next »

[Mannapi]

So glad you think it is clever and great job on finishing the level! Woo!

Thanks for the input. I have been thinking of a "flow direction" effect in a similar vein to what you suggest. The problem with your particular approach is what to do with the corners. Remember each section is really just a triangle, which can connect in a number of ways.

My best idea was to have something (points?) in the center of the path that would move along slowly and always sync up with the next section. Anything off center would not work. I made a ms paint explanation below:



I hope I can explain how this affects the gradient you drew.

I agree with the static walls. I have it hooked up in code and the palette has a specific color to indicate the unflippable tiles. I just haven't made any puzzles using them yet. They will make things easier, since the player doesn't need to solve those tiles.

[Artifice Machine]

Flip Flow is a good title. Much better than just Flip.

Instead of points moving through, how about small faintly colored arrows?

[oahda]

From watching your GIF's, I cannot for the life of me figure out how the game works or what the logic or patterns behind how which path expands how and whither when somewhere is clicked. Nobody else in this thread seems to have admitted the same problem, tho, so I wonder if I'm just slow.

[Zorg]

I'm still thinking that some sort of waves would be possible. 


Edit: Concerning Prinsessa's problem, maybe it would be easier if you add a point at the center of the flipable walls, like a hinge? So you can see around which point the walls will rotate by 90 degrees.

[Lo-Phi]

I think zorg's ideas are great. In my opinion this game is suffering a bit from a lack of visual feedback. Thankfully, all that takes to fix is a little creative problem solving. The game idea itself has a lot of potential!

On top of zorg's suggestions, I would also recommend having some sort of mouseover effect. An example would be a semi-transparent line that shows up on the square where the mouse is currently located. The semi-transparent line would show exactly what line what would change if the square was clicked.

Another option would just be to divide the grid like I explained above (so all of the squares are divided into 4 equal parts). Then you could just make the parts that aren't flipped transparent.

Finally, it might be a good idea to "bold" the square that the mouse is hovering over. It's just very difficult right now for the player to determine where the pivot point is. I think the grid you put underneath the lines helps with this, but I think going the extra mile would be very beneficial for helping your audience understand the game's mechanics. Put another way, the player needs to know what the consequences of their actions will be. There is too much left to the imagination at first because the game doesn't really indicate what options the player has.

As long as you're concerned about informing the player of their options, I think the game will turn out really neat!

EDIT: Maybe zorg's idea addresses what I brought up as long as the point at the center of the walls shows up only where the mouse is. I still think the actual square should also be highlighted or outlined in some way as well though. Otherwise the point just seems like the entire line will flip instead of just the segment in the highlighted square. Let me know if I need to clarify.

[PetterBergmar]

I love this! The visual style is so nice. Tried your demo and it's fun to play too! Reminds me of the Witness.

[Mannapi]

Wow! So much feedback. This is so inspiring!

Flip Flow is a good title. Much better than just Flip.

Instead of points moving through, how about small faintly colored arrows?

I guess I have settled on Flip Flow then.
I hadn't actually thought of displaying the flow direction without animation before. This is an interesting idea.


A title screen something like this?

From watching your GIF's, I cannot for the life of me figure out how the game works or what the logic or patterns behind how which path expands how and whither when somewhere is clicked. Nobody else in this thread seems to have admitted the same problem, tho, so I wonder if I'm just slow.

You are not the only one to have this reaction. It probably my main problem right now. Did you try the demo puzzle?

I'm still thinking that some sort of waves would be possible.  

Thank you for taking the time to make the animation! I will definitely look into having a wave effect. It will need to fit into the minimalistic feel of the game though.

...
Another option would just be to divide the grid like I explained above (so all of the squares are divided into 4 equal parts). Then you could just make the parts that aren't flipped transparent.
...

I am not sure I understand what you mean by dividing into 4 parts. Can you elaborate? Thanks for the nice words

I love this! The visual style is so nice. Tried your demo and it's fun to play too! Reminds me of the Witness.

Thanks! I saw your paternity leave level editor app and thought it was genius. I still haven't played the Witness yet.

I will focus on solving the feedback issues. The problem with a hover effect is that it will not work on mobile. Perhaps I can have it togglable so that when you get used to the mechanics you can disable it.
I am not sold on the hinge idea. Having it as a hover effect might work though.

