Question Any advice for level design specially for cities?

While I've been obsessed with level design, making all kinds of terrains vast landscapes with small villages around it is really my First time working on a city while fortunately and unfortunately to be placed in top of a team to make a huge city that has that cyberpunk feeling as the game take place in somewhere around the 2050 not that extra miles from here.

And to be honest with you, for someone who've been sculpting landscapes and placing foliage easily with the paint tool, and adding smaller buildings with narrow passages to create a village in the middle of nowhere,
I was overwhelmed by the amount of assets I got, not to mention that its just half of them, even the smallest parts, like a tube or sign or a light bulb, including walls building parts that I have to use to make a verity of buildings including collapsed buildings and what not and the only thing I got is a city map including key regions and streets, highways, places to highlight the most (Kinda unfortunate I can't share it with you) .
some of the assets are ripped from the High city and another one that I recognize and the others are made by the designer with much details.

My question is HOW TO MANAGE THIS? even with concept arts and some 2D painter of how the city will looks as i said it feels overwhelming to think about placing every small part in its place correctly and snap it even with Octave tool. any tips and tricks to make my city feels alive and engaging ? any tools that I can use even paid ones to accelerate the work as we have a deadline (pretty long one I guess but stills not that much for a city that scale).
I really need any piece of advice that could help me out.

Thanks in advance

 

Quantum Theory's Rome tutorials, which anyone can access, are good to see how to put together buildings from small parts. Studying demo scenes and constructing your own prefabs right inside the demo scene is also a good way to get up to speed. Cyber City is a cheap asset with a good, if unoptimized, demo scene. High City is made by the same artist, so you should definitely check out the demo scene for that if your team is using it.

Like, if I were you, I'd fire up the demo, pick one of the dead-end streets, and start extending it.

 

This is one of those problems that should have procedural help, really.

 

To be honest I would completely ignore the amount of assets and consider the most important thing first: gameplay. As in: is it fun? Is it interesting to explore? Etc.

Think about how to create unique, interesting, challenging quests etc within this environment. This should drive how you scape the city. You can and should use the simplest assets for this, aka grayboxing. Once it feels right, you can fill in the details and make the scenery feel alive. That’ll make it both easier and fun once the gameplay is approved. Expect several passes for both, and the latter should only be done if you are 90% certain that gameplay will work within the environment.You may even be surprised how quickly the visuals come together once you‘ve got the gameplay fleshed out.

For instance, if it is a racing game you‘d want to focus on the roads, and shortcuts, perhaps jumps and certainly static and dynamic obstacles (ie train crossings). And playtest a lot.
If it‘s an RPG it‘s most important that the world feels believable. You wouldn‘t want the police station and the mob hideout next to each other. You‘d want well defined quarters with recognizable architecture and visuals so the player instantly knows where she‘s at.
Things like that.

 

Agreed, but don't forget that you can do concept art for your visuals while you're grey boxing your gameplay. That's cheap to iterate on, so you can revise it as your grey box evolves, and when it's time for artists to start adding detail you'll have clear targets for them.

 

I totally agree With you, but same time I have nothing to do about it , I got hired to get the job done, I know a lot about the game They are going to make, I've seen them already doing some gameplay mechanics testing on "GYM ENVIRONMENT" or so they called it, like basic shapes walls to run on targets to destroy or stuffs to to collect.

I don't know, there is too many things and the details about the game story are scars as I'm hired to get a certain job done, Under supervision even though I'm a head of a whole team, you know you sign those NDA's and whatever contracts and full secrecy, I do understand why but its a bit absurd.

anyways, I'm not here to complain about the comp i work for, my point is that the amount of information fed to me are a bit scars and while I know a LOT about the gameplay, and i heard almost every single music available made by the composers and the audio SFX made by the sound and sound engineers, I feel like the lack of details about the game story is a little pain in my back, as mentioned before I have concept arts and the concept artist literally shows no mercy in the amount of details he add and I'm supposed to match it with very smaller pieces, god even air conditioners are made in smaller pieces and you literally can create and least 15 verity of them and yet we still can make more, they aim for the level of details such us in the megacity demo as well as the gigantic city that's why I have posts there asking about how to open that project so i can have a look around it (already figured it out) and to be honest it feels overwhelming to re-create at least 10 times the size of the city in the demo all that from small chunks just with that drag and drop technic, I do have OCTAVE, it really make it easier for us as I put 2 designers with me to make variety of bigger parts using those assets provided so we can later combine them like LEGO parts in the main project, still, I'm not so sure if it is even possible to finish it in deadline, even though we just started, its been month and a half now.

 

Should talk to your team if you think there is a real problem.

 

Don't start with the detail. Start broad, and as people are happy with the broad stuff you're doing work in extra levels of details. I don't know the correct terminology to describe this from an art perspective, but essentially: start broad, grow specific as you find things that work.

It's fine for your concept artist to have 15 variations of an air conditioner, but you're not going to start with that in the world. It just means that when your artists decide to assemble the air conditioner props they should have 15 fantastic examples of what they should look like. It also means that when they make other props they've got great examples of what changes and what stays the same, because that detail study has already been done with something.

If the team is sane then that should not mean that you're expected to start your world development by jumping straight to implementing details.

One thing to note: when you're populating a large world it's typical to do it largely with tools rather than by hand. Placing thousands of air conditioners is a huge amount of work for a person, but pretty trivial for a computer if it was thought about early on.

 

any tools you suggest except octave ,

 

Finally getting back into making our game...completely different game than we had planned, this was a lot of help. I have a fabulous illustrator on my team and instead of writing about the world and what it is all about, I can take advantage of the artist's concept art to show off what is coming. It is good to be back on the forums. Definitely what I needed to inspire myself and my team.

Missed you guys and gals.

 

If you are a new dev I would avoid open world maps like the plague.

There’s no out of the box solution here that I’m aware of.

You need to juggle:

-map streaming

-centering the map so distant units don’t get wonky

-implement methods to clear ram and throttle cpu usage

-toggle physics (you can’t have everything a rigidbody at all times)

-LOD and make an imposters system

-Probably AI LOD, you can’t just have navmesh agents retarget every frame

-culling

-LOD and culling management if you want a zoom

-day, weather, and possibly season cycles

-And of course general systems such as police agro which compound with each system

There’s a reason Bethesda and Rockstar are some of the few companies doing it, and they have half a billion dollar budgets.

 

While that's true, those are largely tech implementation concerns, which aren't necessarily the responsibility of the person doing environment / level design. In fact, I'd expect that the tech crew would be dictating and/or negotiating some limitations, rather than having the design / art team tell them how to manage or implement this stuff. Of course that needs to go both ways, but I'd expect it to flow somewhat from High Level Game Direction -> Tech Implementation -> Level / Environment Design + Implementation.

 

I'm probably late but here what is the general workflow, if that's of any use to anyone. Having lots of small pieces is great, you will have to work along two track:

1 - get the lay out and structure for the large scales elements such has building, streets, etc ... find the metrics to design module.
2 - create module out of the lower scale elements based on the metric found in the large scale elements, and on metrics of lower scale issues.

I would verify and normalize all objects positions such as replacing the reference swap the objects. This will allow to make template module from which you can generate variation or adapt to context quickly, module should be swappable with similar module, if you know a bit of code, that will help here.

Actually doing the large scope and filling details later is probably too time consuming, by creating module with details you do both at the same time, and only customized part that need it, that's less work.

Regarding the concept art, I would use it and break it down its visual language into pattern, to make the modules.

Tl;dr: build bigger bricks from the smaller bricks, make the bricks swappable easily.