Talk:Game engine

Wiki Education Foundation-supported course assignment
This article was the subject of a Wiki Education Foundation-supported course assignment, between 16 January 2019 and 22 May 2019. Further details are available on the course page. Student editor(s): Dwylrcbc.

Above undated message substituted from Template:Dashboard.wikiedu.org assignment by PrimeBOT (talk) 22:01, 16 January 2022 (UTC)

Game programmers hate writing engines
I removed the following text:
 * A practical definition of a game engine is what most game developers would prefer not to write (as opposed to the things that make their game seminal, such as levels, artwork, textures, animation, sound tracks, etc.). The extreme realization of this idea has been for some game developers to take existing games and simply modify their art resources.

This isn't actually correct. Programmers prefer to work on the engine. E.g. Carmack couldn't care less about the art and levels. :) Level designers, artists, composers, etc. like to work on other things. Seeing how the link actually pointed to game programmers, I don't understand how what was written could actually be true. Paranoid 21:24, 10 Jun 2004 (UTC)


 * I don't think all game programmers have the same preference: some may love working on engines, others may only want to work on AI or other game-specific features. That being the case, the statement which was removed is clearly POV and should have been removed.    &mdash; Frecklefoot | Talk 18:44, Jun 25, 2004 (UTC)

Game Generations
I disagree that generations are indistinct. The feature sets and, especially hardware/DirectX generations are quite distinct. Moved games around: Paranoid 18:58, 19 Jul 2004 (UTC)
 * UT2004 is very similar to UT2003 and even UT.
 * Battlefield is not 4th gen, Vietnam is the same engine as 1942.
 * Placed back Stalker - someone is going to write an article on it sooner or later. :-)

The concept of generations in engines is used a lot in the press and by developers. Quake was clearly a revolution and so started the 2nd generation. Far Cry, Doom 3, Half-Life 2 are very distinct from previous games and deservedly get called next-generation shooters all the time. Unreal 3 is clearly yet another generation. The only issue which is not totally clear is the border between 2nd generation and 3rd generation. Also there should be no problem distinguishing a highly modified version of a 'old engine' from a brand new engine.

Regarding Battlefield, compare Battlefield 1942 and Battlefield Vietnam screenshots. They are not very different in quality, because it's the same engine with some tweaks and more eye-candy added because after 2 years average hardware can handle it. Paranoid 22:24, 20 Jul 2004 (UTC)

These choices of years and generations seems rather arbitrary. If this information can be supported, great. If not, its just making up groupings of games that are often significantly different. I placed the games based on there release date, and the spans of years that were already listed. Greyengine5 23:54, 2004 Jul 20 (UTC)


 * First, by reverting you removed other changes that I made, which is not nice. Second, the generations are far from arbitrary, as I explain 3 paragraphs above. And you apparently have nothing against the grouping per se, why then you don't accept calling these groups generations? Paranoid 08:17, 21 Jul 2004 (UTC)

Ah sorry I missed adding your changes back in! I had meant to go back and look through to see what new stuff was there. Anyway, I am against labeling these as 'generations' of games, as it makes its seem that these are common names for engines (which they are not), among other points.

The issues is that while one developer can indeed say he has a 'next-generation' game, its only what he is calling the the 'next-generation' of there engine. So while HL2 is certainly a 'next-generation' after the first, there are a whole bunch of engines in between. If we chose a different companies engine generations then HL and HL2 become the ones stuck in between.

Since the next major release of a engine often starts with the core of the old one, the amount of changes that constitute a 'new engine' is a matter of naming. Where a game goes becomes very hard at the boderline.

For example, Nuke Dukem was released in 1996 the same year as quake, yet it was placed with first generation engines. The game was significantly more advanced the doom. The same issues arises with the battlefield games- bfv was indeed the next iteration of that engine and incorporated significant new features. Granted, it was more evolutionary in its changes- but it was released in 2004, which placed in the next section of your categories.

When arbitraty categories are made for something that does not have good breakpoints there are problems at the borderline, which is the case here. Also, I cannot find this categorization scheme anywhere else, google searches come up almost nothing.

Which game goes where is going to be POV for borderline cases, whereas the release date is not. This is going cause problems, just as it did for us.

I do like breaking it up loosely though, and, certainly there can be indistinct era's in technology, but I don't think the exacting labels work. Another idea for organization would be to list 'firsts', i.e. to list the first game to release with a specific, new technology.

Well thats all, I will try and add back in some of the changes you made before that were lost. Greyengine5 14:38, 2004 Jul 21 (UTC)


 * The term 'next-generation' is rarely used in a sense 'next generation of our engine'. Doom3 and Half-Life 2 are very often grouped together as 'next-generation' games and, in Doom3 example in particular, that can't mean 'next generation of our engine', because id had Quake in between.


 * This is rarely true. For big leaps (such as from one generation to the next) the core of the old engine is usually only used during early stages of developement.


 * Duke Nukem engine was not significantly more advanced than Doom engine. The game was, but the engine was extremely similar (with a few improvements). As for the Battlefield Vietnam, the year when it was released is irrelevant (just as with DN3D) - only the features and image quality are important. There are games released today that are still stuck in the 2nd generation (some poorly done budget titles). The year is not a good indicator.


 * The only problem is the distinction between 2nd and 3rd generations, 1-2 and 3-4 are very clearcut. Though I can agree with your argument that calling the sections 1gen, 2gen, etc. may be somewhat misleading. Descriptive names like you suggested before may be better. But I strongly disagree with your claims that there is only gradual evolution with no distinct generations, because that simply isn't the case.


