Saturday, January 1, 2011

civilization V: comments on the december patch

I've now completed a couple games with the new Civ V patch (v. and feel like I can make some informed comments.  Overall, I am very pleased.  The game is significantly more challenging, and a number of balance tweaks were introduced that make the civilization-building process more rewarding.  Obviously, there is still room to improve, but Civ V now ranks among my favorite strategy games of all time.

Some details:  Prior to the patch, I had played three games to completion, two of which were on Emperor difficulty.  I won all of them easily:  one via culture, one via diplomacy, and one via space-race.  The wartime tactical AI was atrocious and could be outmaneuvered by a single powerful horseman.

Since the new patch, I have played one game to near completion on Emperor difficulty as China.  I went for a cultural victory but it became clear to me around 1900 that I was going to lose to Hiawatha, who had researched the Apollo program and was way ahead of me on tech and military.

I dialed things back a little to King difficulty and played as Russia.  Definitely my most entertaining game of Civ V yet.  The screenshots I include in this post come from that game.  Standard map size, continents, 8 civilizations, standard game speed, no mods installed.  I eventually won a space race victory but Napoleon was leading by points for the majority of the game.

Rather than bore you with an AAR, here are some of the positives I noted in these patched games:

1.  AI is better at "bringing it" in warfare and stock-piling defensive troops, especially if it has a tech advantage.

In my 1st game, I mentioned that I realized in the early 20th century that I didn't have enough time to complete my 5 social policy trees before Hiawatha completed his space shuttle.  So I decided to quickly research the Manhattan Project and try to disrupt his progress with an atomic attack.  As soon as I declared war and lined his border with my troops, he responded swiftly and strongly.  He had a large military advantage at that point in the game, and within 10 turns he had already conquered one of my cities and was marching towards my capital.  Nuking him had little effect.

2.  AI tactical military decisions have improved.

 A couple things I noted here.  First, when attacking AI cities, it knows to focus its bombardment and counterattacks vs. your siege units.  Second, if it has researched Flight, expect to be air-bombarded constantly while in its territory.  Third, it is making better decisions about whom to attack with what unit.  For example, in my 2nd game as the Russians, I rushed to knights before attacking the Ottomans to my south.  After I invaded, I noticed he had a number of pikemen defending his cities and he used them quite effectively against my knights.  My longswordsmen and siege units were much more important in victory than my knights.  The screenshot below shows him defending one of his cities with a pikemen poised to attack an incoming knight, and a knight ready to charge against a siege unit.  Fairly clever.

3.  Unit promotions don't occur until the turn after the unit earns the XP.

This is a welcome change, since previously you would almost always heal your units, making it very difficult for the AI to kill off your wounded.

4.  Mounted units and tanks get penalties vs. cities.

I love this change, since it really forces you to research and produce infantry and siege units to support your cavalry and tanks.  You can't simply horse-rush an opponent anymore - their cities will hold you off indefinitely (especially with the improved city-healing).

5.  Diplomacy is improved, but still a little wonky.

I like how the AI's now reveal why they feel a certain way about you (see screenshot below). And it is possible, albeit challenging, to maintain positive relationships with your neighbors for long periods of time.  Denouncements throw an interesting twist into the game.  In my game playing as Russia, I was denounced by the Persians because I founded a city too close to their borders after promising them I wouldn't.  This ruined my relations with the Ottomans, who previously held a neutral attitude towards me.  This made it so that I couldn't effectively trade my luxury resources early in the game (since these were the only 2 civs I was in contact with), which made happiness and cash accumulation very difficult.  I really enjoyed this challenge, right when my empire was poised to blossom.

There are times when the AI will denounce you for reasons you can't understand, and it can be quite difficult (and therefore frustrating) to repair relationships over the centuries.  Given that there must be a rule "under the hood" that the AI is following to determine its denouncement, I suppose I'd appreciate even more feedback regarding their reasons.  For example, India might denounce me not because of anything I specifically did to them, but because they've been bribed to do so by another civilization.  From my standpoint, it appears random - but it would be useful if India said something like, "Some of my allies have convinced me that your foreign policies are threatening to world safety."

