Cities: Skylines

LightningHalberd

Edit history:

LightningHalberd: 2015-03-29 08:11:00 am

LightningHalberd: 2015-03-20 08:10:33 am

LightningHalberd: 2015-03-20 08:08:13 am

LightningHalberd: 2015-03-19 05:14:59 pm

LightningHalberd: 2015-03-19 05:10:16 pm

LightningHalberd: 2015-03-19 05:08:02 pm

LightningHalberd: 2015-03-19 04:25:27 pm

2015-03-19 04:24:25 pm

twitch.tv/immersed13

Speedrunning Cities: Skylines? Why not!

CS has a couple of things that would make it a decent speedgame:

- It has an end goal: reaching the final milestone, which is Megalopolis at a population of ~~80000~~ 75000; this goal is also reachable within a reasonable amount of time (I'm guessing 2 hours max RTA)
- Because of the fast forward ability, it makes some RNG such as the inflow of traffic from the highway almost unnoticeable. The fast forward ability also makes the game a lot more execution intense and therefore more fun
- All 9 starting maps always stay the same, so this means there are 9 different scenarios and each scenario has different optimal strategies

Mods and custom assets will not be allowed obviously (unless the specific mod makes the game more enjoyable and does not drastically reduce the difficulty)

Some initial observations:

- Grid layout = best layout
- Additional transportation besides car doesn't seem beneficial at all, it only helps to solve traffic problems which shouldn't exists in the first place if the layout is good
- I'm unsure about whether the best industry strategy should be either letting the population stay uneducated and using generic industry (and if so, should you go into specialized industry later on?), or getting education and building lots of offices (and if so, should you build high schools and universities?), or a mix of both.
- Going from low density residential/commercial to high density when it is available seems to have no downsides besides traffic congestion, which shouldn't exist in the first place
- Removing low density residential and replacing it for high density residential seems like a nice way to increase the population in the long run. However I don't know what having lots of high density residential means for the nr. of births since those only occur at low density (don't know this for sure though). So if this is the case then this may decrease the number of kids and the city will only attract adults from out of town, and I don't know if adults coming into town can be educated or if they need the education at our town
- Airports should be an easy way to quickly gain more citizens

Anyways, I hope to start a nice discussion about the potential of Cities: Skylines speedrunning and I'm looking forward to your strats Grin

EDIT: Leaderboards are up:

http://www.speedrun.com/Cities_Skylines

Thread title:

Speedrunning Cities: Skylines? Why not!

CS has a couple of things that would make it a decent speedgame:

- It has an end goal: reaching the final milestone, which is Megalopolis at a population of [s]80000[/s] 75000; this goal is also reachable within a reasonable amount of time (I'm guessing 2 hours max RTA)
- Because of the fast forward ability, it makes some RNG such as the inflow of traffic from the highway almost unnoticeable. The fast forward ability also makes the game a lot more execution intense and therefore more fun
- All 9 starting maps always stay the same, so this means there are 9 different scenarios and each scenario has different optimal strategies

Mods and custom assets will not be allowed obviously (unless the specific mod makes the game more enjoyable and does not drastically reduce the difficulty)

Some initial observations:

- Grid layout = best layout
- Additional transportation besides car doesn't seem beneficial at all, it only helps to solve traffic problems which shouldn't exists in the first place if the layout is good
- I'm unsure about whether the best industry strategy should be either letting the population stay uneducated and using generic industry (and if so, should you go into specialized industry later on?), or getting education and building lots of offices (and if so, should you build high schools and universities?), or a mix of both.
- Going from low density residential/commercial to high density when it is available seems to have no downsides besides traffic congestion, which shouldn't exist in the first place
- Removing low density residential and replacing it for high density residential seems like a nice way to increase the population in the long run. However I don't know what having lots of high density residential means for the nr. of births since those only occur at low density (don't know this for sure though). So if this is the case then this may decrease the number of kids and the city will only attract adults from out of town, and I don't know if adults coming into town can be educated or if they need the education at our town
- Airports should be an easy way to quickly gain more citizens

Anyways, I hope to start a nice discussion about the potential of Cities: Skylines speedrunning and I'm looking forward to your strats ;D

EDIT: Leaderboards are up:

http://www.speedrun.com/Cities_Skylines

LightningHalberd

Edit history:

LightningHalberd: 2015-03-22 07:04:55 am

2015-03-22 06:56:06 am

twitch.tv/immersed13

Just found this really cool video which discussed traffic patterns in this game:

When trying out different city designs I usually end up having many traffic problems near the off ramp of the highway where traffic just keeps piling up, so using the idea that he discusses at 7:26 and onwards could be helpful.

Also one more thing that I've found works: use multiple off ramps from the highway to your city once it becomes large enough, and build an industrial zone near every offramp. Each industrial zone should have it's own on and offramp seperate from the one going to the city to manage the flow of traffic going only to and from the industrial zone.

Foxxy

2015-03-22 08:37:04 am

Hey, this game should be very interesting to speedrun, I am kinda excited for it.

I have found few little things.
- if you run out of money right from the start and go down to -10k balance, you will get basicaly 50k for free. It should help to speed up everything if done correctly.
- taxes could be set to 14% and citizens won't complain (tested up to 20k pop city)
- it is possible to boost income by ramping up taxes all the way for short period of time before citizens start leaving the city
- I haven't noticed anything worse than few burned out buildings if you don't build any firestations, which may help with budget (burned out buildings could be demolished, which could be too micro heavy for consistent runs) but there could be other consequences which I haven't encountered yet

The traffics system based on dividing evenly everything throughout the city is very promising. Basically leading traffics out of the city around its borders could be viable if city plan could fit in it or rebulding roads is not too costly to rate at which new inhabitans move into the city. Building very tight city is fast and efficient, with correct traffic flow there should be no problems. I think that by using one way roads as much as possible and evenly distributing traffics should be the way to go.

Quote from LightningHalberd:

Each industrial zone should have it's own on and offramp seperate from the one going to the city

Great idea but also it could backfire because other than industrial AI may use that one exit if it would be most optimal route for them? (needs testing)

There is still a lot of stuff to optimize. What are all factors that improve population growth, what are optimal roads, how high is impact of bad traffic on the city, what services could be omitted, how large area do you need to reach milestones (with low or high density residential zones), etc.

Also you can't keybind everthing, yet. For example type of brush for zoning, curved road or road upgrade. There are mods for it but I hope it will be implemented in later patches? We will see Cheesy

Hey, this game should be very interesting to speedrun, I am kinda excited for it.

I have found few little things.
 - if you run out of money right from the start and go down to -10k balance, you will get basicaly 50k for free. It should help to speed up everything if done correctly.
 - taxes could be set to 14% and citizens won't complain (tested up to 20k pop city)
 - it is possible to boost income by ramping up taxes all the way for short period of time before citizens start leaving the city
 - I haven't noticed anything worse than few burned out buildings if you don't build any firestations, which may help with budget (burned out buildings could be demolished, which could be too micro heavy for consistent runs) but there could be other consequences which I haven't encountered yet

The traffics system based on dividing evenly everything throughout the city is very promising. Basically leading traffics out of the city around its borders could be viable if city plan could fit in it or rebulding roads is not too costly to rate at which new inhabitans move into the city. Building very tight city is fast and efficient, with correct traffic flow there should be no problems. I think that by using one way roads as much as possible and evenly distributing traffics should be the way to go.

[quote="LightningHalberd"]Each industrial zone should have it's own on and offramp seperate from the one going to the city[/quote]
Great idea but also it could backfire because other than industrial AI may use that one exit if it would be most optimal route for them? (needs testing)

There is still a lot of stuff to optimize. What are all factors that improve population growth, what are optimal roads, how high is impact of bad traffic on the city, what services could be omitted, how large area do you need to reach milestones (with low or high density residential zones), etc.

Also you can't keybind everthing, yet. For example type of brush for zoning, curved road or road upgrade. There are mods for it but I hope it will be implemented in later patches? We will see :D

LightningHalberd

Edit history:

LightningHalberd: 2015-03-22 04:59:45 pm

LightningHalberd: 2015-03-22 04:48:42 pm

2015-03-22 04:45:01 pm

twitch.tv/immersed13

Quote from Foxxy:

Quote from LightningHalberd:

Each industrial zone should have it's own on and offramp seperate from the one going to the city

Great idea but also it could backfire because other than industrial AI may use that one exit if it would be most optimal route for them? (needs testing)

I've tested this idea a lot and the off ramp is never an issue cause very little people actually leave your city. They can easily get off from the off ramp in the industrial zone. In the worst case you can always add another off ramp from the city just for that traffic flow.

Besides separate on and off ramps to industry I also like the idea used in this video:

The traffic flow is really visible around the 5:30 mark and it's going really smoothly. Might not hold up well when there is more traffic and unfortunately he switches to a different and more complex system later on.

I currently think that all services should be added just because it appears to attract more citizens and when the city gets large it may cause a lot more problems if it's not included. I do think that the number of services shouldn't be that high and that they can be added a lot later than when you get them (except maybe for education, garbage and leisure). Also leisure should be abundant in your city because it definitely helps to attract more people with relatively low costs.

[quote="Foxxy"][quote="LightningHalberd"]Each industrial zone should have it's own on and offramp seperate from the one going to the city[/quote]
Great idea but also it could backfire because other than industrial AI may use that one exit if it would be most optimal route for them? (needs testing)
[/quote]
I've tested this idea a lot and the off ramp is never an issue cause very little people actually leave your city. They can easily get off from the off ramp in the industrial zone. In the worst case you can always add another off ramp from the city just for that traffic flow.

Besides separate on and off ramps to industry I also like the idea used in this video:

https://www.youtube.com/watch?v=6raNfPhol2Q

The traffic flow is really visible around the 5:30 mark and it's going really smoothly. Might not hold up well when there is more traffic and unfortunately he switches to a different and more complex system later on.

I currently think that all services should be added just because it appears to attract more citizens and when the city gets large it may cause a lot more problems if it's not included. I do think that the number of services shouldn't be that high and that they can be added a lot later than when you get them (except maybe for education, garbage and leisure). Also leisure should be abundant in your city because it definitely helps to attract more people with relatively low costs.

Foxxy

2015-03-22 05:29:58 pm

Yeah, if you separate industrials like that.

Well after some point you will make excess amount of profit, so it will be matter of execution placing every service / leasure objects into the city.
And if the overall happiness / quality of city atracts more ppl, it should be faster to educate inhabitants. Also less plain industry means less polution, which could translate into faster higher grade buildings. And every factory imports and exports stuff, which congest the traffic even more. (using trains helps a lot)

Offices are not transporting anything (afaik).
Buses help with traffic if designed correctly, there is % value of how many citizens used bus instead of car.

It seems that it all revolves around saturating RCI zone demands and using remaining finances to help attract people faster.
I am still curious if there is something that would really break the game, haven't found anything thus far.

lillbrorsan

2015-03-26 03:21:39 pm

Truljin @ Twitch

Just tested doing a run today on Green Plains and it took just under 4 hours and 20 minutes, lost loads of time due to designing one part wrong so had to remake it which dropped over 10k, hope to do some more tomorrow!

lillbrorsan

2015-03-27 10:33:37 am

Truljin @ Twitch

Did another test run today and saved over 30 minutes, can speed it up even more since it was slow at 55-80k

LightningHalberd

Edit history:

LightningHalberd: 2015-03-29 08:05:31 am

2015-03-29 08:05:14 am

twitch.tv/immersed13

Just saw Bersi's great sub 2 run in Diamond Coast: http://www.twitch.tv/bersi__/c/6421379

My comments/suggestions:

- He's using a block size of 1x4, but I think that using larger block sizes like 1x5 (or even 1x6 though I doubt larger sizes will work) might still maintain a decent traffic flow while reducing road costs slightly
- That very simple loop to connect the on and off ramp and then not keeping industry and residential/commercial seperated seems to work amazingly well. I also like the idea of keeping the block just above the on and off ramps not zoned to make traffic flow better on that road
- There still seems to be a massive traffic jam at the first on ramp which is slowing down the inflow of traffic. I wonder if, because the industrial zone is spread out along the highway, building many on ramps from the highway to the city might work divert the massive traffic flow of the primary on ramp
- Going for that 2nd city below the first highway is probably what caused the dip to occur between 42000 and 60000. I think having two seperate cities, one with all the necessary services already in place and on with nothing, means that it's very unlikely for people to move into the one with nothing. In Diamond Coast (and probably others too) there is little room for expanding without having another highway nearby. In these situations what might work best is just expanding the main city to below the highway by building many bridges over the highway so that the new part is still connected to the main city. Then only later on you build another on and off ramp to this part of the city

Anyways seeing a sub 2 run is very nice and with sub 1:30 being a real possibility I am very excited to see more runs Cheesy

Just saw Bersi's great sub 2 run in Diamond Coast: http://www.twitch.tv/bersi__/c/6421379

My comments/suggestions:

- He's using a block size of 1x4, but I think that using larger block sizes like 1x5 (or even 1x6 though I doubt larger sizes will work) might still maintain a decent traffic flow while reducing road costs slightly
- That very simple loop to connect the on and off ramp and then [b]not[/b] keeping industry and residential/commercial seperated seems to work amazingly well. I also like the idea of keeping the block just above the on and off ramps not zoned to make traffic flow better on that road
- There still seems to be a massive traffic jam at the first on ramp which is slowing down the inflow of traffic. I wonder if, because the industrial zone is spread out along the highway, building many on ramps from the highway to the city might work divert the massive traffic flow of the primary on ramp
- Going for that 2nd city below the first highway is probably what caused the dip to occur between 42000 and 60000. I think having two seperate cities, one with all the necessary services already in place and on with nothing, means that it's very unlikely for people to move into the one with nothing. In Diamond Coast (and probably others too) there is little room for expanding without having another highway nearby. In these situations what might work best is just expanding the main city to below the highway by building many bridges over the highway so that the new part is still connected to the main city. Then only later on you build another on and off ramp to this part of the city

Anyways seeing a sub 2 run is very nice and with sub 1:30 being a real possibility I am very excited to see more runs :D

lillbrorsan

2015-03-30 08:32:16 am

Truljin @ Twitch

Managed to get my time down to 1:37 but 1:15 should be possible!

lillbrorsan

2015-03-31 09:06:32 am

Truljin @ Twitch

Down to 1:25 now, have some tweaks to do.

lillbrorsan

2015-04-09 04:09:28 pm

Truljin @ Twitch

So it turns out that the game will actually run slower if you stream it using OBS (or anything with x264 encoding) and go higher than 2 speed in the game, I guess it would be better to go by the date in the game as a timer?

pRo9aMeR

2019-04-13 08:06:16 pm

Hi, I currently hold every World Record that's been attempted in Cities Skylines. I'm bored; I'm hoping to find more competition here since it seems the previous competitors have since drifted to other games. A few maps still have unclaimed world records if anyone wishes to have an easy world record Smiley