There are currently 30 players online.
 
Warsow

 
 

backThe eternal story of BefCa1 - "Bobot Factory"

Forum index
Community Blogs
The eternal story of BefCa1 - "Bobot Factory"
Posted by
Post Scroll to bottom
SolidFake

mapping noob
(4551 posts)

The eternal story of BefCa1 - "Bobot Factory"



Hello dear Warsow community.

If you don't know who I am, then this blog probably won't interest you in any way. If you do then you probably have heard that I work on a ca map for Warsow, since over 3 years. (yes, three years)
The map is still not finished, and I occasionally work on it. The purpose of this blog is because I want to tell the story of the map, and what failed so horribly that caused that extreme delay.
The reasons for you to read this blog is either:
-You are curious why making that map needs so long and want an explanation.
or
-You are a level designer yourself and want to know what could go wrong at making maps.
This is not about mapping skills like knowing how to use the editor, it is more about stuff like workflow, motivation and observation.
I learned a lot while making that map, really a lot, mostly stuff that you won't find in tutorials or walkthroughs and stuff noone warnes you beforehand.
Anyway, let's start.


How it all started
It was shortly after the release of Warsow 0.3, with the addition of the new gametype named ClanArena (or short CA) a new wave of optimism came above me and made me make it my main gametype (which was prior duel). I played a lot of games and joined a ca clan.
The clan sadly vanished fast so it was time to search a new one. Luckily the revival of an old famous Warsow clan was overdue, the name of the clan was BEF, which is short for "BisEinerFlennt" (which is "untilonecries" in german).
To name a few members: Milchshake, Durzan, Wildkater, Pikku, Hatebreed, Grimmgore and for sure myself, solidfake.
We played dozens of hours Clanarena, had a lot of fun, but sadly there was only one official wca map which was also not so balanced regarding the op camper class back there.
I mapped already before, but only a tdm map where I played around and two concept-gametype maps (you can find all my finished maps here) so I thought I should start filling the hole which was the missing of proper ca maps.
I asked my clan if they would help me to balance it and they all agreed and where filled with the same optimism as I was.



The first mistake: Overestimating yourself
So there I was, ready to make a ca map. But that wasn't all, I wanted to make more. I wanted to create the 'perfect' map. I also started to make my first own crappy textures back then, and I quickly decided that I want that the map becomes 'full custom'. Say everything should be custom-made, all the textures should be made by myself as well. Also the map should have perfect gameplay, and especially perfect visuals and a distinct style.
I put the map on the pedestal.
I already had some basic thoughts of workflow back there, so the first thing I made was gameplay concepts. The maps should have perfect gameplay, so I thought I should create the gameplay first and then create the perfect visuals, so that the gameplay doesn't get influenced by the visuals.
These were the concepts of the map:



the ground floor



the middle floor



the highest floor

Yes, I know that there stands turky instead of turquoise, my english was a lot worse back there, these are still the original concepts I made at the 7th february 2008.
A lot of stuff changed from the concept to the final gameplay.
So there I was, with my concept, my mediocre mapping skills, my nearly nonexistant texture creation skill, my underwhelming understanding of the qfusion shader scripts and with my heart full of optimism on the way to create the map of the maps.
Was that clever?
HELL NO!
All I had was a hammer in my hand and an idea in my head, trying to create the 8th world wonder.
At first I just didn't want to see that I wasn't ready for this map. I kept going faster and faster, learning more with everyday.
The intitial gameplay was build rather quickly, scarcely filled with blank walls showing my first own crappy textures. We talked about the gameplay, made adjustements and slowly talked as well about the visuals and the scenes the brushwork could fit in. Great ideas were gathered, to name a few there was the idea of a canalisation, of a factory and of a superstore, the last got rejected quickly cause I knew I was not able to create the models and textures for it (luckily my brain hasn't ceased complete function at that point) and the canalisation was also rejected cause we thought it is nothing special.
So we have chosen the factory. What would we produce? The idea was to make the factory to produce bobots, a legendary model that was as well released with the latest version.
I made fast light settings using shader trims and asked my clan friends which color for the map they want. I initially wanted to make a clean factory lit mainly by white light, but the clan decided for the blue setting, so i decided to make the map more into the dark, cyberpunky direction.
It was somewhere in 2009, when I suddenly saw the obvious, this map is getting out of my hands.
I put more and more time into more details like polishing textures, and because I ever got better and better at my work, I also reworked my textures more than often. I also got a lot better at mapping, and I wanted to make great stuff in the map, revolutionary stuff that was never seen before. I wanted more and more.
Which leads us to the next mistake.


