daily menu » rate the banner | guess the city | one on oneforums map | privacy policy | DMCA | news magazine | posting guidelines

Go Back   SkyscraperCity > World Forums > Architecture

Architecture news and discussions on all buildings types and urban spaces
» Classic Architecture | European Classic Architecture and Landscapes | Public Space | Shopping Architecture | Design & Lifestyle | Urban Renewal and Redevelopment



Global Announcement

As a general reminder, please respect others and respect copyrights. Go here to familiarize yourself with our posting policy.


Reply

 
Thread Tools
Old August 16th, 2011, 09:04 AM   #1
Andros_Pelekanos
Registered User
 
Join Date: Aug 2010
Posts: 14
Likes (Received): 1

Implausible architecture/infrastructure in games

Does it bother anyone else?

Ok I'm not good with techy stuff and thus an image which is in order isn't in this post to help demonstrate my point but...

The point is that the layout of the subway system in GTA4 is completely non-sensical especially when you consider where the stops are. As if the system was put in place purely as a gameplay system rather then for practical and functional reasons.

Other examples include Doom 3, Half Life 2 & Dead Sapce where you're in closed indoor areas and on the roof and on the walls you'll see piping. But following the pipes leads to literally nowehere, they just stop.

Am I the only person who's bothered by the fact that game developers don't hire real architects for games?

With games costing over 40 million to makes these days, surely hiring one legit architect firm wouldn't completely break the bank... or would it?
__________________

mrsmartman liked this post
Andros_Pelekanos no está en línea   Reply With Quote

Sponsored Links
Old August 16th, 2011, 08:29 PM   #2
Gherkin
actual gherkin
 
Gherkin's Avatar
 
Join Date: May 2005
Location: London
Posts: 13,796
Likes (Received): 515

I'd imagine games developers want to create alien worlds where physics is not the same as on earth, to make levels more engaging and to encourage exploration of these worlds. Stretching our imagination is certainly not a bad thing.

Games designers do hire people with architectural backgrounds when it comes to making (for instance) war games or those based on historic events. These require historically accurate buildings down to every last piece of steel/brickwork, whether it is intact or in pieces after a bombing.

Concept artists are a better choice for creating alien environments but architects are better when it comes to creating future environments whether they are post-apocalyptic/utilitarian/eco or not.

Factory Fifteen have architectural backgrounds and produce some amazing work:
http://www.factoryfifteen.com/
Gherkin no está en línea   Reply With Quote
Old August 17th, 2011, 11:22 AM   #3
Yatta
Registered User
 
Join Date: Jul 2011
Posts: 9
Likes (Received): 1

Well, as an amateur level designer, I can give some insight into why this might be the case. It is partly to do with the workflow, partly to do with the nature of gameplay, and partly to do with computer/console restrictions (optimization and rendering- I'll explain these later).

You see, when you design any level in a game, you'll pass through a series of phases. Typically they are as follows:
1) An "untextured" (everything will look grey, orange, etc) version of the level with 0 detailing- just the basic layout of the area. We're talking literally a box stuck on a plane with a hole punched in the wall for a door, and maybe some walls inside that box for makeshift rooms. This is simply to get an idea for how the level will play. Furthermore, depending on the technology the game uses, how well the level will run on a computer is constantly considered. For instance, due to the way the Source engine works, you do not want to be able to see from one end of a map to the other. I'll get to this later under the technical ("optimizing") section.
2) From here, depending on the situation, this basic level might be detailed slightly (walls are "painted" with textures, the floor is turned into grass, etc).
3) Seeing as the emphasis at this stage is still purely on gameplay, not surprisingly, this step is testing how the level plays. At this stage, no one cares about aesthetics except for basic stuff to prevent the play testers from getting lost.
4) The level designer is given feedback and the process jumps back to step 1 until the map is sufficient to whatever standards are in place. Usually there will be quite a few iterations of this (fairly) un-detailed, UGLY level before the level continues onto the next step.
5) Finally, we get to see some detailing. The map is passed to an artist, who will give recommendations on how to decorate the map. Depending on the situation, the level designer might be the artist or the artist might give aesthetic input at earlier stages of level development, etc. But typically, aesthetics are reserved for the later stages. Gameplay comes first- bad gameplay can ruin a game title and endanger the profitability of that game.
6) The level is optimized. Again, I'll talk about optimizing in a second.