 * The first generation is extremely distinct. The move from 2D/2.5D to 3D and from sprites to polygonal objects (and from software to hardware rendering) is as distinct as it gets. Ditto for the move from pre-2004 games to the next-generation engines with extensive shader support. When you place UT2004 and FarCry next to each other, it's very clear that the engines are very distinct (to say nothing of the games). The leap that happened to a large extent due to programmable shaders is almost as huge in technical terms as the move from 1st gen to 2nd gen.


 * And I will add those that you will miss for one reason or another. :) And I still strongly disagree about your claims about indistinctiveness. This is a crucial matter in evolution of the game engines. The developement cycles are quite long, most devloping companies can't afford having two teams working on different engines, instead they finish one game and start the next generation one (with few addons in between to cash on the name). There is a limited number of development houses that innovate in engines a lot. FPS games depend very much on the hardware available (so the move from CPU to first simple texturing 3D-cards, to more advanced cards and then to programmable shaders). There are distinct generations, whether you like it or not. Paranoid 19:07, 21 Jul 2004 (UTC)

Few things I agree with there, few things I don't- the key issue remains that any attempt to draw distinct categories of game generations has serious problems at the borderline. To decide it becomes a bad mix of POV about how advanced the engine is (whether it 'qualifies') and about specific features (such as those that you mention). While these features either exist or do not, such as pixel shaders, etc. (and I do like it!). they cannot individually be used to determine a 'next generation' since they appear in different palces. To use this bfv we have been talking about as an example- BF:V should probably not be in the same category as say HL2, but, it can be with Farcry. What feature sets it apart 'vehicles?' bit-mapped vegetation?. That said, Farcry is probably more advanced overall and could more easily be grouped with HL2. But then where does say, Joint Operations fit in- or painkiller? Which brings me back to my point- the feature set and improvements in feel, are to gradual.

The idea of listing each game that was the first to retail with some feature in a duplicate list or in that one could help aviod this 'less advanced' 'more advanced' problem. Once again, Im well aware of the distinctions that exist between engines- but it just doesn't work out for borderline cases when comparing all the games. I think that groupings would have been popularized by now if it worked well.

In any case, its not so important if you don't want to change back to that numbering system. Still, I have no interest in believeing incorrect things (as Im sure you don't). I of course agree with much of the information you present (the facts are what they are)- its just the conclusion we differ on. So if you have some good articles on game development that has some good support for the 'distinct' groupings things Im not above changing my mind. For now though, it just doesn't seem to work out well when describing these things. Greyengine5 00:22, 2004 Jul 22 (UTC)


 * Just my $.02. Why not get rid of the whole "Generations" thing and present a table with games on one axis and features on another.  Earlier games should come first, progressing to more current ones.  That way the readers can see how many more features recent games have compared to older ones.  If you really want the "Generations" thing in, you can have thicker borders between the generations in the table to emphasize the leaps. Peace. :-)    &mdash; Frecklefoot | Talk 14:53, Jul 22, 2004 (UTC)


 * I saw similar attempts in the past on the Web. The resulting table would be enormous - at least 30 games with at least 40-50 features. As such it would have a number of problems:
 * It would be a bitch to compile
 * It would be permanently incomplete and out of date
 * Only a small part of it ("the diagonal") would be relevant, the rest would be obvious. Wolf3D obviously didn't have reaslistic water and Quake 4 will have floors as different height.
 * But I agree (finally :) ) with Greyengine5's point that calling the sections "generations" can be a bit confusing (though it does have other benefits, IMO). I still disagree with his claims that there are no distinct generations and I will try to improve that part when I feel like it. :) Paranoid 19:11, 22 Jul 2004 (UTC)

The current Generations organization would work better if things were moved around to their DirectX or OpenGL equivalent feature level. UnrealEngine 1, HL, Q2, and earlier Q3 games (the widespread usage of hardware rendering and colored lighting) are from the DX6-7 era; Battlefield, UnrealEngine 2, Max Payne, RTCW and later Q3 (hardware T&L, complex particle systems) from DX8; CryTek/FarCry, Source/HL2, Doom 3, and X-Ray/STALKER (hardware pixel/vertex shaders, stencil buffering) to DX9; UnrealEngine 3 and unnamed future engines to the DX10 generation. BonzoESC 00:00, 6 Aug 2004 (UTC)

Content Disagreement
NrarerN, Paranoid, you guys should come to a consensus because you've undone each others changes and I like this page too much to see it at the top of my watchlist every day because of some petty squabbling :(

Personally, due to the way the industry is moving, it's not really worth the effort to focus on FPS game engines, since the big players are versatile enough to not only work in an FPS game, but RTS, MMORPG, and adventure games.

BonzoESC 04:47, 28 Oct 2004 (UTC)


 * Actually, now that I think of it (after reading all your comments on Talk), I realise that I became too focused on FPS engines, even though in the very beginning was supposed to be more general. I think historically the FPS engines were much more important in terms of graphics and realism impact, although other genres' engines were of course just as important in terms of actually making the games. I don't agree that FPS engines are already versatile enough, although it's definitely happening (Unreal for Lineage, RTS mods for UT2004, and Strife (RPG/adventure) using Doom engine, AFAIR) and will happen in a few years.


 * As for removing the games from the lists, the question is whether we should build a cross-engine historical narrative, explaining what changed, as all genres were developing (between 1995 and 1998 FPS got this, flight sims got this, driving sims got this, RTS games got this, for example ...) or create separate subsections (with different narratives) for other engine types.


