Losing confidence in the dev team

[Goosecreature]

There seems to be some kind of misconception going around that we're on vacation until June 1st. BjornB mentioned that there would be no Dev Diaries or streams last week, and no Dev Diary this particular Thursday (May 25th) since that is a public holiday. The Thursday after that is June 1st, so that's when you can expect the next Dev Diary.

Somehow, because this is the internet, that apparently got turned into all of Paradox being on vacation as of last week all the way until June 1st. I'm in the office right now, and I can assure you that we're not on vacation...

 

[AgentOrange]

There seems to be some kind of misconception going around that we're on vacation until June 1st. BjornB mentioned that there would be no Dev Diaries or streams last week, and no Dev Diary this particular Thursday (May 25th) since that is a public holiday. The Thursday after that is June 1st, so that's when you can expect the next Dev Diary.

Somehow, because this is the internet, that apparently got turned into all of Paradox being on vacation as of last week all the way until June 1st. I'm in the office right now, and I can assure you that we're not on vacation.

Hello! I just wanted to thank you for your continued contact with the community. I'm sure the majority really appreciate you staying in touch with us.

A game with the complexity like Stellaris is bound to have bugs and even really serious bugs with patches every now and then (Such as Prethoryn not being able to colonize planets in 1.5). But I assure you that the vast majority of people playing the game have faith in Paradox to be able to fix them. Just because this particular thread (As well as the bug section) is dedicated to calling out the negative sides, doesn't mean all of us currently have a mostly negative opinion about Stellaris or Paradox and its dev team.

Stellaris was back in the top 10 sold games on Steam last week, a ton of mods are still being created and people are expressing how much they like the game. This is also why this thread is two-faced. People complain, but they do so because they love the game. If they didn't care or truly had no confidence in the team, would they really be bothering talking about it on their own forums?

Some patches go great, others do not, such is the nature of programming and complex games in particular. Stellaris is still a very fun game with a galaxy worth of potential on top of it.

I am very much looking forward to what Paradox can bring for Stellaris in the future!

 

[MirEgal]

I'm a little bit late to the thread but still want to post my 2 cents.

I know both perspectives. As someone who has been working in software development I know of large ctulhulike codebases (and even with best intentions, in a small and dynamic fast-developing team the codebase at some point approaches this state) in which it is extremly easy to introduce a bug and extremly difficult to find it. Especially for some highly interwoven systems like in stellaris.

From my own experience, if you are working a week to fix the one bug deleting the database (e.g. deleting the savegame in Stellaris) in 0.1% of all save processes, you tend to weight the bug "I need to refresh the window so I see the correct values" on your internal priority list quite low. It took myself some serious mental acrobatics to place myself in the customer perspective again and to go the extra mile of improving user experience - because that's what the customer sees and criticizes.

Of course, by using correct refactoring of the code base, a replanning of the architecture with every introduced feature, automated calculations of all interwoven systems to test strange outcomes, it would be possible to mitigate those problems. E.g. automatically run 1000 games without user interaction every night and check for certain properties, it would also be possible to discover those problems. However, for a company the size of Paradox Interactive and with current selling statistics as well as current prices, the inclusion of such systems is not realistic. Either they need to sell more or we need to pay more. In the worst case, this results in a downward spiral. Luckily for us, they had an upward spiral over the last decade with all corresponding problems, best seen in this thread (larger, but less error-tolerant fanbase).

In every smaller software company (a few developers, a few full-time-qa shared between different teams, maybe automatic code testing but no automatic qa), you will have the same processes and problems. Especially bugs introduced by late-minute-changes before releases are prone for this. Even in large-scale companies, see http://thedailywtf.com/series/errord for examples.

Coming from this background, my own tolerance for errors is quite high, especially if I can see that the team has the motivation to fix the problems.

...

That said, it feels like the team has to recheck their properties. My pet peeve for this is the precursor event chain. It is something everyone will have in every playthrough. I've never been able to finish it and I assume I'm not the only one. It is not gamebreaking but it's tiring and also quite bad for my mild OCD to never empty my list of open quests.

In 1.6, I wanted to get boxed in on purpose with my tall, peaceful empire. The goal was to outtech everyone and once the crisis arrives, be the cavalry coming in. For playing tall, voidborne is too late for habitats in my opinion as I sat on my thumbs for some time. I'm ok with this - not very playstyle is supported in every way, but I think that earlier habitats would improve the game. As expansionist, they are still worse than just settling the planet. (Also, please give habitats spaceports).

At least the war-declaration bug explains why the unbidden did nothing for quite some time, I assume.

This, together with a large list of small things reiterated in this and other threads (why can't I have a macro builder for fleets? Why can't I have a macro builder for spaceports in sectors?) brought me back to EU4 until hopefully playing makes more fun. Don't be wrong, I had and still have a lot of fun, but EU4 just makes more fun at the moment.

 

[Nomed[GD]]

am I only one enjoying 1.6.1 ?
Currently testing (again) Hivemind and it's really ok.

 

[Drowe]

@Rundstedt75
It's a misconception that longer work hours mean a better work ethic or that having fewer days off increases revenue of the company. Henry Ford figured that out in the 1920s, as one of the first companies to adopt a 40 hours week. The increase in productivity makes up for the decrease in work hours.
http://www.history.com/this-day-in-history/ford-factory-workers-get-40-hour-week

The value created per hour worked is higher in countries with fewer work hours.
http://payrollservices.expertmarket.co.uk/worlds-most-productive-countries

 

[krios41]

Hey, is Endless Space 2 any good?
Dont know anything about it except some people here seems to like it, so maybe worth a try?