As you can see, this workflow doesn't allow for much input for art consideration in the form of architectural consultation. You can say that creating a level is more akin to designing a performance car over an actual city space. Technical and gameplay (car analogy: performance) trump aesthetics. The aesthetics are applied near the end of level development.

On to the topic of the nature of gameplay, you need to remember that there's a large range of people who play video games. Levels need to be very intuitive to the player and not have too much "extra" to prevent confusing the player. Plenty of players will, if they see an accessible side passage, run down it only to get lost, even if it's a simple hallway with nothing interesting in it. People get lost very very very very easily, even in the most linear of environments. Yes, it sounds silly, but it's the truth. As such, nonsensical aesthetics need to be applied time to time: lack of extra doors, lack of side rooms and passages, lack of any "fluff" for the most part. Furthermore, having "real" settings can cause a lot of other kinds of negative fluff: too much walking, not enough cover (or too much cover), etc. It'll break the game flow. Imagine getting into a really gnarly firefight, then having to cross a big patch of grass with nothing going on. The gameflow is sacrificed. If you want a real-game example, the first actual level of Deus Ex does this. You fight/sneak your way to the top of a tall building, complete the objective, then you have to walk... and walk... and walk... and walk... and walk... until you get back to the bottom floor with absolutely NOTHING going on, and then you have to walk... and walk.. and walk... until you find your headquarters. Booooring! So as you can see, "real" architectural/urban planning design can create "unfun" environments in which to play in.

Finally, optimizing. This is a big one. Fluff in a video game, that is, any unneeded gameplay areas as well as unneeded decoration, is very bad. It is very very bad. Compare a level to a cookie. And the console/computer, your mouth! And when you eat the cookie, you need to eat it in one whole bite. This is an analogy for loading up a level into the computer. You need to load it from the hard drive/disc into RAM in a big chunk. So naturally, when you add fluff to a game, it's like adding more to the cookie. Yes, some is good, but it can quickly get to be too much. Soon enough, if you try to load that map, or eat that cookie, you run into problems. The cookie is too big, and you choke. The map is too big, and it takes forever to load, which is something players do not like to see. This is why games with large, meandering maps are so rare. It's due to this loading problem. The act of reduce the size of the map, whether it be by removing unnecessary detailing or unneeded rooms, is a function of optimizing. This is what I meant by the last stage of development. The level designer might literally be thumbing through his level, looking for things the player probably won't notice, and deleting them. This makes the map "cheaper," easier for the computer to "digest." Have you ever noticed the lack of large forests in video games? Yeah. This is why. Trees are expensive (hard to "digest"), and forests are typically meandering, which is also expensive.

These same restrictions apply very much so to how a level is laid out. Some technologies handle the user seeing a lot of stuff better than others. This is why you can't see very far in Source games, whereas in Unreal games, you typically see far more. Depending on the technology used, this distinction can result in extreme layout and aesthetics restrictions (lookin' at you, Source engine!). This is also why you see buildings, telephone poles, etc randomly pop up as you progress through a level. This is all to make it so the computer/console can load and handle everything in a map without slowing down. This is the other half of optimization: telling the computer what to render (make visible) and what to keep invisible. Since a computer/console can only handle so many objects at once, proper optimization is key.

Naturally, following that explanation on technology, optimization, and the effects of loading and rendering, you can imagine why there is a lack of architectural firms working with video game developers. Any areas designed by architecture firms would need to be conducive enough to optimization. Else, you'll risk having an unplayable product.

Overall there's very little wiggle room for artistic input as you can see. Unless the architects are familiar with the technology being used, any architecture firms hired would be a waste of money with obvious exceptions (historic recreations). You're better off just using concept artists to decorate pre-designed levels instead.

As I said, designing a level is more like designing a car over designing a city space. Almost entirely technical work.

