People so easily forget that every game has an "engine", even 2D games. It's just that 3D games are much more complicated to put on a screen, and so the various pieces of code that enable all the yummy stuff aren't worth rewriting for each game.

The description of engines as getting "long in the tooth" description shits me as well. Code doesn't decay. Sure, you can make an assumption when building an engine that turns out to hold you back (for instance, no-one's going to have more than 64 megs of RAM), but any competent developer keeps adding to their "engine" to provide a flexible, stable platform to get the game up on the screen with a minimum of fuss. Nintendo's been using the same "engine" for its Mario and Zelda games since Mario 64, and they keep adding to it, and the games have generally excellent art direction (Mario Sunshine not withstanding) so people don't notice that they've reused the water physics from Mario Sunshine in Wind Waker, and the diffusion lighting from Twilight Princess in Mario Galaxy. That's how it should be done: an engine, done right, is just a set of tools that let you get a certain aspect of the game for free.

I was thinking about this today. This is a point I cannot agree with strongly enough.

If you can stand it, look at the wheat and chaff separate itself on this NeoGAF thread on the Source engine:

http://www.neogaf.com/forum/showthread.php?t=342761

Raving bullshitters everywhere who aren't paying attention to those that know better, especially the ones that say "none of you know what you are talking about."

I am glad you wrote this; now there is a perfect post to just point to whenever someone starts bullshitting about the technical aspect of development.

Love this post for exactly the same reasons Mr. Walbridge gives.

Some friends of mine worked on "Anachronox" back in the day, a flawed little gem that never had a chance because, as the press at the time never failed to mention, it was built on the "aging Quake 2 engine". Never mind the fact that it had nice art and level design, that the engine itself had been heavily customized / upgraded, or of course that, despite its problems, it was fantastically funny and well-written. Nope, old engine. Do not pass go.

Fortunately I think we are in the last days of engines being truly relevant to the experience of playing games. They'll always be around as ways for companies like Epic to make money and for many others to (theoretically) save money. They're just not as central to the aesthetics or kinesthetics of games as they once were. Which is really only a good thing for both gamers and creators.

You raise some good points about the severe lack of technical knowledge most gamers possess.

I've been playing games most my life and I find that my experience seems to provide me with an implicit, although mistaken, sense that I understand the rudimentary technical aspects of games.

In reality, I can barely grasp html, let alone any of the languages used to create even the simplest games I play. I think gaming critics and enthusiasts need to either educate themselves in the technical aspects of games or honestly admit their ignorance and keep to firmer ground like thematic analysis.

Starting a review or analysis with the simple question: "Is this enjoyable?" might help avoid some of mistakes non-technical gamers make.

The comment on source is amusing because Gabe Newell talks a lot in interviews about how the Source engine was built, essentially, to be the last engine Valve would ever need.

I've seen a lot of comments lately that seem to assume that the graphics of Left 4 Dead don't have as much detail as, say, Bioshock or Crysis because the Source engine can't handle it. The thing these comments miss is that power comes from hardware, not engines. Bioshock and Crysis won't even run on my old computer*(which has 2 gigs of RAM, an AMD 64 2.4 GHz processor, and an nVidia 6150le onboard graphics card). On the other hand Left 4 Dead does run at minimum settings at respectable framerate(It gets a little choppy when a horde of infected show up). The problem Valve was working against wasn't getting Left 4 Dead to run on Source, but getting Left 4 Dead to run on cheap computers.

*Ok, not entirely true. I can get about 2 frames per second out of Crysis at minimum settings and Bioshock will do maybe 10 but won't bother to render certain unimportant details like Rapture.

Whilst I agree with almost your entire blogpost, and I must admit I haven't read the article it's linked to. the comment 'the geometry count is low' however seems to be a valid complaint.

At least, assuming the person is trying to talk about polycount :)

Thing is, even when trying for a certain aesthetic which can be achieved with a low polycount, one can miss the mark. Even if one is trying for a game with millions of enemies onscreen at the same time, if one notices that the polycount is low on the characters or environment, one has miscalculated the aesthetic feasability of the game one is making.

To put it like this: I'm far from a graphics whore...but I am an easthetics whore. I love good art direction; Homeworld, Anacranox, even now Morrowind and Oni (well, not entirely Oni's bland sameness of buildings, but definitely the main character) hold up very well in that respect. But if Oni came out now, I too would comment on the lack of polygons, because even as you stated that a game is made up of the technical and the artistic (and the gameplay mechanics and the sound design and the UI), if one of those is off it warrants mentioning.

The converse is also true. I'm reminded of Strike Commander; a beautifull game which brought pc's to their knees even a year later (and that reminds me of Crysis :) ). And that is bad art too; if it's too simple or too complex and thus ruins the game by unplayable framerate or if the aaesthetics suffer due to overly ambitious mechanics (too many enemies onscreen, too many AI calculations), the overall game suffers...even if the gameplay can still be fun, it is a valid critisism to mention that the polycount is too low or too high.

I strongly agree with you, Matthew.

I've been reading your blog for a short while now and I'm enjoying it. Keep it up. :)

@ shMerker

I'd say so. They lift such detailed system statistics from their users (with the surveys they do every once in a while), it'd be a waste if they then ignored those.

This was an excellent summation of a common problem in game writing, but I would make a distinction in the "half-baked attempt[s] to speculate about the underlying technology" that you deride. While sweeping generalizations about underlying technologies are pretty much always the sort of fluff you describe, there are legitimate points a non-technical person can make about certain technologies that become tiresome through overuse or poor implementation. HDR lighting, for example; the lens flare of the current console generation.

We should not obsess about graphics and sounds. Many of the best games were played on systems like the Nintendo, Super Nintendo and Genesis. I never notice technology during good action sequences: They force me to think about my actions not the pretty pictures. Mega Man 9 and Cave Story were, despite the N.E.S. level technology, the best games I had played in a very long time and very satisfying to defeat.

Another simple yet elegant point Matthew, I agree. I like art comparisons with other contemporary games, especially when it comes down to consoles, and certainly compared to previous ones of the same generation (eg; Saints Row 2 vs. Saints Row), but going on the basis of the engine for art quality is absurd in writing about videogames.

I'm not at all surprised by the variety of sites brought up, which I always cough at reading any news linked to on there when anything technical comes up (or rather, is brought up in a silly way by them), be it interviews, reviews or previews.

I honestly hope developers physically grimace about being asked about the engine in interviews, the amount of times it's misworded and misused.

@Hater: Cave Story is not "NES level technology" by any stretch of the imagination. The number and level of detail of the sprites in that game would have been completely impossible on the NES. Not that I'm saying the game is intended to be a visual feast. I'm just saying that sometimes processing power is being diverted to things other than pretty pictures, which I guess was your point in the first place.

@MacD: It's sort of beside the point whether a comment like "the geometry count is low" is accurate or valid. Even if it is it's not terribly useful to someone who just wants to play a game and doesn't care how any of it works. It seems to me like it should be enough to say "the graphics aren't very detailed." or if you wanted to be more specific "the characters have noticable edges". (seriously, can you even complain that a game looks blocky any more?)

@Matthew: Somebody pointed me over to your blog post. I completely agree with your point. However, after a few days of thought I have one objection. If the the correct thing for journalists to do is ignore the technical stuff that they don't know about, then publishers need to stop using them as talking points.

I don't want to flip to the back of a game box and see references to polygon counts, game engines, or frame rates. I imagine that licensing certain engines requires a mention somewhere on the box. If we don't empower people with knowledge that they may not understand, then we don't have to worry about how they use it to communicate.

Comparing engines to engines is always going to end in confusion. Unless you are the master of every facet of every engine out there (and completely unbiased) how could you know what compares favorably against what engine-wise?

I think it is far more sensible to compare how a developer uses said engine (facet) versus how another developer uses theirs.

Arbitrary example:

Compare Illuminate Labs' lighting engine used in Mirror's Edge (which uses Unreal 3.0 as well) with Gears 2's Unreal 3.25/3.45 lighting. The results each game displays with the two different lighting schemes is inarguably different - neither categorically better. And, definitely, each lighting engine in these games does things differently. But that's not really my point.

The striking difference in the use of light in these examples comes from how the lighting tools are used and what content they are used in conjunction with. Seeing Mirror's Edge and its ultra stark, sharp lighting and its ambient-occlusion friendly non-noisy coloration, its easy to say that DICE have used the Illuminate Labs toolset more successfully by playing to the strengths of this engine and not fighting with it.

Looking at Gears 2 and the use of the Unreal 3.25 lighting system, it is hard to compare to Mirror's Edge due to the overt difference in approach to on screen 'busy-ness' in terms of poly count and texture noise in G2. Early on in Gear's 2 it is difficult to pin down exactly where the contrasty lighting is but I will say that they do have some exceptionally well implemented lighting rigging in certian stages of the game - the overt contrastiness of the lighting is just not as prevalent as in Mirror's Edge due to the nature of the content and the way that the content is delivered. Neither is in any way wrong and both approaches compliment each other (in the wider spectrum of the overall market) very well.

Hey, I like your blog and I realize that it takes some balls to be a good, respectable hater and that it opens one up to lots of hate in return. Anyway, I just recently discussed the difference between "composes" and "comprises" with a friend. And, as one fan of grammar to another, in the first paragraph, "comprised of" should be "composed of."