 6.  AI's seem better at managing their economies.

This allows them to engage in far more research agreements, which they will spam across the entire game.  If you don't pay attention, you'll fall behind in the tech race quickly because of this. After taking over a number of enemy cities (and puppeting them), I noticed that most AI governors were set to maximize cash.  If this is the case across the board, it's slightly problematic - since at least one or two cities should be specialized for production.

In both games, I encountered civilizations that had acquired massive fortunes.  See Napoleon's screenshot below.  I suspect this money is being accrued via conquest and then raping enemies in peace treaties - a game mechanic players have been known to abuse as well.  This again suggests that the AI is "doing well" in some respects but it also indicates that the AI doesn't know how to spend its surplus.  With this much money, it should be very easy to pay off every city-state in the world and win a diplomatic victory - or insta-purchase dozens of science buildings for an unstoppable tech advancement.

However, I didn't get the sense that Napoleon was going for that.  He had conquered most of the world, and was far ahead of me in points.  It was the 20th century and I had researched the Apollo program. I assumed he was going for a military victory, and he was within reach since only he and I had not yet lost our capitals (see screenshot below).

If he had sent his entire military strength against me, he could have crushed me in the late 19th century.  However, I was across the ocean and the AI still seems very hesitant to engage in wars on different continents.  Also, I believe he could have researched the United Nations, payed off every city-state, and won a diplomatic victory before I finished my 1st spaceship part.  But he just sat there, threatening me, until I won the game.  This was the most dissatisfying aspect of the game, and indicates that additional work needs to be done on intercontinental invasion AI.

7.  Some random notes.

I like how science production is now more dependent on the size of your cities, making larger cities more valuable.  Also, it's probably a good thing that libraries don't have a specialist slot anymore, since Great Scientists are very powerful in the early game.

I love the idea of a revolt happening when your happiness dips to -20, but I doubt that's ever going to happen to me or the AI's.  I think this should be tweaked to something much more threatening, like -5.

I'm not sure how I feel about not being able to defer social policies.  Previously, it was an interesting instant vs. delayed gratification choice.  Do I take the bonus to happiness now, or stockpile and wait until the Renaissance to burst my way through the Rationalism tree?  The biggest advantage I can see to the new system is that now we (the player) are forced to behave like the AI's, since they never seemed to weigh the advantages of stockpiling.  Regardless, the option can be changed in the set-up menu if you prefer the older system.

The AI is engaging in a bit too much ICS, settling small cities in inappropriate locations.   

Overall, as I said, I'm very pleased.  I find it interesting to read the various forums, on 2K and Steam, and hear so many people complain about so many different "stupid" things the AI's are doing.  You begin to realize that individuals are having greatly varying experiences:  for some, the AI's are overly aggressive and don't leave you alone - for others, the AI's are too passive and allow cultural victories without nary an invasion.  Such variety, I suspect, is suggestive of a deep, complex, and highly replayable game.  I look forward to subsequent patches, some creative mods (Dune Wars for Civ V please!), and big expansions.

Friday, December 24, 2010

favorite PC games of 2010

Blogging really took a hit this year, and it's quite likely that Aili and I will just let this thing die next year.  We have a baby coming and neither of us is particularly motivated to write on here nowadays.  I suspect the internet won't notice our absence.

Regardless, I did want to do an end-of-the-year round-up of the PC games I played and enjoyed in 2010.  I should note off the bat that most of these were not actually released in 2010.  Like most gamers, I tend to wait a while before purchasing and playing - save money, wait for patches, read player feedback, and play games that better match my system's aging specs.

So, here are the 5 PC games I enjoyed the most in 2010.  I spent more hours playing these than I'd care to admit.

