Traffic

123 views
Skip to first unread message

farina.a...@gmail.com

unread,
Apr 1, 2012, 12:30:39 PM4/1/12
to syne...@googlegroups.com




So , i love this game. And i have played it only today :p . But i built a quite large city and i want to have a little bit more life to it.

So i think traffic would be a good addition ,it does not need to be a objective traffic system , but just some boxes in roads that roam around.Atleast to have some life you know?

Synekism

unread,
Apr 1, 2012, 12:39:23 PM4/1/12
to syne...@googlegroups.com
Traffic is one of those things that has the potential to take the
glory from all other features of the game - all for itself. Therefore,
the other features were concerned (jealous?) so we had to sign an
agreement that we will give them a head start before starting on
traffic.

Ok, jokes aside for a moment, I share your enthusiasm for traffic and
it will come sooner than later but we need to address some serious
problems with the core of the game before we can move on.

farina.a...@gmail.com

unread,
Apr 1, 2012, 1:19:22 PM4/1/12
to syne...@googlegroups.com
Okay , and congrats for starting such awesome game ;D 

Jarren Butterworth

unread,
Apr 2, 2012, 7:18:39 PM4/2/12
to syne...@googlegroups.com
I really enjoy cities in motion's traffic system.  Although cars and traffic are of course present, the focus is on people them selves.  Rather than limit pedestrians to specific networks they just go anywhere they like.  Through a park, between buildings, anywhere.  Simcity failed to present pedestrian traffic because they tried to simulate pedestrians the way way they simulated traffic.  I doubt this game is going to actually track each "sim", specially not to the detail that Cities in Motion shows them.  When you're designing your pathing system, I'd consider implementing a system that simply maps where pedestrians CAN'T go rather than where they can.  Roads, parks, open land, and even many buildings would just be treated as open pathable land.  Only things like a river or specific types of buildings/zones would block the pedestrian's progress. 

Or if one was to go very very simple (if I was making a city sim, the traffic pathing would probably be the #1 headache) just ignore pedestrian "traffic" all together and use a radius system.  If source building A is within X meters of target building B (as the crow flies), a pedestrian trip is assumed to be made.  Would save much processing and programming time and, unless one has built the most extreme awful American style suburban disaster, would be a perfectly fine abstraction and realistic.

Transit could work to a similar extent.  Is there a metro stop within Xm of my home?  If there linked metro stop within Xm of my destination?  If yes, a metro trip is taken.  Don't worry about calculating exact paths,  just abstract it and keep it simple.

Just some ideas on how to keep it simple and do-able on an "indie" level.

Synekism

unread,
Apr 2, 2012, 7:55:29 PM4/2/12
to syne...@googlegroups.com
Those are some good suggestions. Now we just need to figure out how to
assign each person a home and a job and not have them too randomly
distributed over the entire city.

You're right in the fact that we are pretty much going for a
statistical model instead of the more "accurate" but much harder to
get fast agent-based model. We choose this right out of the gate
because we want to go macro (really macro). Hopefully accuracy won't
suffer too much.

edin....@gmail.com

unread,
Mar 19, 2013, 9:19:25 AM3/19/13
to syne...@googlegroups.com
I thought about that, and I got an idea. Why wouldnt you put numbers on citizen and combo of number and letter on buildings. So, first citizen that moved to city is citizen code 000000001, living in residential building ra001, working in industrial building ia001 and shopping in ca001. Max number of residents in low density res building ra001 is 10, workers in low density ind ia001 is 50, shoppers in low dens commerc building ca001 is 100, so ca001 covers ten low dens resid buildings ie ra001, rb693, r696 etc. Every third person in city need to work, so to fill ia100 with 50 workers, you need 15 low density res buildings. I think this is much more realistic. Just attach info about citizen id to buildings and forget about them.

So for example, citizen 000167689 lives in high density res rv445, works in low dens ij100 and shops in cb065. There is metro line crossing the city. Line 1 is passing through station ms001, which is in radius of rv445 of 100m, and station ms045 which is in radius of ij100 so citizen 000167689 will board train on line 1 and get to job. Shop cb065 is in radius of 1000m of rv445 and that is the reason why citizen is attached to it. To get there he can walk or take bus line which is calculated the same as metro line. Cars are used only when there is no any public transportation available. What do you think so far? In case that any of buildings is demolished, entire attachment system for a citizen has to be recalculated. But it would work flawlessly in my opinion. Just assign one core for this and the rest can do rest. You dont have to animate citizens on streets, that can be random. Butisong those attachments, you still can have immersed experience + every building would have some meaning.

