Part 1

Rambled by Jeff Walters

What is the purpose of this article series?

The one thing this article series will not be is a tutorial. The idea here is basically just to talk about things I've learned the last few years editing levels - little tips, preferred ways of doing things that level authors might find informative - NOT a 'this is how you should approach editing' article. One person's way of editing is always different from the next person. No-one has all the answers and at least some of the answers each person have are wrong.

There will be at least two parts to this article series - possibly more depending on if I can think up anymore things to rant about ;) They will also follow no strict format as these articles are more like editorials rather than a formal objective articles. This series is aimed mostly at first-time editors, but a few things I'll be discussing (like architecture and texturing) will be aimed at those with experience as well.
" No-one has all the answers and at least some of the answers that each person does have - are wrong."

So let the ranting begin....

How to start learning level editing?

Everyone has a different way of learning. Some like reading all the tutorials and getting all their problems fixed via editing messageboards, whilst others (like me) like to teach themselves most things by picking apart the existing LEC levels, seeing how everything works in this medium and then learn from your own mistakes. Raiding the editing messageboards is a lot quicker but you don't learn as much via that route. Won't preach too much on this point though as it's best to learn the way that you feel most comfortable.

The most important key when starting out is patience. Level editing can be....scratch that...is frustrating to learn and no amount of tutorials you read will entirely prepare you. But I can assure you it's worth it in the end. Once you have experience - the biggest limitation is imagination. That might sound like propaganda bulldust, but it's true. When I first started creating my Dark Forces level, Mt. Kurek, I had serious doubts about creating a realistic volcano in the DF engine, but with persistence it worked out. Same with Jereth Kok's Stars End level which featured a fully operational spacecraft that you could board and fly out of the atmosphere....if these kind of things were able to be done effectively with the Dark Forces engine - the limit's to what you can do with the JK engine are astronomical.

One of the fundamental areas of designing a level that most authors seem to struggle with is architecture. A rash of SP levels have been released recently with inventive gameplay features, storylines and cutscenes......but inventive architecture? Anyway, I'll stop that though right there as architecture is next month's big topic for this article series. Now, back to setting out on the JK level editing road.....
"....biggest limitation is imagination."

Which level editor? JED, ForceBuilder, JKedit? Decisions, decisions.

As many people out there know, I'm a JK level editing junkie. After editing Dark Forces with no 3D preview (texturing hell) or map-rotation (architectural hell) - I think editing JK is the best thing since sliced bread...no scratch that, I hate bread. The best thing since some smart bloke married the worm with the tequila.

So, one question you might be asking is JED, FB or JKedit......which is the most advanced editor? The most easy to get used to? Well, if you want to be a serious level author - only the first of those two questions really matters. The fact that an editor takes a few less hours to learn than another is immaterial if it's not as advanced. Currently JED is the benchmark and the program most widely used.

What's wrong with FB and JkEdit then? Nothing gravely wrong :) You are not going to be struck down by lightning for choosing an editor not called JED. But the other two choices have a few question marks that make me hesitant to recommend them. ForceBuilder has the makings of a very capable editor but it's main problem is that the future of the program is very uncertain. It has already been pulled out of circulation once by the author and even though it's back in construction now - the author has given no guarantee about further releases or that it won't be pulled again.

Though I sympathise greatly with Mattias Welander over the reasons behind his actions regarding the editor - which was a steady inflow of insulting email. As someone who's done a lot of webmastering, I've faced the same problem. But still....it's impossible to recommend to first time level authors to use an editor that may be taken away from them while they are in the middle of a level because of the action of a few other users with no respect for the program's creator.

JkEdit? Well - it's crippleware. You are going to have to fork out money to use it - something I definitely don't support considering it's an editor for a commerical game. Still, this editor is growing at a steady rate showing that it's creator is dedicated to improving it. So I'm sure you won't be left hanging if you decide upon registering it.

The case for JED is pretty simple. It is the most advanced, has a wide support network of messageboards and tutorials for users, it's creators are responsive to feedback and would never pull the editor off the net. Not to mention the ability for users to make custom additions to the editor via Plug-ins added in the latest versions. This assures the editor will continue to grow long after the Code Allaince release the final version.
"Naivety is the downfall of most new editors."

Gee, should I start an editing group now that I worked out how to load JED?

Argh! I wish I could say that subject title was entirely sarcastic, but hell - I've heard of people starting websites for multi-level TC's (Total-Conversions) before they have even downloaded any level editor. They then barrage new sites with requests for members and are surprised when no-one responds. Naivety is the downfall of most new editors. They take on projects they aren't prepared for, get discouraged when few people jump aboard their projects and are then never heard from again.

So a plea to all first-time level editors out there - please at least try your hand at level editing so you can see the effort required to make quality levels before jumping in over your head with outlandish ideas for editing groups. Another reason to do some some level editing work before forming a group is that very very few people will join a group headed by someone with no previous experience unless they themsleves.....have no previous expereience. I can count on one hand the number of semi-decent mods put out by editing groups out of the hundreds in existence since JK's release last October.

