Victoria 3 – Dev Diary #183 - Great Wave Post-Release Thoughts

Happy Thursday, and welcome to a somewhat more somber development diary for Victoria 3 as I share my thoughts on the release of The Great Wave and Update 1.13, what went wrong, and what our next steps are.

As you’ve likely already understood from the previous paragraph, I do not intend to sugarcoat how I view this release: This expansion and update released in a state of bugginess and lack of polish that is frankly not okay, and we should be better than this. While I absolutely think the core features and content of 1.13/The Great Wave is a great addition to the game, the bugs, lack of QoL and especially the state of the AI was not. 

As you are reading this, following the release of hotfix 1.13.4, many of the most significant issues in terms of bugs should have been ironed out and several QoL features added, but we still have problems to address in the AI and overall balance of the update. Another larger hotfix is planned for next week, and we have some plans beyond that, but before I get into that I want to answer the question of why this update was released in the state it was.

There’s no one single answer, but if I have to pick a single reason I would say it was overconfidence, particularly on my own part. Many of the issues (particularly the bugs) stem from features coming in late, then being changed and tinkered with almost all the way up until release as we tried to get them into a good shape for release day. We especially underestimated the number of knock-on issues we’d have to deal with for troop transportation and supply (we did think these features would be complex, but still underestimated just how complex and error-prone they were).

So why then did I decide to go ahead with the release knowing that it would be hard to get everything into releasable shape? The answer is kinda simple: Because we’ve done it successfully before. Charters of Commerce had very similar issues with features (particularly the treaties) coming in late and being in a poor state of balance basically straight up until release, and that release went extremely well. Hence, the overconfidence. We did manage to get a huge amount of issues fixed right before release, but this time there simply wasn’t enough time to address all the most critical problems.

On top of all that, the barrage of last-minute fixes meant pretty much all our testing time was taken up verifying issues and not playtesting enough to discover all the new bugs created by our fixes. By the time we were putting together the known issues list it was the day before release and it was genuinely too late to do anything about it. I will also admit to a bit of wishful thinking at this point, hoping that all the last-minute heroics pulled would be enough to see us through, the way it had with Charters of Commerce and 1.9.

The rest you know. I don’t say all this to excuse myself, but I wanted to offer something beyond just apologies and platitudes and also to say that the mistake made here is something I am going to take to heart and try my very best to avoid doing again. I also don’t want to push responsibility for this release either downwards or upwards. The team did everything they could to get this release into a good shape, and that’s what we all wanted, but it’s not what we got. 

If I’d acted a few weeks ago, I could have done what I did for Sphere of Influence and delayed the release. Knowing what I know now, that’s certainly what I should have done. At the time though, the release wasn’t looking too bad, and in many ways we actually appeared to have fewer problems to deal with than Charters of Commerce: It just turns out that those problems we did have to deal with were complex, not easily solved properly, and prone to create all sorts of knock-on issues. All things I should have taken into account, but did not.

In the future, we will aim to ensure that we do not end up in the situation where QA simply doesn’t have enough time to playtest the game. Whether this means better scope control, allocating more development time or being more proactive with delays, or (more likely) a combination of these, we simply cannot have a repeat of what happened with 1.13.

For now though, let’s move on to our more immediate plans. Of course we are going to continue to release hotfixes to fix bugs, address balance concerns, improve the UI, bring the AI up to pre-1.13 standards and make it able to use the new features and manage its fleet correctly. Beyond that, however, one of the key issues I’ve identified in our development is that we simply do not have enough focus and knowledge in the team on how to make a competent AI that is able to function as both an interesting opponent and ally to the player. While the AI has made great strides prior to 1.13, particularly in economic development, there are longstanding issues in other areas, particularly its geopolitical decision-making, such as its occasional willingness to sacrifice the wealth of its entire empire to impose a law change on a minor Indonesian principality.

For this reason, I am changing our post-release plans to include a new free ‘AI Update’ sometime after we’re done with 1.13 which will be almost 100% focused on AI improvements and addressing longstanding issues with the AI. The secondary goal of this update will be internal - to spread knowledge within the team on how to better implement and maintain the Victoria 3 AI in the future, and to avoid the systemic AI problems plaguing 1.13 happening again in future updates. This update will not aim to add a bunch of new features, any changes outside of pure AI work will be AI-related, such as balancing or tweaking systems to work better for the AI.

I also want to emphasize that fixing the 1.13 AI issues is not the focus of this AI Update, as that will be done as part of the regular 1.13 hotfix process, while said update will naturally come sometime after we are done with the regular work for 1.13. We have not currently decided exactly when that is, but there will be at least one dev diary with proper details in the upcoming weeks, so stay tuned.

That’s all for today. There will be no dev diary next week, as our focus is currently entirely on 1.13. We’ll of course keep you updated on our post-release process and hotfixes as usual, but the next dev diary will likely be the mentioned AI Refresh overview in a few weeks. See you then!