I can work on this idea, and even help you with coding it. Just tell me what lang you are using.

Synekism

unread,
Mar 19, 2013, 10:43:25 AM3/19/13
to syne...@googlegroups.com
I could certainly see your system work reasonably well. I would argue against letters in any sort of identification as numbers are trivial to store and search compared to mixed letters and numbers but other than that there's no specific flaw I can see at the moment.

That said, regrading actually implementing such a system in Synekism, I'm a long ways away from being ready. Since I have absolutely no reason to rush this project to some sort of "completion" I really want to focus on bringing the current features in the game up to respectable standards instead of adding more features. Minor features will of course be sprinkled as development continues, but new major features such as traffic simulation will have to wait until the current, arguably plain, core gameplay is up to par.

Since your thoughts are mostly about specific implementation details, although much appreciated, I can not consider them too closely because implementation will significantly depend on future architectures in Synekism that I cannot predict.

Regarding help with coding, I am not and will not be any time soon in a position to accept direct product help. Synekism is closed-source and, although free for now and the foreseeable future, I'd like to keep my options open regarding monetization. The problem is therefore mostly about liability and not about your individual skill - but I'll certainly make it public if this status changes.





--
You received this message because you are subscribed to the Google Groups "Synekism" group.
To unsubscribe from this group and stop receiving emails from it, send an email to synekism+u...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.



Message has been deleted

edin....@gmail.com

unread,
Mar 19, 2013, 10:54:15 AM3/19/13
to syne...@googlegroups.com, farina.a...@gmail.com
I understand you. I just thought to help you because I am really craving for a good city building game... I truly want to help, so, if you want to share the info, but what is next you are planning to put into game?
I kinda have an idea for procedural buildings too... I can even send you a basic code for that too. I just want to help and money doesn't interest me that much honestly but I see where that concern is coming from. 

Synekism

unread,
Mar 19, 2013, 3:11:24 PM3/19/13
to syne...@googlegroups.com
I'm currently still working an a graphics engine rewrite in order to better support future changes to essentially all components of Synekism as the graphics engine drives almost everything. The UI rewrite will come next because its getting harder and more confusing to keep relying on keyboard shortcuts for new feature and the current UI infrastructure is absolutely terrible. Finally, I need to rewrite the game saving code, again, in order to more easily add depth to the game in the future (and be able to save it).

Actually, just think of this as a full rewrite of the entire game with minor feature additions and bug fixes along the way. This is a very long-term project...

As for help, again, I really appreciate it and trust me when I say that I'm equally starved for a good city building game (Synekism is my proof, by the way), but you will have to find other ways to spend that energy - at least for now. I'm working hard (not very hard these last few weeks, but "in general") to open the game up for modding as soon as possible.



--

edin....@gmail.com

unread,
Mar 19, 2013, 3:21:29 PM3/19/13
to syne...@googlegroups.com
Okay. Fair enough. Just another thing. Have you considered making game modular? Is it even possible? Would it be easier to add modules later? Modules like Public transport engine which can be one huge function with entire systems (buses, trams, metros) as individual submodules? Or another module which represent utilities (water, sewage, electricity) as another function, that can be used by main game engine (population, buildings, resources). What do you think? Would it be easier later to fix bugs, so when trams get crazy or plants dont deliver electricity, you just check modules and not entire game.

Synekism

unread,
Mar 20, 2013, 3:02:05 PM3/20/13
to syne...@googlegroups.com
It's still very much a debate whether I modularize Synekism or not. I've been trying lately to stuff as much as possible inside the main Synekism executable (for example, all the textures for the UI are inside the exe). It's also much easier to work within a single module as long as it's not too huge, which Synekism is not.

Of course, single module pattern is not very scalable so I will at some point in the future start splitting it up. But for now, it will be just the one exe.

On Tue, Mar 19, 2013 at 3:21 PM, <edin....@gmail.com> wrote:
Okay. Fair enough. Just another thing. Have you considered making game modular? Is it even possible? Would it be easier to add modules later? Modules like Public transport engine which can be one huge function with entire systems (buses, trams, metros) as individual submodules? Or another module which represent utilities (water, sewage, electricity) as another function, that can be used by main game engine (population, buildings, resources). What do you think? Would it be easier later to fix bugs, so when trams get crazy or plants dont deliver electricity, you just check modules and not entire game.
Reply all
Reply to author
Forward
0 new messages