I hope this made sense!
__________________

Dunzdeck liked this post
Yatta no está en línea   Reply With Quote
Old August 17th, 2011, 04:51 PM   #4
No Change No Future
Registered User
 
Join Date: Oct 2005
Posts: 1,865
Likes (Received): 736

^Great post. Do you think game technology itself should just change? Games haven't changed that much since the jump to 3d, of course graphics, physics, and gameplay has improved, but we still have to face this technical limitations even in today's games.

I've always fantasized about the existence of a different system of three-dimensional rendering that doesn't use polygons and instead uses particles, or atoms, or something more akin to real life, something round with no sharp edges, and where there are no limitations to it, it doesn't seem like the polygonal system will really take games to full blown realism, we've been using this for 20 years and games are still full of blocky geometry because of limitations, I know our real life illusion is based off of geometry too, but to duplicate it, it would take infinite amounts of polygons, there has to be a better way.
No Change No Future no está en línea   Reply With Quote
Old August 17th, 2011, 11:16 PM   #5
Gherkin
actual gherkin
 
Gherkin's Avatar
 
Join Date: May 2005
Location: London
Posts: 13,796
Likes (Received): 515

This was posted in the UK forums recently, which I presume you're referring to:

Quote:
Originally Posted by wjfox View Post

Yatta I'd love to know your thoughts on this. Great post by the way!
Gherkin no está en línea   Reply With Quote
Old August 18th, 2011, 07:15 AM   #6
Manila-X
PINOY MOD
 
Manila-X's Avatar
 
Join Date: Jul 2005
Location: MANILA
Posts: 14,384
Likes (Received): 2570

Quote:
Originally Posted by Andros_Pelekanos View Post
Does it bother anyone else?

Ok I'm not good with techy stuff and thus an image which is in order isn't in this post to help demonstrate my point but...

The point is that the layout of the subway system in GTA4 is completely non-sensical especially when you consider where the stops are. As if the system was put in place purely as a gameplay system rather then for practical and functional reasons.

Other examples include Doom 3, Half Life 2 & Dead Sapce where you're in closed indoor areas and on the roof and on the walls you'll see piping. But following the pipes leads to literally nowehere, they just stop.

Am I the only person who's bothered by the fact that game developers don't hire real architects for games?

With games costing over 40 million to makes these days, surely hiring one legit architect firm wouldn't completely break the bank... or would it?
As for GTA-4 since Liberty City is based on NYC, the game does have a subway system despite the player relying mostly on the automobile to get from point A to point B
__________________
Manila X-Perience, My collection of images around Metro Manila

Representing The Pinoy Community here in SSC!
Manila-X no está en línea   Reply With Quote
Old August 18th, 2011, 03:41 PM   #7
Yatta
Registered User
 
Join Date: Jul 2011
Posts: 9
Likes (Received): 1

Quote:
Originally Posted by No Change No Future View Post
^Great post. Do you think game technology itself should just change? Games haven't changed that much since the jump to 3d, of course graphics, physics, and gameplay has improved, but we still have to face this technical limitations even in today's games.
A fundamental change in the game technology used wouldn't do much. Ultimately technological limitations are caused by the very nature of modern computer architecture. We'd need a huge leap or fundamental change in hardware to allow us to greatly downplay/skirt around the technological limitations to a point where they don't matter much at all, which we're working towards. Everything virtual is still restrained by the real, physical world, unfortunately. Like I said, we're working towards bettering our technology, which you can see in the jump between 2D and 3D.

Now, I believe the jump from 2D tech to 3D tech was associated with an increase in CPU word size from 16-bit and 24-bit to 32-bit and 64-bit. In laymen's terms, that means consoles were suddenly able to take in, work with, and spit out about DOUBLE the information as their predecessors, if not more! This was a big jump. Hardware wise, we're seeing something similar in terms of technology today with the introduction of powerful multi-core processors. But really, are these improvements benefiting game technology? (One can't help but wonder- technology seems to be meandering towards network-based solutions rather than remaining client-based.)

