Two rants on level design, need opinions

[Hourences]

Right, I wrote two rants on level design. one about the future of the level design community and one on companies who split up level design

Im using this forum as a testing ground to see to see how other people and level designers think about it and react ont it before i might post it somewhere else.

Prolly creates some hot discussions

Some discussions have already taken place partially in the past.

Please correct me if Im wrong with the valve half life 2 contest example, I believe the thing I said was in the rules but im not sure.

Whatever the case, lets see what gives :

First

The future of the Level Design community

Level design is becoming increasingly complex. The increasing polycounts and texture sizes as well as the introduction of materials and advanced shaders such as normal and parallax mapping are great for the quality of games but they also raise the difficulty level sharply. It is becoming more and more difficult for beginners to get a grip on level design as a whole. Level design is in the process of becoming too complex for the average joe. Every element of level design such as architecture or lighting is starting to require very specific skills and a lot more required time and effort making it only harder for beginners to start with it or even for intermediates to work on.

And that is endangering level design as a whole and especially the hobbyist level design community. Level design is going towards a situation where all user created level are made up out of nothing more than prefabs. Prefabs are bad. They may help your level but on the long term they are going to decrease the overall quality hugely. Not only will all the levels created look alike they will also prevent new people from learning some very important aspects of level design. Architecture will be forgotten, as will be texturing and materials and atmosphere, style and theme wont be up the designer to decide anymore as he can only use what he has available. A light version of Level Design would start to emerge, without some of those very important elements all there will be left are hundreds of standard looking environments.

This happened to a great extend when Half Life 2 was released. The first year most released levels all looked exactly like the game itself. Same textures, models, theme, and so on. While such levels might be interesting the first month you will get really bored with the theme after half a year and start ignoring the mod community.

All of that has wider consequences too. It is going to cause a brain drain in the community. The skilled and experienced people are going to leave after a while because hundreds of beginners who can do nothing more than placing prefabs in the same old style and feel while will be of little interest for them. Without that group of people a serious amount of skill will disappear from the community as well as the possibility for those beginners to get help on advanced subjects and learn from the older more experienced people. Which in its turn would lead to another decline in skill and quality.

People should be motivated to use new content to create their levels with if we want to try and minimize the effects of this. Motivation can be given trough contests, tutorials, correct reviews of levels, and so on. If people's effort of learning how to create custom content and using it does not give them an advantage they will stop doing it. Valve made that mistake for example. After Half Life 2 was launched they organized a level design contest of quite a lot of money. While that was good for their level design community the only problem was that custom content was not allowed. They were sending out the signal that using prefabs is good and can even pay off and that is exactly not what you want.

Relying too much on prefabs is going to cause long term problems and a strong decline in quality and skill. Which in turn could hurt companies behind games such as Valve. The modding communities are a huge potential employee farm for companies. Companies hire lots of people from modding communities because of the ever increasing demand for more employees. If the skill level and quality in the community lowers however it will be a lot harder for a company to find skilled people who truly understand what they are doing. And that again is going to hit back at the community. They will experience less chance and less attention to what they are doing.

The higher the overall quality and skill level the people in a community have the more outside interest they will create. And the other way around.

Second

Splitting Level design Rant

More and more companies lately split level design in to multiple smaller aspects. Some only do this in a minor way by just splitting gameplay, sound and everything else over multiple people while others split everything they can resulting in different people for concept, geometry, texturing, lighting, gameplay, sound, special fx and so on. I am against that.

Level design is one of the most complex elements in a game. It is a combination of a wide range of different things such as architecture, materials and textures, lighting, gameplay, sound and special fx. It is that combination that makes it complex. You need to control every aspects on itself plus keeping the overview on the whole. All elements must be in harmony with the rest of the level. Making a good looking and well playing level isn't that easy. Any element in a level is directly and indirectly influenced by another. Any choice you make will influence the rest of the development. For example you cannot texture a world with blue textures when you intend to use lots of yellow lighting because those two do not match each other. The balance and harmony between every element is one of the most important and also most difficult things in a level.

It is exactly that very important balance that is running the danger of being destroyed or disturbed when too many people work on one level. Every person has his own opinion on something, they all have their own techniques and they all have their own desires and wishes for the level they are working on. Even if the lead or leads are directing their artists and designers greatly, and even if there is enough communication between everyone as it is still bound to go wrong at some point. You will always have someone somewhere who wants to add gameplay situation X when the next person is modeling the environment with his own interpretation of that gameplay in mind. No matter how well you communicate it is always bound to give some misunderstandings somewhere or worse: fights.

