Hourences Level Design Survey Results Posted

[hal]

A recent survey conducted by Hourences, asked questions of more than a thousand Unreal, Source, and Quake engine level designers and developers. Questions in the survey centered around level design from the perspective of both amateurs and professionals.

Level designers seem to enjoy today's more complex level design more than "oldschool" level design. 33 Percent of those surveyed said that it used to be more fun when games and the tools were simpler, whereas 61 percent enjoy today's technology more.

Surprisingly enough, Level designers do not seem to believe that level design communities are past their prime because of today's consoles era (70 percent). I personally believe that the various modding communities have shrunk considerably over the last few years, largely because of the lack of big PC releases.

 

[pine]

Nice survey. By far the most interesting result to me was far, far more level designers got into the industry by working on community mod projects than by graduating from a university with a relevant degree. Kind of flies in the face of those claims you see on ads from game design university programs.

 

[TheIronKnuckle]

I'd have to say that i prefer oldskool level design. Everything becomes 200% harder when you are expected to know a modelling program, texturing program AND the level design program. UT mapping - UT2k4 mapping is a great example. UT mapping only required you to know Ued if you wanted to make something unique. But 2k4 mapping demands that you include custom smeshes to be really unique and that requires knowledge of another program or five.

 

[Larkin]

Community level design not past its prime? Really? Come on, its pretty obvious it is.

 

[Armagon917]

Community level design not past its prime? Really? Come on, its pretty obvious it is.

Hourences is right heres proof.

i have a dozen high quality concept pieces and the guy working on high poly has unused zbrush work for a custom Gears of War map. we were going to create a complete new set of assets to use for it. Windows LIVE and problems with the game that shrunk/killed the multiplayer caused us to back out.

i'm not saying you're wrong. i'm just saying that developers important to the mod community have created flops for the PC recently. i will be releasing something for UT2004 in the next months hopefully.

the planter got modeled and looks fantastic in its finished state. i just wanted to show my team is very willing if there are some solid releases. EPIC has released garbage IMO as of late. i had a really great team together on this. this is my work here and i was on texturing as well. we targeted out strengths. i can't stress how great this could have been. blah. what worries me is the trend towards consoles because developers come from the mod community and get noticed through their work. if we wanted to make our Gears of War map it would have likely ended up as a nice portfolio piece instead of a busy kick ass multiplayer map for the masses to enjoy... as well as a portfolio piece but myself and the team are more concerned about entertaining people and that's what drives us. if that isn't there then motivation goes right out the window.

mapping now is very difficult, but there are people like me 100 percent devoted to it who just need a solid foundation and a sucessful game/ and environment for our work to thrive in. oh, i just love the move to not release GOW2 on PC. i think its great to keep the updated engine off PC. we would have a blast with destructible environments. as game engines get more complex and require more people some in the industry i believe are not giving the talent out there a chance and that really upsets me. i won't work my ass off for a game that sucks. COD4 is next up on the list for us as it has that foundation.

I love the poll and maybe it will help rectify or spotlight that the user community i believe is chomping at the bit to get to work. good job Hourences. i have to put in a shameless plug. buy both his books. =D they really are great.

 

[xMurphyx]

I'd have to say that i prefer oldskool level design. Everything becomes 200% harder when you are expected to know a modelling program, texturing program AND the level design program. UT mapping - UT2k4 mapping is a great example. UT mapping only required you to know Ued if you wanted to make something unique. But 2k4 mapping demands that you include custom smeshes to be really unique and that requires knowledge of another program or five.

I agree with you that 3d programs are the preferred method, but I made my map (CTF-Wasnuni) completely in UnrealEd 3.0 (except for texture-manipulations of course but that wasn't any different in UT99). It might not look quite as good but it's possible.

 

[Slainchild]

:tup:

 

[raziel31]

Thats a very interesting survey, gratz

 

[virgo47]

A large majority of those who did know (78 percent) believed that a level designer creates both the visuals and the game play for the level they are designing. In fact, very few professional level designers still have the opportunity to work on both visuals and game play, despite this popular misconception.

But there was no "they create visuals or gameplay"... I think that both areas are level designers' job. I even think that many ppl think the same and if I was asked "does proffesional level designer create the whole level alone?" I'd say... "ehm... NO".

The whole survey is nice work though, some results might be obvious, some not, but I don't know about any other survey like this. Thanks/grats

 

[TheIronKnuckle]

I agree with you that 3d programs are the preferred method, but I made my map (CTF-Wasnuni) completely in UnrealEd 3.0 (except for texture-manipulations of course but that wasn't any different in UT99). It might not look quiet as good but it's possible.

Yeah, it's still possible to make something good, sure. It's still possible to make something that looks good. But to make something that looks unique you need a 3d modelling program.