As you can see, raw computing power isn't a problem when it comes to visual technology and optimization. So why do we still have this problem of optimization and limitations on graphics? The problem is that game developers are using increasingly demanding components for their creations (more polygons, more special effects, etc. And to be honest, I think we'd have the same problem with voxel tech...). As a result, graphics get prettier, yet the pressing need for optimization is ever-present. I'm sorry if I gave the wrong impression before. This is just such a complex subject at times, especially if you want to get into the nitty gritty of it.

Anyways, regarding modern graphics and fluff, take a gander at Crysis 2. Lovely, isn't it? Requires a multi-core processor, it does! Now, what's really interesting is the technology in that. It's important that I make a distinction right now: in graphics technology, you do not emulate the real world. You merely fake it. Take a cobblestone wall, for instance. You do not create a rough surface made out of stone. Well, you can, but it's very inefficient. Rather, you take a flat, easy-to-digest-for-the-computer surface and stick cobblestone "wallpaper" on it which has simply been altered so it merely looks like it's got grooves, grout, etc. This is called bump mapping (or some variant of it). This is a technology to "fake" a surface: it's cheap, and it's effective.

Back to the video. By using this "faking" technology, Crysis has managed to allow for a lot of visual "fluff." The foliage is most likely flat "plates" with a bunch of leaves attached to them bolted onto a branch. It just looks realistic, swaying in the wind. The shiny plastic wrap on some objects is all pre-"painted" to simply look like it's shiny when it's infact not, I'd be willing to bet. Very little of that shimmer is actually done ingame. Stuff like this. It's all FAKE! By using this faking technology, Crysis has managed to allow for more detailed environments. It's smart use of limited resources. Yet despite this, optimization is still needed. It's needed to fit more pretty into less space. And in the case of the clip I've demo'd, optimization comes in the form of reusing certain things over and over again in creative ways to make them not appear alike (building sidings, small structures like portapotties, etc.). Not to mention "faking" technologies. I can't help but wonder how many of those windows are real, and how many are parallax mapping or some similar technology. But really, I've never worked with the CryEngine, so *shrugs*.

But this is all for polygon-based technology. What about Voxel technology, as was hinted at by No Change No Future and mentioned by Gherkin? I really don't know much other than that the technology is confined to very small (hobbyist?) developers and is still in its infancy as far as game applications are concerned. As for Unlimited Detail, if you're familiar with indie game devs, Notch wrote a very interesting blog post on Unlimited Detail, touching on modern voxel tech as well. Basically, Unlimited Detail is, in his words, "a scam." They're dressing up a technology to make it look far better than it actually is. From what I gather from his post, voxel technology at this stage is very good at making lots of the same detailed statue lots of times and nothing more, which is what we saw in the demo.

Now, assuming technology does improve to a point in the future where optimization in any technology is made redundant- computers have become SO beastly that we don't need to care about how much data is being handled at once, I still think architecture firms would need to be very specialized to be employed in game development for the reasons I outlined in my previous post.

But who knows for sure? One day, if we're lucky, game development might look a lot less like this and this, and a lot more like this. Here's to hoping

(And since it's now 5:40 AM, I think it's time to hit the sack. Sorry if any part of this didn't make sense- I'm somewhat exhausted right now! )

Last edited by Yatta; August 19th, 2011 at 01:39 AM. Reason: Mislabeled 16-bit systems as 32-bit and 32-bit as 64-bit... *facepalm* what a stupid mistake!
Yatta no está en línea   Reply With Quote
Old August 19th, 2011, 11:45 AM   #8
Andros_Pelekanos
Registered User
 
Join Date: Aug 2010
Posts: 14
Likes (Received): 1

Quote:
Originally Posted by Manila-X View Post
As for GTA-4 since Liberty City is based on NYC, the game does have a subway system despite the player relying mostly on the automobile to get from point A to point B
Yes it does have a subway system but for the life of me I can't remember what it is about the metro that simply didn't make sense from architectural point of view. I'll install the game again soon when I have some time and maybe put my grievance to paper this time.
Andros_Pelekanos no está en línea   Reply With Quote
Old August 19th, 2011, 12:13 PM   #9
Andros_Pelekanos
Registered User
 
Join Date: Aug 2010
Posts: 14
Likes (Received): 1

Quote:
Originally Posted by Yatta View Post
Well, as an amateur level designer, I can give some insight into why this might be the case. It is partly to do with the workflow, partly to do with the nature of gameplay, and partly to do with computer/console restrictions (optimization and rendering- I'll explain these later).

