• We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.

Stellaris Dev Diary #328 - New Year, New Beta

Hello everyone!

I hope that you all had a pleasant holiday season, and want to start off by thanking everyone that submitted feedback regarding the Technology Open Beta. That data we gained from these experiments was invaluable, so let’s get right into it.

Summary of Results​

As expected, the players that responded to the survey were overwhelmingly passionate players that have a ton of experience with the game. Nearly 70% of responses come from players that have over 1,000 hours played in the game. This is somewhat natural for an opt-in beta over the holidays with an intimidating feedback form, so I wanted to thank you all again for filling it out.

There was a strong consensus around the military changes (ship cost and upkeep), so we’ll likely be keeping those mostly as-is.

The technology changes were naturally more controversial. Roughly 80% of responses believed that technology (especially at higher tiers) was overall too slow in the beta, but a majority still thought that the changes were beneficial to the game overall. Several of you pointed out that so many simultaneous changes compounded too strongly, and we agree. I was happy to see that your feedback matches our expectations - we expected that the Open Beta was tuned too harshly and that we would want to pull back from it before release.

The Open Beta also revealed several technical issues, including some major performance implications from how Breakthrough Technologies interacted with diplomacy.

Next Steps​

Overall, I view the Technology Open Beta as a great success, and as such am taking the opportunity to update it and let it run for another few weeks, after which we will decide whether or not we want to continue experimentation, integrate it into 3.11 (or 3.12), or discard the initiative.

We concur that the original Open Beta went too hard on technology. We liked some of the things we were seeing, such as tier 3 and 4 technologies becoming more valuable for an extended part of the game, but felt that it delayed other critical parts too long. Breakthrough Technologies were interesting as a slowdown mechanic, but if kept would likely need some sort of temporary (non-technology or unity related) bonuses as some form of reward for the frontrunners. The excessively high costs for late tier techs pushed some critical technologies such as Ascension Theory or Mega-Engineering too late in the game, and certain undesirable behaviors (like ignoring research entirely) were too effective.

The updated Technology Open Beta should be up on stellaris_test now, with the following changes:

[Feature]
  • Difficulty Adjusted Technology Costs slider added to galaxy generation. This slider adjusts technology costs based on tier and game difficulty.

[Beta]
  • Removed Breakthrough Technologies.
  • Reverted base technology costs to their 3.10.4 values - the increased cost between tiers is now handled by the Difficulty Adjusted Technology Costs slider.
  • Removed the majority of Researcher Upkeep Modifiers introduced into the Open Beta.
  • Reverted changes to Knights research output from the Open Beta.

[Balance]
  • Tweaked the tiers of technologies that increase naval cap and fleet command limit.
  • Reduced the amount of Naval Cap granted by technologies.
  • Significant changes to Bio-Reactors:
    • Bio-Reactors are now a tier 1 rare technology instead of a tier 0 technology, and are available to all empires.
    • Bio-Reactors now reduce the food output of farmer jobs and give them a small amount of energy output.
    • Added a tier 2 Advanced Bio-Reactor technology and building.
    • Advanced Bio-Reactors further reduce the food output of farmers in exchange for a small amount of exotic gas output.
  • Decreased the amount of research produced by unemployed pops with Utopian Abundance.
  • Event options in the Knights' quest that improve their capital have been buffed to be better balanced compared to the options that improve knight jobs.

At player request, we have kept the older version of the Technology Open Beta available on stellaris_test_old. It will remain there until the release of 3.11 “Eridanus”.

Difficulty Adjusted Technology Costs​

One of the frequent points of feedback was that there was concern that newer players would be hit especially hard by the technology cost changes. We also recognize that different players have different desires for the pacing of the game, so we’ve added another slider to galaxy generation.

The Difficulty Adjusted Technology Costs slider adjusts the base cost of technologies based on the difficulty of the game. Higher tiers of technology are affected to a greater degree than lower ones, so this slider essentially affects “tier width”. While this does overlap with the Technology Costs slider to a degree, it does so in a different way, so we consider each to have valid reasons to exist as separate sliders.

New Difficulty Adjusted Technology Costs setting

Disabling Difficulty Adjusted Technology Costs will cause them to follow the 3.10.4 / Civilian difficulty curve. As with many other galaxy generation sliders, the Stellaris team will be balancing the game around the Normal setting.