 * There is nothing wrong with mentioning Mafia and GTA 3 somewhere on the page and they definitely made a very strong contribution to the industry, but they simply weren't FPS (just like MP, but at least Max had a very FPS-like gameplay), so they should not be included in the section clearly titled "FPS engines". Adding Battlefield Vietnam, RTCW and Halo 2, on the other hand, is not justified, unless it is stated clearly what was their significance in terms of engine, compared with other games released at the same time as well as their predecessors (and not just adblurbs from their boxes). As for FEAR and STALKER, they are eagerly awaited, they have clearly superior graphics quality (comparable to or better than current heavyweights HL2 and D3), and they will be released soon. Why remove them? Paranoid 16:45, 31 Oct 2004 (UTC)

Well we have a disagreement. Lets handle it this way: I will leave yours and you leave mine. NrarerN 20:45, 31 Oct 2004 (UTC)


 * That's not the best way to handle a disagreement. The way it is usually done on Wikipedia is that both parties give their arguments, discuss and eventually reach a consensus. But we can't discuss when you refuse to participate and just keep reverting. I gave my arguments several times in edit summaries and here on the Talk page. The only thing I ask is that you respond to my arguments. Why do you think that your changes are warranted, and why do you think that my arguments were invalid? Paranoid 22:29, 31 Oct 2004 (UTC)

Your arguments sucked, and I guess you cant read either because I did respond, both in edit summary and here. If you actually knew anything about game engines then I think you'd understand why your arguments aren't even worth responding too. I don't think were going to convince each other of anything, and since which games are listed is somewhat subjective just let all the choices be listed and save ourselves the trouble of talking about this. NrarerN 16:37, 1 Nov 2004 (UTC)


 * NarerN, I know that claiming that other people don't know anything is an attractive way to argue your point, but you too should seriously get your facts straight and chill out a bit. Don't revert changes to make a point because quite often you are incorrect, and there are good reasons for the changes being made. BonzoESC 18:11, 1 Nov 2004 (UTC)

Yea it is attractive, just like exaggeration, which is something you seem to enjoy doing as well. I give you the matter of the 96/97 thing in the interests of ending this, but it should be noted that quake still made use of sprites in some cases. Even if bfv or rtcw have the same core engine as a older game, there nothing wrong with putting it again as that was already being done (like with unreal, unreal 2 being listed). Just like if your going to include FPS/TPS games like max payne you should include the others too. Also, halo is referred to as a fps game. I think you people are the ones who need to chill, your changing back suggestions that fit with whats already on the page. NrarerN 01:10, 2 Nov 2004 (UTC)


 * Personally, I think it should be mostly focused to engines with multiple titles from different developers running on them, otherwise it'll just be a list of arbitrary games selected by us editors :) Since Take Two (Publisher) never used their Halo engine licenses (part of the deal ending their contract with Bungie after they were purchased by MS), I'd kick it off the list.  Max-FX (Max Payne) was only really (to the best of my knowledge) used outside the game once, in 3DMark 2001 (both Remedy and Futuremark employ several Future Crew coders), and I'd hesitate before putting the Serious Engine on the list, because it's only seen one non-Croteam title to the best of my knowledge (the wacky but forgettable Korean "Nitro Family").  I would put the different versions of the Unreal technology up because there are clear boundaries between their featuresets, and the developer does consider the three currently available branches (2.5, 2X, 3) separate products.


 * I would say that edits shouldn't be based around what's already on the page, because for a page about game engines, we should probably focus on the engines themselves and not stuff that's been made with them. BonzoESC 02:21, 2 Nov 2004 (UTC)

Well you make some good points and some points I disagree. I had been following the lead of what already was done here, such as engine only used in 1-2 games, and the focus on mentioning games. I would disagree that there is not a focus on engine's, becuase a game title is the actual manefestion of a given engine. That said, the page could and should be even more focused, such to list a engine, then the games that use that engine underneath it. It is very easy to disagree as to what constitues a 'new' engine, so I think a format that shows lineage and specific features would be better. Overall though, what it comes down to is that the fps game engine's need there own page and have out grown this. This page can focus more on the different types of game engine and there interconnectedness and the other topics can be free to grow from based off there current composition. NrarerN 18:01, 3 Nov 2004 (UTC)

Engine Before Doom?
Were there any commercially viable game engines before Doom?

Yes, at least id's Commander Keen (2D engine(s) the first trilogy engine was used in Duke Nukem (aka Duke Nukum)) and the so called Wolfenstein 3D engine (earlier versions where used in Hovertank 3D and Catacomb 3D, later versions in the Blake Stone series, Rise of The Triad and some others. Between the Wolf3d and Doom Raven Software licensed an 3d engine by id Software's John Carmack for Shadow Caster.

