Dev Blog 81: Testing, testing and postponing

August 22, 2017

Hey there and welcome yet again to another development blog, it’s the 81st in the order and this time we’ll skip the traditional format and jump into what anyone reading this is really interested in.

As some of you may know, we recently initiated a test program where we invited a number of dedicated friends as well as family family. This has been extremely helpful in our work of uncovering bugs and getting some valuable initial feedback about the playability aspects of the game as well as the feel of the UI. Unfortunately we did not have the best start and early on hit a number of critical and game breaking bugs which we had not been able to uncover by ourselves. We learned a valuable aspect about testing and that is that it doesn’t matter if a function has been tested 100 or 1000 times by a developer, it is still not really tested until somebody else gives it a try. To give a practical example of such a bug is in relation to the security check-points, a system that was developed early on and has been tested at least over a 1000 times in our different test airports running over several in-game days and weeks working flawlessly. To our surprise, our testers started reported several stuck passengers and security breakdowns. This was followed by some confusion regarding the designs of the build system, contractors and passenger getting stuck in various situations and the baggage system not working properly partly due to difficulties with understanding how it works as well as some lacking design logic on our part. This was not at all expected and initially escalated quite quickly. After a marathon coding session over the weekend, clocking close to a normal work week, we have managed to get it under some control and are now receiving fewer and fewer critical bug reports.

We had an internal launch date set for the 25th of August and are currently left with two extremely bad choices:

  1. Go ahead with the release in August despite our knowledge of the flaws and serious bugs still exists. This option could lead to bad reviews on Steam (the reality is that many will not care about the fact that it is EA, a broken game will still result in a negative review) and losing out on future sales.
  2. Postpone the release and work hard with the serious bugs to get to a stage where we feel confident in a public release. This option will disappoint a lot of our most dedicated fans and will also, of course, be a disappointment to ourselves.

It is very clear that we have to take an almost impossible decision here. Remember, we have worked tirelessly on this project for just now over two years, nobody wants it to be released as soon as possible more than us. It is the ultimate goal that we have targeted since the initiation of the project but looking at the current situation we also don’t want these two years to go in vain or be wasted just because we got stressed and started taking irrational decisions.

Us thinking and discussing the options over and over again has led to our decision to postpone that initial launch date and until we, together with our testing crew, feel that Airport CEO has a playable and stable build to launch from. This means that we most likely will not make the August window.

That sucks, yes. Yet again, the release date is pushed forward and that is on us as usual. If you’ve had enough of this then that is completely fine and we will readily deal with that criticism, you are welcome to voice it on any of the ACEO channels such as the forum or the subreddit. But before you do that, we’d just like to say that the reason behind this push is the greater picture. Any potential income that ACEO generates will be re-invested in the product, i.e. allowing us to develop the game full time, bringing an artist on board and perhaps expanding the team with additional resources, all with the purpose of making sure that ACEO lives up to its potential. That potential is extremely dependent on the initial launch of the game, if it’s well received then it is more likely that more people will purchase it which in turn will allow us to create a better product which in turn again comes back to you and your investment. For us, that is what an EA purchase is, it is an investment in the future product that ACEO will become. This means that the logical decision is to deliver a qualitative product which people can enjoy from the start, rather than having to put out fires and act customer support to an angry mob. We asked ourselves what we’d think in a buyers position, would we be the most upset if a game developer continued to delayed the game or if a game developers sold us a broken product? The first one sucks (and honestly after these two years we have an even stronger respect for game and software development) but the second one is even worse. We know that we have such a fantastic community that will support us in this decision and for that support we will repay you by delivering an as qualitative product as possible, if you’re still interested.

