The tech arms race in AAA games and why I'm abandoning it

Game Engine Architecture(non affiliated)

http://www.amazon.com/Game-Engine-Architecture-Jason-Gregory...

Telling the AI people that their efforts are wasted on an arms race with little effect on the game play is a much weaker argument than 'hey, I need to revisit what _I_ am working on'.

Look, I get it, it's cool to see hardware get pushed beyond what people previously thought was possible, it demonstrates ingenuity and can produce a captivating spectacle. But it doesn't make the games better (beyond the wishy washy 'immersion' stuff).

To be honest, this sort of 'pushing the hardware' focus makes me look forward to the end of Moore's Law. When games can no longer meaningfully compete just by looking prettier, when there are no new console generations, I hope there'll be a renaissance of games as a form of expression and interesting experiences.

It's not that. There is a lot of optimization going in there and often the reference target is actually the common denominator of mainstream game consoles (lower than what each of them individually-taken have). Only on top of that there are optionally-added (and activated by-default, out of marketing reasons) layers that attempt to use more resources and which gives you your stated impression. And yes, it's true what you say about "the end of a platform generation" because the optimization efforts (including platform-specific ones) are stacked trough time.

I find something quite hilarious about the idea of a MinMaxing Avatar of Virtue. Minmaxing often breaks the mechanics of a game, but it's rare to see it break the morality of a game the way it breaks Ultima IV.

Brilliant game.

Games are an interesting storytelling medium, but stories are much more than variations on the order in which plot points are hit. The biggest issue is motivation - why characters (maybe including you) act as they do. In most free-range videogames the characters are very transactional; they want something and enlist the player to get/do something. In turn, players collect items or set game flags that allow them to defeat antagonists in contests of skill or force. But it's not really practical to interact with characters by appealing to love, greed, loyalty, arrogance etc. in order to motivate them to behave differently, so videogame stories can't do dramatic comedy or tragedy at present, although you can of course have slapstick or situational comedy, and you can have sadness or gloom. But the general lack of permadeath and the reflexive nature of AI are still a limitation.

JRPG's don't tend to try and put you into the shoes of any characters, and instead you just learn to care about the characters the way you would reading a book or watching a movie.

* a game with a huge development and production budget

* a good game

* a challenging game

* a game intended to be played competitively online

* a game that features adult themes like sex and violence

My guess is that it actually means the first, but game industry marketroids speak as if all these criteria were coterminous with one another.

This is what happens when you try to make too much inferences of a trend in a short time within a step function.

The primary ingredient that fuels continued performance growth in semiconductors, including GPUs, is process shrinks. They are fundamentally what gives AMD and nVidia the additional transistors that they can use to increase performance. The foundry 20nm processes were a bust, meaning that GPUs effectively skipped a shrink. This means that the last 3 years, after 28nm became available, the only way they have had to increase performance was slight efficiency increases and growing the chips bigger.

GPUs will move to next gen foundry processes, the Samsung/GlobalFoundries 14nm and TSMC 16nm late this year or early next year. This will provide the most dramatic single change in the underlying manufacturing process of the history of making GPUs. Assuming no major architectural advances, expect twice the performance.

Of course i am assuming there is a story line or nice gameplay..

To give an example: my favorite game of all times, Ultima Online has at __best__, __meh__ graphics..I remember i got hooked before i even installed it, not because of the screen shots but the way the game magazine editor was telling his own experience how he got murdered while fishing quietly in a small town..

But it was pure, frustrating, junk from a gameplay perspective. The controls were clunky and shit was always popping out of nowhere and killing you. I kept dying at the first "boss", a skeleton thing on a throne that looked like it was made of some bigger creature's jaw. What I didn't know at the time was that there was a gryphon I had to defeat by punching the crystal orb it was bouncing; defeating the gryphon would temporarily grant me the power to shoot hadoukens, and the skeleton thing was only vulnerable to the hadoukens. (If you used up your hadoukens before defeating the skeleton thing, well, sucks to be you.)

And it's just full of stuff like this. The game doesn't tell you ANYTHING about how to play it or what its goals are. Plus it was a pioneer in unskippable cutscenes: moving from one place to another -- like going in a door or something -- entailed staring at a still image while adventure-game-style flavor text scrolled by and no key or button press could dismiss it. And once you die, that's it. You have to start all the way from the beginning. The game could be cleared in half an hour -- IF you knew where everything was. I guess it got replay value by surprising you with deadly enemies and obstacles you couldn't see coming and making you start over each time a new thing bit you in the ass.

But we all remember Shadow of the Beast -- indeed Amiga users look back fondly at it -- SIMPLY BECAUSE OF ITS GRAPHICS AND SOUND.

Sword of Sodan was the same thing: clunky and repetitive, but WOW LOOK AT THOSE HUGE SPRITES.

So yes, "good graphics = good game" is a thing, and it's because of the market, not because of the execs.

Of these, Zork: Nemesis is by far the worst game, even though it put a great deal of effort into its graphics. It was another victim of the Myst effect, but five years later, when everyone should have known better. The plot seemed shoehorned into the set, rather than the locations built to support the gameplay.

Return and Inquisitor felt like Zorks, but Nemesis felt more like a glorified fetch quest. I don't care if the graphics hold up or not after a few years, because the game itself was not particularly fun to play. The only factor preventing me from re-playing the other two is that I can still remember the solutions to their puzzles after 18-22 years, but all I remember from Nemesis is "bring me more red pages... er... I mean pure elements".

Therefore, if graphics matter, they cannot matter more than gameplay.

But lets talk about Legendary ones: might & magic series, planescape torment, baldur's gate, day of the tentacle, neverwinter nights, monkey island series, ultima online, warcraft 1, starcraft 1, diablo 1, heroes of mm & Ultima 1/2/3 etc...

The games you are listing can be highly rated, but they will not be remembered..

That doesn't mean that it makes the games less enjoyable, necessarily, but certain art styles are just better at conveying timelessness.

At some point we'll probably reach a certain peak level of graphics that can simply be considered "good enough" for most purposes henceforth (Have we already?), but in general investing too much in graphics for production purposes over tech demos and research, is often shaky.

In that case I wouldn't call it a game, but more of a tech demo or an interactive experience.

However, given that I can't see a movie for less than $15, that $1.5B represents only 100M views, and if you theorize that only 1B of the world population is wealthy enough to occasionally blow $15 to watch a movie, that implies 90% of the population is uninterested in VFX demo reels, which isn't all that dramatically different from my engineering estimate, especially given we could cherry pick examples to "prove" either your claim or mine, or play games with $ per unit time to reach either conclusion.

I completely agree that $1.5B is important business, yet it makes my point that appealing to a narrowcasted local maxima of VFX fans also simultaneously means perhaps 90% of the population will be uninterested. And the point I'm trying to make isn't that 10% of the population is irrelevant, but if someone could crack the code and disrupt the industry and instead of narrowcasting, make something of broad appeal that perhaps half the population would be interested in, that mysterious idea would be worth about $7.5B which is actually pretty good revenue for a startup. But it'll never happen if the entrenched oligopoly isn't disrupted.

I honestly have no idea how to exploit the market, but "obviously" there's a huge under served market with possibly staggering revenue, for some future startup that can figure out how to make movies that appeal to more than about 10% of the population. Piles of money are sitting out there, waiting to be harvested...

There are similar analogies and financials with pop music. Entrenched industries have pretty much figured out how to sell simple mass produced formulaic music to each generation's (each years?) teens. There seems no biological or psychological reason a better industry competitor couldn't sell 3 to 7 times as much if they could appeal to more than young kids. Eventually someone will crack that startup opportunity and make mid nine figures.

Finally ditto, kind of, with sports. "Everybody is a baseball fan" but only about 5% of the population actually watches the world series. If you take existing advertising revenue and multiply it by a fraction representing some magic startup pixie dust that gets maybe 50% of the population to watch, that's some serious ad revenue money; in fact its "football" type money, LOL, which is a whole nother topic.

But, I wouldn't be opposed to tinkering with something in unity either; but thats obviously a WHOLE other ball of wax.

and for anyone else who's curious: https://www.kingdomofloathing.com

Didn't knew it's single core only, that's a shame.