Here's an abbreviated list of a possible outline with over-dramatic titles and mostly engines used for multiple games:
 * Console Style Scrolling World (2d smooth horizonal/vertical scrolling): Commander Keen 1-3 (used in Duke Nukem), Keen Dreams, Commander Keen 2, Duke Nukem 2 and others
 * The Pixelated World, 1st Generation (2.5d with sprite actors, square based): Wolfenstein 3D (Hovertank 3D, Blake Stone 1, Blake Stone 2, Rise of The Triad, Noah's Arc), Ken's Labyrinth (Ken Silverman/Epic Megagames), Shadow Caster (Raven Software, 3d Engine by John Carmack)
 * The Pixelated World, 2nd Generation (2.5d with sprite actors, sector based): Doom 1/2 (id software, used in Heretic, Hexen, Chex Quest, others?), Build (3D Realms, used in Duke3D, Blood, Shadow Warrior, others?)
 * The Polygonal World (3d with rough textures): Quake 1 (id software, used in Hexen 2, Half-Life (w/Quake2 network code) Valve, a bunch of mods), some X-Men game that wasn't very good (the X-Men game is not really a shipped game more like a published Quake mod like Malice, others?) and most of its progeny
 * The Detailed World (3d with detailed textures and mainstream use of 3d accelerators): Quake 3 (id software, lots of games), LithTech 1 (Monolith, used in NOLF, others?), UnrealEngine 1 (Epic Games, Unreal, Deus Ex, lots of games), Renderware through the GTA3 series
 * The Fleshed-out World (skeletal animation, hardware shaders for special effects, wide usage of hardware transform & lighting): UnrealEngine 2 (Epic Games, UT2003, UT2004, Deus Ex: IW, Postal 2, Lineage 2), that engine used in Morrowind (NetImmerse?), Renderware used in Burnout 3, Lithtech Jupiter (Monolith, NOLF2, Tron 2.0)
 * The Shaded World (continuous shader usage for model detail frequent usage for environmental effects): Doom3 (id software, Quake 4, Prey), Source (Valve, I forget what games)


 * You're right, it makes sense to place the engine first and the actual games second, given the article topic.

It would probably be best to only talk about engines that have titles confirmed using them, which means no UnrealEngine 3 talk yet. BonzoESC 23:00, 31 Oct 2004 (UTC)


 * First, Unreal Engine 3 was confirmed to be in use for several titles already being developed. Second, the artile is about game engines, not engines used in actual games, so UE3 qualifies in any way. Paranoid 07:14, 1 Nov 2004 (UTC)

Too point in time of writing-centric
It is poor writing, biased and inconsistent to call Doom's sprites "very simple and pixelated" or Quake II's textures/models "low-res, simplistic and crude" and calling the next generation graphics "extremely detailed" and "advanced". People thought Doom looked realistic in 1993. I'm sure people will laugh at Doom 3's (or Unreal 3's) graphics in a couple of years, as will they do if they find a copy of this article. Let's just state the facts instead of making judgements. Fredrik | talk 06:27, 28 Oct 2004 (UTC)


 * I agree. I was disappointed by the writing when I saw it, since I remember being blown away by every id title I got my hands on when they came out.  Let's NPOV the writing. :-)   &mdash; Frecklefoot | Talk 14:29, Oct 28, 2004 (UTC)


 * I agree with the criticism of the style and the adjectives, but it's still important to state the facts (without what may feel like "derogatory" comments). Paranoid 16:45, 31 Oct 2004 (UTC)

Example any good?
An anon user added an entire "example" section. It seems very POV (that is, the code seems to reflect that user's opinion of a good game engine). I've posted it here for comment before it is reinserted, and actually to discuss whether it should be re-inserted at all. At any rate, it needs to be copyedited and probably converted to pseudo-code and should be inserted near the end, not near the top where it was: --Begin pasted section:

Example
The software design for a standard engine looks something like this:

//THREAD 1

//Loop 50 times per second

while (true) { logicTime++; gfxTime++; usleep(20000); }

//THREAD 2

while (keepRunning == TRUE) { logicLimiter = 0;

while ((logicTime > 0) && (logicLimiter < threshold)) {   doLogic; logicTime--; logicLimiter++; }//while logicTime

if (gfxTime != 0) FPS = 50 / gfxTime; else FPS = 50; gfxTime = 0; drawGraphics; } In this way, the game is split into two parts: the logic and the graphics. This method ensures that the logic will run at the same speed on all computers. Slower computers will display less graphical frames per second.

logicTime: is continually being incremented in a separate thread. Whenever it's greater than 0, we need to run doLogic. So, if it were 6, we would call doLogic six times in a row. In this case, logicTime is incremented 50 times per second. You can change this by modifying the delay in the usleep function.

gfxTime: can be used to calculate the frames per second (FPS). If we are ready to call drawGraphics and gfxTime is 10, we know that only 5 frames are being displayed each second. This is because it takes THREAD 1 a fifth of a second to increase gfxTime to 10.

logicLimiter: very slow computers might not be able to update the logic at the desired speed, even if we skip drawing the graphics completely. By the time doLogic has finished running, THREAD 1 has already increased logicTime again. We never exit the logic loop, and never call drawGraphics. logicLimiter makes sure that the graphics are never completely ignored. This slows down the game logic, but that was happening anyways. At least now the user can see what's going on.

CPU Usage: this simple example works OK, but will always use all the processor power available. A fast computer might be able to draw the graphics before THREAD 1 has a chance to increase logicTimer. And so, we have two or more back-to-back calls to drawGraphics. This isn't a big deal, but is a waste of CPU resources. If the logic hasn't changed, then we're just drawing the exact same image twice. It may be desirable to include a check for this, and sleep instead of calling drawGraphics again.

Mutexes: logicTime and gfxTime should be accessed from within mutexes, as they are modified from two different threads. The mutexes are not present in this example to keep things simple.

