It’s another happy Wednesday! I’m Ryagi, Community Manager for your two favorite games (EU4 and EU5 of course) and today we’ll be talking about a collection of things on everyone's minds, including performance in 1.4, complacency, and more. But before that we have something new to show you, the Campaign Timeline Feature!
EU5 is a game about guiding your nation through the early modern period, and while many of the encounters you face are internal, economy focused or otherwise intangible, borders remain a very cool and popular way of sharing a snapshot of your game with others.
That’s why we’ve introduced the timeline feature! So that players can better share the stories they create in EU5. In EU4 the timeline is a feature that comes with the Mare Nostrum DLC, for EU5 we want to be clear the Timeline feature will be available free with update 1.4. (Not a part of Across the Pillars)
You can find the timeline feature either in within the content tab, or conveniently in the top right UI panel below gamespeed
The timeline will allow you to go back to any point during your campaign and view the borders as well as some basic stats. Let's go through and give you an idea what tools the timeline gives you.
Handy toggles let you jump the camera frame to set regions. (This can be extra useful for those of you that enjoy making timelapses)
You can probably guess who the player is here
The centerpiece of the timeline feature is being able to view any point within a save, day by day.
You can nudge the timeline forward/back in increments of 1 day, 1 month, 1 year, 10 years, 50 years, or to the start/end.
You’ll also be able to just click play and let it go at whichever speed you’d like too of course. Whether that’s 100x speed, or a leisurely 10x
In addition to watching your map get painted by a plethora of inevitable AI shenanigans, combined with your own exploits of course, the timeline feature also lets you view a breakdown for some of the primary stats throughout a campaign. At launch this will Include a nation's Population, Economic base, and Tax Base.
Stats can be sorted as pie charts, or direct ranks (Ranks is shown below)
Probably no surprise Ming is leading in pops and economic base...
As you can see the timeline also denotes particular situations that especially can have global impact on taxes/population within the graph itself! Right now it is capable of showing Black death, Discovery of America, Columbian Exchange, and Little Ice age.
We have also explored the capabilities of showing other situations and even in game happenings (Such as wars or major events). For now it’s limited only to the global 4 we mentioned above, but we’re definitely keeping the doors open for the future if it’s a feature people enjoy.
Here is our YouTube short on the timeline feature if you want to see more of it in action:
Now that we’ve shown you the pretty stuff, let’s get into something more numerical.
Performance has been a priority for us since before launch. We won't lie to you, fighting the performance demons is an uphill battle, but the sweetest victories require the hardest battles! For every bug fixed, feature refined, AI adjusted, etc, performance is impacted in unpredictable ways. This means that good performance is not just a matter of improvements, but also maintenance and consistency. So where do we stand on game performance in 1.4? The good news is, overall our testing indicates we have made large performance improvements in 1.4 compared to 1.3, especially mid and late game performance. As a Community Manager, the inner workings of optimizations are one of my least understood departments… so I'll hand the mic over to Johan for this short explanation section. I am happy to personally attest he has been plugging a way at some of these performance improvements himself as of late!
There have been quite a lot of reasons why the game’s performance has gotten worse over the course of a campaign, but mostly it is about badly written algorithms that don’t always scale well. And it is also about how smartly written they are, and while some are useful for the UI, that may not be ideal for the performance so you need to hand-optimize them as well.[/p]
There is also the fun part of making sure we parallelize algorithms as efficiently as possible, where you try to spread out lets say the AI evaluation of building buildings on the core’s as efficiently as possible. How do you do that? If you got 800 countries running, and the PC has lets say 8 cores. Sure, 100 on each? Yeah.. but lets say.. France has 200 locations and can probably start 5 builds each month, while Ulm has 1 location and starts one every 4 years. Then you need to make efficient heuristics that distribute them relatively evenly so that the building AI cores are done in total as quickly as possible, so not seven cores wait on one for a long time..
On average we aim for games to tick 40-50% faster than 1.3.
Here is a graph measuring tick tasks between 1.3.11 (blue), and 1.4 (orange) A good chunk of the optimization has been in how the UI handles things on the backend, which also impacts gamespeed. With this performance, we have managed to for example have a QA play through a full campaign in under 20 hours! As always, mileage varies based on system specs, and software set up. On high end machines we have found exceptional improvements. Here is another particular anecdote Johan was excited to share: With this performance, we have managed to for example have a QA play through a full campaign in under 20 hours! As always, mileage varies based on system specs, and software set up. On high end machines we have found exceptional improvements. Here is another particular anecdote Johan was excited to share:
On my ryzen 7 7800 with 32gb ram and a geforce rtx 0470.. in observer mode it took me 2minutes 45 seconds to go until jan 1339 on 1.3, currently it takes 47 seconds in the latest internal 1.4
For the examples like this exceptionally impressive stat, along with some of the in office anecdotes, the systems are rather high end. So while everyone here is very excited about performance in 1.4, we want you to take this excitement with a grain of salt. Rest assured though, we are always aiming to optimize systems at min spec and above.
We also worked a lot on FPS (This does impact tick speed IF the “Maximize tick speed” setting is toggled off.). But unfortunately, frame rate is something that cannot be meaningfully captured on our nightly observer tests. This is due to the fact that automated nightly tests don’t simulate a real play environment by efficiently capturing how a player actually plays the game (Opening UI panels, building buildings, etc). So sadly we don’t have any exciting stats to show off on the FPS front, (At least not any we can share in good faith) but our limited testing indicates that we’ve seen some good overall improvements there as well.
While the biggest chunk of performance improvements come from the myriad of optimizations we touched on in the section, there is one other big thing we’ve done that has had a pretty nice impact on late game performance especially… the removal of hourly ticks.
As of 1.4, we have removed hourly tick rates. Ticks are now on a day to day basis, as they were in EU4.
The idea behind hourly ticks was to give players more immersion, especially when it comes to warfare. But they also have the side effect of making the game feel much longer.
A consistent thread of complaints we have seen have to do with pacing. And while removing hourly ticks is not the only thing that can be done to improve pacing, they are one of the easiest. And as a bonus, removing the hourly tick has added about 10-15% faster performance as well.
With the removal of hourly ticks also comes the removal of Speed 6 & 7 too. Meaning speed 5 is now once again the Maximum. Pacing in games is important for the feeling of how fun it is. Even if it’s not something you can always actively identify, we hope that the faster pacing 1.4 offers gives a better experience overall. We’d encourage you to give this change a try and feedback during the open beta! It may take some getting used to (And it certainly has a little bit on my end) Hourly ticks were not the only thing we opted to remove in 1.4. However...
When we first introduced complacency, it was met with a fair amount of critique from you, the players. We updated it to be less omnipotent, however it was clear the notion of being penalized for success was a root of the bad feeling around it. It is also a mechanic that you could essentially ignore, while also creating odd incentives (For example, intentionally creating coalitions for the rare instances where it does become relevant.) So after giving it’s updated variation a fair shot in players hands in a full a full patch cycle throughout 1.3, we've opted to remove the complacency system completely in 1.4
Giving nations speed bumps for growing too large too fast is still something we're looking into potential in the future! But it's clear to us those speed bumps should aim to be a little more engaging and well, fun.
That’s it for this week! Enjoy a 2nd happy Wednesday, and we’ll see you next time when we present some of the economy changes in 1.4, along with the new tariff system. We’re closing in on the 1.4 Open Beta, where you will be able to test all these changes in the flesh and give us your feedback for how they feel. Stay tuned here for more information on that when we have it. Until next time!