Tuesday, January 31, 2012

Who Wants To Be A Millionaire?

Short answer: this guy!

Unfortunately I've not stumbled into a large amount of money. I'm instead referring to a Facebook game by the same name as the television quiz show that became popular when I was in University. The Facebook game adds some social steps (the first few questions are answered by a group of players and their values fluctuate based on how many people get it wrong) but at its very core it's a time limited trivia game with 4 multiple choice answers just like the tv show. Instead of actually winning cash (wouldn't that be nice) you accumulate a running total of cash won across all games which is treated as an experience total. Make enough money and you level up!

It's a fun little distraction if you like trivia, which I do. But as with all Facebook games you should ask yourself how they're intending to make money. Well, you're limited to how many times you can play each day. If you want to play more than that you can buy more from the store. On the surface this looks like the only way for them to make money though they do have little rewards for spamming your friends. In short, it's like pretty much all Facebook games. Spam your friends and pay if you want to play a lot...

Of course it turns out there's a second way to spend money on the game and I feel like I got scammed into using it. To explain how we'll need a little background...

The tv show had lifelines where you could get help from outside sources if you weren't confident that you knew the answer. The Facebook game has the same sort of thing. At the start of the game you get to name two of your friends to be your lifelines. Then during the course of playing the game when you need to select an answer you can click on one of those two people. You automatically get credit for the right answer in the game and your friend gets sent the question. If they get it right then both of you get some experience. If they get it wrong then both of you get substantially less experience.

One of my friends, Ike, ran a Millionaire game show for MathSoc when we were in University. I used him as my lifeline the first time I played (I think I only had 2 friends even playing at the time so I had no choice.) and he made a comment on Facebook about how someone was crazy for sending him a 'Real Housewives' question. I sent it off because there's an immediate payoff for sending it off even if the other person won't have a clue! But then I felt bad after he commented so when I played again a couple days later I kept my eye for a question he'd probably know. (By this point I had 4 friends playing which leads to the selection screen above. My other two friends were auto-chosen but I wanted to tag in Ike and send him a good question.) Lo and behold the very first question was about Steve McQueen. (My complete lack of knowledge about movie actors was the stuff of legends amongst my University crowd so I really needed the help.) Turns out Ike knew the answer (hurray!) and commented on Facebook that that was more like it.

Encouraged by the positive reinforcement I went in to play again with Ike as a lifeline to see if I could send him another good one. Here's the screen I saw when I went to change my lifelines...

Notice the difference? I didn't! For those who may not play a lot of Facebook games the little icons on the buttons there say that I can pay Facebook credits (which have to be bought with real money) in order to unlock those guys as lifelines for my second game. The layout is the same. The buttons are in the same spot. The lines start with the same words. I wanted to do the same thing I'd just done so I clicked in the same sequence I'd clicked before. Instead I clicked on a button which wanted money! I guess this is their main way to make sure you spam ALL your friends to sign up... They gave me 6 games to play today but I only have enough friends to lifeline 2 games. But for losers like me I can pay money to pretend I have more friends!

The kicker? It didn't ask for confirmation or anything that I wanted to spend money. It just made a quick cash grab and went on with its life. 5 credits amounts to about 50 cents so it isn't like they walked off with a lot of my money but it really pissed me off when it happened. Nevermind that sending Ike a trivia question isn't worth 50 cents... I feel like they didn't make it clear that's what was going to happen. Worse, I feel like they intentionally designed their UI the way they did in order to trick me. (I have reported them with the 'Dispute an item purchased while playing Who Wants To Be A Millionaire Game' link, so we'll see what happens.)

What may be the worst part of this is Facebook has apparently cached my credit card information. I had those credits on hand (left overs from when I bought some stuff in Galaxy Legion) but it bothers me that an app can snag credits without confirmation and that Facebook has stored the information needed to sell me more credits. What if that button had cost 50 credits? Would it have auto-charged my credit card to buy them?

Monday, January 30, 2012

Intel Extreme Masters

There was a pretty big eSports tournament held a couple weeks ago which featured a League of Legends event. I've been slowly making my way through the replays of all the games and there have been some wacky things going on so far. Support Nunu. AP Tristana. AD Kennen. Dogs and cats living together. It's insane! And interesting to watch...

Check it out here...

Sunday, January 29, 2012

Campaign Manager 2008: A Breakthrough?

I had a game of Campaign Manager 2008 finish yesterday where I built an infinite deck and lost. I'll need to look into the game more to see what really went wrong but I tried to get aggressive with playing out attack cards once I went infinite and lost control of the game. I scored up several states in a row but ended up behind after that. But what is really interesting about the game is my opponent used a two card combo to pretty good effect against me and I think it might have potential to break open the stalemate...

There's one difference between the two decks that I have yet to cover. Each player has three media support cards in their card pool. Two of these three cards are the standard 'pick what happens when a new event comes up' ilk which seem ok but not overpowering. The third one for Obama makes your opponent roll twice on the 'bad stuff' chart when he goes negative. The McCain card means McCain doesn't roll at all when he goes negative.

This is interesting but I rarely see anyone go negative. The problem is 3 of the 4 'go negative' cards actually require you to discard 2 extra cards to play them. I do this every now and then to win a powerful state but the problem is it sets you back a lot in card quantity. The 4th card doesn't require a discard but doesn't really coordinate well with itself. It gives you 1 defense and 1 economy in a single state.

So what my opponent did was he set up the no roll card and then kept cycling through his very small deck to play that card over and over. The support gains might be unrelated but Obama can't counter both of them. Eventually you'll get into a position to just overwhelm him in one category or the other and he doesn't have a counter. Even if he plays the same card he has to roll for bad stuff and 1 time in 6 McCain will net a support so Obama just loses in this situation...

But is it sustainable? Pretend we hit a 2 card deck with 2 cantrips in the discard. Open with the negative. Then play Oprah into a cantrip. Then play a cantrip. Then play a cantrip. We're back in the same gamestate as at the start except we've gained 5 support over 4 actions. This is certainly sustainable!

Now that I look at it a little more this seems decent even without the right media support card in play. Assuming it's an infinite vs infinite game the extra card draws won't help your opponent and the key demographic swap does nothing so 4 of the 6 outcomes have no game effect. And even when the negative roll hurts you the best it can do is give your opponent back the support you'd gained. Tempo might work out poorly for you but by and large this seems like it should allow you to break out of the stalemate! Time to put a higher focus on drafting that card!

Saturday, January 28, 2012

WBC 2012 Trials

Voting recently finished for the trial events at the World Boardgaming Championships this coming year. Personally I didn't get around to voting (I signed up at the event last year and wasn't really sure how to vote without signing up again. I didn't care enough to figure it out so I just let other people decide my destiny.) 22 of the 25 trial events were events at last year's WBC so there isn't really a lot of turnover going on. 2 of the remaining 3 events are returning from minor absences. Goa is back after taking last year off and Diplomacy returns after taking 3 years off.

Diplomacy in particular is an interesting choice since how do you determine the winner? Some games will finish with one clear winner but often a Diplomacy game ends in some sort of stalemate. Apparently the last time it was an event they played 3 rounds and you scored points based on your finish in those rounds based on number of countries alive in the draw and number of supply centers each of those countries owned. I would imagine such a scoring system is rife with people backstabbing each other trying to whittle down the number of survivors in a draw... Then they add up your best 2 scores and advance 7 people to a final? I doubt I'll be willing to set aside time to play in 2 or 3 full Diplomacy games but I think it's a good event to have. Diplomacy is one of the truly iconic board games in my mind and it just feels right to have it at WBC.

Especially since the one truly iconic game they'd been running, Risk, has fallen out this year. I recently read through all of the event reports for last year's WBC and Risk didn't have one. The GM apparently didn't submit a report which killed the event off entirely for next year. If Risk wants back in then it'll have to get voted in as a trial for 2013. I normally read all the event reports each year and I don't really remember ever seeing a non-submitted report. I don't know if they purge them from the system eventually or get write-ups way after the fact or what. But there were FIVE of them this year. Egizia, Macao, Risk, Storm Over Stalingrad, and Stronghold. Egizia would have been a century event so it got bumped to trial status (it did get voted back in) and the other four were just killed off. It's sad. I like reading the event reports even for games I don't play and have no intention of playing so it's unfortunate that there were 5 no-shows this time.

At any rate, this year's trials... 24 have been events before. What's the 25th? Well, it turns out it received the most votes of any game which is pretty amazing. The game? A Few Acres of Snow. This game has quickly become my favourite game on Yucata. My record in the game is pretty good so far as well... 46-2. The two losses were both tied games where I lost the tiebreaker. (Apparently the French win a tie game.) Both times I didn't realize the game was even going to end, sadly, or I probably could have found an extra point somewhere on my last turn. Having it as an event at WBC has made me even more excited for WBC this year  if that's even possible. I have a new goal... Win this event!

Friday, January 27, 2012

Final Fantasy IV: Bugs

One of the interesting things I've been learning as I've been playing through the Final Fantasy games is just how many weird bugs exist in the games. Final Fantasy had a lot of bad pointer related bugs. It had some local variable bugs. Final Fantasy II had a pretty brutal cancel bug. Final Fantasy III had a dupe bug for sure (though I played on the DS which seemed pretty decent). Did they get any better for Final Fantasy IV? Of course not!

Perhaps my favourite of the FFIV bugs has more to do with not enough edge case testing... Here's the situation:

In combat you can change which weapon/shield you have equipped in each hand. You go into your item menu, select a hand, and then choose what you want to put in that hand. But how are you, the player, to know what you could equip? Well, the system iterates over all items in your inventory when you pick a hand to swap. Anything you could equip gets highlighted. Anything you can't equip gets faded out. Then when you go to select a faded out item you get an error. This seems like a reasonable system, right?

When you select a highlighted item should it recheck if the item is a legal option? If you're coding robustly it probably should but then you're adding an extra check in which seems unnecessary. You already validated every item in your inventory seconds earlier. Why bother revalidating? We're talking about a game that had to have plot cut out to fit it onto the cartridge! Any extra checks are probably unwise...

Now, what if you wanted to unequip an item? Well, you'd 'equip' a blank slot in your inventory. This seems like a reasonable case to build in. Let's just flag the empty slots as valid equips. No sense treating them any differently, right?

How about when we derive our current stats? Should we validate our gear then? Seems wasteful. As long as we're robust in only allowing valid gear to be equipped in the first place this should never come up.

So, what's the bug? Well, it's a timing bug. Get into a fight with a monster which drops a weapon or a shield as its common drop. Have Edge input the steal command (which will steal the weapon or shield). Then quickly get to the person who can't legally equip that item. Have them go into the item menu and start switching items. This will cause a validation of your inventory as it stands when you start. While doing this the steal command will resolve and you'll get a new item. It will go into a blank slot in your inventory. A blank slot that has already been validated as a legal equip. Pick the new item! Huzzah! Rydia can put on a shield! Rosa can use a spear.

There's a similar timing bug which lets you use berserk while unequipping arrows which results in duplicating any item you can legally equip. You end up with 255 copies which is great for making cash or for throwing with Edge.

Berserk is also buggy with Edward and his auto-hide ability. Apparently you can set it up so he can't be attacked but is berserked and beating down.

A final timing bug revolves around the spell stop. It takes out all other timers on the person, completely. Have a temporary paralyze? If you get stopped it will never wear off. A virus? There forever. Unless you reapply it then the timer will get fixed. And if you screw with the timers on monsters with stop? The game will often freeze. (I'm glad I never used that spell!)

In the inexplicably weird category... The sylph spell is free if you put it in one spot (upper right hand corner) in the spell list.

In the mad cheats category... At one point you're in a fight in the dwarven castle underground. You fight some dolls and then you fight Golbez. He almost kills you when Rydia shows up and saves the day. Golbez still manages to run off with the crystal. A future plot point is to go get the last dark crystal from the sealed cave... If the first thing you do after Rydia joins your party is cast warp you'll get warped back to the dwarven crystal room. Complete with dark crystal for the taking. Which counts as the dark crystal from the sealed cave! So you can skip that dungeon!

Now for a bug I think I was hit by... The moon is circular and you can find a path to do a complete circuit. But your spaceship only exists exactly where you landed it. So if you do a full circle around the moon? You lose your spaceship. You need to go back the other way the exact number of times to get back to the precise spot you left your spaceship.

There are a bunch of fights with conditional triggers to do things. The biggest example of this is when you fight the 4 fiends in one fight. You beat up Milon for a while and then when he dies Rubicant steps up. They coded these triggers as counter attacks. So if you kill Milon off with attacks that don't trigger a counter attack, he won't tag in Rubicant. You'll just win. One way to do this is to cast wall on your team and then bounce attack spells off the wall and onto him. Cheats!

Finally there's a bug that certainly afflicted me on probably every playthrough but which I didn't really notice. Thinking about it in retrospect I can see it happening but I didn't notice at the time... Weapons have flags on them. Metallic, so they paralyze you in the magnetic cave. Back row, so you do full damage from the back row. And, for some reason, no-crit. These weapons simply can't get a critical strike. The metallic flag works fine but the other two are buggy. If they ever get set on your character they are never removed. So if you ever put on a whip, for example, you can never get a crit again for the rest of the game. And you do full damage from the back row with any weapon. Axes, whips, and the best paladin sword all have the no-crit flag.

That's not the only way to lose the ability to crit. Apparently if you ever start a fight dead you lose the ability to crit. And sometimes the game has battles which run without player input. Tellah vs Edward in the 'YOU SPOONY BARD!' scene. Tellah vs Golbez. FuSoYa and Golbez against Zemus. If in one of those fights the team on the left dies then everyone in your party which occupies an 'empty' slot for the right hand team loses the ability to crit. So after the FuSoYa and Golbez vs Zemus fight at most two of your characters can still crit. The other 3 will permanently lose their ability to crit if they somehow kept it up to that point!

How... Dumb.

Thursday, January 26, 2012

League of Legends: Better?

A couple months ago I posted about trying to get better at League of Legends by focusing on properly farming minions in the early game. At the time I was running at about 60% of a top player in terms of creeps killed in 20 minutes and in termed of how much damage Nasus' Q ability did. (That ability permanently gains 3 damage for every unit killed by the ability itself.)

I've been playing a fair bit with Robb recently and we pulled my rating up about 400 points from its low point which is a pretty substantial amount. Was that just avoiding the terrible match-ups in 'Elo Hell' or are we actually getting better? I've been noticing my minion kills have really gone up in the last little while since I bought the hero Graves but maybe that's just the hero (compared to Vayne and Tristana)?

I haven't actually played Nasus at all recently but last night I made a point to play him in a game so I could have some numbers for comparison. I was probably a little rusty but a lot of the basics are the same...

After 20 minutes
Nov - 94 creep kills, +189 on Q
Jan - 123 creep kills, +237 on Q
pro - ~150 creep kills, +~300 on Q

This is obviously a really small sample size (one game each time) but that's a pretty reasonable boost. Up from 60% of the pro to 80%. Now, I'm also not playing against pro quality opponents so it's not as good as it may sound but I think it's showing improvement from caring enough to focus on it the last two months.

An a not even comparable data point, here's what I had on Graves in a recent game where I laned with Robb as my pocket healer...

124 creep kills after 20 minutes which is about the same as Nasus, but up 70 at the 39 minute mark (when the Nasus game ended). I certainly did a better job of finding anything to kill during downtime when I was playing as Graves. I think part of that might be team role though... People stand back and let Graves kill stuff. No one lets Nasus screw around killing things when they could have them instead!

Wednesday, January 25, 2012

Band Hero: American Pie

One of the things I really like about various rhythm games is when a truly awesome song is available. Often, especially in the earlier days of the genre, it would just be a bad cover of the song. Or it wouldn't be the whole thing for various reasons. Walk Like An Egyptian was in Dance Dance Revolution Universe 2 (and is by far my favourite song in that game both in terms of the steps and the original song) but it was cut down to about 1:38 from the original 3:24. That's great and all, but it would be better if it was the whole thing! (Well, except I can barely stay on my feet for 1:38... I'd probably die if it was a full 3:24!)

I finished off the story mode of Band Hero last night. The reward for beating it was the credits scrolling across a new song... American Pie by Don McLean.

American Pie is my favourite song. And they seem to have the original version in the game. The whole original version. All 8 minutes and 36 seconds of it. A full 1427 notes! It's glorious!

Tuesday, January 24, 2012

Final Fantasy IV: Kainazzo Shocked!

I can remember playing Final Fantasy IV as a kid and having real troubles with the boss fights in Castle Baron. After you break in via the sewers you have to fight Baigan (and his respawing arms) and then immediately afterwards fight Kainazzo (the water fiend). I remember Kainazzo having these brutal AE water spells and having to try him over and over to kill him...

Here's how my fight last night went.

Cecil - attack
Yang - attack
Porom - cure2
Palom - lit2
Tellah - lit3
Cecil - attack
Yang - attack
Porom - cure2

At this point Tellah finally got around to resolving lit3 which killed Kainazzo. He put his water shield thing up once, immediately before Palom resolved lit2 which brought it down. He never got around to withdrawing into his shell. He never cast an AE spell. He didn't heal himself. Lit3 took him from probably around 60% life to dead in one attack. How was this ever hard? Did I not want to cast my best spells as a kid? Maybe I was using Tellah to heal and the twins to use their stupid twin attacks? I likely didn't know I could heal up between the fights and might have spent all Tellah's mana on Baigan? (To be fair it doesn't make a lot of sense that I can go to my old room in the castle and sleep between fights...)

At any rate Baigan is dead and I got to see one of the silliest cutscenes in any game. Kainazzo (like Milan before him) wants a second crack at killing the party. In this case he causes the walls of the hallway to move inward in order to crush the party like some sort of medieval trash compactor. Palom and Porom get the bright idea to turn themselves into stone in order to stop the walls from moving in any further. After the walls stop Tellah tries to remove the petrification but gets an error message. It can't be removed because it was inflicted of their own will.

There are all sorts of problems here! For one thing, neither of them can actually cast stone. Palom learns it at level 36 and we're not nearly that high. Next, why would stone statues of two little kids stop walls which are presumably able to crush human bones? It doesn't even get to the point of trying to crush them. The walls stop independently as if they're simply unable to push stone children. Finally, what sort of magical universe has the rule that a spell can't be undone if it was self inflicted? There's always absurdities revolving around death in games like this (why can I bring Rydia back to life over and over but Anna takes a couple arrows and is just dead) but this one really takes the cake. Under the logic used here if Tellah had turned them to stone (a spell he can actually cast) he could have then healed them of it after the doors open. But since they wanted to get stoned it can't be undone? Bah! Nevermind the fact it actually does get undone, later, by the head honcho of Mysidia...

Monday, January 23, 2012

Campaign Manager 2008: Fair Deck Options

I mentioned the other day that I thought the infinite deck in Campaign Manager 2008 was relatively easy to build and very powerful when built. I fear it may even be unstoppable. Today I'm going to look at all the options available to a fair deck to see if it can possibly have a shot at beating an infinite deck.

As a reminder an infinite deck is a deck which has reached the point where it can play a cantrip card every turn for the rest of the game. It can chose between gaining an economy, or gaining a defense, or shifting the issue track. It can choose the same one over and over again if it wants. Essentially it is a deck which is giving up tempo (it only gets one effect each turn) in order to have complete flexibility.

A fair deck has to spend time drawing cards with the default 'draw a card' action. This means it is taking a few turns off (during which the infinite deck is just churning out economy or defense) in order to build up a string of good turns in the future. In order for the fair deck to have a chance that string of good turns has to be really good. It has to make up for giving the infinite deck free turns. What cards can the fair deck hope to draw in order to pull this off?

The first class of cards are the cantrips which make up the infinite deck's gameplan. I hope it is obvious that these cards can't help the fair deck win. In the best case scenario for the fair deck the infinite deck merely plays the same cantrip and undoes the fair deck's turn. In the worst case the cantrip from the fair deck isn't optimally useful in which case the infinite deck can play a different cantrip for a better effect. To be fair cantrips will help cycle the fair deck into better cards which could help deal with the selection issue of a 5 card hand but as far as power goes they can't help.

Next up are the double effect cards. These cards let you do two unrelated things in a given turn. The potential options are to gain an economy, a defense, shift the issue track, or alter the key demographic. Note that there is a card to shift the issue track two spaces but there is not a card in this class which adds both an economy and a defense. As such your best case scenario is to build up a hand with the right 5 of these and then get 10 effects over 5 turns while your opponent only gets 5 effects. This seems like it should be pretty good but there is one fatal flaw... You can only gain 5 actual support by doing this. The infinite deck has an easy counter... Ignore any issue shifts and just deal with the support you gain. If they spend their 5 turns undoing just your support in economy or defense then what have you got to show for your build-up? 5 demographic swaps or issue track movements. Then you have to spend 5 turns filling your hand back up while they get 5 free support anywhere they want. It doesn't matter how you set up the issue tracks. Because they can take either defense or economy for any number of turns in a row they just get to score up two states while you refill your hand. The end result of playing a bunch of double effect cards? They win two states and you win nothing. Not good.

How about the card drawing spells? Ancestral lets them build up a 7 card hand for a really big sequence of plays but that doesn't help with any of the cards we've seen so far. Same with Oprah. The regrowth style effects are even worse for the fair deck since they cost extra cards to play. Every extra card is an extra turn spent on 'draw a card' and giving your opponent a free support somewhere so what you get out of those extra cards has to be pretty powerful. Just getting to play a card a second time doesn't count since we don't have any cards that help us at all as it is.

There's an interesting set of cards that allow you to get a bunch of effects all at once. Each side has two of them with the Obama set only impacting economy and the McCain set only impacting defense. How they work is you play the card which does nothing on its own. Then you can discard any number of cards to get a support of your type or to shift the issue track towards your type. There's one card for support and one for shifting and you can't mix and match the two. And, as an added drawback, you can only hit each state once. The shifting really isn't very relevant (we saw earlier that the infinite deck will gladly ignore shifts in order to just deal with any support you gain) but getting 4 support from one action seems like it might be what we need to out-tempo the infinite deck. At the most powerful case we've spent 6 turns (5 draws and a turn to play the card) in order to get 4 support. 6 turns for 4 support is actually pretty bad. The infinite deck got 6 support over those 6 turns and they can even put them all into a single state if they want to! In terms of actually helping the fair deck win a state it seems like the best case scenario is actually one where two states exist and you're two support away from winning either one of them. By then gaining a support in each one you put pressure on your opponent. He can only counter one of the moves and then you can play another card to finish off the other state! Yes, we've won a state! On the downside it cost us 6 turns to do so (draw 2 cards and the super-support card, draw the finisher card, play super-support and finisher) and they spent 1 turn countering the secondary effect of the super-support card. They then get 5 more support to do as they will. That's certainly enough to win at least one state. And since they got to choose which state they defended off the hop you're guaranteed to have won the least valuable state of the two while they turned around and won the better one. And that's assuming there are even two states where you're that close to winning in the first place!

What about media support? I haven't mentioned these cards before but they can be decently powerful. The basic idea is every time someone wins a state they get to pick a new state to put into play. Then you flip a random event card form a big deck of them and it does crazy things. Sometimes to the new state, sometimes to all states, sometimes to individual players. (Make Obama discard 2 cards, for example.) If you have a media support card in play then you get to make choices for the event card. Instead of it working on the new state you can choose any of the 4 states in play. One of the events lets whoever controls the media draw 2 cards. Getting one of these into play as the fair deck might give you a little advantage here and there to try to combat the infinite deck. The problem is you simply can't keep one in play against an infinite deck. If they only have 1 card in their discard pile and you play a media support it will kill theirs, putting it in the discard pile. Then they play Oprah and put it right back into play. They still have 1 card in their discard. They still have media support in play. They spent 1 turn. You spent 2 since you have to manually draw the card and then play it. Even if they have 2 cards in their discard and Oprah doesn't hit the media support it will hit a cantrip which gets them the media support to play next turn. Best case you have a 1 turn window to win a state. After they played a cantrip and gained a support. I think you were better off winning the state before they got to cantrip if you actually had any way to make that happen!

What about key demographic cards? These give you a way to get multiple support in a state in one action. This has some potential but it has two major drawbacks. The first is that you only replace neutral dots and can't actually remove anything your opponent owns so they're pretty useless against an established state where the infinite deck has already played some cards. The second is that they're very narrow. Each demographic card works on only a small handful of the states. Often they need the key demographic swapped with another card before they can work at all.

That said, there is some potential here for sure. If your opponent puts a state into play and you have the demographic card in your hand you can often set yourself up to win the state for sure on your next action. If you end up with n-1 support in both economy and defense there's nothing that can stop you. Assuming you have the right cards in hand to follow up you can play a shift + support card in order to win the state no matter what your opponent does to react. Sweet! Unfortunately there are still downsides. For one, you then have to put a state into play and your opponent can guarantee you can't end up in a position to win that state with another demographic card. Also the infinite deck is going to put states into play starting at the bottom value-wise. So even if you can win the state they put out they get to win the next one and it will be worth more. Nevermind the fact that while it's _possible_ to have the right demographic card in hand and the right follow-ups in hand it's nowhere near guaranteed. If you fill your deck up with demographics then you'll be able to win a few states by surprise but then sometimes you'll have a hand full of cards which actually can't impact the board at all. Then you have to waste two turns (one playing it for no impact and one drawing a replacement) while your opponent gets to score up lots of free support.

There's one last type of card. The negative campaign card. These cards let you grab multiple support in a single state in a single action but come with some potential drawbacks. Each side has 4 of them available. One of the cards lets you gain 1 defense and 1 economy in a single state. The other three make you discard two cards in order to gain 2 defense or 2 economy in a single state. If it is defense or economy is determined by the card. Obama has 2 economy cards and 1 defense card. McCain has 2 defense cards and 1 economy card. Every negative attack card has the drawback that you have to roll a die at the end of your turn if you play one of them and your opponent gets a benefit. The potential outcomes are {draw a card, draw 2 cards, do nothing, gain 1 support anywhere, shift a state anywhere, alter a key demographic anywhere}. If your opponent is already infinite then most of these actually do nothing. The only really scary one is giving them a free support since if you didn't win a state straight up with the card they just get to completely undo it with that roll.

Now we're starting to see a framework. Imagine the following set-up... You use ancestral to get up to a 7 card hand. Then you use the super-support to gain an economy in two different states. Then whichever one he doesn't defend against you go negative on for an extra 2 economy. This only requires two states where you're within 3 of winning which doesn't seem unreasonable at all. Downside? You spent 6 cards over 2 turns to do it. You went negative so he gets a bonus. You can only do it when you draw ancestral. And you still only get the worst state of the two.

What about combining the negative card with a demographic card? This costs you fewer cards and wins you whatever state you want for sure. If the state you want happens to match your demographic card...

Also once your opponent sees the negative card once they can work against it. They can start shifting out of the support you can double up on. They can start trading states since it does take 4 turns to pull off a single double card...

Nevermind the true trump they have up their sleeve... They're only dedicating 8 cards to going infinite. They can have 7 more cards in their deck that do things. Often they'll have a negative double card too which they can use to counter yours or to just blow you out on a big state. And while it costs you 4 turns to pull it off it doesn't really cost them anything at all to do it. They lose some of their flexibility from having a full hand to be sure but they're one ancestral or a couple Oprahs from going back to infinite again...

The other negative card (which gives you a defense and an economy but doesn't cost extra discards) just seems bad. It can combo decently with the shift+support cards as you can try to keep your opponent guessing. But generally speaking every time my opponent plays it against me I cheer since it doesn't hurt me at all and I get a free die roll bonus.

Put it all together and what do you get? Well, most of the cards available to a fair deck are completely useless. There are a couple situationally powerful cards that can work together to win an individual state but you end up giving your opponent too many free turns in the process. You're not going to get shut out if you can put together a couple combos but there's no way you're going to actually be able to win the game. I think building a fair deck is basically guaranteeing a loss against an opponent who knows how to build an infinite deck.

Sunday, January 22, 2012

Final Fantasy: Incoming Damage

{Found this post in my draft folder and figured better really late than never, right?}

Monsters follow very precise rules for how they act in combat. The first thing they do is check to see if they're going to run away. To find out they roll a number between 0 and 50 and add it to a monster intrinsic stat called morale. If that result is less than 80 plus twice your character's level they run away. The key here is they only check the guy in slot 1's level, so theoretically if I moved one of my friendly corpses into the first slot monsters would stop being scared of me.

Assuming they don't run they then check to see if they cast a spell. Every monster that can cast spells has a set chance to cast a spell. If this check succeeds they cast the next spell on their list of spells to cast. There's nothing random about which spell they cast. The first time they cast a spell they always cast the first spell in line. The second time they cast a spell they always cast the second spell in line. If they cast all their spells they start again at the start.

Assuming they don't cast a spell they check to see if they use an enemy skill which works they same way as casting a spell does. They have a list of skills and use the first one in line first and so on.

Assuming they don't use a skill they melee attack.

Astos has 255 morale which means he can't run away. (Max character level is 50 so the highest my side of the equation could be is 180. 255 plus a random number is always bigger than 180.)

Astos in particular has a 97 in 129 chance to cast a spell, or practically 75%. The spells he casts, in order, are Rub, Slow2, Fast, Fir2, Lit2, Slow, Dark, Sleep.

Astos has no skills.

If he attacks he attacks once per round with at attack value of 26. His hit is 42. His crit is 1. At level 28 my thief will have around 68 evade and 6 absorb. So he hits me for an average of 33 a swing, 39 on a crit. Out of 201 swings he'll crit me 2 times and hit me 140 times. Overall, that's a little over 23 damage each time he attacks. So his attacks are not scary at all.

In general spells make the same attack roll as melee attacks do, a number between 0 and 200. The base hit chance for a spell is 148 instead of 168 and there is no hit stat. The targets magic defense is used as an evade modifier. At level 28 a thief has 69 magic defense. (15 base + 2 per level)

What do his spells do?

Rub - instantly kills the target - Roll a number between 0 and 200 and compare against 148+24-69. So about 51% of the time I'm instantly killed.
Slow2 - forces the target to attack exactly once when they autoattack. 148+64-69, or 71% of the time I essentially can't win.
Fast - doubles the number of attacks Astos gets when he melees. This makes his attack average 46 instead of 23.
Fir2 - hit for between 30 and 60 damage. Roll to see if it hits. If it hits, do double damage. If it misses do normal damage. 148+24-69. So 51% of the time I take 90 damage, 49% of the time I take 45 damage. 68 damage on average.
Lit2 - exactly the same as fire but a different element. 68 damage on average.
Slow - the same as Slow2 except for some reason this hits all targets. So 71% of the time I essentially can't win. It is possible since this is cast so late in his rotation that I can still win after it gets cast.
Dark - Blinds the target. When blind a target has 40 knocked off their attack rolls made and 40 added to incoming attack rolls. Lands 51% of the time.
Sleep - Being asleep increasing incoming melee damage by 25% and reduces your evade to 0. You can't take an action but sleep is removed on your next action if a roll between 0 and 80 is less than your maximum health. Lands 51% of the time.

The question of how many rounds I get to live is therefore very complicated. I'll have close to 600 maximum health at level 28 and fudging the numbers some in terms of incoming damage modified by dark and sleep and such I probably take around 65 damage per 'attack' be it a spell or a melee swing. So I get to live until his 10th attack or until he successfully lands any of rub, slow2, or slow. Abstract things and say he attacks every 3 spells. Then after 18 spell casts he'll have meleed 6 times and cast 4 damage spells which is my dying point. So from damage I probably get 18ish swings. But to make it that far I have to avoid 3 rubs, 3 slow2s, and 2 slows. I do that .02% of the time. So, uh, not happening.

Fortunately I found that I only need 9 rounds to win. (Well, 26 swings at 3 swings per round.) That's barely once through his cast bar, so I have nothing to worry about in terms of damage taken and just need to avoid 3 death attacks. The last one isn't even that bad, since I'll already have most of my 26 swings in. Really, I have a 14% chance of avoiding the initial rub and slow2 and beyond that I probably win at level 28. But I'm less convinced I can get lucky and win at level 26.

Saturday, January 21, 2012

Band Hero: YMCA

I turned my xBox 360 on last night in order to rock out a little in Band Hero. I don't have a band (or any instruments beyond a guitar) but it at least has a new selection of songs to play on the guitar. Unfortunately I find the band type games tend to be easier than the straight guitar games. I play on the hardest difficulty and still most songs are completely trivial. That bothers me from a challenge standpoint but just having new songs to play once is a worthwhile trade-off.

One of those songs is the old disco classic, YMCA by The Village People. The song itself (on guitar at least) is pretty easy, but it there was an added twist that made it hard to get a really good score. At one point in the song the band yells out Y - M - C - A and as a human being I've been conditioned to make those letters with my arms. There were no guitar notes to hit at the time so there's nothing wrong with jumping around in letters, is there? Well, it turns out that making jerky motions is enough to trigger the star power multiplier. Which was wasted since it was a guitar note light area. Boo!

One nice thing about Band Hero is that they're able to add in songs mostly known for their lyrics instead of just the guitar track. I've never seen an Evanescence song in a straight guitar game, for example, but there is one on Band Hero!

Friday, January 20, 2012

Campaign Manager 2008: Infinite Decks

I've completed 11 games of Campaign Manager 2008 on Yucata with two more in progress with what seems to be an insurmountable lead. Assuming I do win those games my record will be 9-2-2. I'd say one of those games was close and that was the game I played with the default decks. Every other game was either a dominating blowout or ended in a draw. It has me really wondering about the balance of the game and if it's actually worth playing. There are a few questions that need to be answered to work that out...

  • What is an 'infinite' deck?
  • How hard is it to assemble one?
  • Is there anything at all a 'fair' deck could do to compete?
  • In the infinite on infinite match is there any way to get an edge and break a stalemate?
I'm going to define an infinite deck as one that reaches a point where it can take the same basic action every turn for the rest of the game. It should be able to use any of the three core actions (gain an economy, gain a defense, shift the issue track) on any turn. In order to set this up you need to have two copies of the economy cantrip (cantrip meaning it also draws a card in addition to the first effect), two copies of the defense cantrip, and two copies of the shift cantrip. Then you get up to n-1 cards in your hand. It then doesn't matter what card is in your graveyard as you'll always have at least one of the core cantrip cards in your hand. When you play one of them you get to draw the card in your graveyard putting you in essentially the same game state. You may have a different card in the graveyard but you still have at least one of the core cantrips in your hand.

How hard is it to set up? Well, it turns out that it's pretty trivial if you focus entirely on doing it in the draft and practically impossible otherwise. The key lies in the fact you can't take the default 'draw 1 card' action when you have 5 or more cards in your hand. This means that most people with a fair deck have a maximum hand size of 5. Then you have to play a card which will reduce your hand size down to 4. There are a bunch of cards which are cantrips which maintain your current hand size and 2 cards which might increase it beyond 5. The first is the ancestral type card which lets you draw 3 cards. This takes you from 5 cards in hand to 7 cards in hand. The second is the Oprah type card which lets you draw 2 cards and play one of them. This keeps your hand size even unless the card you play is a cantrip which puts you at +1 hand size. 

Assume you drafted both ancestral and Oprah. You can, with a little effort, build a 5 card hand containing both of them. (Just keep playing anything and taking the basic draw a card action until you reach 5 cards with both of them.) Play ancestral putting you up to 7 cards including Oprah. At this point your goal is to chain cantrips back to back to back until you reshuffle and draw your ancestral. If you could run that cycle with a 7 card hand you can definitely run it with your new 9 card hand! 

How do you guarantee that you can cycle to the ancestral? By having at most 7 cards in your deck that aren't cantrips. When your hand is 6 cards and Oprah you play a cantrip if you have one. Otherwise Oprah is guaranteed to hit a cantrip putting you up to 8 cards in hand. With at most 7 non-cantrips in your deck you're now guaranteed to always be able to at least maintain your hand size. And each time you draw ancestral or Oprah it gets bigger! You'll get to an infinite size in no time at all...

On top of those both McCain and Obama have an additional card which can work to draw ancestral more often. McCain's guarantees the ancestral at the cost of 2 cards (so if you pick up ancestral you spend 2 actions to gain a maximum hand size). Obama's is a discard X cards to draw X cards. It's much riskier but if you've already managed a reshuffle you can cycle into it for plus one card as well.

So how hard is it to draft ancestral, Oprah, and at least 6 cantrips/regrowths? Well, both decks have 10 cantrips in them. So you essentially need to be able to grab 8 of 13 cards. Even with the worst luck possible you're guaranteed 5 of them. I don't know the exact odds but in my experience it's not hard at all. In the two games I have running at the moment I have 10 and 8. In the finished games I won I had 10, 9, 8, 11, 11, and 10. 

