My Digital Neuron

Inception       Jan 2005

I had a month off between jobs, and spent much of it in the beautiful and inspring surrounds of Sherbroke in the Dandenong Ranges, a hilly region of cool temparate rainforest 40 kms east of Melbourne. During the time of relaxation, I amused myself by learning to work with J2ME, and chose to build a toy game to drive my own discovery.

Arcadia was born, through when it began it was simply a motionless grid of self-made terrain tiles.

Go for Funding!     February 2005

From February my wife Louise starting trying to convince me to make an application for funding from Film Victoria’s Digital Media Fund.

May 2005          Putting a Funding Application together

Fortunately she prevailed and by May we were both hard at work putting an application document together.

Here’s what we said we were going to try to do:

“Playscape Games is making this application to fund the development of the first episode of mobile phone game-series Arcadia to a feature-complete prototype stage. Our vision for Arcadia is to create a Java mobile game of unusual depth and richness, which innovates in its game design, usability, and technical engineering. It will be one of the first mobile titles developed with a turn-based strategy/adventure format, in a fantasy/mythic setting. It will appeal to gamers who want to play a more deeply engaging game over a longer time period than the current staple of short-duration action games can provide.”

July 2005   Our primary artist Albert Sandaja (aka Alex Huang) signs on to start designing sprites for us.

September 2005     You’ve Got Funding!

…made especially sweet by the fact that the message arrived while I was having a lovely time on holiday staying with Louise’s uncle Phillip on Bribie Island.

November 2005    The first tranche of our funding, approximately $11k, gets paid to us.

Jan 2006      Leave Goldman Sachs JB Were behind to work a 3-day-week at Internet Business System as my day job. Turned out well, as Im still there now almost 3 years later. This freed up 2 days a week to put towards Arcadia (at least in theory – distractions were and are a major problem).

Feb 2006 Attend GDC Mobile in San Jose. An eye opener.

April 2006  Milestone 1

Deliver 40 page Business Plan to Film Vic. Main objective is to become a niche provider of very high quality and innovative RPG & Strategy content, distributed through a mixed aray of channels, into a variety of global markets, including direct online sales. In response, the Full funding program confirmed, and another trnache paid.

January 2007 Milestone 2

This was a tough milestone to pass. We demoed our playable prototype to Amelia King of Film Vic. It had lots of rough spots and while it technically worked, broken game mechanics and poor usability made it only barely playable. We agreed on some improvements to target for the final milestone.

Feb 2007   GDC in San Francisco.

I learned more at this second conference. I started to feel the strength of the competition in the space, but I also saw alot of techniques I realsed I could apply in Arcadia.

August 2007    Final milestone, which led onto…

Sept 2007     Beta 1

I used friends, family and work mates as the beta testers.

Went Well

* Shipped a working mobile game that ran decently without crashing much on a number of different phones, including Nokia, SE, Motorola and Samsungs.

* Main map view looked good and clearly showed what was happening.

* Fantastic bestiary of animal and monster sprites, from snakes and spiders through to tribal orc bands, huge bears, wyverns and cyclops. Thanks are due to Alex Huang, the talented artist who drew the set.

* People able to use the menu system fairly intuitively.

* MIDI Music worked well. Very nice musical selection created by James Langford.

* Interesting resource system based around farming to produce food that allows your village population to grow and evolve.

Do Differently

* Confusing to play: lots of controls, not enough help, steep learning curve.

* The game involved controlling the whole village’s units; an uneasy mixture of Hero characters and their party members on one side, and peaceful farming civilians on the other.

* Control of multiple units concurrently involved lots of flicking back-and-forth as each unit was moved once per turn in a round-robin fashion. This was hard for people to learn. A tendency made stronger, I feel, by the small size of a mobile screen and limited controls.

* The problem of concurrent unit control was addressed by way of an Automation feature, which allowed units to be put on a program The most useful programs were Escort, which sent a unit following and fighting with a designated leader, and Tend Fields, which keep the food supply ticking over back home. This released the need for constant manual control.

