10 lessons I learned in my first 45 days of mapping


(Ifurita) #1

10 lessons I learned in my First 45 Days of Mapping

I first installed SD Radiant on or about 8/31/03 and now, ~45 days later, on 10/17/03, I thought I’d look back and see what lessons I could pass on to new mappers. Keep in mind that my perspective is shaped from what I learned making 2 open city maps, one of which was scrapped and another which is getting beta tested on limited servers.

  1. Just jump in. It is not difficult to develop a reasonable proficiency in mapping, the basics anyways. Find a couple of good tutorials and just start making things and experimenting. I was able to make passable prefabs (bunkers and V2 rockets) within a few days. These led to small maps, then to a large sized map, which was ultimately scrapped, and finally to a public beta or a large-scale city map, Byzantine.

2 Start small and make Legos. Jumping in and immediately laying out an entire map is a pretty daunting task. So daunting, it can really be demoralizing. Instead, start small and make some of the small stuff that will populate your map. You can make prefab bunkers, buildings, and vehicles. Each is a discrete project that has a distinct point where you can say, “OK, done with that, what’s next?” I think this sense of accomplishment and completion is important to a budding mapper because you get to see a finished product. These smaller projects can later be used to quickly populate a new map, once you’re ready to start with an entire map.

  1. Treat tris and FPS like cash. Everything you include in your map comes at a cost. Make sure it’s worth it. Be frugal about where you make your players draw tris and lose FPS. There are great maps out there that never get a second look because they are simply unplayable. The benchmarks I used were a steady 76 FPS on my machine and r_speeds less than 20K. Tris and FPS efficiency up front, lets you add in more goodies at the backend.

  2. Learn scripting up front. Scripting is what brings your map to life. It’ll make your vehicles move, objectives destructable, constructables work etc. It seems like a lot of new mappers think of scripting as an afterthought and really only start learning the basics late in the mapping process. If you learn the scripting early, you can actually build and script all the gameplay components of your map and test gameplay BEFORE putting in all the time texturing your map and making detail. It also helps to know what you can make work, objective-wise before you build a map around something that can’t be done.

  3. Mind the details early. Get in the habit of considering the boring details as early in the process as you can. It will save you tons of time. Making a paned window and forgot to caulk some of the unseen faces? Not a big deal? Wait until you’ve cloned this 200 times for your city map and have to go back and fix the caulking. At least that’s optional. Wait until you’ve realized that you forgot to make all of those brushes detail.

  4. Understand the vis process and the difference between structual and detail brushes. There are a couple of really good tutorials that explain how the game engine determines what can be seen and therefore what should be drawn at any point in the map. This is directly determined by how you build your map. A good understanding of this up front leads to more efficient mapping and better gameplay (e.g., players can maintain more consistant FPS, map plays well on a wider range of machine specs).

  5. Compile and test often. For me, one of the rewards of mapping, is being able to run around in a map I’ve created. It’s motivating and gives me a sense that I’m actually making progress. This also ties into #5 in that I get a chance to check out game flow, the scale of structures, width of hallways, and other critical gameplay issues before I spent the time building everything else out.

  6. Be absolutely anal about saving your work. You will lose work and you will make changes that you wish could be undone. Get in the habit of using the Save As function to save your work under a new file name. I’ve gotten in the habit of saving my file every couple of minutes and, at the beginning of each new day and periodically throughout the day, I’ll Save As my file with a date and time stamp. The same is true whenever I make a major change that I might want to reverse. This way, should I ever want to revert to a previous piece of work, I know exactly where I left off with other files. The autosave and .bak files are useful too, but don’t rely on them.

  7. Make maps with character. I like dark, rainy, beach maps as much as the next guy, but I’m not sure that ET is ready for another one. Make something that’s interesting and different, whether it be the environment, gameplay, or creative use of objectives. Pay attention to the lighting, sounds, and other items, which make a player spend some time just wandering around sightseeing and thinking, “Cool!”.