You see, when you design any level in a game, you'll pass through a series of phases. Typically they are as follows:
1) An "untextured" (everything will look grey, orange, etc) version of the level with 0 detailing- just the basic layout of the area. We're talking literally a box stuck on a plane with a hole punched in the wall for a door, and maybe some walls inside that box for makeshift rooms. This is simply to get an idea for how the level will play. Furthermore, depending on the technology the game uses, how well the level will run on a computer is constantly considered. For instance, due to the way the Source engine works, you do not want to be able to see from one end of a map to the other. I'll get to this later under the technical ("optimizing") section.
2) From here, depending on the situation, this basic level might be detailed slightly (walls are "painted" with textures, the floor is turned into grass, etc).
3) Seeing as the emphasis at this stage is still purely on gameplay, not surprisingly, this step is testing how the level plays. At this stage, no one cares about aesthetics except for basic stuff to prevent the play testers from getting lost.
4) The level designer is given feedback and the process jumps back to step 1 until the map is sufficient to whatever standards are in place. Usually there will be quite a few iterations of this (fairly) un-detailed, UGLY level before the level continues onto the next step.
5) Finally, we get to see some detailing. The map is passed to an artist, who will give recommendations on how to decorate the map. Depending on the situation, the level designer might be the artist or the artist might give aesthetic input at earlier stages of level development, etc. But typically, aesthetics are reserved for the later stages. Gameplay comes first- bad gameplay can ruin a game title and endanger the profitability of that game.
6) The level is optimized. Again, I'll talk about optimizing in a second.

As you can see, this workflow doesn't allow for much input for art consideration in the form of architectural consultation. You can say that creating a level is more akin to designing a performance car over an actual city space. Technical and gameplay (car analogy: performance) trump aesthetics. The aesthetics are applied near the end of level development.

On to the topic of the nature of gameplay, you need to remember that there's a large range of people who play video games. Levels need to be very intuitive to the player and not have too much "extra" to prevent confusing the player. Plenty of players will, if they see an accessible side passage, run down it only to get lost, even if it's a simple hallway with nothing interesting in it. People get lost very very very very easily, even in the most linear of environments. Yes, it sounds silly, but it's the truth. As such, nonsensical aesthetics need to be applied time to time: lack of extra doors, lack of side rooms and passages, lack of any "fluff" for the most part. Furthermore, having "real" settings can cause a lot of other kinds of negative fluff: too much walking, not enough cover (or too much cover), etc. It'll break the game flow. Imagine getting into a really gnarly firefight, then having to cross a big patch of grass with nothing going on. The gameflow is sacrificed. If you want a real-game example, the first actual level of Deus Ex does this. You fight/sneak your way to the top of a tall building, complete the objective, then you have to walk... and walk... and walk... and walk... and walk... until you get back to the bottom floor with absolutely NOTHING going on, and then you have to walk... and walk.. and walk... until you find your headquarters. Booooring! So as you can see, "real" architectural/urban planning design can create "unfun" environments in which to play in.

Finally, optimizing. This is a big one. Fluff in a video game, that is, any unneeded gameplay areas as well as unneeded decoration, is very bad. It is very very bad. Compare a level to a cookie. And the console/computer, your mouth! And when you eat the cookie, you need to eat it in one whole bite. This is an analogy for loading up a level into the computer. You need to load it from the hard drive/disc into RAM in a big chunk. So naturally, when you add fluff to a game, it's like adding more to the cookie. Yes, some is good, but it can quickly get to be too much. Soon enough, if you try to load that map, or eat that cookie, you run into problems. The cookie is too big, and you choke. The map is too big, and it takes forever to load, which is something players do not like to see. This is why games with large, meandering maps are so rare. It's due to this loading problem. The act of reduce the size of the map, whether it be by removing unnecessary detailing or unneeded rooms, is a function of optimizing. This is what I meant by the last stage of development. The level designer might literally be thumbing through his level, looking for things the player probably won't notice, and deleting them. This makes the map "cheaper," easier for the computer to "digest." Have you ever noticed the lack of large forests in video games? Yeah. This is why. Trees are expensive (hard to "digest"), and forests are typically meandering, which is also expensive.

These same restrictions apply very much so to how a level is laid out. Some technologies handle the user seeing a lot of stuff better than others. This is why you can't see very far in Source games, whereas in Unreal games, you typically see far more. Depending on the technology used, this distinction can result in extreme layout and aesthetics restrictions (lookin' at you, Source engine!). This is also why you see buildings, telephone poles, etc randomly pop up as you progress through a level. This is all to make it so the computer/console can load and handle everything in a map without slowing down. This is the other half of optimization: telling the computer what to render (make visible) and what to keep invisible. Since a computer/console can only handle so many objects at once, proper optimization is key.

Naturally, following that explanation on technology, optimization, and the effects of loading and rendering, you can imagine why there is a lack of architectural firms working with video game developers. Any areas designed by architecture firms would need to be conducive enough to optimization. Else, you'll risk having an unplayable product.

Overall there's very little wiggle room for artistic input as you can see. Unless the architects are familiar with the technology being used, any architecture firms hired would be a waste of money with obvious exceptions (historic recreations). You're better off just using concept artists to decorate pre-designed levels instead.

As I said, designing a level is more like designing a car over designing a city space. Almost entirely technical work.

I hope this made sense!
That's an absolutely awesome post man. A big thank you for that well defined explanation.

Not only did you help me understand the processes involved that ultimately lead to skimping on architecture but you helped me realize that most triple A games today use the same paradigm of thought on development.

And from what I understood from you is that it's not the level of technology that's necessarily at fault (although optimization can have negative effect) but rather the system of game construction. In other words the current system of game development does not lend itself well to architectural considerations.

My question to you is this:
From your point of you, can you think of a system of game construction that does not marginalize architectural considerations to such a large degree?

I mean to me it's pretty simple. Architecture is really important and makes a world of a difference to the immersion and subsequently to the believability of an environment. But I'm no game or level designer so my natural unbias (if you can call it that) would not allow me to make a convincing argument.

The fact is that only you and people in the same line of work as you know what the challenges of refining the work flow to incorporate architecture are.

So I'd be immensely interested to see if you could think out loud about what changes and what compromises would need to happen in development workflow – especially on workflow you described - in order to make architecture a success in a modern triple A game. So much so that critics would take notice and increase the gamerankings score.

I imagine better hardware and more ram would help but what else?

Maybe it might be worth mentioning both Red Factions games - Guerrilla and Armageddon – where building architecture that could be self sustaining and was subject to regular gravitational stresses was built into the physics system itself.
Andros_Pelekanos no está en línea   Reply With Quote
Old May 5th, 2016, 08:13 AM   #10
mrsmartman
Registered User
 
mrsmartman's Avatar
 
Join Date: Mar 2015
Posts: 1,366
Likes (Received): 1442

The point is that the layout of the light rail system in GTA5 is completely non-sensical especially when you consider where the stops are. As if the system was put in place purely as a gameplay system rather then for practical and functional reasons.
mrsmartman está en línea ahora   Reply With Quote


Reply

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Related topics on SkyscraperCity


All times are GMT +2. The time now is 03:13 PM.


Powered by vBulletin® Version 3.8.11 Beta 4
Copyright ©2000 - 2017, vBulletin Solutions Inc.
Feedback Buttons provided by Advanced Post Thanks / Like (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.

vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.

SkyscraperCity ☆ In Urbanity We trust ☆ about us | privacy policy | DMCA policy

Hosted by Blacksun, dedicated to this site too!
Forum server management by DaiTengu