Top Ten Signs You've Picked The Wrong Beta-tester
10. Loads Canyon Oasis mistakedly, then reports to you that "gamers just won't go for it"
9. Says you should use that "cool effect were you slide across floors like on an ice-rink"
8. Thinks you should ditch your new incredibly innovative cog feature in favour of a few more TD traps.
7. His idea of weapon placement involves a ratio of 50 concussion rifles to every room.
6. Asks if he can test the level with some friends he just made on IRC.
5. Wants to know why your JK level lacks coloured lighting.
4. Like's your use of that gray texture with the yellow vertical & horizontal striping.
3. Says he's looking forward to the level, as your first level, The Cube, was a "surefire masterpiece".
2. Regretfully informs you that Quake2 just won't load the damn thing.
1. He writes for this magazine.
I dunno - Is beta-testing a good idea?

Never underestimate the difference beta-testing a level well can do. I discovered this fact during the final stages of making Mt. Kurek. I was lucky enough to have made good choices for the level's beta-testing team and the feedback received was one of the key reasons the level was very successful.

There is a great temptation when testing a level to pick your best friends as your testers or people you know will like the level. This kind of makes the process of testing pretty mute eh? Best testers I've found are, granted, people you can trust, but also people who will take an indepth look at the level, point out every possible error or place you can improve on. Constructive criticism is the best possible feedback you can get to a level.

Best person I can give as an example here is Geoffrey Elliott. I still thank my lucky stars I had him as a beta-tester for Mt. Kurek. Hard working, knowledgable and never pulled any punches. The three best qualities a tester can have. Because of his feedback many hidden errors were tracked down and many extra features were added to improve the gameplay.

The best mix you can have on your beta-testing team includes both dedicated editors and players. Overloading either will skew the feedback you get. In my experience most editors don't play the actual game that much because of their work. Hell, some people I know who are great editors haven't played JK/MotS in months (besides to test their creations). So while pure editors are usually great at tracking down design faults, code errors, etc - it's the dedicated players, however, that usually play very frequently who are your best bet for tracking down gameplay bugs/problems.

Another thing to keep in mind is to use testers with varying PC types. Having ten testers with P2 300's and 3dfx leaves a huge question mark over how well your level will play on lesser machines. It's best to have at least one tester with around a P166 with no 3dfx to see how low end users will fair. Even if you have no inclination of fixin any framerate problems detected on low-end machines - it's at least important to know they exist so you can warn low end users in the level's readme file.

"... the process of making C/S cogs is very easy"


What are Client/Server COGs and do I need them?

If the level you are making is multi-player (as most are), then you absolutely cannot neglect using client/server cogs. A good example of what can happen to an otherwise great level that doesn't use C/S cogs is the original JK MP map, Bespin. It is a beautifully constructed level. Beautiful. It is also pretty damn unpopular. Why? Because it's main combat areas feature many non-clinet/server doors and elevators. No fun trying to chase an enemy when you have to wait five seconds for a door to open so you can then call an elevator that takes a further five seconds to start moving. Frustrating to say the least. But if this level was re-released using client/server cogs that reduce the amount of cog-related lag - I'd bet any amount of money it would be as popular as Canyon Oasis (a level you have no doubt noticed - with no doors or elevators).

So what the hell are client/server cogs? Well, I won't spin this article into an indepth tutorial on how to make the things because a) I'm a slug and b) er...one already exists. Written, surprise surprise, by those talentless imps at the Code Alliance ;) Anyway, getting back to the planet of Relevancy, I'll just give a quick rundown of what C/S cogs are and how they can help your level.

Basically they involve the use of two cogs - one for the client and one for the server. As an example let's just say you have a pair of c/s cogs for a door. When you click on the door to open it - a trigger is sent from the server cog to the client cog, containing very little data, that triggers a certain event in the client cog (event is defined in the trigger sent) - in this case the event would be to open the door. As very little data (just the trigger) is sent between computers in MP - a lot less bandwidth is eaten up. To put it as simply as possible - it's a hell of a lot quicker than the normal way.

Another pleasing thing to keep in mind is that the process of making C/S cogs is very easy. Surprised me greatly how quick it was to do when I wrote my first c/s cog. One tip I have for once you've got the client/server processes learnt is that I find it best to write a cog first the normal way - get it working correctly and then convert it to client/server format. Well, works for me anyway :)

That's it for this issue. Hope I've helped answer a few questions for budding authors. Time to go pack for my holiday in South Australia. As for next month......

Next Month on Ramblings......

Will Billy discover Marian's infidelity with Roger?

Will Marian discover Roger's indiscretion with Billy?

Plus some ranting on architecture and texturing in the ad breaks.....

Jeff Walters is sought for questioning in three countries for comedic malpractice.
Back to D A R K N E S S   F A L L S