Search

[warmeup]

I would like to discuss with you the hardest parts in level design.
As for me after 2 years of amateur experience, the hardest part in making a map is designing the map. Especially if you want to stick a theme following the plot or the storyline.
for example, i've spent the last week literally designing a deathmatch map for black mesa and coming up with concepts on a scratch paper, and yet i haven't made anything interesting.
So what do you think? what is the hardest part in level design?

[wallworm]

Excellent question, and one that pops up inevitably on all the mapping forums. I sense that you are grasping for "the best way" to make a good map.

I've seen many answers to the question that are good--but I think that the most important advice is that you don't latch onto any answer as gospel. I think that the only all-encompassing answer is that you find what works best for you in your particular task. In other words, each artist has his own abilities, creativity and perspective to account for--which may actually be dynamic because you may find that accomplishing one kind of environment may require different strategies than another.

I know of some level designers that swear by making sketches on paper first. Some like to plot the blockout on graph paper. I've used that in some experiments. Personally, my brain doesn't really distinguish drafting something on paper vs drafing it inside Max. I personally like to just jump in and block out in Max. Some of the levels I made for CS 1.6 started with a blank slate--and the theme came to me while building. Others started with a theme.

The two biggest challenges for me personally are 1) getting useful feedback and 2) the final polish.

Feedback

A decade ago my personal gaming group was entirely dedicated to playing Counter-Strike. We were a close-knit group where half a dozen of us were active map makers. We ran our own custom-map server with a constantly active player group. At that time, our gaming group was always excited about playing new maps and providing helpful feedback on the game play of map layouts. When we had that, it was really easy to improve the quality of the levels.

That lasted for a few years. And it was a blast for the group and the level designers.

But something changed in our group (and I suspect it changed in a similar fashion globally). I think that moving to Source actually killed our gaming group, in the end--not because Source was less fun but because the glue for our server (map making) became so tedious. Once Source came, the standards for what made an amazing map changed. Now maps needed cool models and custom textures--and the horrendously complex nature of making those assets in Source essentially knocked out all of the mappers in the group (some had families to feed, others were simply turned off by QC files, VMT files and compilers (outside of hlcsg.exe, hlbsp.exe, hlvis.exe and hlrad.exe (which, once set up, the old Hammer simply dealt with))). In our group, the only one that remained was me... but the whole mood changed. Since the others were no longer making maps, their interest waned. In fact, I was on the verge of quitting Source except that I was devoted to finishing at least one level--and then I got hooked on building Wall Worm.

Other problems occurred. One being that each time I made an edit in a level I was working on after moving to Source, some players started grumbling about the need to download "yet another" beta or new models, etc. The length of time it took to make maps became an annoying joke to the players who were more interested in playing rather than developing maps. Eventually, the only answers my group gave were "Yeah, it's great." But the tone was actually apathetic. Gone was the excitement our group had fed on when making maps for Goldsource (CS 1.6).

At this point there are various venues for getting helpful criticism like Mapcore and Gamebanana. But in the end I find them to be less useful and intimate than what our group used to do. I think that a lot of the feedback you get now is generally less creative than it used to be as so many people seem to have a locked formula for how a map is "supposed" to be (both visually and layout)--which doesn't fit well with my own passions for exploring outside the box and making brand new things. I also feel that too many people in the community have a higher opinion of their own aesthetic style that focuses too much on personal taste. Some people love de-saturated environments, others love a lot of color, etc. But some people don't seem to have enough humility to understand that these are simply personal preferences and should not demean other artists because their pallet tastes are different, etc.

The only useful feedback to me is the kind that helps me make my level fun for me and the people I play with. Feedback that isn't related to that is very often of little value. The absolute worst advice, in my opinion, are those comments that push you to an expected norm. Yeah, I think all mappers should be able to accommodate the norm--but if you hug that principle in your design then there is little chance for you to be an innovator. In other words, every norm we have at one point was someone being innovative and making something new (not listening to the conventions).

Final Polish

The final polish of a level is one of the hardest things to accomplish, in my opinion. I think that some artists suffer from Perfectionism that makes it impossible to finish anything--they have an internal ideal that is actually never aligned to reality. Pixel-perfect design is sometimes necessary; but very often it gets in the way of what really matters--how much fun players are having when in your level. I've noticed that a very large percentage of the details and pains designers put into levels are completely ignored by players (and often even other level designers). In other words, it's easy to get caught into the trap of spending too much time on things that don't add to the general enjoyment of the players. It's hard to discipline yourself to draw a line and stop with the detail (at times) because you can keep saying to yourself that it can still look better; the problem is that it can ALWAYS look better. But ALWAYS implies that you will work on a level forever.

The other side of polish is interest. I personally am more interested in layout than modeling. So I often get bored while making the props that fill up a level. This is the opposite of the perfectionist problem above. In this case, you might not go far enough in detailing; and a lack of detail creates an impression for users that can be negative. By detail, I mean both visual, audio and sometimes entities. For example, one thing that really bothers me in a level is when the level designer doesn't put in soundscapes and create an auditory ambience. (I personally have more fun recording sounds and creating soundscapes than I do in making little props, actually). When a level lacks little details entirely (visual and auditory) the player can get an empty feeling--even if they cannot say specifically what is missing.

The challenge can be finding the right balance between adding good details and also not being too much of a perfectionist that nothing ever gets finished.

[Joris Ceoen]

Aside from Shawn's incredible answer, one of the greatest things that have helped me in understanding a great design, in other words understanding how level environment artists like FMPone (who made cs_museum, de_gwalior and especially de_cache), Rick Underhill (cs_agency) or Skybex (cs_workout) or any other topper have achieved such great designs, is by jumping into their levels, single player, and explore every single little detail. Look at how they put things together, look at the lighting and the colors.

You will often see that there is a great consistency across all areas in the map, with just a few main base textures that kind of set up the general theme. Then, all the other items like props, containers and objects can make each area different. Rick Underhill talks about what he calls 'Hero Assets', props that are very special and instantly get noticed by the player. You place them sparingly and often alone in just one area so the player recognizes that part and it gets its own meaningfull place in the map. Details in textures are key, but you need to stick to readability and what is the essence of representing the visual picture.

The greatest thing you can ever do, is decompile these maps. You need to delve into them and carefully analyse everything the mapper has done to achieve this or that. It learns you very good tricks that you can use in your own maps. Never copy! simply carry the experience with you to achieve similar results for your own projects.

However, what it really boils down to is having a good understanding of colors, lighting and texturing. Making detailed textures is very hard, because it requires you to understand image editting and modeling very well. In Source generally Ambient Occlusion maps can be rendered from a high-poly model to make the texture more deep and vivid while adding potential detail textures or even blend textures for displacement textures. That is one great trick to give textures more depth than they actually have in the level. All detail design in Source boils down to creating illusions of depth and details. In the Unreal Engine 4, for example, most of those shaders are being rendered at real-time, so it might look more like a pain to design for Source. However, the same effects apply to both, the usage is simply different.

What Shawn says about the fact that most of the Level Designers do insane work to achieve great results that the player probably never even sees, is 100% correct. You will often be criticized for things that you as a level designer had no idea about, and they don't even mention anything about the great work you did on model X or texture Y (though if it's FMPone level like, they will most often generally applaud you for the overall looks, not anything in particular). That is most important, keeping the whole picture of your map together as you work on all the individual things. Gameplay is all that matters, and visuals extend that gameplay into a believable world. You should always think about what is the most important thing at any point in the map.

Do note that Valve has a great amount of default textures, and by decompiling maps and studying them, I've come to the conclusion that the greatest mappers will never make a texture unless Valve really doesn't have a default one available for profitable use. By that I mean that it's insane to see how, where and when some of these artists put stock textures to good use, often as a mapper not even realising that 75% of those community maps are made from stock content.

[warmeup]

Wow, I'm flattered. Thank you for your answers and for your support, It helps me a lot and keeps me on the right track.

I'm somehow having a transition where i want to make decent maps and commit my life to it. After all, my only objective is just to make people happy and to surprise them.
At the beginning I wasn't taking it seriously. I made a deathrun map just to give it a try. I knew that my work was going to be dumped in the garbage because i made the map randomly, not meeting the standards of a good level design and no color theory implied. But i did it anyways because i was bored from playing the same deathrun maps over and over again...

Until this youtube video came out.

When i saw these people laughing and having fun, I reconsidered what i want to do with my life. So all i need to do now is just practice my skills till i surprise you guys with scenic vistas and amazing gameplay, and your support guys really keeps me motivated.