#5:  King's Bounty: Armored Princess

The follow-up tactical RPG to the awesomely wacky 2008 hit.  In some ways, AP was a bit of a let-down.  It didn't feature the bizarro quests and side-stories that the original had in spades, and didn't advance the gaming system in any significant way.  Basically, it was more of the same.  Much, much more.  Good thing I really like King's Bounty.  The tactical battles never got old for me - every one was an enjoyable puzzle, and the variety of tactical options (via unit special abilities, spells, and pet dragon abilities) was spectacular.  I still fondly remember the power-combo of Stone Skin & Target on my Ancient Ents.  Deciding on which units to use in my army was possibly just as satisfying as spec'ing out my party in Baldur's Gate II. It's that kind of fun.  Paladins, Trolls, Black Dragons, Inquisitors, and Demonologists.  Fucking batshit insane. The RPG aspects are integrated well, requiring you to level up skills and abilities intelligently or else you'll get overwhelmed in later stages of the game.  I played on the "Hard" (but not "Impossible") difficulty level as a Mage and was able to get through nearly everything.  Unfortunately, I couldn't get past the last couple boss battles, no matter how I shifted things around.  But I definitely sunk over 60 hours into this beast earlier this year.

#4:  Batman: Arkham Asylum

A fantastic, tight action-adventure game.  This is one of those ports that almost starts you thinking that console game development is actually up-to-par with the PC.  Visceral.  Learning the punch/kick/throw/take-down combos, and then using them to go through a mob of 50 thugs was incredibly satisfying.  As the critics have said, this game succeeds at making you feel like Batman.  "I am a total bad-ass."  I loved the grappling hook mechanic, which allowed you to escape out of sticky situations and observe your enemy from above.  The plot was good enough, the variety of villains was fabulous (loved the freaky Scarecrow sequences), and I never felt too frustrated to quit.  I didn't get obsessed with collecting all the Riddler's bullshit, but got a dopamine burst every time I found a question mark.  Steam tells me this took 23 hours of my life.  Probably 90% of that was from the actual campaign, but I also played through a significant number of the challenges.  Whenever I needed to relax for a few minutes, I'd see how long I could keep a combo going versus a horde of hoodlums.  There are few single-player games that I consider replaying when I finish (Bioshock is one), but Batman: AA is certainly worthy of a rerun.

 #3:  Team Fortress 2

What more can I say that hasn't already been said?  Best current online shooter?  Check.  Best online shooter ever?  Quite possibly.  I got the Orange Box in 2008 and have been playing TF2 on-and-off ever since.  I quit, I thought permanently, late last year when I saw the "hat" phenomenon starting to obfuscate what I loved about the game.  I missed a number of the class updates - kept track of the changes via blogs and internet-drip - but didn't feel the pull back in until this Halloween.  For whatever reason, I decided to re-install on Steam and see if I had been missing anything, and that's all it took.  TF2's greatest strength is that you get near-immediate pleasure from the game.  It's incredibly easy to find a game, jump in, and start contributing to your team.  It's almost certainly the most noob-friendly online shooter, and that reduces the asshole factor significantly.  Especially this late in its lifespan, most players are mellow, there to have a good time, and willing to give you the benefit of the doubt.  I also love TF2 since I'm now typically one of the better players on the servers I visit.  Picking medic and turning a control-point game around for your team is absolutely thrilling.  I also love TF2's balance system.  If teams are uneven, some players will get switched around - but more importantly, it doesn't matter much since the teams can be randomized for the next map (which on fast rotations, isn't that far away).  Just a great great game that's only gotten better with time.  Over 200 hours played and I still don't fucking know how to play a Spy well.

#2:  Demigod

It's almost embarrassing to admit it, but I suppose I'm over that.  I just loved this game to death.  Flaws?  You bet.  Probably the biggest being the small player base that made getting into a decent game an exercise in frustration.  I'd literally spend 45 min waiting for every 30 minutes of playtime.  And that's eventually what broke me.  The release of Starcraft II took just enough players away to effectively kill Demigod, and around that same time it also became clear that GPG & Stardock were no longer interested in supporting the game.  They did give us 2 more demigods to play with:  the underpowered Demon Assassin and the overpowered Oculus (see above), but the community's continual call for a significant item overhaul went unanswered.  After more than 500 games, it's no surprise that I got pretty good - and when I found myself in a 3v3 with five other solid players, the intensity was unmatched in my gaming life.  Until someone would randomly disconnect.  Ahh, Demigod.  How I love and loath thee.  I *hope* someone resurrects you one day, as Demigod 2: the Rebirthening, with dedicated servers and better item balance.  You could have been so much more.

#1:  Left 4 Dead 2

L4D2 has to win my personal game-of-the-year, since it so captured my heart.  In a time when zombies are overplayed, it's a testament to Valve's design team that L4D2 so perfectly represents what I want out of a zombie-killing game.  Co-op play, where a single panicked idiot can take the whole team down.  An AI Director who changes up the experience each time you play and periodically throws the perfect storm at you to make your life a living hell.  Losing never felt so good.  A versus mode that's highly competitive (and attracts some douchebags) but intensely satisfying.  I never quite got "good" at versus, but I stopped feeling like a total noob.  Tip:  you can aim your Booms up to get greater range and coverage.  The free released content (especially the No Mercy campaign) was fantastic, and allowed us to play as the characters from L4D.  Getting through the entire Dark Carnival campaign on Advanced difficulty (never even got close on Expert) with three other solid players who communicate and help each other throughout is one of the best experiences there is in gaming.

Honorable Mentions

Civilization 5:  a poor AI kept this off my top-5, but the newest patch may have tightened things up a bit.  See my last post for why I think this might end up being the best Civ yet.
Dawn of Discovery:  the game I wish I had more time to play.  A deep, satisfying city-building game with a complex economy.  I hope to get back into this one soon.
Dirt 2:  a hard-as-nails racer. Varied race-types, beautiful courses, tons of unlockables, and challenging events.  Ultimately, however, I just couldn't handle the repetition-to-perfection cycle that's necessary to succeed in games of this ilk.  Someone recently compared racing games to Super Meat Boy and I'd have to nod in agreement.  This is both a good and bad thing.
Solium Infernum:  a mind-boggling interesting game, I wrote about this one in early 2010.  What kept this off my top 5?  Well, I only ever really played 2 full games (against humans).  It was just a bit too hard to organize a game and run it to completion.  The play-by-email format was pleasant, in that you could take your time with plotting, but ultimately the slow pace did get a bit frustrating.  Probably the most "intelligent" and interesting game I played this year.


Magic the Gathering: Duel of the Planeswalkers:  no deck-building in an MtG game?  I don't understand.  Otherwise, a lovely interface and classic gameplay at a decent price.
Battlefield Bad Company 2:  I tried hard to get into this (mostly so I could break my TF2 habit), but it just never clicked.  It didn't help that I really really sucked.  I would join a squad, spawn, and die within 15 seconds.  I also hated how poorly balanced the teams often seemed to be.  Finally, I don't think my PC is good enough to play this at a decent frame-rate.
Tropico 3:  it pains me to put this on the list, since I do adored the 1st Tropico.  But this version was really nothing different.  Plus, the economy was too simple which took all the fun and challenge out of the game.  I need to play through more of the campaign to see if it gets better, but I consider this one of my biggest gaming disappointments of 2010.

Sunday, October 3, 2010

Civilization V: some thoughts

Sid Meier's Civilization games are all about the enjoyment of making decisions.  Yes, there's a narrative that emerges due to the outcome of your decisions interacting with the "decisions" of various AI entities, and that narrative can be quite fulfilling (especially in retrospect).  But I'd argue that the majority of your minute-to-minute happiness while playing Civ is based upon the interesting decisions that are presented to you.  Should I built the Hanging Gardens now?  Should I settle to the north where there is a flood-plain and an abundance of food, or to the south where there is ivory and incense?  The greatest strengths and weaknesses of the Civ franchise lie within these decisions.  Some are compelling:  they present difficult choices, carry meaningful outcomes, and allow you to "feel" like a world leader.  Others are distracting:  they are embedded within the intricacies of the game system and are only meaningful on that level.  They annoy and frustrate you.

Personally, I've found that most Civilization games (I should note that I have experience with II-V) tend to ware me out after I've played a few full games.  I *want* to love these games.  And part of me does, since I have such a fondness for complex turn-based strategy.  But as the game progresses, the number of fun & interesting decisions becomes heavily outweighed by the number of mundane decisions.  Other Civ players have decried this as "excessive micromanagement," but that somewhat misses the point.  Some micro can be fun - if the decisions are interesting, meaningful, and immersed in role-playing.  Some macro can be distracting - if it feels too "gamey" and like pushing numbers on a spreadsheet around, instead of conquering the world.

One reason why I think the latest Civilization (V) might be the best of them all, is that it has gotten rid of a lot of mundane decision-making.  That, inevitably, has led to a backlash among long-time Civ players that it has been "dumbed-down" to be made more "accessible" to "retarded console players."  I'm not even sure what they're referring to when they say this, but the 2K forums are full of posts like this.  Place me firmly on the side that thinks many of the changes made to Civ V are positive - but, of course, there's great room for improvement which I am optimistic will come.

Back to the point.  Using Civ V as an example, here are some decisions that I would consider interesting and meaningful.  I have attempted to rank these from most interesting to least:

  1. What victory strategy will I pursue?  Military domination, cultural, space-race, diplomatic, or score?
  2. Should I go to war?  And when?  When turns out to be a very interesting decision in Civ.  It is almost always a good idea to time your wars to coincide with the development of your "unique unit" for example.  This is especially powerful early in the game, when the number of military units involved will be relatively small.  In my last game as the Persians, I tried to time my wars to coincide with Golden Ages, since Persian military units receive a +10% bonus during these times.  
  3. Tactical decisions during wartime:  how and where should I attack my opponent?  Civ V has made some great improvements in making tactical combat decisions more interesting.   An aspect of the game which was always compelling is now more so, due to a greater emphasis on unit diversity and use of surrounding terrain.
  4. What social policy should I pursue?  I, for one, like the new social policy system in Civ V - but if you prefer, think of the Civic system of Civ IV.  Interesting decisions that carry weight and meaning in the game, and feed into the role-playing nature of civilization evolution.
  5. Where should I build this city? 
  6. What should I do with this Great Leader?
  7. What technology should I research next?
  8. What kind of city is this one going to be?  Production emphasis, gold income, high population with specialists, etc.
  9. How should I spend my excess gold?  To form a friendship with this city-state?  To purchase a new tile for my city?  In a trade agreement with another civ for a luxury resource I don't have?
  10. What tiles should my citizens be working?  Should some citizens be specialists?
  11. What should I build next in this city?  This decision should support my overarching strategy (#1).
I put city-decisions at the end of the list, because while I do find them to be interesting throughout much of a Civ game, there comes a time when I usually stop caring.  Especially in previous incarnations of Civ.  For example, if you can build every building in every city, then the decision becomes mundane.  Civ V has made this decision more interesting because:
  1. You do not have the time to construct every building.
  2. There are high maintenance costs on buildings, so you shouldn't spam.
  3. There is a greater emphasis on designing your cities to serve particular functions (for production, for population/research, etc.).

Here are a couple Civ decisions that I find less compelling:
  1. Where should I adjust my tax slider?  To tweak out a little extra research, or happiness, or culture?
  2. What should my workers be doing right now?  Building roads?  Cutting down trees to speed along production in one of my cities?  etc.
This is obviously a less than complete list but it is telling that these two issues (along with the last two from above) have typically ruined Civ games for me.  Especially at the end of a game, in previous Civ incarnations it really felt like most of your game-time was spent tweaking the tax slider (every turn, to maximize efficiency), assigning and re-assigning workers, and visiting every city in your empire to decide what to build and how to assign your population.  Ugggg.

If Civilization V has done something right (and I believe that it has), it has reduced the amount of time spent making dull decisions.  Unfortunately, it has not adequately filled that time I now have with other, more interesting decisions - especially if I'm pursuing a non-military victory.  As others have said, you will press "Next Turn" a lot in Civ V without having made a single decision that turn. I don't think that is a good thing.