  8. All script routines need {spawn {wait xxx}} and every map needs a script_multiplayer.. I can’t think of a single stupid little detail which has cost me more time then these two. "Nuff said.

Cheers


(redfella) #2

Nice writeup. :slight_smile: Though I disagree with step 2 for the advanced mapper, I think that it may work fine for the newb. Keep it up.

:chef:


(Ifurita) #3

yes, this is aimed at newbs and the rationale for that point is that as you get started with mapping, it’s easier to start off with a smaller project you can finish vs. starting a grand project you’ll never finish.

As I read thru tutorials, I made tons of small prefabs as experiments to learn how to make buildings, constructables, destructables, etc. It’s a really simple way to get started and build confidence and skills


(hummer) #4

Nice :slight_smile:


(blushing_bride) #5

perhaps you could add a further point… don’t rush there’s no hurry. The longer you spend on your map the better it is likely to be


(Ifurita) #6

Yes and no. A lot of it really depends on your level of ‘perfection’, time constraints, or desire to move on. I wouldn’t really want to make that a blanket point of advice.

While I totally agree that one should not release a map just for the sake of releasing a map, one should also no continuously grind away, never releasing one, simply because it’s not ‘perfect’.

Everyone has to find a happy middle ground where they’re happy with the product they’re releasing to the public


(sock) #7

Very nice summary and seems to reflect alot of your mistakes you encountered while mapping. I also think people often learn best by making mistakes. (Like detail/caulk brushes)

But if you are going to rush your work out of the door and not wait a bit then be prepared to take feedback, unlike some people …

http://www.planetquake.com/lvl/comments.asp?id=1608
http://www.planetquake.com/lvl/comments.asp?id=1607

Damn funny in its own little way … :wink:

Sock
:moo:


(Loffy) #8

Hi!
I like these thoughts. Well written.
I guess another important aspect of creative work, is to be able to kill your darlings. To be able to see the map from the player’s perspective and reason: “This part/this idea is not working - it has to go.” And not “hang on to” stuff that is crappy or bad, for some reason, just because you - as a mapper - want them to remain in the map.
// Loffy


(MuffinMan) #9

LOL! sock - you found some nice fellas there, that’s constructive critics :banana:


(Ifurita) #10

Maybe we can hook him up with Sin


(G0-Gerbil) #11

You learnt about structural / detail in your first 45 days?
I R jealous - I still have fond memories of footy taking 2 hours to compile, then going back after it was finished and Str / det it properly (sadly once it had been released!) and it compiling in about 5 minutes :confused:

I’d say, for someone who wants to bugger around mapping, learning structural / detail early on would be most beneficial, simply because it allows you to test / progress on your work as fast as possible, and as we all know, early maps are crap, so the sooner you get onto your second or so, the happier you’ll be :slight_smile:


(Ifurita) #12

Well, if you ever saw my first map, Zeppelin you’d know where I’m coming from. I ran into the max_visibility_exceeded error, in part because everything was structural – including my rubbled buildings and the hanger (and all the railings inside) – OY!


(ConchMan) #13

Anything learned about how not to make a sucky map?

Cause I’m at that point (4 months of brush work into it) where if my map turns out to be disliked by everyone or if it runs to slowly to be playable…I’ll internally feel really bad. I’ve put a lot of work into this thing. Oh man, 4000 brushes, I feel like giving up again. :banghead:


(Ifurita) #14

Actually yes.

I think I actually went about making my Byzantine map in the opposite way that a lot of new mappers do. Keep in mind that I was building a large open map, so that gave me some lattitude to play around with stuff I couldn’t get away with in an underground map.

  1. I started with big blocks of caulk, textured on 4 sides, to use as placeholders for buildings

  2. I inserted some prefabs for a couple of the key buildings (#2)

  3. Made most of my objectives, some of which were prefabs, others were just cloned, and did the scripting for them (#4)

  4. Then I compliled and ‘played’ my map (#7)

I was able to do this all in the space of a couple of hours and it gave me a good preliminary sense for scale, connectivity, vis-issues etc. I then built on this, first adjusting buildings around, adding some detail, adjusting objectives. The last thing I did, was add in the detail on the buildings (facades, doors, windows etc), clutter in the streets.