8 isn't actually a full requirement either. One of the players gets to keep a card in play permanently. If you're that person your deck is only 14 cards so you only need 7 (and to not have your permanent removed) to pull it off. Obama also has the ability to remove one card from each player's deck. If he plays that card he then only needs 7 to eventually guarantee an infinite deck. 6 with a permanent in play!

It feels like the real danger is to have both ancestral and Oprah in the same draft pack. In that case you should take ancestral and just hope you can pound enough cantrips. (At this point you need 8 cantrips on top of ancestral.)

I'll try to answer the remaining two questions in a future post.

Thursday, January 19, 2012


I mentioned earlier that Yucata has two rating systems. The rating system they use for each individual game is something called TrueSkill that broke down into 3 numbers but I didn't really know what they meant. I've had games where I won but both players had their rating go up. I've won games without an increase in my rating while my opponent lost rating. I come from worlds (Magic, League of Legends) that use an Elo system and neither of those things could happen there. I turned to my trusty companion Google to see if I could find out more...

It turns out TrueSkill is a rating system developed by Microsoft Research for use with xBox Live. It's a Bayesian system that had the same basic idea of the Elo system. The system has an idea of what the players have as current ratings. It predicts the likely outcome of the match. Then it looks at what actually happened and uses that information to update the ratings of the players. Win and it goes up. How much it goes up depends on how likely you were to win. Lose and it goes down.

But wait... I've seen losers gain. I've won and stayed even. That doesn't line up... Or does it? I read their paper on the topic to learn more. Part of me is now wishing I've done more to keep my math skills sharp but I managed to fight my way through it. The basic idea is you don't actually have one number associated with your rating, you have two. The first number is what the system thinks your rating might be right now. The second number is how confident the system is that your 'real' rating matches the first number. Then when the system reports a single number as your rating it is actually reporting a number representing a low bound of what your rating might be. (It subtracts three times the uncertainty from your rating. These numbers are a close approximation to a Gaussian distribution with the uncertainty as the standard deviation. Knocking three standard deviations off gives you a number you can be confident with 99.7% certainty to be at least your 'real' rating.

Elo is also only used for 2 player games. Microsoft's whole goal was to make a system that worked for multi-player games and team games (Halo and all the different game types there for example) and this one does. It also converges to your 'real' rating much faster than Elo does.

When it comes to Yucata in particular they set all initial ratings with a value of 1200 and an uncertainty of 400. This means your reported rating for every game on Yucata is 0 before you play a game. (The system has _no_ confidence in your rating of 1200 since you've done nothing to indicate you can play at all.) Apparently they've also set a parameter individually on each game to indicate how much randomness they think is in the game. The more randomness in the game the slower it takes to converge on your 'real' rating. (Maybe you won because you're better than me at Can't Stop. Maybe you won because you rolled better than I did. But when I beat you at Six that's all skill, baby!)

The more games you play the lower and lower your uncertainty number becomes. I've played 441 games of Roll Through The Ages, for example, and my uncertainty has come down from that initial 400 to a mere 19. On the other hand I've played only one game of Pompeii and my uncertainty there is 344. I guess beating Andrew doesn't say all that much about my skill, huh? My estimated rating is actually around the same in both games (1413 vs 1411) but the uncertainty difference (thanks to playing 440 more games) is huge. My listed rating in RTTA is 1356 which is good for 5th of 2710. Pompeii? A mere 378 rating. Good for 709th of 1055.  As for Andrew his one loss puts him at -44. (It floors to 0 but a little math gets the real number.) But with only the one data point there's not any real certainty there either. As far as the system knows I might be the second worst Pompeii player in the world! Andrew might be the second best! Or maybe I just got lucky!

It does explain how both players can gain rating from a game. The loser just needs to have his uncertainty number shrink by more than 1/3rd the amount of rating he lost. I do wish it would actually list the change in all the numbers instead of just the lower bound.

Wednesday, January 18, 2012


Normally I don't pay much attention to foreign politics. Or domestic politics. Or the weather for that matter! In general I'm content to come home from work, play some games, write a blog post, and go to sleep. The Daily Show is the closest I ever get to the news. I'm a pretty oblivious guy with some really unconventional ideas so politics mostly just serves to make me bitter. There's nothing I can do to change anything and realistically I'm not going to be impacted by any changes anyway. I needed to renew my passport so I can go to WBC next year which was a minor inconvenience but that's pretty much as impactful as things get.

SOPA (Stop Online Piracy Act) is a rather drastic departure from that norm. I've been seeing a lot of games and sites talking about trying to stop it from being passed and decided to read a bit about it. It turns out that SOPA would have a pretty substantial impact on this blog! I'll get to how in a second, but first let's go over how I came to this information...

I follow Major League Gaming on Facebook (I have a membership on their website so I can watch League of Legends and StarCraft II events and wanted to get reminded when they were running) and saw what seemed to be a pretty crazy post the other day. They'd changed hosting services for all of their websites. Why? Because GoDaddy supported SOPA. Then they announced they were going to blackout their website on January 18th. I've since learned that Reddit and Wikipedia are also shutting down today.

League of Legends had a link in their client talking about how they were against SOPA. I clicked on it on Friday to read about it. One of the Riot lawyers had a Q&A on Reddit talking about SOPA and what it would mean to the community. The game itself would be fine if SOPA passed, he said, but it would probably obliterate eSports. Buried deep in that chat was a link to this thread on their forums where one of the US Congressmen (Rep Jared Polis, Colorado) posted speaking out against SOPA. (It would seem Polis made a lot of money in the eCard/flower delivery website businesses and doesn't think that would have been possible if SOPA had existed.)

Ok, so there's a lot of people making a big stink about it... But how would SOPA actually impact this blog? Well, I post pictures sometimes. I write about potentially copyrighted information like game plots. I've linked to a music video or two. Over at Ziggyny's Star Trek Trek I have a lot of pictures! But for now let's pretend Square has a problem with me posting a picture of my Final Fantasy solo-ninja game. It doesn't matter why they might have a problem with it. Just pretend they do.

As things stand right now they'd need to send me a cease and desist order. If I care I'll take the offending content down. If not they'll have to sue me. They'd need to build a case for court showing that my post violated the law and that they'd been harmed by it. I'd get a chance to defend myself in court. If it turned out they were right and I was wrong then I'd suffer the appropriate consequences.

Realistically that wouldn't happen. Even if Square has a legitimate beef with my picture it wouldn't really be worth their while to come after me. This site is worth nothing. Personally I'm worth practically nothing. If my post was causing their brand real harm they might still come after me but they'd only come after me to try to rectify that damage and not for any monetary gain.

What about under SOPA? Well, the first thing Square would do is send Google notice that they were going to be filing a complaint against me under SOPA. (I'm a foreign website allegedly violating copyright laws and Google owns Blogger which hosts this blog.) Google then has two choices.

  1. Shut my blog down. Doing so immediately would grant them immunity from being prosecuted for my transgressions.
  2. Side with me. They'd keep my site running and would become co-defendants in the pending lawsuit under SOPA.
Remember above when I said I'm not worth suing? I can barely afford pockets! Google has some pretty deep ones. If they were found guilty there could be a pretty hefty payout. Even if it turns out I'm not breaking any laws Google still has to pay a legal team to build my defense.

Ask yourself this... Why would Google pay some lawyers to defend me? Why would they even bother looking at my site? I know if I was in their shoes I wouldn't bother. Square tells Google that Ziggyny is a bad man and getting sued under SOPA? Google just flips the switch and kills me. 

Ok, but why would Square take offense to my picture of a max level ninja? Here's the thing... It doesn't matter why. It could be some legal intern was having a bad day. Maybe they're going to take offense when I don't choose Final Fantasy VII as the best of the series. All they have to do is tell Google they're going to come after me and I get shut down. Square has no burden of proof at all. I don't get my day in court to defend myself. They just need to look at me funny and I'm dead. The burden of proof is on Google's shoulders now and they need to make their decision before really investigating. Side with me and find out I'm bad? They're screwed. 

I know some of you are going to say I'm overreacting. Would a content provider really come up with a flimsy excuse to shut down some guy they don't like? I should point out one of the entities really pushing for SOPA to come into law is the Recording Industry Association of America which has a history of doing just that. How about the Universal Music Group which apparently had a deal with Youtube allowing it to pull down any videos it wants for no reason at all. Youtube eventually got involved and put the 'offending' video back up but the fact UMG could just take anything down when it felt like it is disturbing. They were somehow given power and were more than happy to abuse it while they could.

SOPA would give them so much power it's insane. I linked to a Youtube music video for the song La Luna after playing that game in Niagara last year. To Walking Like an Egyptian after playing the song in Dance Dance Revolution. I linked to an awesome zombie song a year and a half ago. The artist even came by and commented! But under SOPA that post would be quite the hanging chad. The artist was happy to have the word spread about a pretty cool song, but is the RIAA happy about it? If they were happy about it last year will they still be happy about it in 5 years? If not... I'm toast! (If Blogger even exists at all in a post-SOPA world...) I broke down a strategy for beating a fight in World of Warcraft. Is Blizzard happy about that? They could really dial down how tuned fights have to be if they shut down strategy sites... 

When it comes right down to it there's nothing I can actually do as a foreigner to stop SOPA. But some of you might live in the US. Look into the issue for yourself and then get in touch with your local congressman and voice your concerns if you have any!

Tuesday, January 17, 2012


Diablo III seems to be a pretty popular topic recently. Blogs I read on the internet are posting about it. People I only tangentially know at work are coming up to me to talk about it. Blizzard still hasn't given a release date so it's just a lot of things getting me excited about nothing at all.

I saw an article title somewhere (in an IGN spam email I think) mentioning something about how Torchlight 2 had missed the boat. It needed to get out before Diablo III or completely lose its market and was recently pushed back for additional testing. It reminded me that I haven't played Torchlight but that my brother gave me a copy of it over Steam for Christmas in 2010. My computer with Steam on it was dead at the time and I was too lazy to install it on my laptop. I was going to fix my computer, honest. And then it slipped my mind completely...

I did eventually get my computer fixed and use Steam to play Civ V. Unfortunately there's nothing in the Steam client to indicate I had an incoming gift so I continued to forget about it... Until the Torchlight 2 article! I tracked down the email with the link and installed the game.

It's most definitely just a Diablo clone, but it's a pretty good one. (Potions, identify scrolls, and town portal scrolls all look the same, for example.) There are some definite improvements over Diablo II (shared stash in town so I can easily trade items between characters in particular) and it's been fun so far. I started as the gun using character but have found it to be a little frustrating. The same button is for walking and for shooting so I frequently find myself walking up beside a monster when I really wanted to shoot it from a distance. Having to accurately click to target things isn't really my thing in a game like this. I pretty much exclusively played the sorceress in Diablo II. Who needs to target a chain lightning? Just fire it off in the general direction of the enemies! If it hits any of them it hits them all! Frozen orb? It hits everything! INFERNO?!? Just run in circles!

That said, I think I'm going to start a different character up. There's a melee dude and a caster as other options. Presumably the melee dude doesn't mind accidentally walking up beside the monsters!

Monday, January 16, 2012

Campaign Manager 2008: Mad Cheats!

Let's say you're playing a card game. In particular I'm talking about Campaign Manager 2008 but it could be any game. It could be Fluxx. It could be Magic: the Gathering. Whatever. It's a card game. One of the cards you can play in this game says 'Draw two cards and play one of them immediately, adding the other to your hand.' How would you assume the following interactions work? (For simplicity's sake I'm going to call this effect an Oprah after the Obama card in Campaign Manager 2008.)

- The only card in your hand is Oprah and you play it. The two cards you draw both require you to discard two cards in order to play them. Can you play either one? If not, what happens?

- The only card in your hand is Oprah and you play it. The two cards you draw both require you to discard a card in order to play them. Can you play either one? If not, what happens?

- You have only one card left in your deck and 5 cards in your discard pile. You play Oprah. Do you get a card from your discard pile? If so, could you get Oprah?

- You have only one card left in your deck and no cards in your discard pile. You play Oprah. The one card you draw also has, as part of its effect, draw a card. Do you draw Oprah?

In the first scenario I'd say there's no way to actually get the effect out of a card. You can't discard two cards to pay the cost for either one. Depending on how the cards are worded you'd either get to keep both (probably having to reveal them to prove you couldn't play them) or you'd have to play one with no effect.

In the second scenario it's tricky. When exactly do you add the second card to your hand? The wording of the card says you play one immediately and tells you to add the other to your hand later in the text. I could see  it being ruled either way.

The third scenario is pretty clear I think. You draw the one card in your deck, shuffle the discard, and draw another card. Then you put Oprah in the discard. I can see other ways it might work though. Maybe you should get a shot at Oprah? Maybe you don't reshuffle. (A Few Acres of Snow change their rules such that you don't reshuffle in the middle of a turn, for example.) But in general I'd assume you reshuffle right away and can't draw the card currently drawing you cards.

What about the fourth option? When, exactly, do you end up discarding Oprah. I think Oprah can't draw Oprah because it's 'being played' while you're drawing those cards but is it still 'being played' when you play the second card? Does the answer to the second scenario matter? I'd say if you can't discard the second card then you either can't draw Oprah here (we haven't added the second card to our hand so Oprah has to still be 'being played') or you can draw it off of Oprah itself. If you can discard the second card then I'd say you should probably be able to cycle into Oprah off the second card but I could see it going the other way.

Well, how is it implemented online? You're forced to discard without getting the effect in scenario 1. You can discard the card to get the effect in scenario 2. You do reshuffle in the middle of the turn in scenario 3 (but get no shot at Oprah). Scenario 4? It works both ways depending on how you click...

If you play it out normally you don't get the card. Oprah is in the discard pile when you resolve the second card but the reshuffle doesn't happen. Here's where things get tricky... I didn't put any thought into it and just assumed I should get the card. I thought I must have done something wrong (why didn't I get my card?) so I clicked the 'Reset Move' button to do it again to make sure. It resets back to after drawing the cards off Oprah (and rightly so, after I see what they are I shouldn't be able to undo playing Oprah) and lets me pick which one to play. This time around it forces me to draw a card which has to be Oprah.

Something's wrong here, right? The implementation has to be buggy one way or the other. I did this once in a game last week and didn't even give it a second thought. Maybe I'd screwed up clicking the first time or something but I managed to make it work the way I instinctively thought it should so all is well.

Today I got into a stalemate in a game. We both had all of our cards but one in our hands and were able to cycle the same game state forever. I was wondering about what should happen. Could I have built my deck differently to beat his infinite combo? Did he have anything up his sleeve to stop mine? Did the rules have a chess style draw rule? I opened up the rules and browsed a bit to find the following...

"If a player wishes to draw a card but has no cards remaining in his Campaign Deck, his discard pile is first reshuffled to form a new deck. If this occurs in the midst of a player’s turn, any cards played earlier on that turn should not be shuffled into the new deck."


So the way I thought it worked was wrong. And by finding a way to make it happen I've been cheating...

I reported it as a bug. My opponent also noticed and asked how I did it. I told him I'd reported it and offered a draw since I'd been cheating. Also since the game was unlikely to progress even if I stopped cheating. The odd thing there is we both gained TrueSkill rating for the draw.

Sunday, January 15, 2012

League of Legends: Turrets

One of the new masteries added to League of Legends in season 2 is called Siege Commander. It reduces the armour of nearby enemy turrets by 10. I was wondering about how good this actually was (I have it in my Warwick spec) and tried to take a look at the armour of the enemy turrets in a game as I was walking up to them. The turret went from 250 armour to 50 armour as I walked up to it which was really confusing. It says it knocks 10 armour off! Why did it lose 200? Clearly some further investigation is required...

I did some searching and found that turrets actually have all kinds of weird rules surrounding them. I'd always just assumed they had static stats (possibly that scale as time progressed) and just shot for sensible damage. It turns out there are weird rules for how much damage they do and how much damage they take.

In terms of how much damage they do it seems that every consecutive shot on a single target does an extra 22.5% damage. So getting hit once will hurt but getting hit for a bunch of shots in a row is going to destroy you. Turrets also have 20% armor penetration and do only 94% damage to champions. 

Defensively if there are no enemy minions within range of the turret it gets x3.88 ac and x2.47 mr. Presumably this is to make it harder to 'back door' without a creep wave present and is almost certainly why the tower lost 200 ac in my test. 

And perhaps most interestingly... The damage a champion does to a turret is either her attack damage (the same as with every single other thing you can auto-attack) or 40% of her ability power. This blew my mind. I've always operated under the assumption that casters are terrible at burning down turrets! I popped in a quick replay from a game yesterday where I was Veigar. I had 109 ad and 690 ap. I was cracking a turret, post ac, for almost 200. Assuming the 40% number is right that means I do 2.5 times as much damage to a turret as I was expecting I would. Not bad! A true physical attacker is still going to do more since they're apt to build some sort of attack speed items and I'm really not with Veigar. (Or am I...)

Saturday, January 14, 2012

Final Fantasy IV: Super Rare Drops

My brother commented on yesterday's post about the 'infinite' victory Game Genie code and said that the thing he most remembered about it was getting a large number of rolls on the enemy loot table. There are some stupidly rare drops in the game and this was the first time we'd ever seen one of them. (Mage, I think.)

I mentioned that Rydia learned almost all of her spells via leveling but that's only true of white and black magic spells. Her call spells are learned in a wide variety of ways. She starts the game with one. She learns five more when she rejoins the party after growing up in the land of the summoned monsters. Most of the rest she learns by having the party defeat a legendary monster. Find Odin at the bottom of Castle Baron and beat him up to learn the spell. Brave Bahamut's Cave on the moon to learn that spell. Hit Yang on the head with a frying pan to learn Sylph. There are a couple spells that you can only learn by getting a rare drop from a random encounter. I don't think I've ever had Mage myself. I got Imp once from a monster in the final dungeon on the moon. Are there any other super rare drops? Lets see what gamefaqs has to say on the topic...

First of all it looks like every monster has a 5 in 96 chance to drop an item at all after you kill them. (Dr Lugae's second form has a 100% chance to drop an item as the only exception.) After determining that they'll drop an item there are 4 things they could drop. The first item has a 32 in 64 chance. The next has a 20 in 64 chance. Then an 11 in 64 chance. Then a 1 in 64 chance. The mage monster, for example, has a cure2 potion as the first three items on the loot table and the mage spell for Rydia as the last item. Which means that if you kill a single mage your odds of each outcome are...

5824 in 6144 (~95%) - no drop
319 in 6144 (~5%) - cure2 potion
1 in 6144 (~.016%) - mage spell


How is that a sensible game mechanic? You fight mages in one dungeon. (The magnetic cave.) I can't find random encounter tables but it's a pretty short dungeon and you're probably only fighting like 5 of these guys in a given game. With 5 shots at the item you're at a mere .081% chance to get the item. How many times would you have to play the game to have even a 50% chance at getting it? 852. If you played the game 852 times and killed 5 mages each time then you're barely more likely to see the mage spell than not. How can this make sense? Is it really worth spending developer time on something that will come up that rarely?

This isn't the only item in the game with this kind of rarity, either... Here's a list of all the items that only come up in the 1/64 slot of a loot table...

Bomb spell
Medusa sword
Dragoon spear

RuneAxe axe
Artemis bow
Lilith rod
Cursed ring (I didn't know there was a cursed ring in this game!)
Crystal ring
Zeus gloves

Glass helmet
Imp spell
Mage spell
Pink tail

I've tried to get the pink tail before but now that I actually see the numbers behind it I'm pretty sure it's off the charts crazy to try for it. Estimates placed getting into a fight with the monsters that drop it at 1 in 64. I think you get to fight 5 of them at a time but that's still in the neighbourhood of 54528 fights before you reach a 50% chance of getting a pink tail. Holy crazy drop rates Batman!

Friday, January 13, 2012

Game Genie: Final Fantasy IV

My brother had a Gamie Genie for the SNES when we were kids. It came with a book of codes you could enter to modify how games worked. Infinite lives, access to all the guns at the start of the game... Things like that. I don't know how it worked (I am curious though) and I don't remember ever actually using it on any game. Any game, that is, except for Final Fantasy IV!

I remember there were codes you could enter to change how much experience you earned at the end of a fight. Make each monster worth 1 xp. Or 10 xp. Or 100 xp. And so on. There was an obvious pattern to the codes and we toyed around with making things bigger and bigger. Eventually we stumble on a code which seemed to break the game. A fight would end and your characters would stand around doing the victory dance but it wouldn't list how much experience/gold you'd earned for the fight. Eventually we turned off the Game Genie (it had an on/off switch) and suddenly earned a ludicrous amount of experience. It was like the game was stuck in an infinite loop while it was turned on and it was constantly adding on the experience for the monsters over and over. Hello tons of levels!

We figured this out in the antlion cave near the start of the game when your party is Cecil, Rydia, and Edward. Gaining a ton of levels at once had two interesting consequences...

First, Edward has an auto-hide ability where if he's at low health he runs away from the fight. In the Japanese version he has a healing ability which he can use while hidden (in order to get enough health to actually stick around and fight) but in the NA version he just screws around hidden. Gaining 80 some odd levels raised his maximum health a lot. So high, in fact, that I couldn't heal him out of low health. It might have been a problem if I didn't have a super high level dark knight!

Secondly, Rydia learns spells as she gains levels so when she gains a ton of levels she learns all the spells. All the spells, that is, except fir1. She learns that spell as a plot point at the start of the next dungeon (you need Rosa to convince her to get over her fear of fire and burn down the ice blocking the party from walking to Fabul) so it makes sense that she wouldn't learn it from just gaining levels. What doesn't make sense is she does learn fir2. And fir3. So the scene where she's too afraid to melt some ice with a fire spell doesn't real gel with the fact I've been exploding enemies with fir3!

I don't remember actually playing very long with the stupid amount of experience. Cheating just isn't very fun it turns out.

Thursday, January 12, 2012

Hunger Games: Training Days

At the end of 2010 I was at a local book store trying to spend a gift certificate when a board game caught my eye. It was a board game based off of a book trilogy (the third book had just come out) and was marked down 75% off. A quick look at the game didn't get my hopes up that it was going to be very good and I'd never heard of the book series. I decided to get the first book and then later pick up the other two books and the game if I liked the first one.

It turned out that the first book, Hunger Games, is awesome. Possibly my favourite book of all-time, though we know how hard it can be to compare things across years and years. At any rate, it's really good. I eventually went back for the second two books (sadly only in hardcover when I got them) and the game. The game was gone, however. Oh well...

I was out at 401 last month and saw they actually had a copy of it. I snapped it up and gave it to my sister for Christmas since I knew she liked the books as well. The board game, sadly, was a bit of a disappointment.

Now, without spoiling the book too much (seriously, go read it if you haven't) the basic idea is every year they hold a 'hunger games' where 12 districts each appoint a teenage boy and a teenage girl to enter the games. The 24 kids get put into an arena of sorts and battle to the death. Last kid alive wins food and stuff for their district. There's all sorts of crazy things that can go on in a death arena...

A board game based on that premise seems like it could have a lot of potential. I envision a dungeon that gets built as you're playing the game with players linking up randomly to fight. It would be an elimination game, of course, with different characters and stats. I'm thinking it could have a Scotland Yard style mechanic so you don't really know if you're following along behind someone else or not... There may be some clever way to have the game track that sort of stuff but it could work with a GM character who oversees the arena while the players compete to see who will win. (Maybe give the GM some 'cylon leader' style goals?)

Unfortunately the board game doesn't have anything at all to do with the arena. There is no battle to the death. The key is in the last part of the name... 'Training Days'. The game simulates being one of the kids and training to get ready for the arena instead of the arena itself. So you do weight lifting and archery and junk like that instead of actually fighting each other. The theme was rather disappointing after the expectations I had.

As far as the gameplay goes it's a hidden auction system. Each round you flip up a card for every player and then you bid face down tiles to try to win the cards. You have a 1, a 3, and a 6 to spend each round. On top of your tiles you have a character with stats that you get to add to the card check. So someone with a high agility would be better at the archery challenge and someone with high strength would be better at lifting weights. The challenges are all vaguely related to the book and all of the characters are from the book too so it is still flavourful... Just no killing.

It has some interesting gameplay as you try to figure out who cares about which card and how much you need to invest in a specific card in order to win it. But it has a random game end condition and there's not really any long term planning. There are obviously correct plays every turn and you just need to figure out who is going to deviate from the script to try to steal a card cheaply.

Overall I'm happy I played the game once but I don't have any real desire to play it again. I am intrigued by the prospect of working out a system to play out the actual arena deathmatch though...

Wednesday, January 11, 2012

Campaign Manager 2008

Campaign Manager 2008 is an interesting little game on Yucata simulating the 2008 US presidential election between Obama and McCain. The game is essentially a card driven wargame (you play one card each turn to impact the voting in a given state) but it has a couple twists. It's non-symmetrical in that each player has their own distinct deck of cards. It also has aspects of Dominion in that you build your deck at the start of the game instead of always having the same cards available each time.

As an aside, here's roughly how the game works... There are 4 states in play. Every state has two issue tracks. It's a tug-of-war style mechanic so if red gains a defense on this card he takes over the blue one. There's also a little meter on the left showing what the state actually cares about. If a state ever cares about an issue completely dominated by one of the players they immediately win the state. The demographics on the left are used on some cards to take all the white dots in one fell swoop. Score points for the state and then add a new state to the board. Keep going until someone has a majority of the votes. Each turn you get one action which can either be to play a card or to draw a card. You can only choose to draw a card if you have fewer than 5 cards in your hand.

The first twist isn't actually much of a twist. Each player has 45 potential cards to include in their deck and most of them are mirror images of the cards in the other player's deck. Gain support in economy and draw a card. Gain support in defense and shift the issue track towards defense. Get control of the media. Draw 3 cards. Draw 2 cards and play 1 of them. Campaign targeting women... McCain has a slight bias in his cards towards defense and Obama towards economy but by and large most of their cards are the same things. They let you do two unrelated things. Gain one support and shift the left track. Switch the key demographic and draw a card. But there are a couple cards with powerful and unique effects for each side...

Obama's first such card is essentially windfall. Play it as your action and then discard any number of cards. Draw a card for each card discarded. This can let you take some situationally powerful cards and discard them if they aren't useful to turn them into something else. It can also let you dig for specific cards if you need them. McCain's comparable card is a regrowth effect. Play is and discard one card to pick up a card from your discard pile. This does similar things but lets you double use your best card each time through your deck. It wastes two cards and an action though so it had better be a really powerful effect!

Obama's second unique card lets you discard a card to remove the last card McCain played from the game along with the unique card. He still gets the impact when he played it but it won't get reshuffled into his deck. When I first saw this it seemed overpowered. Killing off your opponent's best card is pretty sweet when you go through your deck over and over. Unfortunately McCain's second unique card trumps it. It also requires you to remove it from the game and discard a card to play it. It lets you take any card from out of the game and add it to your hand. Including the card Obama removed from the game. And since it puts the card in your hand you get to play it again next turn! If there exists a card which is really powerful then McCain gets to use it twice. If all cards are about the same power level then neither of these cards matter...