The second mistake: Evolving own Skill which hurts map constancy

The second mistake was hard to see, hard to recognize, and even harder to accept for me.
I needed about 2 years after I recognized where this is going.
If you remember I talked above about how low my skills in mapping, texture creation and shader understanding where when I made the map.
For sure, while I made the map, I learned more about level design. I learned how to create better textures, and with every texture I made, I got better at making textures.
While this seems to be good for your own skill, this can be a desaster if this happens while an ongoing project like my map.
The problem is actually so simple that it is ironic, while I got better with every piece of work I put into it, the quality of newer made stuff exceeded the initial stuff by a lot.



The most painful room of the whole map, I had some amazing ideas with it, but when my skilled increased the old textures suddenly started to no longer fit together with the newer, more detailed ones I made. Also the detail differed too hard, in the edges there were highly detailed bobot models and well textured trims while the rest of the room was just rough and looked crappy in comparison. I am still trying to adjust this room to the rest of the map.



Three different lamps, where you can cleary see differences in the quality of each one, the trims were first created, then the bulbs and at last the rectangular lights. Also at first the trim and bulb lights had no normal and specular map, everything I had to readd in the and.

And suddenly, before you can even see it, your map is full of unconsistency.
Half of the textures in the map just looked outdated compared to the fresher ones, parts of the map were filled with astounding details while other parts consisted of blank walls.
I played Icarus and ascended so fast, that the brightness of the sun burned away my own wings.
This is one of the most crucial mistakes I made in this map, because this mistake haunts me till today. I am still trying to rework outdated textures, adjust the detail of the maps to be adjacent throughout the map and try to recreate constancy.
And believe me, this work is not only boring, but it is also extremely frustrating for you. You have to look back at your own work, what you created, and you have to destroy it, remake it, just because the quality of the work at start and end differs so heavily.
So, if you plan to make a bigger project, watch yourself. The optimum would be if you wouldn't increase your skill in any way while the map isn't finished. For sure this is not applicable if you haven't hit the limit of mapping yet.
My advice to everyone is that you try to improve your skill OUTSIDE of the map, make concept maps, try out stuff in tiny crappy maps, but keep the constency in your map!
And when you make a bigger project, often watch back to where you started, did your details suddenly change? Does the quality of work suddenly differs? If yes, then you need to act fast, you need to watch your steps, and use your initial work as an anchor. Yes, your initial work, not your recent work. It might hurt when you have to make a map or a work that doesn't really represent your skill, but it will be the best for the map.
Consistency is more important than general quality!
Yes, this sentence is highly debatable, but in my opinion a book that keeps a consistant quality is better than a book which highly fluctuates through the readout.
I learned this the hard, painful way.