Airport CEO Alpha v0.18.1 Changelog

Aviation:
  • Solved a minor path finding issue
  • Solved a stand related deadlocking issue
Economy and Business:
  • Overhauled major parts of the job task system (and still got some work to do here)
Building and Construction:
  • Fixed a bug that caused baggage scanners to spawn as deactivated
  • Solved an issue with holding point rotation upon deserialization
  • Solved an issue where holding points were not save on stands if built on foundation (or if it was removed)
Simulation:
  • Solved an issue with passengers arriving too early at some instances
  • Overhauled parts of the path finding algorithm to improve functionality and performance
  • Solved an issue with the path finding algorithm that prohibited persons to plot in very tight areas
  • Solved several queueing deadlocking issues
  • Solved an issue which caused persons to randomly roam across the terminal area to random points upon activity failure
  • Security stations can now be set to only cater employees, passengers or both
  • Solved an issue with fuel trucks deadlocking on stands
  • Solved an issue where passengers did not process their baggage in an x-ray scanner
  • Small security check-points can now be attached with commercial stands
  • Solved several issues related to check-point deadlocking
  • Improved passenger behavior to allow them to attempt a certain activity twice before considering it failed
  • Solved an issue which caused arriving passengers not to leave properly
  • Solved a minor issue with the path plotting functionality which caused paths to be considered as failed
  • GA spawning on occupied runway is now limited to a queue of five
  • Solved several issues related to object pooling and object re-use
  • Solved several null reference situations
  • Increase breaking ratios on vehicles to mitigate deadlock issues
  • Solved several issues with construction material being out of sync in relation to ongoing construction
  • The number of vehicles in the world is now intelligently limited
  • Solved issues related to idle contractors
  • Solved an issue where employees would not arrive to a closed airport
  • Overhauled aspects of dynamic queueing to avoid deadlocking
  • Queues can now be rebuilt while passengers are interacting with them
  • Solved some minor issues related to passenger baggage retrieval
  • Solved an issue with unexpected passenger behavior on deserialization during boarding
  • Increased randomness of passenger boarding action execution
  • Prolonged the bus waiting time to allow for a better passenger terminal flow
  • Fixed an issue with the weather system
  • Solved a bug that caused the weather system to not scale to the world size
  • Fixed several issues related to baggage occupation on conveyor belt nodes
  • Increased the global night light value from 0 to 0.1
  • Solved a major bug causing serious passenger spikes on decision taking which gained us a lot of performance handling large passenger amounts
UI and Audio:
  • Improved the demolition function as it interfered with other gameplay aspects
  • Implemented validation of save file integrity prior to save to try and mitigate issues with corrupt save files
  • Removed the ability to search for buttons that represented upgrades which were not yet purchased
  • Changed button for room overlay toggeling
  • Implemented several dialog panels to communicate functionality with the player
  • Fixed several issues related to the tutorials and tutorial branches
  • Added several debugging tools and buttons
  • Quitting the game now reminds you to save it
  • Reduced the size of a few icons
  • Added a button for the flight planner
  • Enabled default object spawning on objects with variations
  • Solved a bug that caused the camera to move despite of the application being out of focus
  • Implemented an overlay tutorial window to highlight the most important controls
  • Implemented a new persistent notification panel to highlight really important messages
  • Improved several aspects of UI communication
  • Solved numerous UI bugs that caused unexpected or breaking behavior
  • Fixed a minor issue with object sub-panels not working properly
  • Solved a bug that caused contract images not to load properly

We’re going to stop there and get back to the code. Thank you for your time and patience, when we feel that we have a stable build we will share more details. Fly safe!

Related blog posts

Airport CEO: Helicopters

October 19, 2023

Innovative rotor configurations, sleek cockpit designs, and formidable thrust! These are just a handful of features that define the helicopters in Airport CEO, a new type of...

Read more

Airport CEO: Beasts of the East

October 19, 2023

Unique engine placements, see through nose cones and raw power! Those are just a few of the components that summarize the eastern aircraft, birds rarely seen flying in the west...

Read more

Newsletter

Enter your email to get the latest updates and news.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.