Portal Online

The tl;dr of what I want in a new Portal game can best be summed up by reading the lines in bold. If you’re bored, have a read!

Today HTC announced they’d partnered with Valve for the Vive headset (part of HTC’s “re” product line, so, re:Vive). Obviously the thoughts quickly turn to software for it. With Valve being the partner, Portal came up in a post I was reading. (Imagine sailing through the air in first person.) It always felt like Portal 2’s online component didn’t take off like Valve wanted, so, let me pitch a Portal 3, or at least a Portal Online. The chief decision in my choices is that it should be a game with continued updates, like DOTA2, CSGO, and TF2. And even better would be allowing it to rely on community support for content, but still allow financial opportunities for the developer that seem fair to players.

The important part? Each server is a testing facility. The person who runs the server (herein owners) should get a large series of options to better fine tune the experience for players. And this will present financial opportunities for Valve AND for the community members that contribute to the game. Also important is offering several non-required elements for players to improve their experience. Let’s look at those items first.

1. Maps – Highly rated maps should be vetted for inclusion on the marketplace if the creator wants. A very small minimum price should be required. ($0.25?) This pays for the evaluation of maps, as well as hosting, and should benefit the map maker. However, server owners should be able to run maps from outside of the marketplace, just without a logo of approval in server lists.

2. Player components – Players need to be differentiated. Playing as the people unfrozen at the end of Portal 2, suits and colors are a solid start, but offering new purchases (maybe play as robots, especially if done in components) could do well. Components, suits, colors, sounds, and mini-actions are never integral, but can be used as random drops as well as purchasables. Much like TF2, each server can do its own, or abide by the official list.

3. Personality cores – And here’s where it gets a little more interesting.

Server owners will choose a “GLaDOS”-like AI to run their testing facility. You can start by giving players a few basic free options like some of the personality cores from Portal 1 & 2. Imagine bounding across giant gaps with Rick, the Adventure Sphere from Portal 2, urging you on. To triumph in the face of danger! Or imagine a dark room with flickering lights, with the hiss of the evil red sphere from Portal 1. Then it snarls violently as a piston came from the ceiling and tried to smash you. (Or RuXx Emma Thompsonor even actress Emma Thompson who narrated Will Ferrell’s life in Stranger Than Fiction.) Even allow players to fashion their own personality cores and offer them for free download, the better ones vetted and placed in the marketplace.

Each Personality core recording (that costs money in the marketplace) should have story hooks built in for story arcs that rise and fall that trigger X maps into the server. The important part is different script portions that can be used by map makers and communicate the same information in a way specific to that personality core. The types of script needed to be used in the procedural story would be “introduction”, many “normal chamber” lines, appropriate “going away” scripts, and “re-introduction” recordings as well. Why? Each server starts with the introduction. Then the owner’s chosen chambers progress normally. But with the first Valve Update, GlaDOS takes control of the owner’s test chamber from the owner’s personality core, runs a few levels (introducing updates), then the owner’s AI regains control. Maybe GLaDOS patches in via network. Maybe it’s a copy. Maybe the two AIs are battling; maybe they’re working together. That’s not important here.

It’s only really important that personality cores in the marketplace have the appropriate generic audio recordings that maps will assume and reference. The lion’s share will be the large variety of “normal chamber” recordings, but Valve Updates would include maps that would use rare plot advancement hooks that would be required of all personality cores on the marketplace. Those would be the “going away” and “re-introduction” scripts. During the update we may get a new paint color (a la Portal 2) or world item (enemy units, new world mechanics, etc.). Maybe you could even let users trade off cores with the right audio hooks. Maybe some servers wouldn’t care and just want tons of levels.

Another important factor is that while the server owner decides the order of the chambers, this is not TF2. There is no automatic progress. Each chamber needs to be instanced when a player arrives in it. If a chamber already exists, a player should be joined with whoever is currently running that chamber on the server. In the elevator at the beginning of a chamber) players should be able to go the previous puzzle, or restart the current one. “Next chamber” should be possible in exit elevators, as well as beginning elevators if you’ve already completed the current chamber. Servers should track player progress, like they currently do for non-standard TF2 items, so you can easily get back to the chamber you belong on. But people should be able to backtrack and help, if they want. If you get to chamber 2, and I join the server for the first time, I start in chamber 1. Server voice and text chat could be passed off as communicating on a network.

So, players can buy suits, components, colors, sounds, and actions that spread across servers. Server owners can buy AI personality cores to personalize their server, and maps to use in their servers. And occasionally Valve sends updates that temporarily take over servers and seamlessly advance the world.

That’s the Portal Online I’d be happy with. Also, you get two O’s in the logo. POrtal Online. One blue and one orange!

Armchair Quarterback

Comments (0)


I love FTL

I’m unemployed. I refuse to spend money on games while this is the case. Hearing that, a friend stepped up and bought me a game he was loving, FTL. I can’t thank him enough, because it’s awesome.

Does FTL make you a space-faring bringer of death, or a kamikaze pilot flying a rickety death trap? Yes. Is it worth playing? Absolutely. Dying hasn’t been this fun since Dwarf Fortress. Killing hasn’t been this surgical since Fallout 3’s VATS system, this easy since you first picked up the BFG, or this difficult since you spec’d completely opposite the way you should have for a RPG boss fight. Each play has the potential to serve up a wildly different game due to the game’s elements being so randomly generated, that each play will bring many stories of harrowing success and escape, telling friends how you barely did this, and almost got killed while doing that, and each story will almost certainly punctuated by the full stop of your death. It’s billed as a “a spaceship simulation real-time rogue-like”, but what does that mean? Let’s start with the basics.

The Salad (Before the game.)
In Subset Games’ FTL (yes, it means Faster Than Light,) you fight for the Federation against rebel forces and other nefarious tropes amassing weapons and crew for the big showdown. This is done by plotting your own course through eight sectors full of nodes as a wave of enemies chase you. Each node providing an event. Events range from battle, a social opportunity, an environmental threat, a store, to empty space. Sometimes you’ll even get combination of those things, or they’re made more intricate as the game reacts to your crew, your ship upgrades, or some of the (simple) quest lines you’ve previously opened. As if those combinations didn’t offer enough replay, the nodes of each sector, and the available paths between sectors, are generated freshly with each play through. Each event allows a chance to gain resources along the way.

The Potatoes (The good.)
Ship management shines. It’s what every RPG wishes its management was, simple and robust. It’s so intuitive you can become good at it by accident. You start with one ship available, but more begin to open quickly. You acquire a crew of up to eight from the seven different races, each with special abilities. You choose from six different weapon types, each with strengths/strategies. There are also five different automated drone types you can deploy, not to mention the option to board your enemies’ ships. And on top of it all as you progress you can equip your ship with up to three of twenty-one augmentations that vary wildly in effect.

Upgrading your ship in FTL is micromanagement heaven as you spend scrap (the in-game currency) upgrade your ship. All ships have a reactor. As you spend money to upgrade it, more energy is produced. It’s from this common pool that most of your ship’s systems are powered. Each ship has several systems (weapons, shields, medical bay, life support, etc.) which can be upgraded, and extras that can be purchased. As you upgrade a system, it opens a slot in which you can route an extra cell of energy. It gets frantic when you realize that energy is hot-swappable, meaning you can power and de-power items as needed. So while you’ll likely always want your life support system working, if you don’t plant to run from a fight, you can take energy from your engine and fight your battle with stronger weapons, shields, or what have you!

The Steak (The great.)
Battle is superb for a player like me. FTL offers the perfect amount of control. In most games either the developer has very stupid AI and is too easy, the game cheats to make things interesting, or the player simply can’t match the AI in accuracy/quickness and things are too hard. This is a known problem in nearly any game from Madden to Tetris. This is not the case here. Enemies operate at maximum efficiency (minus a little dumb AI in asphyxiation created by simple logic, not cheating or intentional dumbing down,) because FTL lets the player pause the game at any time with a tap of the space bar.

This makes every meaningful choice available at any time, and you can change strategy instantly on a whim. Pausing time, issuing commands, changing attack patterns, and resuming the battle in a single moment allows you to match the AI move for move. Earlier I mentioned VATS, Fallout 3’s method of giving players a refined control in battle, but it was limited in making the player wait until your endurance was recharged. Captaining your own ship in FTL, giving out orders is expected to come instantly, and pausing time works perfectly to that effect.

Just Desserts (The self-indulgent run-through.)
I load up on offensive slots, going light on reactor energy, and rerouting power from my meager engine and medical bay and give that extra bump to weaponry. Toss a few Zoltan crew members (who add energy to the ship systems they’re assigned to) and you’re boxing above your weight class. My starting ship is the Red-Tail (Kestral B), which comes with one Zoltan (who add one energy to your reactor), one Mantis (excellent fighters), and two humans. It also starts with four single shot lasers.

The first thing I do is invest in blast doors to asphyxiate boarding enemies, expand my crew with as many Zoltans as possible, and invest heavily in laser weapons. Soon I’ve got an opening volley of lasers that rips through enemy shields and begins taking apart a system of my choice. Worst case scenario, I use the lasers to tear at an enemy’s shield and then use a beam weapon to do my damage in a line drawn across the enemy ship with my mouse. Either way, I have a good chance of jetting straight to the final sector in no time as long as things go decently well for me, but I only really get a chance of beating the final boss if I’m lucky and a certain few augments and the right weaponry come my way. To ensure that, I take my time, scouring as many nodes as I can in each sector. I’m a man on a mission, people!

But before I leave the second sector, I die. That happens often. You’ll have bad luck, you’ll not get awesome items by the time you expect to. You’ll need to spend your scrap repairing your ship, instead of buying upgrades. You will die very often in this game. Here, let me go back and underline that for silly comedic effect. And you’ll quit the game, and you’ll start something else, and you’ll instantly want to start it back up and play again. I actually wrote all of this in a relatively short time, I just kept starting the damn game back up and playing again… In fact, I think I’ll go play it now!


Comments (0)


Armchair Quarterback: OnLive

Is armchair quarterbacking useful? I don’t know. Probably not. But it’s fun. So I wrote this long expansive article that I’m not going to bother publishing here because it’s cumbersome. Let me hit the bullet points.

OnLive’s business plan emphasized streaming, which was an open-ended cost, but did not actively make money after a game’s purchase. Their CEO Steve Perlman? He was human, and seems to have made some really bad decisions, like kicking EA’s games off the service before launch, and others. Virtualization was not uniform, meaning some games actually requires one real GPU for each instance generated by a player. Finally, latency was a threefold problem: The perception that the Internet was too slow for this, the reality of broadband penetration, and then monitor/kbm/gamepad latency, which people generally don’t acknowledge, and assume it’s your service. That part really sucks. But the key in latency is that it’s not as bad as you think. It’s actually playable.

I’m not even going to factor in set-top boxes. I’m focusing on PC here, so let me state that now. Retailers generally make about $12.00 per game, but the average gamer, if they did all their gaming on OnLive, would cost about $33.80 annually for streaming. ($0.10 for two hours of HD footage for Netflix as a baseline. Then figure 13 hours a week gaming.) So you’d need to sell three games per year at $60.00, and then you’d only make $2.20 per year per gamer. What does this mean? Streaming will not be a top priority out of the gates.

Controllers can be found for a buck each on Alibaba.com, but we want a good one, so, let’s throw $8.00 at each one. (They cost Microsoft $11.00 in 2006, and I assume the price has gone down. But we don’t want shit controllers, and we want our own brand, so let’s say $8.00. If you can find a hardware manufacturer that wants in with you, then you can save money and feature their brand.

Then you have to worry about salaries, hardware, rent, and other costs of business. I have no idea what that’s going to cost, but, it’ll be a lot.

The Armchair
I like the idea of OnLive. Very much. No more upgrading your PC, no more buying console after console, any device, just games. Let me repeat that. Just games. But we have to build to that. I think the (a?) problem is that they pushed themselves as a console competitor first, with a set top box, when they should have framed themselves as a Steam competitor and expanded from there. GOG launched to much fanfare when they came out, and they had an angle, “Good, old games, DRM free!” Simple and great!

No streaming. Not at first. First I launch as a store with easily navigable social features (Steam hasn’t done that yet) and integration to FB, Twitter, G+, whatever. The client should offer screenshots, an integrated XMPP IM client (Jabber, Google Talk/gchat), and the ability to record videos. Try to be lightweight as much as possible. But the key is to be the Apple to Valve’s Linux. Be stylish.

My OnLive’s sales angle? A free controller with your first $60.00 purchase. Given the $12.00 profit from the purchase, and the $8.00 controller cost, that gives us $4.00 for shipping (which is a little over $1.00, but we’ll call it $2.00). So, on our first sale to each address/credit card/email we make roughly $2.00.

With each controller, pack in a coupon for a free indie game that makes good use of the controller, and has DLC available now, and more coming soon in the pipeline. For each copy you give away, pay the developer $3.00. They make less on each copy, but likely get a lot of interested parties to try to sell DLC to. It’s probably worth it for them. (And it’s definitely cheaper than my other idea for a controller pack-in, a $5.00 coupon on your next $60.00 purchase. A free game with DLC has the opportunity to net you extra cash that way.) So with our second purchase we make $9.00, plus a cut of DLC.

So now gamers have two games in your client, and a spiffy controller. At least one game has DLC on the regular. How do you keep that going?

Cloud gaming, and Mac/Linux/iOS/Android clients. For $10.00 a month, you offer a cloud streaming of any game you own through the client for any platform. Click purchase and play. For $15.00 you get a digital locker too. For $25.00, add a cloud server for any game you want. It doesn’t even have to be one you own with the service. We install it for you.

Maybe, a year into it, we move to the TV with a set top box. Six months maybe, but that’s only if we’re a success. Until then, open a community to help people best make their own TV media servers.

And if you ran OnLive?

Armchair Quarterback
Game Industry
Gaming's future

Comments (0)