 

[elmuerte]

I sometimes think back to the days when I was a level designer... the good old days

 

[Wail of Suicide]

As much as this is a useful thing for LDs in general, I'd like to see a specific breakdown of results by community (Unreal/Source). I think the Source engine LD/modding community probably has a much different view of things because they aren't dealing with the same issues as the Unreal LD/modding community. For example, I would be surprised if many Source LDs felt excessive visual clutter was a significant issue, but for Unreal and Quake LDs I think this would be much more important.

 

[Original9]

I'd have to say that i prefer oldskool level design. Everything becomes 200% harder when you are expected to know a modelling program, texturing program AND the level design program. UT mapping - UT2k4 mapping is a great example. UT mapping only required you to know Ued if you wanted to make something unique. But 2k4 mapping demands that you include custom smeshes to be really unique and that requires knowledge of another program or five.

If you consider yourself an artist, then you should pick up a modeling program and learn it. The former generation and current allow a greater output for artistic creativity. Designing a level hasn't changed much. Sure there might be cover nodes or some other actor you have to think about when placing objects, or kismet, but it boils down to the same basics - making something fun and enjoyable.

I guess what I'm trying to say is that if you really want to design a level and make it look unique, you'll want to learn a modeling program. Probably the hardest part of learning a new program is figuring out the UI and how the controls work. Once you're past that, atleast for most people, you're golden.

 

[Armagon917]

i invite more tools to allow me to add more detail. it gives the artist more options. if you don't wan't to create these detailed rich environments go with a stylized look. in the end people only care if the gameplay is solid and its visually interesting to them. that kind of thing can be achieved many ways.

Portal, and TF2 are good examples.

its easier to learn a modeling app i find if you just jump into it.

 

[Beelzebud (Satanas)]

What is a good modeling program that isn't a crippled preview copy, or doesn't cost over $1000?

 

[elmuerte]

What is a good modeling program that isn't a crippled preview copy, or doesn't cost over $1000?

Blender?

 

[TheIronKnuckle]

If you consider yourself an artist, then you should pick up a modeling program and learn it. The former generation and current allow a greater output for artistic creativity. Designing a level hasn't changed much. Sure there might be cover nodes or some other actor you have to think about when placing objects, or kismet, but it boils down to the same basics - making something fun and enjoyable.

I do not consider myself an artist, i hate art. I consider myself a level designer, and by no stretch of the imagination could my levels be called art .
It's true, to make levels these days requires you to learn a modelling program. I don't want to.
I suppose designing a level hasn't changed much.... true. But building the level has, it now requires more skills. You don't have to use a modelling program, so that makes things easier. But it is better if you do, and that requires more training.

I guess what I'm trying to say is that if you really want to design a level and make it look unique, you'll want to learn a modeling program. Probably the hardest part of learning a new program is figuring out the UI and how the controls work. Once you're past that, atleast for most people, you're golden.

Very true.
Unfortunately I can't be stuffed to learn a modelling program right now. I haven't even figured out how to use smeshes properly. I'd want to push out at least one ut2k4 or ut3 map before i took the plunge and learnt a modelling program.
Right now i'm happy designing for UT99. I don't give a rats ass if not many people play it anymore. I make the maps for myself to enjoy because i am an offliner. I'll still go and release them on NC or mapraider and those places.
But I build levels for "Me" to enjoy, UT99 is my favourite UT, and BSP mapping is so much easier. You can probably see why i couldn't care less for moving on to a modelling program.

 

[zynthetic]

So what's keeping all these self-proclaimed non-artist level deisgners from making bsp based maps in 2k4/ut3, just because you can import complex assets?

 

[TheIronKnuckle]

true. But if i can't get my maps to look as good as the stock ones in 2k4 then i say "whats the point?". I may as well stay with ut99 where i can make it look good AND play good in comparison to the stock maps.

 

[Hourences]

Silo. Feature per cost is barely beatable by any commercial tool.

I agree! Silo is awesome and very powerful and professional, and only costs 160 dollar. One problem is that you cant export to UT with it though, but if you also get Milkshape, for like 30 dollar, you can use that as an inbetween.

As much as this is a useful thing for LDs in general, I'd like to see a specific breakdown of results by community (Unreal/Source). I think the Source engine LD/modding community probably has a much different view of things because they aren't dealing with the same issues as the Unreal LD/modding community. For example, I would be surprised if many Source LDs felt excessive visual clutter was a significant issue, but for Unreal and Quake LDs I think this would be much more important.

Yes that would have been nice but almost impossible to implement. If I need to supply specific results per modding community, per age, per pro/amateur/gamer and so on I would need over 3300 charts to display all of that (and yes I really calculated that). And that is just too much.


I will prolly do a part 2 in half a year or so.