Normal scaling tech cost graph

Tech Curves of basic technologies on “Normal” scaling, at different difficulty levels.

The base cost of technologies is now based on 3.10.4’s formula, y=1000*2^x, multiplied by the difficulty modifier of 1 + (q*x*d), where x=technology tier, q=difficulty adjusted tech cost galaxy setting (0 - 0.10, default 0.05), and d=difficulty (Civilian = 0, Grand Admiral = 6).

Normal scaling tech cost spreadsheet

TierXCost1 technologies at different difficulties, on “Normal” scaling (q=0.05).

For the players that enjoyed the larger amount of distance between tech tiers, scaling can go up to a maximum of “Extreme”, which gives Grand Admiral a curve that is similar to, but not exactly, the Open Beta numbers. Note that technology acquisition will still be faster than the old Open Beta as we’ve removed Breakthrough Technologies.

Extreme scaling tech cost graph

Tech Curves of basic technologies on “Extreme” scaling, at different difficulty levels.

Extreme scaling tech cost spreadsheet

TierXCost1 technologies at different difficulties, on “Extreme” scaling (q=0.10).

Previous open beta tech costs for reference

Previous Open Beta values for reference.

We have a new feedback form for this version of the Open Beta, available here. As with the previous version, you can respond multiple times if you have different thoughts after different playthroughs. Please let us know what you think, and whether you think we’ve gone back too far in the other direction.

Currently we're planning on collecting feedback from this phase of the Technology Open Beta for two weeks, until the 1st of February, but will leave the branches available until the 3.11 "Eridanus" update releases later on in the quarter.

See you all next week!

Please note that the Technology Open Beta is an optional beta patch. You have to manually opt in to access it.
Go to your Steam library, right click on Stellaris -> Properties -> betas tab -> select "stellaris_test - Technology Open Beta" branch.
Please disable mods for the Technology Open Beta, they are likely to break.
In-progress games should continue on the “stellaris_test_old” branch.


Leave your feedback!



Eladrin is talking about turning off your mods, and now the Community Team shows up, telling you to download more mods:

Want a sneak peek at the Legendary Leaders included in #MODJAM2024? Check out the feature video:


Voting will run until February 11th, so there's still plenty of time to play and vote for your favorite submission here!
 
  • 58Like
  • 12
  • 9
  • 7Love
  • 2
Reactions:
Depends on what they are trying to accomplish with it, and how well what it does is documented.

I honestly would like the slider to simply affect the leap between tech levels so that I can make later techs much costlier to research, thus keeping the mid-tier techs relevant a bit longer.

Base level techs should always be relatively easy, with the following levels become more costly each time you tier up.
Do you want a slider for traditions too? So that You can make later traditions much costlier to activate?

I think that can be a bald idea for 2025 next big patch
 
Last edited:
Depends on what they are trying to accomplish with it, and how well what it does is documented.

I honestly would like the slider to simply affect the leap between tech levels so that I can make later techs much costlier to research, thus keeping the mid-tier techs relevant a bit longer.

Base level techs should always be relatively easy, with the following levels become more costly each time you tier up.
As far as I can tell, this is precisely what the slider does. The only issue is it factors in difficulty for some reason; for instance on the lowest difficulty it does nothing at all.
 
  • 1
Reactions:

For those of you who like even more math and graphs, I threw all the tech curves (at 1x tech costs) into a google sheet. There appear to be 16 different curves, since some difficulty / difficulty scaling combinations overlap.

I'm struggling to articulate how much I dislike this. I know the devs talked about not overwhelming new players but as a 1000+ hour player the last thing I want is even more complicated and convoluted settings. It's already hard enough to experiment with settings only to find 10 hours into the game you got it wrong. If we had to have a slider for tech curve I really don't understand why it was linked to a different slider.
 
  • 11
Reactions:
Having played the beta, many of the removals are from the leader traits. The two Physics tech speed techs and Curator agreement still come with increased upkeep cost.

They did remove the upkeep from the Discovery finisher and Technological Ascendancy but I think that's okay. Last beta you had a Discovery tradition (Faith in Science) that gave you -20% upkeep only for the finisher to give you +10% upkeep.