In other news, I did not make a single level while away, but I did manage to fix some editor issues and rewrite some of the bridge code. That is something I guess.
I also remade the color changer graphics:

[PetterBergmar]

Whoa! Keep that title screen it's beautiful!!!

[Mannapi]

A bit of progress based on the feedback I have gotten from you guys.

-I added a disabled color.
-I made the tracing visible. The color fade is controlled by a curve. Right now the curve is a smoothstep kind of curve, but I haven't really experimented with other types. Linear might even be better.
-I added an intermediate frame for the flipped tile.
-I made a custom background shader. Very simple, It just blends 4 colors. The ones in the gif were randomly chosen from a hastily authored palette. (The bg in the previous pictures was a fake - bitmap).
-I tried adding some particles to the end (haven't hooked it up to the delay caused by the slower trace).



Does this help with usability in your opinion?

[rebelholic]

What's your purpose and inspiration to make this game?

I am still don't get it about main objective from this game

[Mannapi]

My purpose for making the game is to get a relatively simple game (I have very little free time) published and learn what that entails. I also want to create a fun, challenging and rewarding experience in the game itself.

The objective of the game is to solve puzzles. I admit I haven´t shown much puzzle design here, mostly test levels to try out my mechanics.

Please try my playable demo and see if you understand what the challenge is.
https://dl.dropboxusercontent.com/u/4396617/flipp/index.html

[oahda]

Does this help with usability in your opinion?

Now all I needed was a single look at this GIF and I immediately understood how the game works, finally—great improvement!

[Mannapi]

Excellent! Thanks for the feedback!

[Zorg]

Yes, much better!

But the difference between solid walls and clickable doors is still too subtle, imho. I'd use a geometrical feature which works without color. (For example like the hinge joint i mentioned before. )

[a-k-]

Have just installed a browser that supports WebGL so I could finally play your demo. I really liked the aesthetics, and also the fact that all changes were reflected instantaneously (I may be in the minority here). You mentioned earlier that you may want to count the number of flips – I would also consider mileage for scoring.

[Xonatron]

Interesting concept! So simple yet complex!

[Mannapi]

I noticed that the fan on my laptop got a bit louder when I traced the flow path after adding the color fading. I decided to do a little profiling and as a result changed all my Coroutines from using

Code:

yield return null;

to

Code:

yield return new WaitForEndOfFrame();

since my color lerping only needed to happen once a frame.

I created a 20x20 board to profile and fiddled with it until it would reach all the way to the end. Behold!

This was running at ~200 fps, being capped by Unity. I guess I need to limit the fps at some point.

When I made a 30x30 board it didn't finish because of my infinite loop checker (simply checked whether the recursion reached 1000 in length).


Obviously neither of these are realistic sizes, but for testing purposes only.

I watched this talk about optimizing for mobile devices which was incredibly informative. I am guilty of many of these things (I talk about my use of Dictionaries earlier in this blog), but I will not optimize for this until I know it is an actual problem. My game is still quite simple compared to many Unity projects out there! The points about string libraries in C# having culture sensitive comparison is about the craziest thing I have heard. I link to the actual place in the video below.

https://youtu.be/j4YAY36xjwE?t=2320

P.s. how do I embed a video?

[Zorg]

Sadly there is no [video] or [yt] tag. The URL is only accepted as "youtube.com", not "youtu.be".

[oahda]

æææææ that video ;;
*hugs C++ tightly and cries*

I feel like the tl;dr of this is basically "don't use anything that makes C# C#; program as tho it were C".

Unity still runs really well despite all this IMO. That underlying engine is doing a good job at the heavier lifting. But it's really odd that Unity C# differs in such things as inlining from non-Unity C# (if I understood correctly)... How come?

[Mannapi]

Exactly! Use C# as if it were C++ with automagic pointers.

No properties, no foreach iterations, no boxing and no dictionaries. It is important to remember that these restrictions only apply to mobile devices where the GC is not run, and probably only when this is happening multiple times a frame. Small temporary structures that never get cleaned up will cause issues if enough of them are made.

Thanks Zorg, I will make sure my youtube links use the proper url