Imagine how bad it is when you have to work on a level that has been designed badly or when you have to light a building that looks really bad. No matter how well you try to do it, if the base is wrong you will never get it right anymore. And that is very frustrating. It is very frustrating to not being able to fix any of the other persons mistakes because after all you are not allowed to since you are only the lighter or the level artist and it is not considered your task to fix or question such mistakes.

And that brings us to the next point. Even if you would report that mistake or in general if you simply need anything from the other person concerning the level which could be anything like a specific model you are lacking, a gameplay aspect that is unclear or a pitch black corridor that urgently needs lighting you will likely have to wait days for it if not weeks. The person responsible for the lighting might be busy working on something else. Getting him to fix it would first require you to report the bug or problem to your lead who than would have to go ask or schedule the other guy in for that task.

Whatever the case, that process takes days or weeks and in any case a lot longer than if you were in control of everything and would be able to quickly fix up the problem yourself, something which would take less than five minutes in case of the corridor light.

Another problem is that most people that do one thing have little skill in other things. For example a gameplay designer is not an artist most of the times. He does not understand how art works, what is important in it or what to look out for. Yet exactly that person will lay out the base for the visuals ? The basic composition, the theme and setting of the environment and so on ? If a problem arises you are in for an hour long discussion simply to explain the other guy why you need thing a. or b. this way. Simply because the other guy does not has understanding of what you do. That is a huge time loss and it is simply plain frustrating. The problem would not arise if you would do everything yourself and if the other people who work on levels have equal understanding of things.

I have been in lots of situations where other people who work on levels have caused me a lot of frustration or time loss. Often I am given a gameplay design that makes little sense, such as a apartment building in the middle of a jungle or designs that are far from possible. Such as giant open fields with dozens of brand new assets to create while we are in a tight deadline or a design that wont let me add occluders in it anywhere or optimize. All of those are directly caused by people who do not understand the way everything works or the overall point of level design. They lost track of the overview.

Splitting level design can let you loose track of the overview. It slows down the whole process. It causes unnecessary frustrations and hour long discussions of trying to explain unartistic designers why certain things need to be changed in order for it to look cool or be made in a reasonable time frame. It makes you wait days or weeks for some very simple bugs or problems to be resolved. You loose large amounts of time.

It often happened to me that I had to wait weeks before I receive the basic floorplan of a building while I have the skill to do it myself too, in just half an hour time...The waiting and the discussions are frustrating and pointless. A good level designer should be able to control the entire level himself. A good level designer should be well skilled in artistic aspects as well as gameplay and sound. This was how it used to work in the game industry until a few years ago some companies started to rename all the jobs. I don't even know what to call myself anymore. Am I a level designer ? A Level Artist ? An Environment Artist ? I have no idea anymore because every single company out there defines my job differently.

Level design is way too complex to simply split up.

While the increasing complexity of level design does require more people to work on one level there should always be someone who makes most of the level and keeps control over the level and the overview. The complexity may increase the time and effort, the increasingly better tools and editors available however reduce that time again. Workflow is finally starting to grow up. No more notepads to create shaders but in game menus that show real time previews of anything that happens. It balances out the time issue.

[Pericolos0]

This happened to a great extend when Half Life 2 was released. The first year most released levels all looked exactly like the game itself. Same textures, models, theme, and so on. While such levels might be interesting the first month you will get really bored with the theme after half a year and start ignoring the mod community.

Thesame thing happened with half-life 1. It took ~2 years for the real custom maps to pop up. In thesame way it'll probably be around this year the more custom styled hl2 maps and mods will be popping up

[kleinluka]

I agree with what you said about there being a lot of problems with designers not understanding the needs of the artists. I've experienced it first hand many many times. But let's face it, it really is becoming inevitable to split up level design into DESIGN and ART. Simply because creating art assets for next gen games is tremendously more demanding and time consuming and requires more specialized skills than it used to.

What I have observed as reason for most of the problems caused by the split is that the design and art side are often far too detached from one another. It is VERY important that design and art actively work together in order to get the best out of the game, designwise AND artwise. This is a big thing.

So yeah while I agree with the base of what you are saying I really don't believe that splitting the job up is avoidable for games with next generation visuals.

[von*ferret]

I've been having companies come here to SCAD and they've been talknig about splitting it up as well. One of the major ones they have spoken about is splitting up lighting. While lighting nowadays is becoming more and more intricate, I think its important that lighting be kept WITH the art team. And I hope the art team stays small and concise to keep the beaucracy down. Light has a huge effect, if not the largest, on mood. If the art team's control over that important factor is reduced then you could quite possibly ruin all continuity in the game.