Here are some changes/additions I would like to see instantiated in Civ V, which I think would enhance the fun-and-interesting decision space:
  1. More intricate diplomacy.  Diplomacy should be like a mini-game, and it should matter.  Achieving a Diplomatic victory should be complex, difficult, and an exercise in juggling different personalities and demands. Every few turns, I should be making a "fun" diplomatic decision (e.g. send ambassador, boycott, increase tariffs, publicly denounce at the U.N., etc.).  The current system, which emphasizes trade relationships and pre-war alliances is fairly mundane and shallow.
  2. Espionage.  I suspect they'll bring Espionage back into Civ V in a future expansion, and I hope they offer compelling decisions that don't bog down the system.
  3. Science and research.  I'd actually like to see additional complexity to science & research.  Perhaps individual cities (or research labs) can focus on particular inventions or "breakthroughs" that are separate from the standard Tech-tree.  Maybe there could be civilization-specific technologies to research.
  4. Enhanced cultural options.  Pursuing a cultural victory still feels quite "gamey" at this point.  There is no real sense of a civilization's "culture" spreading across the world.  In my first complete game, I played as Gandhi and achieved a cultural victory on "Prince" difficulty.  Overall, I enjoyed the experience, especially in the mid-game where I had to step lightly in the diplomatic game to ensure that none of my overly aggressive neighbors attacked me.  However, much of my late-game was spent pressing "Next Turn," waiting to unlock my next social policy. Maybe if the player was given choice of which direction you could focus your culture:  great works of art and music, better consumer products (intersecting with scientific research, perhaps), athletes, movies and television, etc.
  5. Disease.  This could be my personal bias, but I believe that disease has had a major impact on civilizations since the beginning (consider Europe's Black Death).  As such, I'd like to see disease instantiated in the system - with high population density increasing the probability of an outbreak and certain buildings (Hospital) and technologies (Medicine) ameliorating damage.  I didn't much like the "Health" meter of Civ IV since it added more city micromanagement.  Rather, I'd like to see Disease (like Happiness in Civ V) represented on a civilization-wide level.  This would make decision-making simpler and also more impactful.  And just a side-note:  what if you could (purposely even) spread disease into adjacent civilizations?
  6. Natural disasters.  I can only assume these are coming, and I'd certainly like to see them make an appearance.  And, of course, players should be able to turn them on-or-off at the Set Up screen.  Disasters are interesting because they can throw a wrench into your long-term plans.  What if an earthquake hits right when you're in the middle of a war against the Aztecs?  Should you pull out and focus on rebuilding? If a tsunami hits your ally, Japan, will you send them some money to help rebuild their empire?
 Oh, and beef up the A.I.  Because it really really sucks.

    Thursday, May 20, 2010

    demigod strategy: tower-slam rook build

    I thought I had my Demigod habit kicked but apparently not.  No doubt, Demigod can be incredibly frustrating.  On a slow night, it can take nearly an hour to get into a decent game.  Connectivity issues have cleared up a little, but still plague players regularly:  "mystery lag" can slow down a match in which everyone's ping seems to be fine, and your teammate might get disconnected unexpectedly halfway through.  It's enough to make you want to poke your eyes out.  But when it all gels, Demigod is still one of the best tactical, team-based games on the market.  

    After having played ~400 complete games, I feel relatively competent with most of the demigods.  I'm probably weakest with Queen of Thorns and the Demon Assassin, but they're the weakest demigods for everyone.  It look me a long time to get the hang of Rook, but I finally feel like I can help my team dominate a match with the Big Guy.

    I'd argue there are two viable Rook builds:
    • Tower spam:  max out Power of the Tower, load up on some mana items, and annoy the hell out of your opponents.  This is probably the most common build and it certainly can be an effective one.
    • Hybrid tower-slam:  some have argued that the problem with a 100% tower build is that Rook's towers decrease in value as a game goes on.  Enemy demigods will start stacking health and armor, so that even the thickest tower farms aren't doing enough damage on their own.  As such, it can help Rook significantly to have an additional Ace up his sleeve.
    Here's the Rook build I've been working with lately:

    1.  Power of the Tower I
    2.  Archer Tower
    3.  Structural Transfer I
    4.  Power of the Tower II
    5.  Boulder Roll I
    6.  Tower of Light
    7.  Hammer Slam I
    8.  Hammer Slam II
    9.  Hammer Slam III
    10.  Boulder Roll II
    11.  Hammer Slam IV
    12.  Power of the Tower III
    13.  Trebuchet
    14.  Power of the Tower IV
    15.  Dizzying Force

    First, some comments on Rook's weaknesses.
    • Rook is slow.  This is an understatement.  As such, you are at serious risk of getting ganked at almost any point of a game.  Different players utilize different strategies for dealing with this, but one thing is certain:  you need to have a Teleport scroll on you (or the Amulet of Teleportation) at all times.  If I'm flush with a little extra cash around level 7 or so, I'll also spring for the Wand of Speed (1750 gold) which can get you out of jam in a hurry.
    • Rook has low base armor.  I'm not sure if a lot of players are aware of this, but Rook (at level 1) has the same Armor rating (240) as Torchbearer.  Only Erebus (at 220) is lower, but Erebus has incredible speed and life-steal abilities to make up for this.  Rook looks a lot tougher than he actually is.  What this means is that Rook needs to be particular wary of an enemy Erebus, whose Bite (at level 1) reduces Armor by 250.  This would put Rook in negative armor, which means he will take massive amounts of damage from auto-attacks.  Rook also needs decent armor so that he can take out enemy towers, whose damage is mitigated by high armor.  As such, Nimoth Chest Armor is one of the best items you can get for Rook - and I try to get it ASAP.

    So what are Rook's strengths?
    • Rook can dominant a lane.  Most teams like to send their Rook to the health flag on Cataract right away, so that he can set up camp and establish dominance over that +15% health bonus.  Along with a couple towers, Rook can farm creeps, level up, and keep enemy demigods out of the lane.
    • Rook can take down enemy towers in a snap.  Structural transfer helps quite a bit with this, but it's not even necessary. 
    • Rook's towers serve as teleportation beacons for his teammates throughout the entire map.  Thus, Rook makes ganking easier. 
    • Rook can do a shit-ton of damage.  He has tower farms, a good auto-attack that only gets better, and a Hammer Slam that can flatten any demigod.

    Hybrid Rook works like this.  Through level 7 or so, play like a normal tower-spammer.  Try to have 2 or 4 towers up at all times (depending on what level you are), and use them to dominant lanes.  Your goal in the early game is two-fold:  1) help your team get a significant lead in Warscore, and 2) take down some enemy towers.  You are not out there to kill, and you must avoid dying.  Get dirty when your teammates port in to your towers for opportunistic ganks, but don't take any crazy risks.  Use Boulder Roll to stun enemy demigods trying to escape or chasing for kills.  Use Hammer Slam conservatively to take out some creep waves (for faster leveling) and minion armies.  If Oak's spirits are harassing you, a single Hammer Slam will usually do the trick.

    Taking down enemy towers is a relatively simple affair.  It's best to wait until the opposing team leaves you alone in a lane.  Approach an enemy tower with a friendly creep wave just ahead of you.  Build a tower immediately in front of the enemy tower - this will draw it's fire while you crush it.  Start whacking away.  If your health dips (to 2/3 max), use Structural Transfer on the opposing tower.  Then whack away until it's gone.  This entire process should not take more than 10 seconds.  Good opposing teams will know not to leave a Rook alone in a lane, or if they do, they'll port in to protect their tower as soon as they hear the warning that it's under attack.  Keep an eye out for port-swirls 

    Bolder Roll is such a great skill it deserves it's own paragraph and pic.  The true power of Boulder Roll emerges when you get to level 10.  As soon as you Roll an enemy demigod, start your Slam.  The 2 sec stun is enough to ensure they'll get crushed by the full weight of 1700 damage.

    I can't count the number of times that a Rook has caught me off-guard with this move.  Level IV Hammer Slam is the single-most damaging skill in the game - and that includes Level IV Spit, which does 1650 damage over 10 seconds.  The problem with Hammer Slam is it's 1.5 sec cast time and obvious wind-up.  Any observant demigod with an interrupt is going to whack you as soon as they see it coming.  But if you stun them first with Boulder Roll, they're doomed.  No Heal or Shield is going to help them now.

    The way I see it, Rook's biggest strength is that he's deceptive.  Early game, your opponents will view Rook as an easy target.  He's slow and he's big, so enemy demigods tend to target him by default and his low armor means that he really can't handle too much attention at once.  Your tower farms can cause headaches, but woe to the Rook caught outside a farm.

    But then something happens.  Around level 8 (and certainly 10), Rook becomes a monster.  His role transforms from lane-controller to offensive powerhouse.  And it usually catches people off-guard.  Maybe they haven't even seen you pull off a Hammer Slam until that point in the game.  And BAM, they go from 2000 hp to 300.  Maybe it's that pesky Regulus, who thinks he's got your number with Mines.  Stun... SLAM!  Dead.  It's pretty satisfying, and honestly, a more fun way to play Rook that straight tower-spam.

    What about items?

    Favor:  you have a few options here. 
    • Personally, I like Dark Crimson Vial.  Once they patched this guy up a bit, it became a viable alternative to the standard Blood of the Fallen.  I prefer the Vial since it allows you to play dead and surprise your opponents.  That's half the trick of Rook.  Make your opponents think you're weak, lure them into your tower farm, and then pop the Vial.  Now they're running away and you can Roll and Slam for the kill.
    • On bigger maps especially, the Amulet of Teleportation can be Rook's best friend.  It basically guarantees that you'll always have that telie on you.
    • Pure tower-spammers will sometimes take Blade of the Serpent, which can completely solve your mana problems.  This also means that you don't really have to buy any helms and can stack health and armor.
     Items:  in order of importance to my particular build
    •  starting:  Banded Armor & Scaled Helm.  You could get Banded & Scalemail, but you'll run out of juice for your towers too quickly.  And it's key that Rook be able to stay in the field for as long as possible in the early game.  Remember:  you want to dominant a lane, control a flag, and level up.  If Rook falls behind on leveling, things can get ugly fast.
    • Nimoth Chest Armor:  your best friend in the world, until you can afford Groffling.
    • Unbreakable Boots:  your 2nd best friend in the world, and one of the best items for the money.
    • trade Scaled Helm for Vlemish Faceguard:   you're going to need one helm throughout this build, and it's likely that Vlemish will be enough. 
    • Wand of Speed:  especially if you're facing fast opponents, like UB, Erebus, fire TB, etc.
    • Duelist's Cuirass:  this is just a personal taste thing.  The safe bet would be to go with Hauberk of Life, for standard health stacking and auto-healing, but I don't think you need it if you're carrying the Crimson Vial.  Instead, I like the added offensive bonus that the Cuirass provides.
    If the game lasts well past level 10, then you're going to need to make some final adjustments to all this:
    • trade Banded for Groffling Warplate.
    • trade Wand of Speed for Orb of Defiance.

    A lot of Rook players love the Orb, since it can give your teammates time to port in (remember to always build a tower near you!) and reverse a losing situation. It also provides a great +500/500 health/armor boost to late-game Rook.

    And there it is.  It's funny:  in the early days of Demigod, there was a lot of muted disappointment over Rook.  Not surprisingly, he was GPG's poster-child for their new IP.  Massive, towering over the battlefield, dominating all who approached.  He was the assassin you couldn't wait to play.  But then you played him and he was surprisingly fragile.  Slow and methodical, in a game that favored speed and flexibility.  Rook dropped to the bottom of the ranks (just above Queen of Thorns) in win %.  Players called for buffs.

    Fast forward to now.  Rook is considered a "Tier 1" demigod, at least when paired with the right teammates (Oak and UB, for example).  A good Rook can completely dominate a match, tear down enemy towers on Cataract before anyone even hits level 3, and establish a unassailable Warscore advantage for his team.  What happened?  Well, people learned how to play him.  He's not an easy demigod to master.  He takes patience and skill and careful micro-management.  His item builds are vital to his success.  But when it all comes together, the Big Boy is a beauty to behold.