I wanted to share this with you guys because it's taken around 3 years of reading HN to really convince myself to quit my job and build my a business with my friends
Congrats on doing that. Please keep us posted on the progress of the commercial part. Now that the game is out and since it's a good game, i'd be really curious to know how you plan to get people to buy it, and if it succeeded.
This only works if their site gets views and the free demo gets plays. In my experience, apps in the App Store are primarily found through the App Store, meaning potential users won't hit the site first to try the demo.
The standard solution is to give away a bunch of levels for free and charge for additional levels using IAPs.
I'm guessing that demo is getting a lot of plays right now and they should capitalize on the hype that's being built.
Relying on the App Store is a good strategy for development firms that cannot build their own hype. So far, I'm guessing these guys are doing an okay job.
But I have zero data on any of this, hopefully they do and are acting appropriately.
Who says you need completely novel gameplay mechanics to charge money for an app? Surely more polish, hand-built levels and an app are sufficient. Whether OP would make more on ads or IAP remains to be seen, though. An IAP to get more levels would be a good way to hook users.
We felt that out of the gate making it free to download was our best chance for success, but who know, if we start getting into A/B testing in the future, we may find that keeping the game free in the browser and charging for app store installs is the way to go.
Is it intended that any solution will have all squares coloured? I like that property, and I think most of the online levels have it, but I solved level eight with some blanks.
If I had an iDevice I would pay for this. Good luck.
Very nice! I beat the ten demo levels too quickly, and found myself wanting more. Two things:
* An iPad version is needed. The graphics seem to be high res enough to make an iPad version feasible, and if this is on the iPad I'd buy it for sure. I don't really like the idea of trying to trace out the paths on the tiny iPhone screen.
* Charge money for it. I'd definitely be willing to pay personally.
Great game! Played to level 5 on your site, heading to download it onto iPhone now, for more playage later.
Without having looked yet, what is your revenue model, if there is one? Is the plan to release this for free, build an audience, a name, get feedback and all that - to go bigger later with a new game?
Well done. I love that you allow users to try out the game on your site first. I wasn't initially inclined to download the app, but after trying the game I was hooked.
Ditto. Was "meh" on yet another free trial, but one easy click "hey why not" and 10 levels later I'm downloading it, and figuring if addictive enough a buck for more levels is cheap.
I've played this game before (well, not this one, but the same game by a different company). Very enjoyable and your UI looks better than the other. The question I have is: how does one go about creating a puzzle game? I've always wondered. It seems like a lot of work to make something that's difficult-but-solvable.
Yeah, its a clone of an old shockwave game I played about 10 years ago and felt it would be perfect for touch interaction. For this puzzle game specifically, I found that starting with the solution and working my backwards was actually the best way to build levels. I'm not sure if that holds true for other puzzle games.
Interesting. I kept wanting to rotate the cube in your game. Nice to see it in the 3d-logic game the other commenter mentions. Do you plan to add that?
It's simple and elegant. I might have to grab it on my iPhone. And as others have said, it's a great idea to offer it online to give prospective users a taste of the game. Now I can see if it's challenging enough to be worth my time :P
For now the strategy is to avoid ads on iOS and sell in-game level packs, however I'm not sure how well that model works on android, I'd love to know if anyone has thoughts on that, it seems most people just do ad driven games on android
Small tangent, but what is the state of the Lodsys in-app patent? Are mobile devs protected at this point? I also read somewhere that the patent was suppose to expire soon.
I went with two apps and a Lite version. It's a bit of a hassle.
There are so many free things on Android that it's hard to get myself to try paid for games. The exception is games that I get to know through some other source, like your excellent web page with live demo.
I'd pay for this game on android, but paid level pack would also work for me.
Definitely well-conceived and fun! As others have said, very well done on the web-based version. It's too bad it's not on the Play store, though, any plans for an Android version?
[edit: Asked and answered below. As an Android user (and obviously this is anecdotal but it's all I've got), I spend more money on in-game purchases -- reasonable ones, not FarmCityWhatever insanity -- than I do purchasing them outright. For something like this, level packs seem like a good IAP.]
Not to detract from the fun of the game, but I get a strong feeling that this is probably a very studied problem in topology - and that thought keeps on nagging me to write a solver script :)
Yes in fact I've spent the past 6 months thinking about whether i could build a level generator that can verify and solve the levels too. The way I see it, with a simple 5x5x5 cube with 5 colors on it, the number of possible combinations to check is 75^5, which might take awhile to brute force, so there needs to be a intelligent way to map lines.
The general case might be something like that, but you only need any given configuration that fits those parameters (that is, you're not trying to solve complete, abstract topological problems). Maybe if you arbitrarily fix the first path, thereby artificially reducing the solution space? Just thinking out loud. Very cool puzzle!
EDIT: or even take a lower-level solution and inject another path into. So every prior level is a seed to future levels. Rotation of the three faces, even cycling the colors, would probably make it hard to recognize the older level within the new one (at least, for casual use).
Many of the levels have been designed so that if you follow the path of least resistance to solve them, you will always be left with one line that wont connect. Thats the true puzzle part of the game, and that's the aspect that makes me think it's difficult to solve without resorting to brute force.
I think you are right, it's harder than I thought.
The problem can in fact be modeled with a graph in which you need to find vertex-disjoint paths between pairs of vertices (s1,t1), (s2,t2), and so on, but this problem seems not to be reducible to max-flow, instead it is NP-hard.
My thoughts: what if you select the initial 5 colors, then recursively "extend" one of the colors in one of three possible directions. The sum of results is your level-space. Store results in a DB. Then, select all levels that have a)cell coverage of 85%+ b)min path length of 5 on all paths.