I took a picture from a random researcher mid-game on the last beta. I'm not quite far enough into the current beta for a fair comparison (haven't seen the two +output/upkeep% capital building techs yet).
Very glad they removed the upkeep from those two...so much of this seemed like a just blind attempt to smack down researcher efficiency at every turn. I get it for researcher boosts from tech, but researcher boosts from unity sources was just "guess my empire's belief in science makes researchers drink and party all the time".
On specific sources where I think some upkeep adding makes sense, I think for the AI techs and Curator Insight, it should be adding energy upkeep only, while others like Real Time Peer Review should add CG upkeep only; this would lessen the blows to efficiency and distinguish different sources in their effects while being more in theme with the idea of what the sources do.
 
Largely, yes. We considered a different implementation that would work around the performance implications of the current variant, but we (and the community) didn't find the feel of the breakthrough techs to be good as-is. I can think of some variants that might prove satisfactory, but I feel that dedicating Custodian resources towards some other tasks is more valuable overall.
Oh that's a shame, I really liked the concept. I'd love to see it or something similar come back at some point. Getting the occasional free "catchup" tech was nice, even if the tech itself didn't do anything.
 
  • 2
Reactions:
As far as I can tell, this is precisely what the slider does. The only issue is it factors in difficulty for some reason; for instance on the lowest difficulty it does nothing at all.
Which is what I don't want. Let the tech cost slider be independent of game difficulty and I will be satisfied with it.

Basically if setting the tech slider to it's 2.0 setting doubled the costs of each tech tier level 2-5 and not being affected by the game difficulty selected, then I think it would accomplish what I'd like.
 
Devs, when will you guys fix portrait SHAKING bug? It has been in the game for like 2 years. Do you guys even play the game?
Has someone filed a bug report with enough detail to reproduce it on a different PC?
 
  • 1Like
Reactions:
Very happy to see Breakthrough technologies removed is the only feedback I will add.

Looking forward to see what the team has in store for us in 2024. Still praying for Dinosaur species DLC though ;)
Might be better as a custodian expanded orgin/trait/leader update for avians and reptilians to give them more love. not sure there's enough content for a full DLC.
something to consider maybe PDX ;)
 
We really could do with a general mechanic where having intel on other empires with a tech makes that tech easier to research, and the effect is intensified if they are multiple tech tiers ahead (simulating how in real life as we get more tech, the older stuff becomes just common knowledge or household instead of industrial-only etc.).
This could be done with an intelligence-like operation. Let each embassy increase research speed by a small amount (2%) if you are researching something the other empire already has, similar to how a Protectorate works. This would simulate the ambassador or spymaster going out and buying textbooks or copying the local scientific literature. I think I prefer it being the spymaster so that the game doesn't need to total up a large number of 2%s.
 
Gonna get downvoted to hell for this but:

Adding more sliders to game creation is how we ended up in state like this, when everyone bends the rules of the game however they like -> everyone disagrees on what the main issues of the game are. BECAUSE EVERYONE IS PLAYING A DIFFERENT GAME.

Developers just dump all responsibility onto players & modders under disguise of being "inclusive" & "adaptable", instead of taking care of balancing properly.
 
  • 8
  • 2Like
  • 1
Reactions:
Adding more sliders to game creation is how we ended up in state like this, when everyone bends the rules of the game however they like -> everyone disagrees on what the main issues of the game are. BECAUSE EVERYONE IS PLAYING A DIFFERENT GAME.
Are they?

Without any data whatsoever to go on, solely based on the rule of thumb that players like rivers tend to follow the path of least resistance, my guess is that most players use default settings apart from choosing difficulty and galaxy size/shape, and that players fiddling more with the settings are overrepresented in the forum.
 
  • 5
  • 4Like
Reactions:
Gonna get downvoted to hell for this but:

Adding more sliders to game creation is how we ended up in state like this, when everyone bends the rules of the game however they like -> everyone disagrees on what the main issues of the game are. BECAUSE EVERYONE IS PLAYING A DIFFERENT GAME.

Developers just dump all responsibility onto players & modders under disguise of being "inclusive" & "adaptable", instead of taking care of balancing properly.