I also tried to get the map out to people early in the process to get their thoughts on layout, size, objectives, which really helped a lot.

Lastly, and the lesson I probably forgot to include, was connectivity, both from a movement and a fire perspective. This will be different for each map, but you have to find the right balance of connectivity. Too much, and it’s impossible to defend or you will never know from where you are taking fire (problem w/ my first map). Too little connectivity and you end up with horrible chokepoints or people can’t maneuver (e.g., cooridors which are too tight). Early play testing will help you figure this part out


(BXpress) #15

i just tell you one thing: knowing how to use the tools and having ideas, isnt everything you got to have when you want to get a famous Leveldesigner.
You need experience and imagination. you need shedule and dreams.
and its what you make of it…
Im not telling you im a famous Leveldesigner, im absolutely not.
But i just want to say: you cant learn everything in just 45 Days…
*i tell this especially to you Ifurita…
… you have got to learn how to use your ability!
:lookaround:
PS: its not an attack or something. just wanted to tell this to you!
PPS: sorry because of my bad english :frowning:


(Ifurita) #16

Oh, I totally agree. I never intended to make this sound like here’s what I learned, that’s it. There is tons more to learn, experiment with, try out, get better at. However, I think most people got the gist of my “10 Lessons Learned …”

Very valid points. I certainly don’t take it as an attack.

My intent was to share a couple of hard won lessons, things that had I know about earlier, I would have saved myself hours work.

That’s all.


(Machine for to kill) #17

Oh yeah this coming from a guy who copied his entire map directly from pictures and has started it since april. REAL IMAGINATIVE

Give ifurita a break, his map is one of the most creative ones I’ve seen. I think you can learn a thing a two from him…

P.S.

Isn’t sock’s quake3 username firethrottle :bump:


(Ifurita) #18

I don’t think that was his point. I don’t see that as a personal attack on me, though I agree that a “10 Lessons …” article coming from a self avowed mapping newb is unusual to say the least.

So why did I write that?

  1. I made most of those mistake. A map made entirely out of structural brushes - yep, a ton of windows where the window was oversized by 4 units - yep, a map that was waaaay too wide open - yep, stupid scripting errors - yep. I sweated out each and every one of those lessons. If any of those 10 points a) gets a new person into mapping or 2) savings anyone more than an hour, then it was worth the 30 minutes to type those points up

  2. There are tons of how to tutorials. A lot of them are really good, but they focus on the technical aspects of how to make a room, how to make rubble, how to make a constructable. Almost none of them teach a new person how to approach a mapping project. I think, as a result, we get a lot of dark, rainy, beach maps.

Just my 2 cents


(Machine for to kill) #19

You’re too nice ifurita. You don’t have to take crap like that from no kraut. Not many people know The Q3 engine and radiant inside out as to be properly named experts. Most of us are people who are learning as we go along. We might not know everything, but even as noobs there’s always someone more noobish than you. That’s why it’s important that we help each other however little we can so we can all learn a little. Life is too short to try to learn from your own mistakes cause that way you’ll never get anything done. I think it’s great that you posted what you thought were some good tips for people starting out. We need more helpful people like you, and fewer snooty jerks like him ragging on noobs. Totally inappropriate.

P.S.
I’ve seen his map and it sucks. ooohh yeah a bridge over a river…never seen that before.


(BXpress) #20

try reading the whole post okay Machine for to kill?
i just tried to give him something… not trying to say “im da big masta, i know even more than you, you suck, learn of my bridge map muhmuhmuh”…

well Machine for to kill creativity is what you make of it. ive made 7 maps from scratch. 2 maps from RL (Stargate SG-1 for and workin on Remagen).
just try to think positive, not agressive…
i even try to help people.