One of my favorite things in simulation games is seeing your creation come alive and just sitting back, watching all the bustling activity.
In Parkitect it’s mostly all the guests running around who are responsible for creating this effect, and it’s not quite yet where we want it to be.
“The more the better” has been a sort of unofficial motto throughout development for us, whether it be object counts, park sizes or the amount of rides - the more we can add the better it’ll be for this game.
Guest counts are still lower than we’d like though and right now is the last good time to properly solve this, since increasing the amount of guests also affects the money balance and the entire campaign.
Improving guest performance has been a constant task nearly every month throughout development and it’s at a point now where the only significant further improvements can be gained from multithreading. The main hurdle there is that Unity, the game engine we’re using, is making multithreading really difficult. They are currently making very impressive progress to improve that for the future, but it’s too late for Parkitect to use the solutions they are working on. So I have finally started with finding my own solutions for these difficulties. It’s a very annoying task that requires going through the entire guest AI code and making it compatible, but so far it looks like it’ll be worth it!
As a first smaller improvement, overall performance in Beta 7 will be up to 15% better in busy parks.
The really big improvement will not be in Beta 7 just yet, but I hope I can release an experimental version for anyone interested in testing sometime during June. In the first tests it looks like we should be able to have quite a few more guests while at the same time having better performance overall. Here’s a small teaser:
That’s 8000 guests in a pretty detailed and big park (Northern Highlands), with roughly 1500 guests on screen, running at 65 FPS!
This is just a stress test of course and we won’t increase the guest count anywhere near this dramatically, because 1) it turns out “the more the better” has a limit where it stops being fun and 2) the game needs to run on weaker systems as well. Comforting to know though that guest performance should never be an issue again after all of this! You should also actually start seeing guests complain about long queue lines...
Parkitect
Texel Raptor
Texel Raptor
2018-11-29
Simulation Singleplayer
Game News Posts 215
🎹🖱️Keyboard + Mouse
Very Positive
(6061 reviews)
http://themeparkitect.com/
https://store.steampowered.com/app/453090 
The Game includes VR Support
Parkitect Linux [355.71 M]Parkitect DLC1 Linux [14.22 M]Parkitect DLC2 Linux [31.13 M]
Parkitect - Taste of Adventure
Parkitect - Booms & Blooms
Build the theme park of your dreams, and keep it running! Meticulously design that newest roller coaster, place the rides where you want them to be, and construct scenic structures and objects to make the park beautiful. Let your creativity shine, and when you're finished share it with the world!
Features
- Tools to build the theme park of your dreams
- A robust, piece-based roller coaster designer to let you build the greatest roller coaster you can think of
- Detailed guest behaviors. Watch them take the train to that new section of the park, or read their map to find a ride
- Build efficient transport infrastructures to keep your park stocked
- Maintain the illusion by hiding utility structures and the inner workings of the park from guests
- Extensive support for mods and custom content
- OS: Ubuntu 18.04+. SteamOS+
- Processor: 2.3 GHz Intel i3 or comparableMemory: 4 GB RAM
- Memory: 4 GB RAM
- Graphics: Shader Model 3.0 capabilities (anything with a graphics driver that isn't older than ~2016 probably works)
- Storage: 2 GB available spaceAdditional Notes: This will run the game. but for the best experience something closer to the Recommended Requirements is better
- OS: Ubuntu 18.04+. SteamOS+
- Processor: 3.0 GHz Intel i5 or comparableMemory: 6 GB RAM
- Memory: 6 GB RAM
- Graphics: Dedicated graphics card made after ~2015
- Storage: 2 GB available space
[ 5951 ]
[ 3221 ]