Back in 2011 we had an idea that became PR:ARMA 2. We gathered together a great team of developers and eventually released a few iterations of a mod that we do believe was really good fun.
However, from the technical point of view PR:ARMA 2 was a bit of a mess – two separate systems from the era of ARMA 1 held together by duct tape. The original idea was to have 100 players on a server. On the day of the launch we did get 100 players on the server, the only problem was that the server reported 0 “FPS”, or simulation cycles per second if you will. Now, the networking in ARMA is infamous on its own but the scripts did not do it any favour. The amount of traffic being broadcast from the various scripts was staggering, with the zone capture system being probably the worst offender – every time somebody was capturing a zone the server and all the connected clients would be updated with the capture status of all the zones in the mission. Approximately every 0.1 seconds.
Another major issue was that as the mission went on timed events became delayed. The more people on the server the worse the delay. I remember once playing a mission where a helicopter which was supposed to respawn every 20 minutes had respawned nearly an hour later.
With all of this in mind and with ARMA 3 on the horizon Deadfast and I made the decision to start PR:ARMA 3 from scratch. We weren’t naive to think that we could create something that would completely replace PR:ARMA 2 feature-wise. We wanted to start small and create a stable foundation to build upon: infantry-only missions, advance and secure game mode, squad system, kits, rally points. This would then be incrementally built upon.
And we got 99% of it done! The problem is that the past year we’ve been working on the last 1%. This is not an unknown phenomenon in software development. By the time you get this far it’s nothing but a grind.
The ARMA side of PR never really had that many people, and this goes double for people with SQF skills (ARMA’s proprietary scripting language). I think the most we ever had at once were 4. This is because there are very few people with the necessary skills to develop for ARMA.
As we slowly reached completion the work started becoming less and less fun. Instead of implementing features we had to focus on bugfixing in order to bring it to a releasable state. Unfortunately as anyone in software development can tell you bugfixing is very time-consuming, especially in such a complex environment as a multiplayer game.
Some time ago we announced that, due to “real life” challenges we planned to suspend development of PR:ARMA for the time being.
Since we don’t want all of our work to go to waste we have decided to publicly release the source code under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International Public License, thus allowing it to live on. You can find it here on Github.
If you have any questions regarding the code we will stick around the PR:ARMA 3 section to hopefully be able to answer them.
Read more here.