[Hourences]

Very much so.

But if the texturing is crap you wont be able to lit it...

I know very well that in next gen games itll be quite hard to do everything yourself, altho Ill give it a go because I believe you can, it still shouldnt mean companies have to start splitting up the entire process. If they split gameplay from art ok, I could live with that but also splitting texturing,lighting, special fx etc is a step too far, its going to get really fragmentated and that can have consequences for the level and youll only get more meetings and more discussions...

Next gen assets could be created by multiple people but there should only be one guy who puts those assets together in the level and lights them out, not 5.

What do they do at ubi klein ? Im guessing lots of tasks are splitted ? Like lighting ?

[von*ferret]

Its like modeling. If you have the most swank model in the world and everything is PERFECT, a bad texturing job will make it a shitty model. You cant' get around that. In my mind, same with lighting. We all remember the days of HL1 with the bright skyboxes that makes everything fullbright and the hobbyists, like me, had no idea wtf was going on

[KungFuSquirrel]

Another problem is that most people that do one thing have little skill in other things. For example a gameplay designer is not an artist most of the times. He does not understand how art works, what is important in it or what to look out for. Yet exactly that person will lay out the base for the visuals ? The basic composition, the theme and setting of the environment and so on ? If a problem arises you are in for an hour long discussion simply to explain the other guy why you need thing a. or b. this way. Simply because the other guy does not has understanding of what you do. That is a huge time loss and it is simply plain frustrating. The problem would not arise if you would do everything yourself and if the other people who work on levels have equal understanding of things.

The exact same thing can and does happen the other way around, too. Designers say "hey, we need this" and get something completely wrong, or get assets just tossed to them without consideration of how they'll work in the level. Not to mention holdups when something that's critical for one level gets bumped down on the priority list behind stuff for other aspects of the game.

[Zacker]

The future of the Level Design community:

There are indeed many who smack in lots of highly detailed props while they don't really fit and the architecture around them got no proper detail level. It just looks wrong and won't work.

As a level designer the lack of custom content is frustrating, where are all the texture artists and modellers?

Splitting Level design Rant :

Agreed big time, we see it everywhere and I am not too happy about it either. I consider myself as a good level designer because I am capable of having this overview and being able to do multiple things. When they split up the role I loose my forces, I am a jack of all trades - master of none. That is how I like it and I don't think I would have fun if I was just doing lighting all day long.

[Defrag]

I agree with a lot of what you said, but the whole issue is a double-edged sword.

If you don't split it up, it's an enormous challenge to do everything by yourself. People with the ability to do a whole next gen map themselves are few and far between, especially as technology pushes on.

If you do split it up, you have a lot of potential problems.

If you were to split everything up, ideally you'd have a team populated with people who are equally comfortable in every respect of level design (gameplay, architecture, modeling, texturing, lighting, scripting etc) who all appreciate how the different things connect together and don't make totally inept decisions due to ignorance.

The main problem, for me, is that I don't see an obvious way to remedy the situation. Every HL2 map looks nearly the same because the vast majority of 3rd party mappers can't make props or texture them; valve's cryptic documentation and lack of user friendly interfaces just compounds the problem. Even if the documentation and tools were better, would it really matter, though?

I think it may just be a case of an unavoidable shift, as there's no idea solution

[ReNo]

It's sad but inevitable, as klein said. I had a guest lecturer from Codemasters come up to Abertay recently, and at one point he asked all the artists to put up their hands, and then all the programmers to put up their hands. I sort of half put up my hand for both, not knowing which of the two I really belonged to. I'm neither a programmer nor an artist in the strictest sense, but I create work that belongs to both disciplines. One of the things that appealed to me most about level design when I first got into it was the mix it offered - technical work (entity systems, scripting), design work (level layouts, gameplay ideas), and artistic creativity (architecture, lighting). Now that it is being borken up into specialised roles, which is it that I want to do?

I'm graduating in a few months, and I've no false presumtions about my chances of getting a job off the bat. With environmental artist positions requiring solid 2D and 3D art skills, I know fine well I'm not going to be a feasible candidate for these roles. On the other hand, I've not ever really done any single player level design, which detracts from my appeal as a gameplay based level designer. It's quite probable I'm gonna have to either take the QA route and get in through testing, or just grab some other menial job while I work on my portfolio.

[von*ferret]

I graduate in a couple of months too!

Reno you'll get a job. You have a great sense of mood and gameplay that is rare among artists. you'll be fine

[zaphod]

In terms of a professional setting . . .

It really depends on the scope of the project and the amount of resources that are availible for the project. Big budget games will split up job responsibilities becuase it makes sense given the time they have to make the product, the amount of content they need to create, and the amount of resources they have available to get it done. There is simply no way to make a big budget AAA game where a single person has responsibility to create an entire level - design - art - scripting - textures - lighting - and have it fit together with the rest of the game, much less on time. Say you have a game that needs 20 levels, you can try and find 5 top notch people that are very good at modeling, texturing, uv mapping, lighting, scripting, and design (a very rare find) Or you can find a designer, a lighting artist, modeler, texture artist and scripter. It's easy to find those that are good at what they do.

Smaller scope and budget games will still do fine or have to make do with a smaller team of people handling more responsibilites.

I think it's a lot more about logisitics than it is about what will produce the best result. As games get to a certain scope there is just no other way to do it. Pixar is a great example - they have different teams for every minute part of the proess - but they have perhaps the world's best directors keeping it all together.

As far as how it's going to play out in the modding community - I feel like parting out roles in some fashion similar to what professional developers are doing is the only way to get a mod or level out in a timely fashion. I think the days of Niel Manke's They hunger being done by one man with new art are numbered. Even the people that are around that can do every aspect well themselves, are going to take a much longer time to get something done. If you look at a lot of "pro" mod teams you can see this already happening, using people with specific roles to get content together and done. As I see it the only options left in terms of mapping are these:

1. Make maps for some mod team that are feeding you assets.

2. Make custom maps with a team of a few people helping to make models / textures / etc.

3. Make a map with mostly existing assets in the normal amount of time.

4. Make a map with all your own custom assets, taking 4 times the amount of time options 1, 2, or 3 take.

In my mind, 1, 2 and 3 will be the routes most people take, with smaller number of more talented people taking the lengthier route 4.

[TomWithTheWeather]

Where I work:

Level Designer - Designs and makes the maps, plans the gameplay, scripts cinematic moments, is in control of the final look and feel of the map, lighting, etc.

Artists - Make models and textures based on design requirements.

This situation has been very successful for me and other LD's I work with because we work closely with the Artists. In-the-same-office close. Looking-over-your-shoulder close.

But it really depends on what type of game you're creating. The above workflow works great for a game where the same assests will be used through out the game. It gives the game a unified look.

On the flip side, if you have LD's that make their own models and textures, you often end up with levels that don't match, which can be bad depending on the setting of your singleplayer game. Though if you're making a game like UT2k4, it really doesn't matter.

People who can do level design, modeling, and texturing are a rare group. And even then, most of them I know, suffer a little bit from "jack of all trades, master of none" syndrome.

Within the level designer role though, we've been successful in splitting the gameplay/combat guys from the story/cutscene guys. This allows me to work on MapA's gameplay while another works on MapB's cutscenes. Then we trade. As long as we keep to certain standards and make everything fairly modular, everything works correctly.

[Izuno]

Within the level designer role though, we've been successful in splitting the gameplay/combat guys from the story/cutscene guys. This allows me to work on MapA's gameplay while another works on MapB's cutscenes. Then we trade. As long as we keep to certain standards and make everything fairly modular, everything works correctly.

Personally, I think that's optimal. Of course, you really do need people in the same office working next to each other to make this work.

Only thing I'd add is that having a common vision that everyone understand is key to making it work well. That does not always happen, but what seperates good teams from great teams is the ability to recognize when the vision isn't clear and come up with some kind of resolution that keeps everyone working.

More on topic...

If you are a modder and are thinking of doing it all...I really suggest you pick one thing and master that, learn the other stuff as much as you can. Don't try to be all things to all people. The complexity of today's technology doesn't allow the time to master it all anymore, unless you can function without sleep or any kind of life.

[PS_Mouse]

While I have no industry expirience, it strikes me that a good way to do it would be to have small mapping teams, with perhaps 3-4 level designers to 1-2 texture artists. As far as the distribution of workload is concerned, it doesn't make sense to strictly divide architecture and lighting between multiple people when both tasks can be done within the same editor. Admittedly, yes good lighting does require an amount of finesse; but that's no reason why the person designing the architecture can't at the very least 'block in' the lighting before passing it on to a more talented teammate.

[EDIT] To a certain degree I think if you get a handful of level designers working together on a project, it wouldn't take too long for them to realise each other's strengths and weaknesses and distribute the workload accordingly.