It's not possible to "balance" personal preferences on what is 'fun' or not. If two people play the same game and one thinks the tech is progressing too slowly, while another thinks it is too quick, trying to impose what you think is the right answer on them is just going to end up with the one that you don't agree with quitting the game and not buying future DLCs. PDX is in the business of making money by selling DLCs. Driving away half, or a third, or even a quarter of your playerbase by trying to dictate their playstyle for them is not a wise business move.

There is a reason practically every complex strategy/4x game has a healthy amount of pre-start customization options. Go take a look at Civilization 6 (and 5, and 4 etc) and get back to us with how many options are available for the player to tailor the game rules to their tastes before it even starts. It's a staple of the genre.
 
  • 10
  • 1Like
Reactions:
I like the idea for the bioreactor. Much more plausi and it ties in nicely with the other district boost buildings
 
Okay, some final thoughts before I go fill out the survey.

I played Payback on Commodore with mid-game scaling, Normal DAT, Medium Galaxy, 18 empires (1 FE and 1 maurader via the randomizer), 2275/2350/2450. I ended the game with 4442P/4150S/4090E monthly research, ~177% research speed, and 1847 Empire Size (which imposes 297% penalty).

I was initially skeptical at the new curve values. Fortunately, this was unwarranted as T3-T5 slow down enough that even with a bunch of tech (for me, I know some GA players have tens of thousands) and tons of research speed, late-game techs were still taking 3+ years to research. I did not go for the full tech tree like in the last beta, but only made it a few levels into repeatables, which is roughly what I feel should be achievable in a full run. So I think they largely nailed the default curve values, at least for Commodore.

That being said, I can see players of all difficulties wanting both more aggressive and more lenient curves, so I'm not entirely sure tying the main curve factor to difficulty is a good idea. I understand why they did it - they're trying to make the default experience for those not fiddling with advanced settings feel good - but this seems to be the most controversial part of the change. The DAT slider is also clearly confusing as I've been answering questions on the forums and reddit for over a dozen different people about how it translates into their preferred game settings.

Now, there is one caveat. Tier 2 tech feels just like live. Without breakthrough techs, there's no distinguishing barrier between T1 and T2. The cost goes up slightly, but not notably as much as T2 to T3 or future tiers. This meant the first 50 years of my game were similar to the non-beta. Destroyers weren't the top ship for very long. Level 2 ship components quickly got replaced with level 3 ship components. So if there's one big criticism about the current curve, it's that it really only kicks in after the first 50 years.

Additionally, special projects could use tuning. I know other people have called out the space fauna, but I'll bring attention to two projects in particular:

With the Payback origin, you start without three T0 techs: Research Labs, Planetary Administration, and Corvettes. You do have a research lab on your starting planet (along with a couple buildings you can't build yet), but you have 10 fewer pops, 33% planetary devastation (which reduces job output), a temporary planet buff to counter the devastation (though it doesn't last as long), no construction/science spaceships, no mining/research stations, and most importantly, a debuff from the debris field that gives you +25% district/building cost and upkeep. It takes 600 Engineering research to remove this debuff. Since you start with a pair of actual researchers, you really feel that researcher nerf in the early game. I waited until it would take less than a year to finish the special project, which meant my capital was expensive to build on for some fifty years. I think the cost should be reduced to account for the researcher output nerf.

I also went Synthetic Ascension. This took a really long time. Not only do you need two T4 techs (which the agendas can help a little with if you don't mind canceling your other agendas when it comes off cooldown), but it also took 100K Engineering research (108 months) to ascend my 450ish pops with the whole tradition tree researched. When they finished ascending, they had no robomodding traits allocated, and it was another 60 months to assign out all those traits. I didn't finish it until 2365. Now some of that could be a skill issue (I last went synth in 3.1), but that's a lot of time to dedicate to your ascension path, when something like psionic goes way faster, and biological uses up the less-valuable Society research rather than the highly in-demand Engineering research.

Oh, and once you synth ascend, over half of the Society tree loses all relevance for you. I had many draws that were just some combination of +Habitability, Terraforming, +Food %, +Leader Lifespan (the repeatables that only affect years, at least the two policy ones are useful), Gene Clinics, or some Archaeo-Tech component that might possibly have been useful 100 years ago. I would have liked the military techs, unity techs, starbase capacity, Ascension Theory, or even the Strikecraft repeatables over any of those picks, but it was not fated to be.

Some more assorted thoughts:
  • I spent almost the entire game with 2 or more "must-have" techs available on each of my Engineering card draws. I think that's great. I was constantly having to choose and adjust based on what would benefit me the most, and it was nice being actually engaged with the tech tree until 2400 or so.
  • In contrast, Physics and Society really show their weakness with an increased tech cost curve. I was drawing filler techs like crazy starting in the early 2300s. Having to choose between Citadels, Battleships, Mega Engineering, and Tier 2 Forges is way more interesting than choosing between Reactor Booster level 2, Cloaking Fields, Encryption, and Curator Survey Speed.
  • The L-Gate Insight techs aren't absurdly expensive, but I still didn't research a single one. Way easier to buy from Curators or use Minor Artifacts on them and save my precious tech points on all the Engineering techs I needed yesterday. Maybe these should move to the Physics tree? They might be a competitive mid-game pick there.
  • Despite running half the unity income as I did on the OG beta, I still ended up blocked waiting on Ascension Theory. I think I finally got it around 2420, which was after the crisis and the AE were wrapping up. This means I couldn't benefit from half the planetary ascension levels until after it was no longer relevant.
  • I had Cybrex so I can't comment on Mega-Engineering draw timing as it was available in the early 2300s, though I did not spend the 384 months to research it at the time. By the time I did get around to it, the crisis (Unbidden) was wrapping up, so they were mainly only useful against the AE that was going on a rampage.
  • Even with the Cybrex shortcut and Living Metal edict available, Galactic Wonders comes too late to actually help.
  • The AI was much more competitive than in the last beta. Some of that could be because I played Commodore instead of Captain and those AI bonuses kept up, but I had a solid rival or two for quite a lot longer - into the late 2300s. And while the AI couldn't counter the AE's 2 mil fleet power on their own, I was happy to see the biggest players attempt to stand up to the empire, and whittle a couple of those 450K fleets down to 150Kish.
  • The diplomatic voting AI for using the Declare Crisis resolution against an AE is dumb. Every empire gets -80% acceptance because "they don't want to anger a Fallen Empire" even if said empire is currently waging a total war against them and their federation and cracking their planets.
 
  • 4
  • 2Like
  • 2
Reactions:
@Ikael

Here's a full comparison between last beta and current beta researcher upkeep:

Last Beta:
1706087770737.jpeg


This Beta:
20240123214402_1.jpg


Both are on regular worlds with the level 2 capital efficiency tech. 19% upkeep is missing from leaders and leader traits, 10% from the Discovery finisher, 10% from Technological Ascendancy. I don't remember what Real Time Peer Review and New Numbers were, I think they were event bonuses. Match is from Enmity (everyone wanted to rival me, so might as well get buffs out of it).

The Curator scientist has +10% research speed/+10% upkeep if you put him on your council, but I think that one is ok. It's completely controllable and a pretty powerful buff.
 
Last edited:
  • 1Love
Reactions:
Are they?

Without any data whatsoever to go on, solely based on the rule of thumb that players like rivers tend to follow the path of least resistance, my guess is that most players use default settings apart from choosing difficulty and galaxy size/shape, and that players fiddling more with the settings are overrepresented in the forum.
And you know what happens if default setting are bloody terrible?

"Well you should have tuned them yourself, you have a slider!" - perfect excuse for any fault of the game.

For amount of setting you need to tune to make the game playable, they should be paying game designers wage to every player.
 
  • 1
  • 1
Reactions:
Adding more sliders to game creation is how we ended up in state like this, when everyone bends the rules of the game however they like -> everyone disagrees on what the main issues of the game are. BECAUSE EVERYONE IS PLAYING A DIFFERENT GAME.

For amount of setting you need to tune to make the game playable, they should be paying game designers wage to every player.
The game is playable on defaults. Some people will enjoy it enough on defaults to never bother touching the settings. Some people will explore the settings out of curiosity. Some people will un-enjoy it enough on defaults that they will feel the need to change settings on their second playthrough.

Paradox have better data on the proportions than you do.
 
  • 2Like
  • 2
Reactions: