My two cents for two years…

Wow, I missed my two year anniversary here at killtenrats on February 27th. Has it been that long already?

2007 was a bit of an iffy year for me, pretty chaotic with a few surprises (good and bad). 2008 looks “interesting” thus far. Hard to believe we are getting close to the end of the first quarter already…is that “new car smell” of the new year already beginning to wear off?

I expect I’ll be continuing to point my finger at the game industry and yell “noob!” frequently this year. There is just so much going on (or not going on, depending on your perspective). Yes, sometimes I state the obvious, but hey, that’s what I do. I don’t always make people happy with my commentary (even my gerbils have gotten death threats), but my response to that is “get your heads out of your asses and make better games so I don’t have anything to rant about”.

So…will the IPhone SDK and the $100M venture fund for IPhone applications have far reaching impact on a) IPhone sales and b) the games industry (specifically casual mobile games)?

Will Kotick announce a massively overfunded FPS shoehorned into MMORPG architecture? Will it get equally overhyped and fail to deliver like the “industry revolutionizing” Tabula Rasa?

Will the sneaky guys over at Interplay ever get their $75M Fallout Online project launched (I bet Duke Nuke’em Forever comes first, or at least a crazy “how fast can we blow $75M without making anything” party)?

Why has it taken 38 Studios so damned long to license a client/server engine for their MMORPG? Wasn’t the company founded in 2006? What the hell have they been doing? I’m sorry, but shouldn’t you be figuring out the most basic of technology decisions right off the bat before you start hiring your developer and production staff to make whatever it is they are making? Couldn’t they be more original than the Unreal/BigWorld approach?

Will anything worth the time and effort happen with the NASA MMO effort?

Will artificial life and artificial intelligence guys quit screwing around and do something mind-blowing for a change?

Will alternate reality games and augmented reality technologies finally converge in 2008, or am I thinking too far ahead? Let me know if I get too forward looking for you guys. I’m still pissed about the lack of flying cars. Never mind a real “next-generation” mmorpg or virtual world. Why is it that 2D flash based social “mmo’s” and second life (in all of its craptastic glory) are so popular? Did someone turn on the “let’s devolve” machine?

Speaking of which, is Spore losing the excitement and anticipation that it had when it was first revealed? Has it been washed down and stupified for the general public and launch on mobile devices? Will the final launch be cause for celebration or a huge collective “eh”?

Who will be the next big online games failure? Plenty of venture capital has been thrown around (in all the wrong places I still insist), and it is just a matter of time before another heartbreaking disappointment occurs (I’m looking at you Stargate…the recent preview trailers were…weak).

The only thing that lets me sleep at night is the assurance that there are other people out there like me (and like many of you regular KTR readers) that still dream about a wonderful, engaging, immersive, interactive, and social future. A place where we both create incredible worlds and share them with each other. Where new stories and epics come to life, where that hot elven chick is really a hot chick…

and of course, flying cars. Where is my flying car dammit! Maybe I’ll settle for a jetpack, or VR glasses that tap directly into the optic nerve. Yeah, that would be cool.

11 thoughts on “My two cents for two years…”

  1. Popular Mechanics, and just about everyone else, has been promising us flying cars for generations.


    Imagine drunk driving accidents, but times the damage caused by 500.

  2. Regarding 38 Studios… I don’t find it unusual that they’d wait so long.

    First, you need to hire the people experienced with the tech to MAKE such a decision. Next, you need to develop your core design and make a NEEDS ANALYSIS so you get the tech that will support your concept (and don’t overinvest in tech that’s absolutely not needed, too).

    That core design isn’t a simple term paper. You don’t necessarily need the whole design document fully fleshed out, but you need a HUGE chunk of it in place. Once you do, you need to identify the risk areas- parts of your design that really are new/novel/untested in ANY engine. You need to pick your candidate platforms and do a rough prototype “proof of concept” to be sure that the tech is possible.

    While that is going on, you want others developing anything that’s not platform-specific. Once you agree on a platform, you want to hit the ground running- every day lost makes your tech 1 day older. You want the concept art done. You want reference models done. You want art assets that can be ported to any of the candidate systems done. You want as much story written as possible. You want development tools in place, and you want every developer working on the project as familiar with the environment and workplace as humanly possible.

    …Because from here on out, it’s about producing as much quality material as quickly as possible so that the platform you’ve chosen isn’t dated by the time you enter the market.

  3. I agree and disagree Chas. You are right in that a lot of groundwork needs to be established sooner rather than later, but the technology platform is one of those key pieces. Much of the design, the production pipeline, and technology decisions can’t be made without first determining what tools you will be working with, as well as the limitations and features of the client/server side.

    You can’t waste resources on art assets or creating development tools before making this choice either…otherwise you end up trashing everything and starting over. Sure, many of the client/server technologies out there are similar in a lot of respects, but there are also a lot of differences.

    I would also disagree with your point about doing a rough prototype of tech on different platforms. That is a huge waste of time and resources…there is a learning curve to get proficient with the tools and tech of any of the MMO engines (complicated when you are also integrating a client side tech and a server side tech that are not part of an integrated platform).

    All I’m saying is, the choice of client/server stuff should be one of the early decisions. Heck, publishers and funding sources (like VCs) usually ask you what platform you are going to be using as one of the first questions. If you can’t answer that, it makes them wonder if you know what you are doing. Unless of course, you are a famous sports star and they are all more interested in your autograph than your business acumen in the game industry.

  4. 1) WHO must make the decisions on the tools? The people you have to hire. Thus, the tech engine must come into play AFTER you hire.

    2) WHAT determines the needs of the game? THAT will determine which engine is right for you. That is identified in the design document, written by the people you hire. Your concept art and reference models will similarly impact which platform you choose.

    You have two choices: pick a platform and design to the limitations of the platform or design what you want and pick a platform that meets those needs. If your goal is to “make a game” then by all means, choose an engine before you know what you want to make. If you have a distinctive art style in mind or gameplay mechanic or innovation you’d like to see implemented, then it’s absolutely reckless to choose a platform without defining your vision and determining if the platform will meet those needs.

    Platform prototyping is CRITICAL to choosing a commercial platform. DO NOT TRUST THE MIDDLEWARE VENDOR that the product supports a design element. If it’s tried and true feature, you only need to look at other products using the engine to verify it. If it’s something unique, TEST IT or expect to drop it when it won’t work the way you want to.

    Make something very rough and very ugly, but demonstrates the core questionable tech or you’ll likely find that the core tech doesn’t work MUCH later in the process, requiring SUBSTANTIALLY more cost than if you did the upfront test.

    Art Assets are easier than you may think to pre-produce. Most candidate engines will have easily-listed definitions on what authoring tools they support and what finishing techniques work best on the system. An art department can begin developing quite a bit of raw art- getting the foundation in place, but not optimizing any asset for any one engine.

    Same goes with story- a great deal of story development is getting the right production tool (not necessarily the ENGINE one, but one that can take text from storyboarding to a data format to be used by an engine) and authoring the content. In an MMO, just doing the basic world overview is a HUGE preparation task that can be started long before the engine is in hand.

    Is this really abnormally long? Look at other games:

    Bioware Austin was announced March 13, 2006. They already had several core developers in place when that announcement was made and even had the MMO concept well under way before they decided to open the Austin studio. The engine was selected April 2, 2007. Over a year later… and that’s with a large chunk of development history in the company.

    Green Monster Games has no history. They didn’t even lease floorspace to begin planning until 2006. At that time, they were a business core setting up the infrastructure. Their “first round of hires” was announced in a November 14, 2006 press release. The announcement for the bigworld license was February 26, 2008. Three months longer than Bioware’s… for a company’s first game.

  5. Hrm, I think you are missing my point. Yes, you need to have someone evaluating the technology and platforms, that goes without saying. What I’m trying to point out, is that you don’t need 20 people or whatever working on a game for a year and a half or two years before getting to that point. Established companies might allocate 2-5 people to get the basic stuff in place to get the core plan and designs, but not more than that. The same goes for a new studio with no background…you simply do not waste resources on a large staff without making those key technology decisions early on.

    By the way, story has absolutely nothing to do with the platform decision. Zero. The same goes for pre-production concept art and reference models. Do I really need to break down what exactly the different components of a MMO engine do?

    “You have two choices: pick a platform and design to the limitations of the platform or design what you want and pick a platform that meets those needs. If your goal is to “make a game” then by all means, choose an engine before you know what you want to make.”

    Of course you should know what you want to create before you choose an engine, that goes without saying. My point, again, is that you shouldn’t launch into development and production without having made that choice. You will end up having to rework, redo, reengineer, and recreate.

    The third choice is: Determine the game’s requirements and specifications in accordance with the basic premise and design treatment, find a technology platform that meets your needs (cost, functionality, features, tools, pipeline, scalability, extensibility, modularity, etc.), and then start staffing up. You shouldn’t be developing add-on proprietary technology/mechanics or starting production without making decisions. Creating a MMO is similar to any other software development and engineering process. You simply don’t start until you know what you are making and you know what your specifications are.

    Then again, maybe this is why MMO budgets keep skyrocketing. If you just throw people into a room and tell them to start making a game, while putting off the critical technology/platform/tools decisions for later, you are just asking for a mess.

    I’m not attacking 38 studios here or trying to single them out, but it isn’t a good sign for them to be making this level of decision only now.

    As far as your example…there are plenty of others where the platform decision was made very early on. Red5, Cheyenne, Hero, the list goes on. To me, this isn’t very different from making other decisions like which database to use, naming conventions, source control, bug tracking, language, etc. There are a lot of things that need to be decided before you ramp up. Otherwise, you are just wasting time and resources.

  6. Don’t mean to be pushing the issue, I just don’t see how the timeframe you’re mentioning is at all unreasonable.

    In the traditional sense, you might do what you’re talking about: Gather a core team, create a vision, select the tech, hire the bulk developers, start building, test, etc.

    In agile product development… you look for the unnecessary bottleneck. In the previous example, “select the tech” is a bottleneck, since you don’t start the bulk of the production team until that’s done. That’s a shame because the production team can always use more time.

    To improve the process, you can either move the “select the tech” earlier, or look at whether it is a NECESSARY bottleneck. In this case, it isn’t.

    As you stated, story and concept art have nothing to do with the engine. Even moreso, many engines identify the data formats they support, so building within the common data formats of candidate systems is also very do-able without locking in to a specific technology.

    Production line analysis would thus show that you can begin these core elements long before the engine is selected. This form of agile development would allow your team to get a head start on these core elements that aren’t engine dependent while allowing you the time to test any “risk areas” of engine compatibility with core designs.

    As for your examples:
    Red 5 Studios was founded in mid 2005. Their first press release is September 2005 but they were registering the domain names back in June and had assembled their core team at least at that time. Their engine announcement was 2/14/2006. Under a year response either way, granted, but we’re only talking 1 year 3 months for GMG too.

    Cheyenne’s first site press release refers to hires on July 1, 2005. Stargate was announced Jan 25, 2006. While Bigworld was selected just a few months later (Mar 1 2006), the unreal engine took half a year (Aug 28, 2006). That’s from TITLE announcement to selection of an MMO. Given that they’d planned to produce an MMO from company launch, they should have been researching engines for quite some time earlier. If we use the same metric as GMG and go from the first company press release about hiring (July 1 2005), they’re running about on par.

    Hero… sorry, I kept getting Hero engine and Hero’s Journey and Hero Online mixed up in the google searches. My search-fu is weak.

  7. Choosing an engine (and immediately trying to make a game) before knowing what you’re going to make with it is one of the mistakes that has lead many companies to fail. It’s better to find an engine that suits your needs than to make a game that suits an engine choice.

    We’ve done more planning than most companies ever get a chance to, which has allowed us to make an extremely informed decision about engines. That’s why we’ve picked up BigWorld (the best server technology) and Unreal (the best tools/client technology).

    Whether that’s original or not is completely unimportant. It’s “unoriginal” because they are currently the best solutions, and choosing something else for the sake of being different is simply unintelligent.

    Note also that what you do with an engine is what really matters, so there’s no reason to talk about how “all games on Unreal 3 engine look the same!” First, they don’t. Second, they don’t have to, which I can be 100% confident in saying because our game currently looks unlike anything I’ve seen on the Unreal engine (or anywhere else, for that matter).

    Now that we know what game we WANT to make, we can focus on determining what game we CAN make (and what we have to change about the engines to make it). Hence the engine choice and moving to prototype. After we’ve determined what game we CAN make, we will actually make it.

    Anyway, the important piece is this: we never ran into a wall during the concept phase because we didn’t have an engine officially selected. If we had run into a wall, then it would have been a mistake to select an engine almost a year after the company formed. Since we didn’t, it leads me to believe that the decision was, at the very least, not a bad one.

    Our concept phase was longer than most companies because of our focus on quality. You can also see long concept phases at other companies that produce high quality games (Blizzard, Valve, etc.). It’s better to plan well than to dive headlong into something without knowing where we were going.

    The longer concept phase lead to an obviously longer period of time before we finalized our engine selection, which was wonderful because we could be extremely picky about our engine to make sure our decision was right.

  8. Eh, I never said anything about picking an engine and then jumping right into development. I am also in favor of long conceptual and pre-production phases (I think it saves time on the backend). What I do have a problem with (and as I said, I am not targeting 38studios specifically here) is jumping into early production and development and then making the engine decision. It should be done sooner rather than later.

    Ryan, I assume you are implying you are with 38 studios?

    It is arguable whether or not BigWorld has the best networking and Unreal has the best tools/client tech. Personally I agree about BigWorld’s networking technology, but Hero has the best tools, and the best client graphics tech is probably Offset. (No, I haven’t seen the new Unreal tech yet). The BigWorld/Unreal combo is a good one, and this is proven by the sheer number of companies going this route, but that doesn’t make it the best.

    Long concept phase does not equal quality.

    “Now that we know what game we WANT to make, we can focus on determining what game we CAN make (and what we have to change about the engines to make it). Hence the engine choice and moving to prototype. After we’ve determined what game we CAN make, we will actually make it.”

    What? Thats like saying “now that I know what I WANT to eat, I can focus on determining what I CAN eat. After I determine what I CAN eat, I will actually eat it.” Goes without saying, doesn’t it? Then again, maybe I’m being a little critical here.

    “It’s better to plan well than to dive headlong into something without knowing where we were going.”

    You mean like spending time building a game before knowing your tech specifications?

    All that aside, I’m irritated about the artificially inflated costs for MMORPG development, and the incredible amount of waste or less than efficient practices. I don’t care if a team spends five years in conceptual development before doing much else, but you should get your technology sorted out before you jump into production and development. Its fine if you disagree, but that’s what I think.

  9. Yep, we’ll just have to disagree. I’ve experienced both ways, and the better way seems to be the direction 38 is taking. That said, I’ve only worked on a few games myself, which is why the decisions are in the hands of people who have worked on far more games than I have.

  10. Darn straight. And, there are different paths for just about anything. Even when designing zones, I take a different approach than everyone else and we end up with the same quality. That expands out to entire games, which you can take different approaches to building and still end up with gold.

Comments are closed.