The second twist is that you build your deck before the game starts. Each player has 45 potential cards and builds a deck by drawing 3 cards and keeping 1 of them. Repeat this until you have chosen 15 cards and play with those cards. Want to build a deck focused on the economy? Draft a bunch of economy cards. Want a bunch of powerful but situational cards? You can do that too. It's a really interesting mechanic and one I think has a lot of potential. You build your deck in a vacuum without seeing anything your opponent does which I think hurts the strategic depth. I'm wondering if it would be better to show your opponent what you're taking? Or to maybe do a solomon style draft!

At any rate, I've played the game 4 times. Once we used the default 15 card decks and the game made sense. It was back and forth the whole way and I won by like 1 vote in the end. The other three games were complete blow-outs. McCain got into a situation where he seemed to be unbeatable. In the game I played against Andrew today I ended up with 14 of my 15 cards in my hand. Then every turn I was able to play a card which gave me a defense support in a state and drew a card. (My deck had 3 or 4 cards with the effect in it.) If he ever did anything that forced me to do more than that I could play one of my other cards and cycle back into it in two turns!

The problem is almost every card does 2 things. If one of those things isn't draw a card then you need to spend a full turn drawing a card. So over 2 turns you can draw a card and get 2 abilities. Alternatively if you play two cards which do a thing and draw a card then over 2 turns you can draw 2 cards and get 2 abilities. If you can afford to wait the second turn for the second ability you're up a full card. Even worse, the cards which do 2 things do unrelated things. If you don't need both abilities then you're behind a card and an ability over two turns!

But how do you build up a 14 card hand? If you could always draw a card as your action on a turn you'd just spend the first 10 turns drawing cards. Unfortunately you can only use the draw a card action when you have fewer than 5 cards in hand. So instead you need to make use of the "draw 3 cards" card and the "draw 2 and play 1" cards. This will slowly increase your hand size as long as every time you play a card it draws a card. This is yet another reason why you want your cards to replace themselves. On top of it just gaining a card every 2 turns you're setting up a bigger hand which provides more options each turn.

Why is it always McCain that pulled this off? I'm not sure. The "draw 3 cards" card seems to be the best card in either deck by a pretty large margin. The two times I played as Obama I nailed it with my exile a card card but they always just brought it back and got to play it again immediately. Coupled with the regrowth it seems pretty easy to build up a big hand. Possibly the windfall card from Obama can set up the same thing if you get proper discard tempo? I want to play again as Obama to see if I can put the deck together... But even then, if you just have two brutal decks going against each other is the game going to be interesting? It feels like it should just end up in a draw?

Tuesday, January 10, 2012

"Elo Hell" Revisited

Two months ago I posted about the rating system in League of Legends and how my rating at the time was making the game unfun. I've been asked since then by several people about if I ever got out of "Elo Hell". I figured I should revisit the topic to see if two months of play has changed anything.

The first thing to mention is I still often see people in games complaining about being stuck in Elo Hell. It doesn't seem to matter what rating you're at; there's always people who are stuck at that rating and think they should be higher. I'm pretty sure this can just be explained by Illusory Superiority. Most people think they're above average. Pretty much by definition some of them aren't. Those people are going to reach their rating equilibrium point, believe it should be higher, but get stuck. And then since this is the internet they're going to whine about it. That people complain about Elo Hell in this manner doesn't mean it exists at all.

The second thing is that games have gotten better recently. There are still people who whine. People who make mistakes and play badly. Blame gets thrown around. But by and large everyone is still trying to win. I had one game recently where someone stated point blank that we had to give him solo top or he'd feed. (We gave him solo top and then he was bad and fed anyway...) Beyond that one game though I haven't had anyone threaten to play badly at the start. I haven't had people who intentionally join a game just to lose. Or who don't really join at all. Games, by and large, are fun.

As far as rating goes I had fallen as low as 1001. I started playing games almost exclusively with Robb and focused on playing a couple of champions well. In particular I stopped playing Vayne (8-16 record) and switched to Tristana (13-11 record) as my ranged physical attacker. I'm up to an overall winning record (120-111) and my rating is 1294.

Apparently Riot also implemented an automated banning tool which suspends accounts of people who routinely go AFK. Maybe that would help the matches down at the 1000 rating zone? I have my doubts. The Tribunal thing getting some press for actually banning people might be helping too. I posted about people flat out not caring about the threat of being reported and that may have changed? I kinda doubt that too.

But yes, I am out of "Elo Hell". Games are, by and large, fun. I'm sure my rating should be higher than it is since I'm obviously above average at the game but I'm sure that will come in time...

Monday, January 09, 2012

League of Legends Tribunal

8 months ago League of Legends brought in their tribunal system as a method to try to foster good behaviour amongst the player base. The basic idea is players can report people in their games and then a jury of their peers would vote on in the player behaved in a way that broke the rules and should be punished or not. I'd heard about it when it first came out but didn't pay much attention to it. I report people all the time but I never really cared about what went on after submitting a report.

A couple weeks ago I finally got around to signing in as a jurist to punish some bad players. It was interesting to see what actually gets submitted when a player gets reported. The jury gets to see the entire chat log from the game (but not the champion selection or game recap screens) with the offender's chat in a different colour to make it easier to find what they said. It also lists their final items, their kills/deaths/assists in the game, their incoming and outgoing DPS, and what level they finished the game at. It also lists every reason they were reported after the game and differentiates if it was a teammate or an opponent who reported them. It doesn't say which player specifically made which complaint.

The great part is it lists all the games they played where they got a complaint. You can browse all complaints to see if there's a pattern of bad behaviour or if people are just being really oversensitive or bitter when reporting.

I did a few and got an email on Friday saying that I got 8 cases right and they were giving me 40 IP for my help. I don't remember how many I did. It doesn't seem there's any way to see how many I did, so I don't know what my winning percentage is. I do know I pardoned more people than I was expecting...

I did some poking around today to see if I can work out what's exactly going on behind the scenes here. They posted some interesting stats last month... Apparently only 1.4% of players have been punished by the tribunal. But of people who get presented at the tribunal 94% get a punish vote. Half of those people get sent to the tribunal a second time and get another punish vote. Apparently most people who get reported are on the losing team and get reported by their teammates. (That makes sense since you're more likely to get angry and start raging at someone who you think is making you lose a game.)

I also dug into the forums and found some facts about how the system works. They refuse to give out specifics since they think offenders will be able to abuse the system if they know details. I guess that could be true but as the saying goes... Security through obfuscation isn't security at all. I'd rather they have a system that couldn't be abused but I guess they don't feel like they got that far. What they have said, however...

  • One report won't get your case presented at the tribunal. It takes several games with reports to get presented for voting. The percentage of games played compared to games with reports matters.
  • Long term or permanent bans have to get approved by a Riot employee but everything else is automated.
  • Your first time found guilty you get a warning email that doesn't tell you what you did wrong, just links to the code of conduct.
  • The second time you get a short term ban (1-3 days or so) but still don't get told specifically what you did wrong.
  • Even if you get permanently banned you don't get told what you did wrong.
  • Riot employees who spot check random cases think the voters are too lenient. With a 94% conviction rate. Sounds to me like they need to start sending more borderline cases to the tribunal since if the conviction rate should be more than 94% I can't imagine it makes sense to actually review the cases. Punish everyone and move on!
  • You can get reported for unskilled play but those reports don't get added to tribunal cases. The idea is to give people a way to vent without actually hurting anyone since people being bad isn't an offense. (With the rating system they should eventually fall out of your play range if they're actually worse than you!)
  • The system had a lot more oversight built in at launch but they're pretty happy with how things are working so it's almost all automated now. The computer picks who gets put on trial. The computer tabulates the votes and decides when someone should be punished. The computer punishes them.

So it is entirely possible that, through no fault of my own, I could get banned. If enough of the people I play against randomly report me and enough of the jurists blindly vote to punish I'll get hit without any oversight. On the other hand it would have to happen in a lot of games. And since I really don't do anything bad and the jury does pardon 6% of people... I really have nothing to worry about. It's theoretically possible for the universe to align and screw me but it's really not plausible. 

It seems sketchy that they don't tell people what they did wrong, but I've been a moderator for an online game. People who are apt to misbehave are also apt to whine and argue and try to rules-lawyer their way out of their punishment. I didn't really have all that much going on so I'd typically indulge them but it almost never went anywhere productive. They'd want to be able to swear. Or trade in the play areas. I'd always tell them specifically what they did wrong but the vast majority of the time they already knew what they did. They knew we didn't allow it. But they thought we should allow it so they were going to do it anyway. 

The real key is they would gladly debate the facts for hours on end. They'd try to convince me I was wrong. And from their point of view it was just them wasting a couple hours. But from my point of view, when you get several of these people every day... Suddenly my whole day is wasted. (In reality I'm very good at multi-tasking chat and my side of the 'conversation' was pretty limited so I'd mostly just let them chat themselves out.) Even then, we had more chat moderators than Riot does. And fewer players. We provided better customer support for sure, but I don't know that Riot is doing things wrong. It doesn't make financial sense for them to hire a bunch of people to debate with people who are almost certainly going to get banned soon.

I do think it would make sense for them to provide a list of the reasons on their tribunal case. Add in a disclaimer about how they were reported for some of these reasons and be done with it. At least then people can rest assured that they weren't banned for unskilled play, for example. Getting an email warning saying you've been bad and to read the code of conduct is great and all but getting an email warning saying you were reported for 'AFK, AFK, AFK, AFK, AFK, Negative Attitude, AFK' gives you a pretty good idea what your problem is!