Search

[Joris Ceoen]

Hello everyone. As WWMT helped me across all of my level designing blocks and allowed me to realize numerous projects, I decided to post them here for everyone to understand the power and utility behind the tools Shawn provides us. I realised during my developments that Wall Worm acted as a decisive bridge between 3DS Max and Source and without it, nothing here below would have existed, or barely coming to the results I could achieve so far.

ZE_CATHEDRAL (CS:GO)
URL: http://csgo.gamebanana.com/maps/168170



Made on request, it's my very first project to have ever been finished and released. It's in critical condition, deserted and the performance is below average, but the entirity of the cathedral you see is modeled inside of Google Sketchup, imported into 3DS Max for texturing and then exported with the Wall Worm Model Tools (WWMT from now on). Without WWMT I would probably never finished this project, and maybe I would never have continued level design. Cathedral was my first step into the Level Designing world and still remains (for me) one of the most iconic maps.

DE_WINTER_FOREST (CS:GO)
URL: http://csgo.gamebanana.com/maps/171244



A smaller project finished in 5 days, featuring a large-scaled outdoor, winter-like environment. It was the first map to use custom trees that used the Sway shader (By copying the Urban Tree MVT values at the time) and all of the models where used in conjunction with WWMT. Again, I have to mention that I would not be able to make these without the WWMT. Yes, I could have used custom models that were already compiled and exported for the Source Engine. Then again, CS:GO was just released at the time so there was no way people released foliage models that were compiled for the CS:GO version of the Source Engine (as you have to recompile models with the studiomdl.exe from CS:GO). I still had to learn about the VTF settings and alpha maps and pixel art in general, so the map received a huge update after release, fixing all of those old issues.

CS_ARDENNES (CS:GO)
URL: http://csgo.gamebanana.com/maps/173968



Ardennes was my first ambitious project where I wanted to take my artistic endeavours to the next level. The map is an attempt to prove that the Source engine does handle large environments, with lots of foliage. It is not the best optimised map however, and still has many issues gameplay-wise. I learned some new techniques with this project which allowed me to delve deeper into the 3DS Max tools. It took me 3 complete months to finish the project. All of the texturing and modeling was done in 3DS Max, Sketchup and exporting with WWMT.

This time I also wanted WWMT to take a very important role in exporting such as using the illumination origin option, using the origins of models to perfectly align across brushes (I calculated all the brushes their dimensions and then I made models onto those in the 3D modeling program Sketchup to have them perfectly align, then export to Source and the results are outstanding). I created and used fully working custom breakable props, which I found otherwise unable to create, all thanks to WWMT!

Ardennes is also the first map where nearly every single model is made with the WWMT, without Wall Worm there was absolutely no way I could even start this project. It was, however, designed in Hammer and Sketchup, something which would fade away after in future projects.


AWP_RATS_ELECTRIC_ORGAN (CS:S)
URL: http://csgo.gamebanana.com/maps/173968



A special project where I recreated my electric organ at home. It was released for a contest, and although it never won any spot, I remain confident it was one of the most original of all entries. The stops, keyboard and pedals were all photographed, modeled, textured and finally exported with the WWMT. It was a lovely project to work onto, unfortunatly it's clear that it never saw the light of day at any point in time during and after release, and is now entirely abandoned.

It was my only work on Source (as opposed to CS:GO).

HIMEJI_CASTLE (CS:GO)
URL: http://csgo.gamebanana.com/maps/177542



At the time, Himeji was a huge leap forward compared to all my previous works in nearly every aspect. It was extremely ambitious and required me to learn many new tricks on a timespan of 35 days for yet another contest (Famous Landmarks). Originally I wanted to create the Atomium from Brussels... but after 5 days of trying and getting frustrated it just didn't work. It was then that I saw an image of the Himeji Castle on Google and I was instantly grasped by its beauty and detail.

I brushed out my knowledge of modeling and texturing, using Sketchup to model and 3DS Max to texture. This was the last project to ever use Google Sketchup and Hammer to solve independent tasks of my projects. Himeji took exactly 35 days to complete with considerable time spent on the map, each and every day. Having time over at the last 3 days I added more detail and gameplay elements. Later after release it received a huge update and a better attempt towards competitive play. The map is still unfinished and is one of the few projects that I want to finish later down the road.

Himeji took first place, as can be seen here: http://gamebanana.com/contests/winners/42 Having been the first map to ever win a place it boosted my spirits to continue.

DE_CATHEDRAL (CS:GO)
URL: http://csgo.gamebanana.com/maps/181217



Despite its barren gameplay and experimental graphics, Cathedral holds a special place in my heart because it is the first map I have ever attempted to finish and release publicly while being made entirely inside 3DS Max with the aid of the Wall Worm Level Designing tools (Wall Worm Anvil, WWA from now on). WWA and WWMT allow you to essentially make a fully fleshed out map in 3DS Max without ever leaving the program or environment, or even entering Hammer for that matter. You can design brushes, models and textures all in 3DS Max, so why would you even want to switch between all these programs to essentialy waste a lot of time for the same results.

Such filosofy allowed me to change my workflow and transfer towards Wall Worm for creating all my levels. There are so many essential tools in 3DS Max which simply aren't present in Hammer. If you count that on top of all the tools designed by Shawn Olson such as the 3D Skybox converter, Sky Writer, Corvex, Carver, ShellVex, Detailer, PropLine and finally Wall Worm Pro, you're saving houres and houres of work.

Admittedly, while it holds a special place, it is nearly impossible to properly play on because of scaling issues and shading problems. I was faced with the hard reality that gameplay is, and remains the key element of a map.

DE_PONTIFICI (CS:GO)
URL: http://csgo.gamebanana.com/maps/185072



Pontifici was the first attempt to improve the gameplay element of a map, taking the basics of the competitive spirit to play by balancing timings for bombsite and cover, while still trying to go for a relatively open map. It should give you the feeling of being in a medieval town that uses a canal for transport, connecting both sides with a bridge.

While it succeeded in gameplay partially, it can be improved, and already underwent considerable art-transformations that are unreleased as we speak.

SPACE MARBLE (CS:GO)
URL: http://csgo.gamebanana.com/maps/185591



Space Marble is the direct result of my temporary retreat of the classical gamemodes such as Bomb Defusal and Hostage Rescue. It is a prime example of a map that hugely benefitted from Wall Worm as a whole. The map simulates a Marble Track by letting spheres roll around with purely physical interaction. A very tough task to ask for the Source Engine if you consider that physics are very dependent on material type and object weights.

In order to create the tracks, I used splines for crazy turns and creative routes that at times also diverge into branching paths. In Hammer this would have been a hell of a nightmare to design considering that you'd have to create the tracks and then export them, and then reposition the new track everytime, hoping that it is attached seamlessly to the previous one etc etc...

Space Marble took first place, as can be seen here: http://gamebanana.com/contests/winners/61

SURF_GREATRIVER_2015 (CS:GO)
URL: http://csgo.gamebanana.com/maps/185591



Greatriver is probably one of the most popular surf maps of all time. CS:GO currently has a hard time to really grasp CS:S players to over to the new game, so I decided to support the Surf Community by remaking this iconic masterpiece. Black is the original designer of the concept and first version, and Xdream (clan) has made considerable improvements over the map.

I co-operated with a professional surfer who has a lot of knowledge about surf that I simply didn't know. He adviced me to keep the map as close as possible to the original, as well as removing some elements which simply had nothing to do with the gameplay. I intend to bring the ultimate greatriver edition by focussing on Surfing as much as possible, rewarding those with great skill, as well as making the map just a little bit noob-friendly. The most controversial change is undershotty being removed (a connection on a sloped surface on the back of the ramp that starts from the 3 little platforms with the shotgun, hence 'undershotty').

Greatriver 2015 boasts a ion-like design and tried to guide the player on the ramps by using linear textures. Reflection and river-like elements are important, so the map is mainly tainted in blue.

[wallworm]

Thank you for this great write-up! I will watch your progress with great interest.

I am glad you shared all of this!

PS. At the moment, the images are broken.

[Joris Ceoen]

PS. At the moment, the images are broken.

Ugh, I know why, my bad. Gonna fix that in a minute!

EDIT: Fixed!

[Neil]

Really cool :D

[Infidel]

WOW, great work. I am impressed with the overall quality of the images shown. These maps look fantastic. Thanks for sharing them with us and I look forward to further updates.

;D

[Joris Ceoen]

WOW, great work. I am impressed with the overall quality of the images shown. These maps look fantastic. Thanks for sharing them with us and I look forward to further updates.

;D

Well I recently finished the entire inside (I still have minor work on the organ that I'm building now). Once that one is finished, I'll post new pictures so you can see the 'finished' inside. I still will have to add furniture and carpets/altars/cathedral stuff xD But that's something for later when the level design is done, as it is part of gameplay instead of level design, which I intend to do later on.

I'm also starting the outside, which I believe to be even more challenging, partly because it's something new over the older version, as well as being difficult due to lots of details that I'll have to calculate in terms of polygons, and value.

Thanks for the encouraging comment!

[Joris Ceoen]

I updated the post with my latest map de_himeji_castle

[wallworm]

Great job! The map looks great. I will add it to the Wall Worm server this weekend!

[Redskull]

I know this an old thread but I just wanted to say your maps look like professional quality maps, good job! Do you mind me asking what process you took in deciding to design your levels? I always find myself starting a map but never finishing it because either doesn't look right or it bores me to death after play testing it.

[Joris Ceoen]

I know this an old thread but I just wanted to say your maps look like professional quality maps, good job! Do you mind me asking what process you took in deciding to design your levels? I always find myself starting a map but never finishing it because either doesn't look right or it bores me to death after play testing it.

Well, when I first started with cathedral I was tempting to do Zombie Escape but it ended up being a failure gameplay-wise so I'll keep that one out of consideration, but generally speaking for all of my maps, the first step I take to make my levels is to have a concept or a particular idea that I want to recreate, possibly accurate or partially.[/li][/list]

For instance, my de_winter_forest map, I always wanted to have a large open-area map with foliage in it. At the time I was just learning modelling so the foliage in the map isn't that great, but the idea was there. I then had to consider the gameplay and alter my original idea of that forest and large-area by adding bombsites and a somewhat manageable layout. As the map was large, snipers where a total candy whereas shotguns are a nightmare unless being in the house.

My cs_ardennes map which followed is probably the most unique map in level design approach. My goal, just like it has always been until the organ, was to create 'impressive maps' by using foliage and being able to use it in a map for Source. This is a challenge because foliage is hard to make correctly and hard to implement in a map competitively (actually competitively it will never work, but you can use it in maps such as hostage rescue if it doesn't dominate too much. Personally in the last operation (Breakout), the forest-like map with different spawns and a loooot of fog is a nightmare, there's just too much vegetation and you can barely see T's). For the idea, I always wanted to recreate a real-existing place I've been to many many times in my life. The link can be found here: https://www.google.be/maps/@49.813257,4.99167,3a,75y,299.97h,88.85t/data=!3m4!1e1!3m2!1sjx7pdqoKpPw4dLxhqjlZlw!2e0!6m1!1e1 (The images have been updated ever since I looked at it, at the time it was summer on Street View...).

The approach and design is different, because I literally wanted to recreate something existing. This has given me so much joy in level design it has haunted me until de_cathedral (http://gamebanana.com/members/submissions/maps/1241752). Literally recreating a location or items helps you immensely in learning approaching objects on images and look at them out-of-the-box. Not only that, but recreating something existing in real-life directs you to references you can base on, whereas making something out of nowhere requires more than just images. Therefore it's really an easy method to make a level, as you already have an example of what you want to have at the end of the road.

The problem, the BIG problem is that if you really want it authentically, you have no layout choice. You will just end up with a map that should more or less 'work out' for players, and generally this is just never the case. There is hardly a real-existing battlefield or location in the real world that authentically matches all of CS:GO competitive scenery. It doesn't mean you can use several parts of it or just a few elements to dress up your own layout. It also doesn't mean you can't choose this level design approach at all, it could just end up being a very fun pub map!

I would say that for such levels, hostage rescue is preferred as that gamemode is extremely flexible and with just good placement and a few tweaks can quickly end up being a good hostage rescue map. Not to mention it learns you so much about insight and recreating props that you might need for later, more advanced projects.

One of the biggest dissapointments and the best example of a map that didn't 'work out', not even as a pub map, would be my organ. I entered it for a contest and thought to hit high with originality. Not only did I lose but I didn't receive any feedback whatsoever about how the map did etc, and I never, ever heard of it or even saw it ever on a server. It was literally a vanity project for the public, but I did learn a lot on that project, especially on how to turn a real-existing object into a whole battlefield. What's better to choose than a rats styled layout for that? But, as you can understand, the concept of an organ in CS:S just didn't work out. It learned me that a lot of time can be spent in a map recreating something real, but the risk is so big that it's better to stick with something that could work.

This led me to my first somewhat 'succesful' map. It's gonna sound weird after all of this but again, this map was nearly as good as 100% accurate! The theme of the contest this one was entered in was to recreate a real-existing monument, it could be any around the world, and output it in a map. Preferably having the monument part of the gameplay and the gameplay itself doable. I had no inspiration whatsoever, which led me to the inspiration page that was released at the time. It contained images that showed monuments, plenty of which I never heard or saw before. I did the old method again: Choosing what I knew personally the best. As I live in Belgium, this was the Atomium (http://tempora-expo.be/img/p/37-412-thickbox.jpg). The idea was cool, but with the spheres it started to become a nightmare after 5 days. Having spheres in Source is one thing almost undoable, but then having holes in them and connecting them is total apocalypse. I ended up with corruption in my files after even just a few of them, and I had to scrap the idea. As soon as you start to realise something is not gonna happen or at least very, very unlikely, quit the idea or save it for later when you're not under timepressure/deadline.

I then went back to that page I talked about, and saw Himeji Castle. I looked it up on Google Maps and saw images. Now, not only did it look amazingly cool, but I was always interested in Japanese architecture (and I also frequently followed anime series with the same style such as Rurouni Kenshin) and it had lots of details to impress with. As I developed the skill of recreating over  about a year, I wanted to hit the first place and take it seriously.

This approach might really interest you, depending on what you want to do, but here goes:

• First, I saved a lot of images from all perspectives of the castle. Just browsing this on Google will yeild you the same results.
• Then, I wanted to have a 3D scale of the model. As I understood with little research that it was very famous, I knew Google Warehouse had to have a file somewhere. Indeed, there were many version, and I took the one that was actually used on Google Earth: https://3dwarehouse.sketchup.com/model.html?id=8ca606490df33edb90d97ba3324988bf. I imported it in 3DS Max and simply outputted a scale of the entire thing as a Source model with WWMT, somewhat proportional to a Source player. Note that at the time, I still used Hammer and WWMT only for models (silly me, I could never think about that ever again!).
• I put the model in a fresh new Hammer file, then started reconstructing the brushes around it so that I would have a nice copy in brushes from the entire castle, just like it exists in real-life.
• After that, all I did was carving some new hallways inside the castle which I could hardly completely know as those didn't exist in the 3D file, neither were there enough images. There's something you gotta consider when doing real-life examples and that is not to have all of it 100% the same in terms of inside, as it might cause panic among people that think it could be used as a terroristic attack. It sounds silly, but I understand that and there have been incidents with these situations in other games in the past (COD airplane attack).
• Afterwards, it's really just analysing pictures for textures and colors, and props. I had to use some typical props from the environments. As rice fields were nearby, some rice bags where a good idea to recreate. Then finish up some details and the map is ready to go. To my favour, the layout of the map carries a lot of potential, and even up until today it can heavily be improved.

brag in 3... 2... 1... I won the contest *big fireworks in the background suddenly pops*. I take it was because of the detail and time I spend in it + the requested features of the contest where all there, the quality was there and on top of that it looks like the real thing.
[/list]

It learned me so much about how to perceive models. The ricebag example is perfect, as I never made ricebags before. Just looking at pictures and trying to recreate it in 3DS Max is amazing, then seeing the textures come to life and easily exporting it to Source with WWMT is nothing but joy!

This is where this thread ends, but I will update it to include an extra map that I recently released: DE_CATHEDRAL (I'll relink here: http://csgo.gamebanana.com/maps/181217).

Now this map is currently what I consider the pinnacles of pinnacles across my whole line of level design, even though when you'll probably open it you're gonna be like... Really? Because of a few reasons (which I point out just before I continue):

• It was released for yet another contest, and because of time constraint, the map is more than just rushed: Many textures are missing and the gameplay is just empty.
• Basic lighting and normal problems (which are all already fixed by now).

Now that we got that clarified, I think it's really the most interesting project for me till date because it contains a vast amount of techniques and tricks that I have never been able to do/achieve in Hammer. Never ever could I make a 3D Skybox like cathedral contains because in Hammer precision to the Skybox like that is impossible unless using WWMT. Texturing brushes the way it can be done with WWMT and the speed at which you do it is impossible.

Let me go into detail: Obviously after this whole text, if you're still reading it, you'll understand how my approach works. This time around my idea, a dream I always wanted to realise, was to make a big fatass cathedral with all the pletory of stuff and details and just put it right there in CS:GO. I searched some cathedrals up and the one that caught my eye was the Saint-Vitus Cathedral in Prague. I started to recreate it, as a cathedral can hardly be different than what it always has been: a crossform designed buidling that in the gothic style relies on the vertical element.

I knew beforehand that hundreds of thousands of triangles where at stake, so I had to have an idea to optimize the map beforehand. Usually optimization can safely come at the very last phase of your map, but in my case it was different: CS:GO has no LODS anymore and relies on memory more than polygoncount. Yet the polygon count was high enough to consider a plan that would naturally optimize the map with little effort. I wanted the whole structure to be visible at all times, yet making sure that whatever can't be seen on the other side would be culled. The only way to realise this is with a good skybox. However 3D Skyboxes are a nightmare in Hammer: You have to scale everything down 16 times (models included which all have to be done manually in a seperate program if using Hammer) and then put those pieces at the relative correct position from a point entity called sky_camera. You can imagine the time and effort it takes. With the precision necessary in my plans, I could only have done it with WWMT which does this all automatically.

When you scale brushes in Hammer, textures don't scale along, and even if you scale them, their offset gets lost, meaning you can't syncronise textures with 3D Skybox elements and normal elements alike. But with WWMT, this was possible, and this made my idea of cutting the whole map horizontally in half achievable.

Cathedral is build so that the upper part of the cathedral is all 3D skybox and everything below (just below the roof) is all the playable world. This makes sure that whatever is at the other side gets culled, even without using any hints or areaportals. The thing is, all the elements in skybox and playworld are all the same size and perfectly offset with textures matching perfectly.

Once that was done, I pretty much just had to focus on the environments of the cathedral. I purposedly chose for an italic environment as it fitted the cathedral its colors and all of those assets already existed in CS:GO making it not a 100% custom content map which otherwise would quickly rack in megabytes. Unfortunatly because of the deadline (which was btw 90 days :O) I WASN4T Able to even properly start the details on the environment and the gameplay, ending up in the current wacky version with broken textures and no gameplay.

However, it was the first map made entirely with Wall Worm Model Tools and its Level Design tools, and it taught me one thing: I will never touch Hammer anymore. With the tools it offers, you can do so much you just can't do in Hammer, just not/noppes/nada. You can create a 3D Skybox on the fly, in other words during the process of making the actual playable level as if all of it was part of it. Usually with Hammer, this only comes once you have everything, and then you'll quickly realise how many things can't even be put correctly on the Hammer grid on a scale of 1/16. You can create models on the fly, with textures built entirely in 3DS Max. In Hammer when you don't have a model, it's like awwww not possible or you have to search the internet for an eternity. In 3DS Max it's like, ok let's create this. Render the texture, export all the files and the level and its there! Not only that, you can texture brushes so much more quickly and the selection tools are endless. Considering that Valve actually never uses Hammer but Maya internally with their own tools (they went with Maya, their choice) it makes it all the more clear that level design in a program like 3DS Max is much more evident and useful. Hammer is a program created for those who don't have the time to invest in custom content or in other words the possiblity to do whatever you want and just realise it on the spot. It's definitly not bad, but for the best, Wall Worm Model Tools is just the way to go.

In the end, what keeps you going entirely for a project, in all aspects, is a good layout that feels right, one that works with the chosen gameplay and that doesn't imbalance either teams. Visually it has to be pleasing, and most importantly interest you, in other words choose a style that makes you enjoy level design. If you don't want yet another warehouse themed map, don't start it. I can't bang my head enough against the wall about people who still choose this more than overly abused typical style in fighting games, but yea if you really would want that for example, do it. If you really want something cool and unique, do so! Don't make a dust map because people like, make a dust map if you want it! Then, you can start thinking about how to make it good for players too. Working on a project with a style, idea or design you don't like will never get you running, and you'll end up being bored indeed.


I hope that this was an interesting read, and hopefully it wasn't too long. I really think that everyone should start with 3DS Max instead of Hammer, because all of the above projects, dating since my first map (ze_cathedral) all used Wall Worm Model Tools, and I couldn't have realised them without it. I'm currently using WWMT for 2 new other projects and I have also been using it for Aperture Tag which is going to be released soon: http://store.steampowered.com/app/280740

Let me know if you have any more questions and I'll be glad to answer them :)

[Redskull]

Thanks for the info I know that took some time to write. I've been basically doing the same thing except in tf2 but never thought bout using Google warehouse. I know you can use sketch-up thought to export maps but I know trying to make brushes with it is very ineffective.

I've been researching ways to properly design a level using the world of level design website. So, I appreciate all the info, will take some time to digest the information. I think maybe next time I make a map I'm going to try and strictly use wall worm. After seeing that cathedral map theirs no denying that theirs just no way you can achieve that kind of detail accuracy in hammer alone. Especially with how your skybox  is setup, that would be a pain.   Just looking at it again I can definable see the detail you added, you did a great job.

Also the reason why I'm bored with the current map I'm detailing, is because i choose another warehouse themed map.  At first, I didn't dislike the layout but after adding a little detail the whole idea just became bland to me. I had another idea in mind but was just doing that one because I decided I would take a different approach to mapping. I normally never plan my maps or use references so on that attempt I actually used references and planned out the layout. :/

[Joris Ceoen]

Updated the whole OP. Will always update in the future.