--End pasted section. &mdash; Frecklefoot | Talk 15:34, Mar 24, 2005 (UTC)


 * Not good. It is better to discuss design than list arbitrary examples. Fredrik | talk 16:15, 24 Mar 2005 (UTC)

That example is awful. 206.169.113.230 (talk) 02:03, 5 November 2011 (UTC)

Removed extra links
I removed the following links because I thought they bore little relevence to the topic and were mostly spam:
 * IDV's SpeedTree website
 * Great Game Graphics... Who Cares? Includes overview of graphics development in video games till 2003 (Powerpoint)
 * The List of areas of a typical game engine
 * Gamemaker home page
 * List of Adventure Game Engines
 * PTK 2D game engine

If anyone disagrees, state your reasoning and add them back in. Peace. :-)  &mdash; Frecklefoot | Talk July 4, 2005 00:51 (UTC)


 * None, NONE of these links add to the article, but they could. Thanks for putting them here...
 * Nothing here or in the article refrences the a text based game engine made by Infocom, called the Z-Machine. Pbbbbt! —Preceding unsigned comment added by 69.232.210.165 (talk) 11:21, 4 February 2011 (UTC)

Computer and video game terminology category change
Hi! I've removed the cvg "terminology" category from this article as part of the portal project discussed here. The remaining categories cvg "development" and "game engines" cover it I think. Glad to discuss! -- Sitearm | Talk 20:06, 2005 August 20 (UTC)
 * Thanks for the explanation. Rl 20:15, 20 August 2005 (UTC)

Prices?
I've done some investigating of prices. I haven't seen any "high end" game engines for $10,000....can someone name the one in question?

Here's the webpages on the licensing of a few specific popular "high end" engines:


 * http://www.idsoftware.com/business/technology/ (a little outdated)
 * http://www.unrealtechnology.com/html/licensing/terms.shtml ($350k + royalties, or $750k flat)
 * http://www.stainlesssteelstudios.com/EngineLicense.html (lowest price: $250k + some amount of royalties)
 * http://www.touchdownentertainment.com/home.htm ($10k "per seat" + $10k upon game release)
 * http://www.renderware.com/
 * http://www.valvesoftware.com/business.html ("very competitive licensing terms")

--ChristianB


 * Well, there is also the Torque Game Engine for $100. There are also some free open source ones.  The prices for high-end commercial engines vary, though.  They are very often tied to the expected number of units to sell (not strictly a royalty), whether or not the customer wants tech support, full source code and other considerations.  So, one really can't say, "The Slavegirls of Phobos engine costs $5000."  The cost is contingent on a number of factors.


 * Also, sign your posts. I added your signature above, but you can add a signature with 3 or 4 tildes (  or ~ ).  The latter is preferred, since it also adds a timestamp.   &mdash; Frecklefoot | Talk 17:00, August 26, 2005 (UTC)


 * When i was shoping for an engine at Jan 2006 prices, Warcraft 3 (hence i linked with my edit) was costing $US 3 3/4 Million, Oblivion was not open to licences, and the market rate for UE3 is $450,000 per seat or unlimted seats for $1 Million. I DO NOT WANT MY NAME HERE

Any prices need to have a citation. I have never seen the Warcraft 3 engine for sale any where. id Software makes their prices obvious, as does Epic.

Web-based Games
For a web-based game (specifically Hattrick), could the web server be considered the "game engine?" -DMurphy 00:40, 15 April 2006 (UTC)


 * IMHO, no. Unless, maybe, it is only a game web server and not a general purpose web server. I'd get some other people's opinion on this... &mdash; Frecklefoot | Talk 15:28, 17 April 2006 (UTC)


 * The Web Server is NOT the game engine, its the middle ware that the Web server serves that is the game engine. I know of NO web servers that stream game engine content, however, online MMRPGs have servers that deliver their content, but the engine runs on the client. —Preceding unsigned comment added by 69.232.210.165 (talk) 11:16, 4 February 2011 (UTC)

Game Engine VS 3D Engine
Currently 3d Engine redirects to Game Engine and also this A game engine is the core software component of a computer or video game or other interactive application with real-time graphics.

I don't think this is technically correct. A 3D engine is only responsible for graphics while a game engine would be more complete. It may be common for people to use the two inter-changably to mean one or the other but maybe a note should be added to point out the technical difference. —Preceding unsigned comment added by Quantum1024 (talk • contribs)


 * Be bold. &mdash; Frecklefoot | Talk 20:10, 22 September 2006 (UTC)


 * I don't get it either. A RT 3D Engine can be seen as a part of a Game Engine (or as a part of a software of other kind, not game, which is misleading). I think we should have a separate article about 3D Engines and a little subsection under game engine.

--Krahd 20:58, 7 November 2006 (UTC)


 * I agree too, is a complety (althoug inclusive) diferente term. I 3d engine "IS not" a game engine at all, is just abstrac the graphic part only. A 3d engine however, "is a part" of a game engine.

Is like a engine from auto redirects to auto, is two diferent stuff. New arctile must be form. —Preceding unsigned comment added by 200.108.215.226 (talk) 17:39, 9 December 2008 (UTC)

NPOV
"""A very popular and inexpensive game development environment was created beginning in 1999 by Mark Overmars. His Game Maker object oriented interpreter makes development of 2 dimensional games extremely easy. The system can support many game formats including role-playing games (RPGs).""" —Preceding unsigned comment added by 24.108.200.84 (talk • contribs)


 * Thanks for pointing that out. &mdash; Frecklefoot | Talk 16:36, 28 November 2006 (UTC)

13: Mugen Game engine
I have a question why isn't this labeled as a game engine. It is used to create 2D fighting games. —The preceding unsigned comment was added by Agentsix (talk • contribs) 09:07, 22 February 2007 (UTC).