Another problem: Time
This is not really a mistake, cause I couldn't really tell this beforehand.
But this happens to probably every artist at least once.
You start a big project when you have time, but a big project also means a lot of time to finish it, and you often can't say if you will have time in the future to work on it as much as now.
Same with this map, I started the map when i had quite a lot of freetime. The general other timeeater was school, but I never really ever cared for my notes. I am not stupid, but I just never find myself learning specific things till you memorise them perfectly, I read stuff and learn cause I am interested in it and how it works, once I know it it often becomes boring to me and I don't invest more time to perfectionate these knowledge.
I would say I am intelligent, but I am just either too lazy to learn stuff general or I have some sort of attention deficit (where both of it again doubt my intelligence ^^).
Well, when i started the map, I had a lot of time, but my school leaving examination was nearing very fast, and I was forced to dig deeper into knowlegde of things. A lot deeper.
I succeeded and thought I would have more time now. Ehhh, wrong, I had to go to the armed forces for 6 months afterwards, leaving my freetime again short. After that I decided what I want to study, and to earn a bit money till it starts, I worked for 3 months, again no time. And now I am studying and, well, I have more time than usual but still less than I used to when I started that map :S



The third mistake: Making a project over the course of various program version

Another very painful problem.
The map started at Warsow Version 0.3, and we are close to reach version 0.7.
In between a lot happened, while my map was barely playable in the 0.4 version due to the extreme high amount of detail which was uncommon there for Warsow maps, it got well to play after 0.5 and the graphic changes/improvements.
This was good for sure, but the version changes brought more problems than they were good for it.
Bugs.
Dozens and dozens of bugs.
My map uses a lot of stuff that many maps doesn't, stuff like mirrors or high tris counts, so many bugs seemed to only appear on my map.
And the stuff that was bugged gone from okay, 'I can deal with this' to 'wtf is this?'.
Stuff about spawning entities emmitting purple dynamic light, stuff like rotating objects slicing suddenly through the whole map and killing the player at touch.
It became sort of abstract in some parts.
For sure this wasn't only because of new Warsow versions, but also due to the switch from gtkradiant to netradiant, and the different netradiant versions.
But the main problem wasn't even the bugs, the main problem was the changed lighting of Warsow.
I used mainly shaders who emitted the light in my map, to get a better 'natural' feeling than using simple light entities, as a lot of stuff with them looked bad.
Now they changed it, but hey, they also changed the whole lighting system, which meant that shaders emitting light suddenly behaved completely different, at first it resulted in my map becoming nearly completely dark.
I tried ages adjusting the shaders to the new lighting system, but I quickly recognized that I couldn't simply adjust them, i would have to redo all shader lights in my map.
But my map was build upon the light trims, so I had to find another way.
So I commented the light emitting of the shaders out and instead now try to create the whole lighting with light entities. This not only makes the lighting look quite natural but also gives me a lot more power about lighting different areas brighter or weaker, an option i didn't have when I worked with shader lights.



I am still experimenting a lot with this and it isn't even partially done, it is very time consuming because you ever have to check how the lighting works out with full light compile and bounce options.

So, there is nothing more to say here, just that if you make a map or any other projects, get ready to come upon a lot of problems when your programs suddenly release new versions (I think source developers can sing a song of this).



Another problem: Motivating myself

I think every artist knows that problem, I will still talk a bit about it.
I generally have large problems to motivate myself todo stuff that has to be done, I guess just because I am a lazy person. But in the terms of this map I also knew how much of a long way I still have before me, and I learned that making a texture and polishing it till it fits into the engine includes a lot of boring stuff.
The more you know about your work the more you know what parts are boring and enjoyable.
As said, the gameplay of the map was quickly finished and the rest of the time was mostly spend upon polishing the looks and the performance...which is boring as hell.
I as a mapper enjoy it the most to create the basic gameplay, create brush details, add fancy stuff like triggers and moving objects, but when it comes to polishing the consistency and the looks of the overall map, I am at the end of what means fun to me.
I have basically done already all fun parts in this map and now I am stuck with the boring, repetitive stuff. But well, can't change that.
Another bad thing I made was making a list of what I have todo or want in the map.
At first I thought this will help me not to forget amazing ideas, but it quickly became the hate list for me and everytime I checked the list a big sigh halled through my empty head.
But I also learned a bit more to motivate myself in these stuff, I should rather not think about the actual part of the work, but rather think about the result. Thinking about what you have todo is not often motivating, but thinking about what the final result might look like can help a lot.



