Letter Stack now on the App Store!

krisis

Our first game, Letter Stack, is now available on the App Store for iPhone, iPad and iPod touch.



App Store: https://itunes.apple.com/us/app/letter-stack-free/id657295290?mt=8



Please check it out and let me know what you think!

Facebook: http://www.facebook.com/LetterStack
Twitter: http://twitter.com/LetterStack_pp
Web: http://pushpoke.com/letterstack/

Built with Gideros! Using BhWax, Parse, Facebook, Flurry, Chartboost, Crashlytics.

Thanks to the Gideros community for the great support!

hgvyas123

Kongrats and it looks very good

vitalitymobile

Cool GUI! Kongrats!

bowerandy

@krysis, well done. Graphics look great (and its nice to see that my "soft hands" have found a home somewhere). Started a game (joetest2) but, I ended up playing against the master so I'm not doing too well.

Best regards

krisis

yeh, they worked great, thanks!

krisis

@bowerandy - your libraries and help on the forums were invaluable in building this app. Thanks!

Nascode

@krisis Cool! congrats on your first game!
What backend do you use? like Parse or something like that

bowerandy

Btw, is that TNT Particles you're using for the backdrop?

Best regards

krisis

@Nascode - Yeh, using Parse.

@bowerandy - yep, just wanted something subtle to add a bit of life to the menus.

phongtt

Very cool @krisis. Kongoratz!!

@bowerandy:

Scouser

CongoRatz

mysps

Looks great. I'll download it later today.

Good luck!!

Platypus

Congratulations, @krisis. I can't wait to play it.

atilim

@krisis CongoRatz!

ArtLeeApps

Well done !

OZApps

@krisis, it is fun and addictive since it also introduces the competitive nature by comparing the scores against the other players.

If you would want some feedback read on, or you can skip it from here on.
1. While moving the blocks sometimes it disrupts the other blocks even if the block being moved is the top most

2. Why does a block turn orange? Specially if it the bottom most?

3. Pull to refresh changes the text, but if the screen is moved back within the safe zone, the text does not revert back, it still says release to refresh, but does not refresh.

It is indeed fun because it makes one think in reverse to make the word, so jest has to be thought of as T-S-E-J or J-E-T-S which is a good brain exercise and good that your dictionary caters for multiple variations, like STOP, POTS or OPTS which are all valid.

ArtLeeApps

@anyone I've always wondered how you store an entire dictionary of words into a mobile app?

unlying

@ArtLeeApps i have a .txt file with all words. I read it on game start and store in array of words. Works fine for me.

OZApps

@krisis, if you do read this, your word list has issues.
TAKE is a valid word as is TEAK (the only one accepted by your app)
There are times when the words are all stacked up but are not recognized till you nudge them a bit (throw them into the air) so must be a physics collision thing.

Would you want to consider adding more words (5 letters, etc?) the 4 letter ones get a bit repetitive after a while.

krisis

@OZApps - I appreciate all the feedback!

1) Yeh, I spent a lot of time tweaking the stickiness with beta testers. Some people have suggested locking the blocks when you lift them but its lower on the priority list.

2) after 10 seconds it turns the first letter of the word orange as a hint. However, you can form any word with the letters so you may have solved with a different one then the one the game chose.

3) thanks, hadn't caught that edge case.

TAKE and TEAK are in the dictionary, but as you mention it occasionally doesn't recognise the word without nudging them. I've found it happens very rarely though.

I do have both 5 and 6 letter dictionaries in the game but most beta testers found it too hard so they are turned off. The 4 letter list is also filtered down to ~2500 words (out of ~5000 possible) so as to only chose words that most people would know. It lets you solve against the full dictionary though (I.e. if JETS was in the 2500 but you put JEST). I need to tweak the randomiser a bit as there are certain words which get repeated more often using the current method. I may make the 3rd round 5 letters or something in the future.

What is your username in the game? I'll take you on in a game

Thanks again for the feedback.
- krisis

krisis

@ArtLeeApps - there are many different ways that have speed / memory trade offs that you need to consider. It also depends on how you plan on using the dictionary. I.e. in my game I need to both choose a random word as well as check if a given word is valid (and quickly). Depending on how many words you are including it can take up considerable memory unless you store it optimised.

I wrote a script that converts a dictionary txt file to a Lua table with each letter as a key. E.g.
FourLetter = {a={r={g={h=1,s=1}}},b={a={t={h=1,s=1}}}}
In order to store the words "argh", "args", "bath", "bats".

It has O(1) lookup and is much more compact memory wise then storing each word separately. It also means it is preloaded on startup rather than parsing a txt file.

If you wanted something more hardcore you could look at something like http://en.m.wikipedia.org/wiki/Directed_acyclic_word_graph

- krisis

OZApps

@krisis, you can try to have an array of the words and mark the array with a frequency, so you can try to choose words with a lower frequency (more chosen random words)

BTW, your game is fun as we had everyone downloading it and playing madly against each other and once in a while cursing that the app did not accept the word

BTW, the Play Again button is a bit misleading, and one ends up clicking it to start a new game that they cannot get out of other than by resigning the game. Where as one would expect the main menu button to be there by default.

phongtt

My English word skill is too bad.

OZApps

@krisis, you are too good and FAST The blocks getting stuck is frustrating and so is losing

specially with double the points to you @krisis...

OZApps

It is a bit frustrating when the words do not give points due to the silly physics bug, here's the proof of it happening more than once

krisis

Play Again is there on purpose to increase engagement, but I know it's a bit sly.

Blocks getting stuck as in the word not registering? Are you seeing it happen often? I personally haven't had it happen to me in the last hundred games but it may be the method I use to stack. I do have a potential fix but not sure of there will be a performance impact.

I usually let people beat me or only just beat them to encourage them to play again but you had a good score so i thought I'd give you a higher score to aim for

- krisis

OZApps

@krisis, this might be a specific issue with the way I might be playing and I would agree with you that while developing and testing, we learn how to *best* use it as we are aware of how the code works. I recollect that while creating Thwart for a client, but the moment they took to ß testing, it would crash totally baffling me. Ultimately found the problem it was the way they were playing it.

The scoring is based on the time taken to create the word, right? or is there something else 'coz achieving a 30,000 is realistic but 60,000, how do you manage that? What are the tips to do that... :-O

Tips would be welcome for all players and maybe even on your website. This is a good showcase app like Game42 made with Gideros.

bowerandy

@krisis, I'm interested in your choice of Parse. At first glance the "basic" (free) plan seems quite generous, with 1million pushes per month and 1million requests per month. However, the Pro plan at $199 per month seems quite a jump.

I'm thinking about using Parse purely for the simplicity in implementing push messages. But, if I understand it correctly, each push message needs several other API calls to make it work. You are obviously using the API calls much more in Letter Stack- what do you think your installed base will have to be to require moving to the Pro plan?

best regards

ArtLeeApps

@unlying, @krisis

Thanks for the explanation, that would be some task to get as many common words as you deem worthy of the list too. I'm guessing if you want to support multi languages, you would only download the dictionary for that country, based on the users settings.

krisis

@OZApps - yes, scoring is based on the time it takes you to complete the word, with a minimum of 2000 for each word completed.

I was able to get 100,000 once using all the powerups, but generally get 50k-70k without. Maybe I'll put together a video showing how I do it, but basically I can figure most words out before they hit the floor, then just drag them up and down onto the stack as quickly as possible (don't just release it, or throw it against the wall). The more you play, the more familiar you get with the words.

krisis

@bowerandy - I think Parse's free plan is quite generous. Based on the first day of data, I'm doing ~34 API calls per active user per day, which would be ~1000/month. So I guess it'd only support 1000 active users/day before you'd need to start paying.

They have partial usage pricing for each tier, so you don't necessary need to jump straight to $199/mo, though it does include other useful features and functionality. Free plan you pay 7c/1000 requests or 7c/1000 push notifs when you hit the limit.

If you consider the cost of running and maintaining your own infrastructure to support the same number of users and API calls then I think it's a pretty good deal.