Licensing Engines
I couldn't find anything about Warcraft III being licensed for 3.75M USD ... Citation or more details please

Generic example
Uploaded a textual example of the process a game engine does during an event. Being too lazy to sign up, i couldn't add the image which is available at imageshack hereimageshack, be nice if someone would add it. If there is a better place for this type of data, I couldn't find it. —Preceding unsigned comment added by 80.230.255.60 (talk • contribs)


 * I had to remove this. It was poorly written, and full of assumptions based one possible design of a game engine.  I don't think it improved the article at all.  Quite to the contrary, as a matter of fact.  If someone disagrees, please post your reasoning here, and we'll go from there. &mdash; Frecklefoot | Talk 16:32, 15 May 2007 (UTC)


 * I was just passing through while looking at the IP's contributions, and I have to agree. In fact, if it wasn't removed, I would have done it myself. Not only is it full of assumptions, poorly written and the like, I found it to be less like an encyclopedia entry, and more like an essay: "The following paragraphs" e.t.c.
 * So, I have no issues. --Dreaded Walrus t c 13:52, 28 May 2007 (UTC)

Recent trends
Removed references to Apple and Iphone since Iphone does not support game development using managed languages.

Middleware
So exactly what's the difference between game middleware that doesn't specialize in one thing and a regular game engine? Per the info stated under Purpose, it seems to be just a synonym. If so, I challenge the need for its own section. --Renegade78 (talk) 13:02, 7 March 2011 (UTC)
 * Havok is not a game engine; recent versions of the Unreal Engine use it. It's not a synonym. It's quite possible that the article doesn't make that clear (so that part needs to be improved). In fact, there are so many problems with this article that I don't know where to start. But that doesn't change the fact that there is a difference between Game Engine and Middleware, at least in the Video game industry (in the broader context of Computing, a game engine would be considered to be a type of middleware). -- Nczempin (talk) 13:18, 7 March 2011 (UTC)
 * So you're essentially saying that there are two possible definitions for game middleware?:
 * A piece of software used by game engines
 * Game engines themselves within the broader context of computing
 * If so, there should be an effort made within the article to point out these different meanings. --Renegade78 (talk) 16:23, 7 March 2011 (UTC)
 * "Game middleware" only has the first meaning; something that is added to a game engine. "Middleware", is a broader subject however, of which "Game engines" are an example. Yes, there should be an effort made to point out the different meanings; hopefully even including references. -- Nczempin (talk) 17:29, 7 March 2011 (UTC)
 * You changed "some middleware..." to "middleware...". I don't think that's entirely correct. What it's trying to say that there's middleware that is fairly generic, such as a complete physics engine like Havok, but there are also examples of very focused kinds of middleware, like one that only does trees (and bushes). While it is true, I'm not sure whether this falls under OR or not: Do we need to say it? Is this specific type of middleware treated in any particular way in reliable sources? -- Nczempin (talk) 19:20, 7 March 2011 (UTC)
 * Well I don't think it falls under WP:OR; it is sourced. And there definitely seems to be a distinction. Would it hurt to leave it in? --Renegade78 (talk) 19:44, 7 March 2011 (UTC)
 * What source are you referring to? In general, I am in favour of stating facts or "widely held beliefs" that eventually get sourced, with a tag until a reliable source can be provided. I am certainly not a reliable source :-) So, if I came up to the article for the first time, and what you added was the only thing I noticed, I would think "citation needed" and "possible or without citation", even though I believe it's true (since I just said it here myself) -- Nczempin (talk)
 * Quote from Middleware: Middleware sits "in the middle" between application software that may be working on different operating systems. It's a muddy term, but that's what I mean by "a game engine is middleware". Interestingly enough, that article has a disambig link to our middleware section. It would be very useful to find a source (gamasutra?) that specifically mentions this distinction, otherwise it's still OR. (something seems to be wrong with signing right now, I hope it works this time -- Nczempin (talk)
 * Oh, I thought you were referring to the SpeedTree thing, lol. Well, I definitely don't have anything against using a cit. needed tag until we find a reliable source mentioning the distinction. By definition alone though, wouldn't game engines fall under the middleware term? (Just an FYI, the wiki code/signatures weren't working because you opened a nowiki tag without closing it). --Renegade78 (talk) 18:04, 8 March 2011 (UTC)

First Person Shooter Engines
Seems like an orphaned section that has a bit of out of date ramble that doesn't deserve its own section and has no content worth integrating into the rest of the article. I'm suggesting deleting the whole section and moving the page link to the bottom of the article. Caspar esq. (talk) 03:21, 15 April 2011 (UTC)


 * To the "See also" section? No objection here. &mdash; Fr&epsilon;ckl&epsilon;fσσt | Talk 12:56, 15 April 2011 (UTC)

Changes to the list of game engines and the list of "graphics-only" engines
I moved JMonkey out of the graphics engines list and into the game engine list. I also thought that Gamebryo and RenderWare look like they should be in the graphics engine list instead of the game engine list, but I did not move them since I know little about them. I made that judgement based on the Gamebryo and RenderWare wikipedia pages, which make them sound like graphics engines. Could someone who knows more about those two engines make that decision? Thanks. 150.155.129.251 (talk) 17:48, 6 May 2012 (UTC)

My page, Game Integrated Development Environments and Card Game Engines
It would seem like my page, Game Integrated Development Environments should be taken down, and the card game builders/makers/engines and other 2D game builders moved to this page. This will take some research to write up the various tools. The distinction I was trying to make was between building/making a game and running a game (you need an engine to run a car, but motorized tools and robots to make an engine and a car. You can't use an engine to build a car). But perhaps at this point, the distinction is irrelevant.

We need to include the card engines and the grid engines on this page somehow. — Preceding unsigned comment added by 64.5.67.75 (talk) 00:27, 5 December 2012 (UTC)

My thought is the difference between a workbench and an engine. But certainly, the engine (the model) must lie below the workbench (the view into the engine). We made a distinction with that with our IDE as well. The workbench wouldn't work without the engine, but the engine could work without the workbench. Would it be possible to define an interface such that different workbenches would work with different engines? Something to think about! — Preceding unsigned comment added by 64.5.67.75 (talk) 00:42, 5 December 2012 (UTC)


 * Could that be folded into this article? It's bare-bones as it is, and includes very few sources. -- Consumed Crustacean (talk) 20:59, 17 August 2014 (UTC)

The Source engine is open source...
The Source engine is aviable for steam users using the Source SDK, All basic game codes & Source engine are download-able using this Tool, You can download Source SDK Using the Steam. — Preceding unsigned comment added by 79.183.193.153 (talk) 18:21, 13 February 2015 (UTC)

External links modified
Hello fellow Wikipedians,

I have just added archive links to 1 one external link on Game engine. Please take a moment to review my edit. If necessary, add after the link to keep me from modifying it. Alternatively, you can add to keep me off the page altogether. I made the following changes:
 * Added archive https://web.archive.org/20131017044354/http://adlcommunity.net/file.php/36/GrooveFiles/Games%20Madison/report%20Middleware.pdf to http://adlcommunity.net/file.php/36/GrooveFiles/Games%20Madison/report%20Middleware.pdf

When you have finished reviewing my changes, please set the checked parameter below to true to let others know.

Cheers. —cyberbot II  Talk to my owner :Online 08:44, 30 August 2015 (UTC)

Several issues with the recent rewrite
I'm referring to this rewrite:. Since the author now wants a reassessment based on this new version, I feel I ought to reply.

I'm trying to focus specifically on incorrect and unsupported facts, as well as organizational choices; I don't think I'd actually keep any of the text that's present in the article at this moment as the style isn't summaryish enough. I don't mean to be rough with a new user. Mind, I was considering eventually boldly gutting the old article myself, I really did not like it either.


 * 1) The lede
 * 2) * Seems to imply they are, but game engines aren't necessarily strictly data-driven, even if you want to make some degree of data-driven architecture part of the definition.
 * 3) *And, there's at least one source in this article that provides a very different definition of "game engine" (the Carmack video is one). I'm sure I could dig up others. It oughtn't be defined too narrowly if our sources can't even agree.
 * 4) *Referring to game logic as "assets" is strange.
 * 5) *Suggests that assets get fed into a build system and voila, game. This is skipping out on the code side of game development.
 * 6) *Stating that game engines are a series of libraries is wrong, and it's even directly contradicted by the Carmack video, where he's specifically comparing game engines (he provides idtech and UE as examples) to game libraries (bringing up Criterion and Argonaut).
 * 7) *The "modular" vs. "monolithic" line is big. It's also a synthesis. You can't link to the documentation for UBT to support this.
 * 8) *Yes, Engines have APIs. They're software. A framework is not an API, they are not interchangeable terms. The term framework is thrown around later on in strange ways as well.
 * 9) History surely needed to be trimmed, but this seems to have gone a bit too far and dropped anything that might have been useful from the old version.
 * 10) *Again the assets are all you need to make a game thing
 * 11) Agents
 * 12) *Uses in-line external links.
 * 13) *Isn't correct. Unity's GameObject is equivalent to UE's AActor; Unity's Object is equivalent to UE's UObject.
 * 14) *Doesn't cite any sources or provide any other clarity on why this should be its own section or why the reader should care. This ought to be gracefully integrated into some other architectural section, especially considering the last bit on composition vs. inheritence, which seems out of place under this heading.
 * 15) *Nystrom, Bob. "Component". Does not even attempt to support the assertion it's being used for.
 * 16) Main Game Loop
 * 17) *Generally, again, a little odd sitting out here, it needs build-up and placement inside an architecture section or something.
 * 18) *"The simulations are therefore numerical" - what does that mean?
 * 19) *"To make the simulated world seem to move as quickly as the user's world" - What is the user's world? Reality? The rendered view?
 * 20) *"Frameworks will usually do the main game loop for the developers." - Frameworks again. What do they have to do with anything here?
 * 21) *"agents not detecting collisions will result in a lack of collision response, which could result in inaccurate simulation." is not why you need to fix physics, and the Semi-fixed timestep section doesn't say why you need to either.
 * 22) Optimization
 * 23) *Generally. Why does this exist? Also a very confusingly-written section.
 * 24) *The RAM statement generally. Also disregards the nice points that its provided source actually makes RE: CPU speed vastly outpacing RAM latency improvements.
 * 25) MySQL and Unreal's Editor are equated at some point.
 * 26) I stopped here. The rest seems to be stubbed out so there's not much to discuss other than the layout / organization, which I'm not sure about yet.

I do appreciate that a lot of the synthesis and irrelevancy has been stripped from the article.

I'd like to help more, though my current thoughts on "help" do amount to undoing much of what's new in the article. -- Consumed Crustacean (talk) 02:52, 23 September 2015 (UTC)

Thanks for your feedback. I have tried to make some improvements, but I agree the style isn't quite right. Feel free to revert Alanhaugen (talk) 06:47, 23 September 2015 (UTC)

This article needs a section on the state of the art
Speaking as a gamer with a casual interest, and therefore a Wikipedia reader not qualified to contribute usefully to this page, I was disappointed there was nothing that identified the most popular current game engines, even as a jumping off point for further reading. I'm aware of Unreal, Source, CryEngine, but there's probably a handful of others that underpin highly popular current games, or a large number of them. But from this article, it's impossible to know. I appreciate going into this raises a page maintenance problem to keep it current, and might introduce further challenges with NPOV and puffery. However, an article on "Game engines" that doesn't tell the reader what the most important current game engines are is of limited usefulness.217.155.66.190 (talk) 11:03, 14 November 2016 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified one external link on Game engine. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20110108073502/http://www.m-trends.org:80/2008/01/mobile-and-wireless-trends-for-2008.html to http://www.m-trends.org/2008/01/mobile-and-wireless-trends-for-2008.html

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 15:31, 7 January 2017 (UTC)

Proposed merge of Free game engine into Game engine
Very little if any independent coverage of "free game engines". Searching WP:VG/S, general google searches, Scholar search, the term isn't really discussed by reliable sources. Much of the article is unrelated sourcing that is handling WP:V for individual engine details, not the topic in general.

The few pertinent details can be merged into a paragraph or two of the main article, if suitable. But there may be little available to merge. For example, the first paragraph of "Advantages and disadvantages" is presented as applying to "free game engines", but in actuality is sourced to an article about Godot. The pros/cons are present as being about Godot, not all free game engines. The second paragraph? Almost entirely unsourced/OR. -- ferret (talk) 19:01, 12 June 2022 (UTC)
 * Merge Per nomination. The section on "Advantages and disadvantages" reads almost like someone was trying to make Free and open-source software fit specifically for video games, but could only source it to Godot. I concur that what content there is that is actually sourced, could be merged into Game engine in a paragraph or two. Sideswipe9th (talk) 19:58, 12 June 2022 (UTC)
 * Merge. Sources do not describe libre game engines as independently notable from the general concept of game engines. Should be covered in summary style and only split out when warranted by an overabundance of sourcing. czar  23:37, 12 June 2022 (UTC)
 * Merge per above. ~ Dissident93 (talk) 07:21, 4 August 2022 (UTC)

No opposition after nearly 2 months, I have redirected free game engine. There was no specific content to merge that had suitable sourcing, as noted in the nom statement. -- ferret (talk) 13:03, 4 August 2022 (UTC)

replace image of Blender Game Engine
I don't think it's a good idea to use Blender Game Engine as the only pictorial representation of a game engine on this page. It just distracts people to click on the link to find out what the heck it is, only to find that it was removed from Blender back in 2018. It would make a lot more sense to use an image of a contemporary leading engine (or a groundbreaking early one). How about an image of UE or Unity or something? DKEdwards (talk) 00:56, 31 July 2022 (UTC)


 * I agree. I did a quick search on Commons and found an image of the Godot editor. I've now the Blender image in the article with the Godot one. Sideswipe9th (talk) 01:07, 31 July 2022 (UTC)
 * I disagree with replacing it with Godot's editor image. Godot Engine is not widespread nor representable in the game development industry in contrast to other game engines out there. Therefore, I agree with @DKEdwards's idea to use an image from a game engine like Unreal Engine or Unity. If you disagree with adding a screenshot from engines like Unreal Engine or Unity because of blatant advertisement, then adding Godot can also be considered as such if we follow this kind of logic. People who are interested to know about game engines should better be exposed to industry-standard tools used by professional game/engine developers and game designers, and not something which has alleged potential or attracts cult-like followings caused by popularity bias, which is typical to Godot community. If Godot's editor image was picked because it's open-source, then this criteria is more ideological rather than pragmatic.
 * Taking into account all above, I propose to add several screenshots: one from an open-source game engine (Godot), and another from a commercial one (Unity). I guess this is the reason why "Free game engines" page used to exist in the first place. The "Game Engine" page should not discriminate against open-source nor commercial game engines. Xrayez (talk) 11:42, 9 September 2022 (UTC)
 * No opposition. I added an image of Unreal Engine (Blueprints). The rationale is that Godot removed visual scripting starting from Godot 4.0. Xrayez (talk) 16:38, 11 September 2023 (UTC)

Wiki Education assignment: Digital Media and Information in Society
— Assignment last updated by Stevesuny (talk) 14:20, 16 October 2023 (UTC)

Editing possibilities
Hey everyone,

I would like to let you guys know what I am planning to edit this article to see if we are all in the right page. First off, i realized that programming languages are barely talked about in this article, so i would like to add more about the different programming languages that game developers use. I would also like to dabble a little bit more in Virtual and Augmented reality, to see what game engines are used for those realities. One final thing is I would like to improve the Game engines as an Industry section, since there is only one paragraph there and i would like to expand that.

That is all the plans I have for right now, I might add more, but i will let you guys know. Also let me know if you want something else re-edited in the article, I can have a look.

Thanks for reading,

Hunter Kaminski (talk) 14:18, 1 November 2023 (UTC)

Edit finished
Hey everyone

I wanted to let you guys know that I have finished editing the article.

I added onto the game engines as a industry, plus added a new section discussing the future of game engines

If you would like too add on to it, or completely get rid of it, be my guest, this is a open source after all. Im just letting you guys know

Hunter Kaminski (talk) 01:32, 8 December 2023 (UTC)