The fourth mistake: Denying possible Feedback

A mistake I could punch myself in the face for it.
As said, i intially made the map with my clan, we had a forum and I updated often and asked for opinions and stuff.
We also playtested the map sometimes or the people were just bugsearching.
Sadly the bef clan didn't last long enough as well.
More and more members vanished, and soon there were just 3 pseudoactive members.
The clan vanished once more and with it the visits to the forum.
But I still worked on the map, something was missing but I didn't really recognize what.
It was FEEDBACK.
I used to gain feedback for our clan forums, and as it was a 'clanintern' map I decided to keep the map secret for the official Warsow forums and then one day 'surprise' everyone with the perfect map.
This was just horribly stupid, I kept working on the map, and working, and working, without asking anywhere for feedback.
This created the constancy problems I talked about in the second mistake, the quality of textures and map details differed more than heavily and I just didn't recognize.
In my school, we once learned a psychological state which is called "company blindness".
It basically says that if you work on a project without getting feedback from 'outside', then you tend to not see obvious problems and flaws. This happens also in greater teams, as everyone in the team tend to adjust to the same behaviour.
This is why feedback from a multitude of people is the most important thing for an artist.
I then suddenly realized when I released a picture of the control room on the mapping forums, and people started saying that the lights don't really fit with the rest and the room itself is too overloaded with gfx.
I thought "wtf, what are they talking about, it looks awesome!" but my ignorance soon crumbled under the truth of feedback.



The fifth mistake: Cutting visuals and gameplay completely off
Do you remember what I wrote about the workflow of the map?
I first made the gameplay and afterwards the visuals.
I wanted the visuals to take no effect on the gameplay of the map, they should be completely cut off each other.
I needed a long to time to see that this was not the best thing todo.
While it is the best for gameplay, it is definitely not the best for the visuals.
Why? Because you have to make compromises, you have to hurt the visuals in regard to keep consistency in gameplay.
I wanted this map to have a certain distinct style, with amazing detail and yet nothing of it should influence gameplay.
This is not possible.
You can't have both, perfect gameplay and perfect visuals, you have to make compromises regarding one.
The gameplay is more important? Then cut visuals down shorter, make details only where they don't influence the gameplay. A great example for this is Kimza's last screenshots of my good old wda1, he didn't change the gameplay but yet managed to give the map good detail.
But you can't have both, gameplay and visuals harmonize with each other and work in a flow, it is hard to see that flow sometimes, but it has to be recognized.
I tried to reach both at this map, and all I got was a lot and lot of work to try to fill the map with detail without making it look bad or hurting gameplay.
This is in a ca map more of a challenge than in a duel map, because in ca you can reach every single place in the map, you have to ensure that details are clipped out properly and usual brushwork stays pretty much the same.
Now when I make a map, I decide beforehand what style it should go, I think already while planning the gameplay how I could make this into a detail and maintain the gameplay.
You should neither build the visuals around the gameplay, nor the gameplay around the visuals, you have to find the middle and make compromises to both.
I could write a whole tutorial now about it. ^^



So, what now?
I can't exactly say what I will do, but more what I will try.
I am studying atm and my freetime is quite variable, but it is there.
I will try to work occassionally on the map, not too much thought because I have other things in plan as well (skyrim mapping e.g.) and also not too much to keep my motivation up and my frustration low.
I might make a thread in the mapping area with the last map file and ask for feedback (I need to redo the lighting first), not completely sure yet though, but I will definitely post more in the wip thread.
Maybe one day I will finish this map and release it, the result will probably be completely underwhelming, as it is with all things that need that long to come out.
But hey, Duke Nukem Forever also got released one day, eh? :D



Here are just some screenshots of the map and me explaining some of the features:
Everything still highly wip (after 3 years, yeah)!