However, most beta players were not able to discover and use Automation feature, because of unintuitive controls.

After Beta 1, it took 13 months of development to reach Beta 2 in October 2008. Having finished our Digital Media Fund program at Beta 1, this period was marked by greater fiscal dicipline. I’ll try to cover the Beta 2 release and the present state-of-play in a future post.

At least five things got us to Beta 1

• Talent: do what you love, do what you’re good at, do it well and with passion.
• Help from others, eg my lovely wife, and the Digital Media Fund.
• Luck, eg meeting Albert at the start of the project
• Hard Work,  lots of evenings and weekends spent on the project.
• Tenacity:  You have to Suffer for your art. When things get tough, just Keep Going. Dont give up.

Playscape Games mobile Strategy-RPG Heroes of Arcadia (renamed from ‘Arcadia’) has been 3 years since inception to date. I suspect some friends and observers of the project, quite understandably, wonder (a) what we’re doing all day, and (b) if we’ll ever finish.

Well, I’m hard at work. Making a quality RPG, on any platform, involves a large investment of effort is a wide variety of areas.

To convey a sense of the many little details that go into it, here’s my notes for a source code commit spanning 9/3/08 – 31/3/08, about 6 man days of development time on my part-time schedule. Apart from demonstrating a lack of discipline when it comes to regular source control checkins ;), I’m quite happy with the steady progress we’re making.

As for a “release date” – sorry, I cannot estimate that accurately enough to speculate on. However, we’re aiming for a second beta release mid year.

Numerous enhancements and fixes (too long between commits):

Hero class removed, replaced by ExtendedUnit
HeroType removed, heroes become regular UnitTypes

Added/updated sprites:
-Animated structures
-Increased size potions, fixed tombstone
-Shadows removed

TargetSelectionRequestor
-Merge with its interface, removal of dead code
-becomes a Runnable task
-Less filtering of invalid target applied by UI
-MapViewInputHandler greatly simplified by moving target selection logic to TargetSelectionRequestor

Remove World methods delegating to Zone

MapView
– Health bar goes yellow/red when wounded
– Fixed bug where sprites faced wrong way in combat
– Made ranged/melee attack code pathways more uniform
– Removed turn counter from the HUD
– Removed “Cannot enter” warning when attempt move to impassable location

IOManager
-Fix: Zone added to read table when deserialized
– Better exception handling when content script init fails

Textwindow
-Globally enforce rule requiring use of task queue to display text boxes

Scripting:
Completed northern wolves quest
Completed forest shrines quest

Removed remaining resource rules code

Party leader passes baton on death, new leader selection dialog

Big speed up in Faction invoke code
-NPC factions serially called from Zone
-Zone runs from Task queue, World dosnt anymore
-NPC Factions not intersecting player faction are only run periodically
-Faction class no longer runnable, PlayerFaction remains runnable, code diverging alot.

Relationships
-Added 5th alliance state Neutral, renamed old neutral to “wary”

Pathing & Movement
-Replaced Breadth-First Search with AStar pathfinder
-AI now prefers not to paths thru other npc factions

boost AI aggression to pevent units fleeing too much

Introduced consistent unit targeting rules and added AI
-Melee: defender nominates defender
-Ranged: attacker nominates defender

New AddTrigger command for scripting

Remove obselete Abort- and Contiue- TargetSelection commands

New LocationFunction types: Location -> int, used in pathfinding

Fixed bug in unit’s cached event history causing party escorts not to
attack

Fixed divide by zero bug when finding minimum threat locations

Quests now award XP on completion

I have just become aware of Game-Developer/Philosopher Adam Russell through his thought provoking article on Situationist Game AI in Game AI Wisdom 4. I look forward to reading more of his work in the near future.

Reductionism vs Constructionism

The article lucidly describes one of the key design tensions in designing any game with story:

• Reductionism: You want your world to be filled with interesting, autonomous agents that the player can interact with
• Constructionism: You want to choreograph these agents in a very controlled, predictable way when required to tell the story

These tendencies are conflicting; for example, through interacting freely with an agent you might move, re-align or even kill it, which compromises the ability to use the agent to advance a pre-conceived story. He gives a nice exposition of how this tension has been handled, or more correctly, “worked around”, in various games to date. Its the best explanation Ive yet read of a tension that has plagued me during the developement of Heroes of Arcadia.

The beginnings of a solution

Adam’s response is to advocate a hybrid approach dubbed Situationist game AI. Unfortunately, this solution is much less precisely articulated than the problem. But he starts down the road with vigour. He draws parallels with debates about local vs global influence in sociology and psycology, and finds middle ways in both fields where an autonomous agent’s environment strongly conditions and shapes its response.

He then sketches how these ideas might be applied in game animation, perception of the environment, and in layering of multiple action-states in an agent’s behavioral representation. Little that’s directly usable at this stage, but I do hope it spurs other more practically-minded developers to make advances in this area.

Check out Audiosurf, an incredible and inspiring indie game and rags-to-riches story that has gone from a IGF finalist at GDC 2008 in February to the number one most downloaded title on Steam in March! Because it’s so good.
Theres a free demo through Steam. An unmissable experience for gamers.

Most turn-based videogames use non-simultaneous turns, where one unit of one player or side moves at any time.

Simltaneous turns are more difficult as they raise lots of concurrency issues, and I will not discuss them here.

See also this blog article at Gone Gaming for a broader discussion of turn-based game mechanics.

Non-simultaneous turns come in at least 3 common flavors:

Units on each side Chunked together

1. Each side moves its units in turn. Individual units make a move/action once per turn. Examples: Advance Wars, Ancient Empires, Many Board Games.
2. Each side moves its units in turn. Individual units can interleave their moves & actions arbitrarily within the turn, changing focus in between. Examples: Civilisation, Arcadia

I see option 1 as really a lite form of 2, where the mechanics are simplified because a unit is never in a partially moved state, and there’s less need for controls to cycle between active units.

Individual Turns

3. Each unit in the game moves separately from others on its own side. Different side’s unit’s actions are interleaved. Examples: Final Fantasy Tactics Advance.

Variable Action Frequency

What Indivudual Turns allows that Chunked approaches cannot, is the ability for fast units to act more frequently than slow ones. Their turn can come up more often. I really like it, its quite realistic and better matches real time games. However, I dont think it scales well to losts of units that need to execute team maneuvres, because their individual turns are scattered through time. Hence, that mechanic tends to be used in Tactical style games featuring a small number of participants. In Strategic games like Civ, where you control many more units concurrently, having all your unit moves happen at once makes it easy to execute large plans of action.

Chunked approaches often struggle to model different unit speeds of in a meaningful way. Typically, units can move different amounts but are allowed one attack, spell or similar action per turn.

Movement Points In Arcadia

In Arcadia, we are trying to meet the challenge of variable action frequency via variable number of Movement Points (MP) allowed to each unit.

• A unit can attack, cast spells or any other action multiple times per turn, if it has any points remaning. (So MPs are really Action Points.)
• Different actions have different costs; movement in the open costs 1; a melee attack costs 3.
• If taking an action would result in a negative MP balance (eg Unit with 1MP attacks, costing 3MP), we allow that! But, we store up the debt and repay it, in effect an overdraft, so that next turn the unit will be able to do less.

Where this approach is really nice is its robustness to MP-reducing effects. For example, imagine a Slow spell reduces a unit’s MP-per-turn to 1. Without an overdraft, it would struggle to every attack, as attacks cost 3. With an overdraft however, it can attack ever 3rd turn, which feels just.

I have loved the Civilization game series since 91 – Civ 1 was my first truly engrossing game experience, at age 18.

I consider the latest incarnation, Civ 4, to be my all time favorite game.

Unlike so many sequels, its emphatically superior to all predecessors and derivatives (AFAIK), and displays a profound level of richness and subtle game balance. And its still really fun to play.

Follow Through

To me, Civ creator Sid Meier and SimCity creator Will Wright are both game design geniuses. But the diverging trajectories of SimCity and Civ 4 reveal some interesting differences in their characters and illustrates the great value of Follow Through (something I myself often struggle with, but are determined to improve through my life’s learning):

Sid Meier & Civ

Sid Meier created Civ in 90/91 and has continued to guide it’s evolution since. Each Civ title has been a refinement of the last without varying the fundamental formula.

• Civ 2 managed simultaneously to be massively and incrementally better than 1.
• Civ 3 introduced key new concepts Resources and Culture, both very cool, but strayed a bit with rampant Corruption.
• Civ 4 threw away stinky features like Corruption, Civil Disobedience and alot of Micro-management, and added Religion, Promotions, Great People and better Displomacy.

Its amazing how the Firaxis design team stuck at the problem and got the decicions right time after time, so that the resultant game has kept evolving better and better. Sid could have gone and done something completely different (he did: Railroad Tycoon, Colonisation, Gettysburg), but he kept on following through with his original creation. A+

Will Wright & SimCity

Will Wright, on the other hand, didn’t follow through as well.

• Simcity was brilliant, but a beginning, like Civ 1 was.
• Simcity 2 was great, the highpoint of the series.
• SimCity 3 felt like Simcity 2.1, technically better and yet a dissapointment.
• And then, along came SimCity 4,which is a nice looking, but really awfullly designed game that is not much fun to play! Not much Will Wright left in it by then.

Wright was no doubt already consumed by Spore. Im sure Spore is going to be brilliant; it’s recently infected Civ4’s lead designer Soren Johnson, who moved over to EA from Firaxis. But surely Will could have spared a day a month to keep an eye over his firstborn, have ensured whoever led design on SimCity 4 was competent, and have taken corrective action when it became apparent the game was on the wrong track. Now it seems like there wont be anymore SimCity, which is sad because the idea of a City Simulator is still as awesome as ever. More Follow Through needed. A-

Documentation for players of videogames (ie Manuals, Help) is often bad, full of holes or missing entirely. There is a practice in games of leaving gameplay, interfaces and product features undocumented that would not be acceptable in other types of software. For simple, intutive games, little documentation should be needed. However many games are complex, and include lengthy paper manuals, and yet the content  is still patchy.

Patchy Documentation Examples

Elder Scrolls Morrowind

Its documentation does not include

• What your magic spells actually do
• What a permanently visible on-screen dial is actually showing: the red dial beneath the active spell icon in bottom left corner. I spent an hour Googling for it, didn’t find out, but it appears everybody else is wondering too.

SimCity 4

My retail copy  of this 4th generation city simulator came with only an install manual.

The online site has a “Tips-n-Tricks” section, but no coherent high-level description of how the interface and game works.

Black and White 2

A complex Simulation/Real-Time-Strategy game that comes with a 40 page but very scrappy manual. The player grows towns and their people, but manual does not mention anywhere how to create new people, nor is there any visible control/button to do it. You need to drag one “breeder” person onto another breeder, and then they procreate and multiply. (Obvious, huh! …only it wasn’t for me.)

Civilisation IV

As similarly complex as a 3D tool like Maya or a programming tool like Visual Studio. To its credit, it comes with a 205 page manual and an in-game “Civilopedia” help. Evident effort was spent on documenting the game. Despite this, there are holes in the documentation where key aspects of the game design go unmentioned:

• Civ allows you to group units together and issue orders to the group. First mention of this is in a keyboard shortcut reference chart in the appendix! No description of how Groups “work” is given.
• Coverage of multiplayer internet play is half a page. You just have to try it out and then try to figure out what is going on. This is despite the fact that Civ has some really innovative and unusual multiplayer features that a player is very unlikely to understand at first.

Let players write the docs?

A result of this trend is that fansites that spring up around successful titles to compensate for their own lack of decent documentation, such as WowWiki,  CivFanatics, Simtropolis or TES3 . The quality of the material on many of these sites is amazing.

Is it a cop out to let your player’s do your work for you? Well, to a large extent this player-driven documentation proves that players desire such material. Possibly frustration at not understanding the game has motivated them to contribute. And what tends to happen in many decentralized fansites is that fact, opinion, misinformation and spam get smeared together in an unstructured ball, and it takes a great deal of effort to dig out the answers to questions of interest.

Good Docs => Informed Choice => More Fun

Like most players, I really enjoy figuring out how to play a game well. I greatly value what Damion Schubert dubs Tactical Transparency, the principle of knowing what your options are and making an informed choice. When I simply do not understand how a game or interface works, and dont have a ready place to get help, both Tactical Transparency and my enjoyment are reduced.

What Good Docs look like

Its a stark contrast to the level of documentation I enjoy as a software developer, as a momentary look at the docs for Hibernate or Spring reveals. They describe both the Concept/Model behind the software, and the details that you need to know once you get the concept. I guess we programmers value knowing precisely how a piece of software works.

Time to eat my own Dog Food. Yum!

At Playscape Games, we shall ensure that Heroes of Arcadia’s game mechanics and interface are well documented online soon after the product becomes available.

I played Lord of the Rings Online (LotRO) for about 2 weeks during the free beta. Overall, I enjoyed the experience, yet felt the game fell short of what I hoped for. Over 2 years after Blizzard’s World Of Warcraft, Turbine have created something that feels like a clone of WoW’s world, interface and gameplay, yet with less humour and slightly clunky in the little details.

The Good

• LotRO copies WoW’s user-interface, and that works well. Consistency amongst game interfaces is a good thing; there’s no need to re-invent the wheel. I found it very easy to immediately learn the basics of the game.
• The environment of the Shire is quite cute, humourous and fun. Beforehand, I thought it would be a boring region of the game, but Ive enjoyed it.
• I tried a night’s Monster Play (playing a monster on Sauron’s side) as an Uruk in the Ettinmoors and greatly enjoyed the stylistic and gameplay contrast.

The Bad

• I have a level 10 Elf Loremaster. Loremasters abilities seem to suck. They have the armour and weapon restrictions of a magic-using class, but at level 10 Ive only got 2 main combat-orineted spells, Burning Embers & Summon Raven, which I use endlessly. The other abilities are limp Buffs and Heals that have little attraction.
• My first hour of gameplay was poor: despite choosing to play an Elf Loremaster I was corralled into Thorins courtyard running errands for Dwarves.

• • When entering my first monster battle in minute 3, entirely scripted and unavoidable, I pointed-and-clicked my default weapon at the appointed opponent, and we stood there hacking until I died. Dying in your learn-to-fight battle is not fun. If you’re going to have a compulsory first learning battle, make winning near certain. (Likely explanation: difficulty was not calibrated for Loremaster weaklings)
  • After first battle, all characters and places I’d learned about were scrapped, and I got teleported to big empty elf ruins, containing one elf, who told me to leave ruins and go somewhere else. There seems little point of having the ruins if you do nothing but leave them…
• I find the terrain ridiculously and needlessly hilly: in Ered Luin I repeatedly fell off cliffs and had to take boring back-tracks to return to the road. The creases in the terrain force you to take tiresome round-about loops when trekking cross-country.
• What dissapopnted me the most was the heavily constrained geography of Eriador. Unlike Frodo and co, in LOTR Online, it’s not possible to slip out of Hobbiton at dusk to go tramping over hill and dale. The Shire is a walled zone with 2 exits, one to Bree and one to Ered Luin (in the Beta, at least). All other directions are impassable. Just think how it would have made those Black Riders jobs easier; post a wraith at both exits and wait! Turbine’s middle earth has a graph topology: nodes connected by edges. Its a sharp contrast with the solid earth, open skies and freedom of 4 compass points that one feels while reading the travel sequences of the novel.