In order to cut down some of the search, you can make the recursion smarter (e.g. extend one color at a time, early cycle / impossibility prediction, etc)
This is an ideal problem for constraint solving. If there is one thing constraint solving does well it is make NP-Hard problems more manageable. I wrote up a quick solution to the first puzzle in sabr (1), you can see what the puzzle result looks like here (2). The key insight is: "each color must be surrounded by exactly two of its color, unless it is on an end, in which case it is surrounded by exactly one" after that it's just a matter of coding it up. It's not much extra work to go from here to a generator, which I may make later if I have time. Cool puzzle game, I like it ;D.
Nice game. Please post any monetary results, once you have some data. I thought making serious money from a game in the app store today is very difficult.
For the web based version, I might save the last color they clicked on and then use that when they click on a white square because I've found the game a little harder to play with that feature missing on a laptop with a trackpad.
Question: how essential do you think it was for you to quit your job to create this game? How much longer would it have taken you had you developed it on the side/weekends?
I'm in no way trying to argue your decision, I'm genuinely curious how much work it was to create a game like this. Also, how much of the conceptual work for this game had you done prior to leaving your job?
I would say pretty essential, but that's really because of my personality. I've started and stopped more projects then I can count over the past ten years while holding a steady job. Without saving up money and leaving my job, I'd probably have a half finished game sitting in my projects folder right now.
Great work. The fact you can play it straight away on the website is great, I'm not really a games player but I quickly got hooked.
Out of interest, do you make the levels by hand, or do you have a script to generate them? Would be interesting to know the rules for a valid level against an impossible level.
We made the levels by hand, writing a script to generate and validate levels is still an interesting math problem I would love to tackle if i had the time, its just very difficult from the limited attempt I made.
I really like your landing page. Do you know of any other game devs that have a playable demo on their landing page or did you come up with the idea yourself?
Honestly, it was just because I built the initial prototype in html5 and the level builder is also html based, since the effort to make a simple 10 level demo was so low we decided to throw it in.
Your game struck me as pretty similar to "Crazy Cube", a fairly popular flash puzzle game. I'm sure it's crossed your mind(s), but if you used an entire three-dimensional cube like that you could probably get even more permutations for levels. Just a thought :)
Edit: Ah, didn't realize you increased the density of squares on the cube as levels progressed. Very nice.
I'm impressed with your take on the level list. It's so nice to see something more than a grid of colored, numbered boxes. It may be a simple puzzle game, but approaches like that make it feel like an adventure.
340 comments
[ 2.4 ms ] story [ 324 ms ] threadDoes the website demo work on an iPhone (I don't have one to test)?
The standard solution is to give away a bunch of levels for free and charge for additional levels using IAPs.
Relying on the App Store is a good strategy for development firms that cannot build their own hype. So far, I'm guessing these guys are doing an okay job.
But I have zero data on any of this, hopefully they do and are acting appropriately.
[1]: http://www.addictinggames.com/puzzle-games/3dlogic2.jsp
Free sells.
This one has much better controls on the web app, as well.
[1]: http://www.kongregate.com/games/AlexMatveev/3d-logic
If I had an iDevice I would pay for this. Good luck.
I'm not sure how making a game for free is a business, but still, best of luck!
* An iPad version is needed. The graphics seem to be high res enough to make an iPad version feasible, and if this is on the iPad I'd buy it for sure. I don't really like the idea of trying to trace out the paths on the tiny iPhone screen.
* Charge money for it. I'd definitely be willing to pay personally.
Without having looked yet, what is your revenue model, if there is one? Is the plan to release this for free, build an audience, a name, get feedback and all that - to go bigger later with a new game?
I'd rather if my move didn't get set until I stop holding down the cursor (may be different on the mobile version, don't have an iPhone).
ETA: Thanks for no ads. Easy-priced paid content is worth not cluttering my head with more noise.
I went with two apps and a Lite version. It's a bit of a hassle.
I'd pay for this game on android, but paid level pack would also work for me.
Just grabbed it for my phone for free but I would have easily paid $0.99 for it. :)
How do you intend (or do you intend) to ramp up revenue to match your previous salary?
[edit: Asked and answered below. As an Android user (and obviously this is anecdotal but it's all I've got), I spend more money on in-game purchases -- reasonable ones, not FarmCityWhatever insanity -- than I do purchasing them outright. For something like this, level packs seem like a good IAP.]
But that's what makes it fun. It's the perfect combination of familiarity & challenge!
EDIT: or even take a lower-level solution and inject another path into. So every prior level is a seed to future levels. Rotation of the three faces, even cycling the colors, would probably make it hard to recognize the older level within the new one (at least, for casual use).
The problem can in fact be modeled with a graph in which you need to find vertex-disjoint paths between pairs of vertices (s1,t1), (s2,t2), and so on, but this problem seems not to be reducible to max-flow, instead it is NP-hard.
In order to cut down some of the search, you can make the recursion smarter (e.g. extend one color at a time, early cycle / impossibility prediction, etc)
1) https://github.com/dbunker/SABR/blob/master/module/other-tes...
2) https://github.com/dbunker/SABR/blob/master/module/other-tes...
I'm in no way trying to argue your decision, I'm genuinely curious how much work it was to create a game like this. Also, how much of the conceptual work for this game had you done prior to leaving your job?
edit: It seems that on reload, things are working fine, there is some glitchyness with the intro help text not always disappearing
Out of interest, do you make the levels by hand, or do you have a script to generate them? Would be interesting to know the rules for a valid level against an impossible level.
Edit: Ah, didn't realize you increased the density of squares on the cube as levels progressed. Very nice.
The game is beautiful, too. Wonderful job!