I created a lot of custom textures for this map, including:



completely custom lightflashes and halos



Bobot production lines, completely with custom made normal maps.



Dozens of gfx effects, containing stuff like blood and oil stains, signs and screens, cables and servers.




One thing I am still proud of is the bobot statue I created, it is basically the 0.4 bobot model with the rocketlauncher model, and with a custom marble texture.






A lot of nice looking effects, like broken grates with pipes where steam is coming out of.



Production lines, which weld the different bobot parts. This was quite a challenge, for which I used a combination of shader functions, func_bobbing, func_wall, trigger-relay-loops and custom particle sparks made with particle studio. It is still full of problems, like shaders and entities tend to unsynchronice, so I thought already about changing these with func_trains.



There is one place where you can watch through a grate and cables to the floor beneath you.



The gameplay is highly inspired by wca1 and soheica1, but includes also a lot new, interesting architecture.



Sadly these are all old Screenshots, the new map is completely dark due to the changed lighting.



I also plan (or ever did) to release the textures as a free texture pack once they are done.



The pain room, this is an old screenshot, I already defused it a lot regarding too bright lights and I will add more detail and another floor instead.



A bloodstain on the wall, also a custom texture.



An old screenshot of a production line, the lighting is pretty hard fucked up and needs a lot of work, also the brushwork and detail.



A videoscreen featuring custom communities, like #warsow.mapping and racenet.



This jumppad is actually a completely flat texture :P



This is the stuff I love at mapping, at first this was a simple one brush ramp, this is what I changed it to.



The bobot statue from a close view (the marble texture gets probably changed).



This map was the most perfomance-heavy map before the engine changes took place, yet at low graphic settings it had better performance than wca1.



Tried to make stairs looking more pleasing.



BEF for ever! (extremely old screenshot, also without per-pixel lighting)



Thank you for Reading!

adem

Translate WSW on goo.gl/aZo1pt
(2126 posts)

thank you SolidFake for the intresting post. I'm not a mapper/artist but your advices looks useful :) imho - your map needs feedback extremly! I like the main style of the map, and hope somedays you'll finish it. Also i think there are a lot of unnessecary details on the map. Perfectness doesn't mean complexness :) gl and hf.

idontcare

Member
(205 posts)

WOW! very intresting indeed, I love the bobot statue. Hope you'll finish the map at some point ;)

m0t0

Member
(948 posts)

adem wrote:
Also i think there are a lot of unnessecary details on the map. Perfectness doesn't mean complexness :)


Indeed i totally agree about this.Often when map has right amount of details the details are much more visible and this way alot stronger.In a bit same way than in painting art.
thats why often less is more !
(updated 2012-02-12 14:22:24)

homart

Player
(715 posts)

Great :D hope you will finish a day ;)

Pikku

Member
(609 posts)

Still remember that map, great post, thx for letting us know the story behind it and the whole bef-thingy, beter times :)
Seems you learned pretty much for yourself, even if its finished or not!
Go on Soliddood <3

Mef!sto

Member
(738 posts)

release or it didnt happen....

crizis

Mad Hatter
(4570 posts)

heh, actually my latest bomb map test (factory_beta) is also named 'the bobot factory'... coincidence, I swear :p

But when it comes to the map, geometry-wise there's nothing 0.62 and especially 0.7 can't handle with perfect fps - but I think your gigantic abuse of multipass shaders is what will bring fps down here and hard...

SolidFake

mapping noob
(4551 posts)

crizis wrote:
heh, actually my latest bomb map test (factory_beta) is also named 'the bobot factory'... coincidence, I swear :p

I think there can be more than one bobot factory :P

crizis wrote:

But when it comes to the map, geometry-wise there's nothing 0.62 and especially 0.7 can't handle with perfect fps - but I think your gigantic abuse of multipass shaders is what will bring fps down here and hard...

The biggest problem I had before 0.6 was the excessive per-pixel-lighting and the relief mapping, these are pretty much gone since.
About the multipass shaders, well, I don't understand a lot stuff of the engine, and I applied the usual thinking "if it works, it can't be bad" ^^
There are a few shaders with a lot of passes, but I think I could make them easier without making them look worse.

And thanks for the rest of the answers!

Huksi`

Hat Matter
(3509 posts)

Gotta admit the story was TL;DR, but the map looks very nice :D

m0t0

Member
(948 posts)

I think the mistakes/problems you have handled during making this map are pretty common.This:The third mistake: Making a project over the course of various program version is more like outcome than reason tho.oh and what comes to textures the best way is always to fully make textures first and then map instead of doing both simultaneously.Or atleast then you should really be carefull how much you change textures because then the project can be easillly delayed.

talking about my own experience here and seems that solidfake has experienced similar issues :)

RuKeN

Member
(20 posts)





hypnotism... @_@

but all your work is rememberly and i desesperate for play it
(updated 2012-02-13 19:09:46)

SolidFake

mapping noob
(4551 posts)

m0t0 wrote:
oh and what comes to textures the best way is always to fully make textures first and then map instead of doing both simultaneously.Or atleast then you should really be carefull how much you change textures because then the project can be easillly delayed.

talking about my own experience here and seems that solidfake has experienced similar issues :)


yes, definitely, but I didn't know that back then, I didn't know what a good texture pack needs, so instead I mapped and when I needed a texture, I made it.
As you said, that made all textures look quite different from each other.

It seems that every level designer comes to realize these issues or at least similar ones with the time ^^

KoFFiE

Member
(1898 posts)

I don't really agree on your all conclusions, especially one - and that is the skill advancement. This happens with any skill, the more you use it, the better you get at it, nothing you can do about it. The best way to learn something is to just dive in head-first and be so naive you can handle it - as you clearly did.

This is how I learned coding, and I now know why some of the projects I attempted failed, and I'm still amazed how I got a few of them working as well as I did. If I take a closer look on how I did it however, it's horrendeous.

I just want to say with this: experience isn't always positive, it also restrains you from trying something seemingly impossible, but something that you could actually manage to pull it off if you just tried and never give up. By being this naive you won't be afraid to fail, this is something that never even enters your mind. End-result is, even if you fail, that you will have learned a lot. And if you pull it off, you will, at one point in the future, be amazed of what you did. Experience only puts a brake on you in this regard. You get a sense of the amount of work it is, and how hard it will be. You'll quickly see what area's your skills are lacking. Experience tells you what exactly is needed to reach your goal, and this can become quite overwhelming. Sometimes you just have to act as a fool in this regard to get it working.

I think the main problem you had was accepting that it was time to let go, wrap it up and finish it, start something new from scratch, with the experience and skills you learned. This is very hard to do, and the longer you're at it, harder this becomes, since you've invested so much time in it already.

Once you are skilled and experienced, one of the things you will have developed is a workflow. Once something interrupts a workflow, you should re-evaluate everything if your goal is perfection. In any other case, you'll end up with a quick hacky thing, no matter what skill we're talking about. Note that this is not always a bad thing, as long as you can keep yourself motivated.

Experience tells me that if I have a deadline, these hacky things you come up with can be brilliant when faced with changes or challenges. And that if you don't have a deadline, that it becomes very hard to motivate yourself in this situation.

Also, the last 10% of a project is the hardest. These are the boring timeconsuming parts, testing, debugging, stupid details, ... The "good enough" approach is your enemy. This last 10% makes the difference between an "ok" end-result and a brilliant end-result.
(updated 2012-02-14 17:43:19)

Durzan

Member
(61 posts)


Scroll to top



RSSRSSRSSRSS

Copyright © 2014 Chasseur De Bots

Warsow in social media



Warsow is an indie game developed by group of hobbyists. Please consider giving us a small donation: