Tale, and others - a very popular genre.
• With a boosty type of game, you could then add occasional coins to pick up or points for pass obstacles.
• Keep the simplicity of the interface.
• Add text to say 'Tap SpaceBar' or 'Tap Up/Down to move'

Congratulations on your first project! You are now a game developer .
Gigi

PS - Consider posting a picture in your thread, so it stands out as a new game to review.

 

@Gigiwoo - There's of course still a lot of work to be done. I appreciate your quick feedback!

With the Intense and Impossible modes of gameplay, you will press the space bar to gain small boosts of altitude, similar to that of Flappy Bird (which is the inspiration for this game by the way). I will soon be implementing a touch interface where you slide your fingers up or down to move the fish or you can tap the screen in Intense and Impossible modes. The mode the current version of the web player build is considered the Normal mode which is meant to be easier than Intense and Impossible which is why I intend to stick with the easier controls for that one. I will also have about a 1/4 second delay between the time you press the key and the time that the fish actually moves, so this will make it more realistic and natural.

I would like to add more obstacles to dodge, such as ship wrecks or clams along the seabed and a jellyfish that will shock you if you get too close, but I will have to work out a way to do that. I tried that in the past and I ran into several problems with collisions, so I will have to play around with that some more. Hopefully I can get it to work!

I intend to do just that. By the end of this week, I plan to have the web player updated with the new and improved UI. It will include a pause button on the gameplay scene.

I will have little text bubbles that will say something like that. I just want to get most of the main aspects of gameplay down before I work on the stuff that only the customers would benefit from.

I will add new screenshots when I show off the new UI at the end of this week. There's no point in uploading screenshots that will be outdated in a week!

 

I like the graphics. It's got a nice, warm style. The objects all look like what they're supposed to, and the bubbles are cool. The light shafts in the water is a nice touch as well. The music fits the theme pretty good I'd say, but it gets a bit repetitive. Maybe throw some pauses in between looping samples to break it up every once in a while. I wanted a sound effect when you get hit. The controls are good, and the fish is easy to move. Some more variety in the types of objects you have to avoid would be nice. Maybe some of the hooks could move up and down in the water? A bit of variety would hold my attention longer, make me see what's coming up next. Different modes would be fun. There's lots of stuff you can do as a fish I'd imagine.

 

Thank you for your feedback!

For the bubbles, I just used Unity's built in bubble particle effect which works very well!

For the repetitive music... there isn't much I can do about that. I got the file from an open source website as I don't do audio recording/editing, so I am sort of stuck with how it is. However, I suppose I could try two things: (1) Have a few seconds of silence in between loops, or (2) as the music starts up and nears its end, have it progressively fade out. Unity 4.x has good audio editing options that I need to learn how to use.

About a sound when you swim into a hook... I would like to have some sort of metal clunk sound play. I need to find an open source sound for it though, or I would need someone to make the file for me as I don't know how to make audio files.

I intend to add more variety in terms of objects to dodge. I would like to have some things along the seabed, such as ship wrecks or clams, so you can't just stay along the seafloor. I also have plans to add a jellyfish (comes in a few different colors) in Intense and Impossible modes which would be the hardest obstacle to dodge! Also, I plan to actually have the hooks bob in random directions at random times only in Impossible mode. The reason for why I won't do that for every mode is because I'm not even sure if it would be possible. There are tight gaps in between some hooks that are just wide enough for the fish to squeeze through, but to have the hooks move when they're that close to each other may present some problems, especially when the fish is going at high speed. I'm sure you will like the jellyfish idea!

There is a lot more to come!

- Chris

 

There's Bosca Ceoil if you want to try your hand at laying out some midi tracks. I learned that making terrible music an in fact give you a headache

I use BFXR to make sound effects. The results sound pretty retro and "video gamey", so I'm not sure if it will work for what you want, but it's pretty easy to get useful stuff out of.

Sounds like you've got a plan for filling out the content and you're off to a good start! Hah, jellyfish was one of the first things that popped in my head.

Oh, how did you get the web player to run from google docs? I was monkeying around with it earlier in this thread, but couldn't get the plugin to load the .unity file.

 

Hey Chris. Good to see you & your game here.

I just played it again. I think what you have is very solid. Just needs some additional things added and what you posted above sounds like they will be just the additions to make it stand out.

Right now it seems very solid and very basic. Adding the things you mentioned will add a little "oomph" to help it keep the player's attention. Especially, if you introduce the new things at different points. Like if I made it 45 seconds and had just seen my first jelly fish that is incentive to play again because I would wonder what else lies ahead.

Keep up the great work!

 

Hmm. Maybe it's not a Friday thing at all?

 

BFXR looks good. I will give it a try soon!

About the web player in Google Drive... Build to web player in Unity if not already. Open the new folder and copy the HTML file within there and paste it in your Google Drive. In Google Drive, under the Details tab for the file, edit the sharing visibility settings and change it to public. Then click the link under the HOSTING subtitle under the Details tab and your game should open in a new tab. If the file is in a folder, the folder must be public as well. Just remember, you don't load the .unity file, you load the .html file. I hope this helps!

 

The jellies will spawn randomly, so you may encounter a jelly at 30 seconds one time and encounter another at 53.751 seconds. There will be a set minimum distance between each jelly to ensure they are never too close. Every obstacle is spawned randomly within a certain distance, so there's no predicting where you may encounter a jelly or a hook.

And now that I think about it, I think it would be great if I had some fish swimming in the background (they won't be obstacles, just extra visuals). There would be a lot of variety between the fish. I definitely want to try this now!

 

Ugh, figured it out. Your post isn't the first time I've seen hosting mentioned under the details tab, which doesn't show up on my drive. I had to get the doc id and add it to a hosting URL to get the browser to open the page as html instead of in the docs preview window, then the unity plugin wouldn't actually do anything. After poking around a little, website hosting is broken in the new Google Drive (which I was using). Reverted back to the old interface and it works now!

Thanks.

 

Every day is Friday 'round here.

 

So where does that leave Pizza Friday? Cause I thought that was happening.

 

The problem may reside within your browser. Either your browser is total crap or it is no longer supported by Google. I haven't used drive in IE, Safari or Opera, but I've used it in Chrome and FF.

 

I don't think so...I use Firefox pretty much exclusively, but checked it out in Chrome too. For example using Chrome v.38.0.2125.111 m with the old Drive interface, this is the bottom of the details tab:



This is what we want, the link here works and the game loads fine.
Now if we click this button:



The details tab is now this:


No more hosting section, and using the doc ID the build a URL works in loading the page as HTML, but the Unity plugin never loads the game.