In my personal opinion it's worth picking up

 

[Red192]

Maybe you should play it less and observe how AI plays more? Food bug, pacifism bug, building bug are obvious and easy to notice if you use observe mode.

I must say he is right

 

[TheAtreides84]

Hey, is Endless Space 2 any good?
Dont know anything about it except some people here seems to like it, so maybe worth a try?

Very good, if you can put up with some bugs and not stellar (ehehe, I made a pun) performances. I think that between it and Stellaris, all other 4X games (apart from Distant Worlds universe for a certain kind of players) will have to bite the dust for a long time.

 

[permanently_afk]

One of the threads which use overly harsh language to bemoan things they neither understand nor are able to change. And then they feel themselves on the right side. *sigh* Nevertheless, it turned out well.

@Rundstedt75
It's a misconception that longer work hours mean a better work ethic or that having fewer days off increases revenue of the company. Henry Ford figured that out in the 1920s, as one of the first companies to adopt a 40 hours week. The increase in productivity makes up for the decrease in work hours.
http://www.history.com/this-day-in-history/ford-factory-workers-get-40-hour-week

The value created per hour worked is higher in countries with fewer work hours.
http://payrollservices.expertmarket.co.uk/worlds-most-productive-countries

This goes double for jobs were creativity plays a role. And despite all efforts, programming still is one of those.
For my Bachelors' thesis, I compared several studies on avoiding projects and programs being late and/or error-ridden. Turns out, not overworking your project team is one of the key factors. From that data and asking around, I have the impression that the average usable "programming time" averages at 4-6 hours, with the rest of the workday spent on preparation and coordination. This view is supported by the agile systems (SCRUM et. al.) who posit that sustained lower pace is better than overwork. This comes from the fact that the latter requires more total work (from fixing bugs, throwing out system(parts) and so on). And it is further sustained by the fact that after work hour 6 - 8 the amount of errors made will increase exponentially (which will then require fixing). And thats not getting into the side of finding a proper division of problems.
This does not mean that teams can't work 10-14 hours a day. It just means that they must not do so all the time. Linear correlation of work time to productivity is only true for machines or shift-work manufactory. For everything else, there is an upper bound. And thats the reason why computer scientists, architects and skilled workers are rare. Ask a watchmaker (you know, the kind which makes the expensive, swiss kind) how many hours he works on an average day.

 

[GAGA Extrem]

Very good, if you can put up with some bugs and not stellar (ehehe, I made a pun) performances. I think that between it and Stellaris, all other 4X games (apart from Distant Worlds universe for a certain kind of players) will have to bite the dust for a long time.

Go try Star Ruler 2, then edit your post accordingly.

 

[sr999]

- Game runs in North Korea mode, where starving pops do not have an effect on the empire in any way

I don't know, even Kim whichever-this-one-is is feeling the pinch. My evidence is a Private Eye photograph of Kim talking with his senior generals, titled "Tensions Ease Over North Korea".
General: What happened to the dogs of war?
Kim: I've eaten them.

Oops, OT. Please accept my craven apologies.

 

[Jamor]

Hey all,

Given the option of me and my devs either a. hanging out on the forums filled with questions we can't answer until official announcements, or b. working actively on the game, I think in the long run most people would prefer "b."

So, please don't interpret lack of voluminous and immediate replies as indifference. We do follow the threads but don't always reply. Our priority is on the game.

I just want to reaffirm that 1.7.0/Bradbury, the new multiplayer beta, is done by a separate team and doesn't take any resources away from my game team for bug fixing or new development. It will go live when we've had a full beta and testing cycle and are confident that it works well.

Because I don't want to have two betas out at the same time, confusing people, gameplay fixes will happen separately from that. So, the plan right now (subject to change, of course) is to release an optional beta with gameplay fixes separate from the full Bradbury MP release. I'm not going to rush to put something half-baked in your hands too soon. When? "When it's ready."

Your concerns are heard and are being acted upon. If you are upset, please direct it at me. My developers are incredibly hard working and passionate people doing a complex job with a lot of moving parts. Any failure of planning is mine.

 

[TheAtreides84]

Go try Star Ruler 2, then edit your post accordingly.

Tried it. I'll put it in the same tier of Galactic Civilizations III. Good, but not *so* good.

 

[Razzlie]

Hey all,

Given the option of me and my devs either a. hanging out on the forums filled with questions we can't answer until official announcements, or b. working actively on the game, I think in the long run most people would prefer "b."

So, please don't interpret lack of voluminous and immediate replies as indifference. We do follow the threads but don't always reply. Our priority is on the game.

I just want to reaffirm that 1.7.0/Bradbury, the new multiplayer beta, is done by a separate team and doesn't take any resources away from my game team for bug fixing or new development. It will go live when we've had a full beta and testing cycle and are confident that it works well.

Because I don't want to have two betas out at the same time, confusing people, gameplay fixes will happen separately from that. So, the plan right now (subject to change, of course) is to release an optional beta with gameplay fixes separate from the full Bradbury MP release. I'm not going to rush to put something half-baked in your hands too soon. When? "When it's ready."

Your concerns are heard and are being acted upon. If you are upset, please direct it at me. My developers are incredibly hard working and passionate people doing a complex job with a lot of moving parts. Any failure of planning is mine.

Is there any chance of a second pass at a QoL/Bugfix focused patch on the radar?

It feels like we still need polish on seemingly small things like trading star charts essentially punishing players when it should be explicitly beneficial, and these don't feel like they get much focus due to so much of it being directed towards new features and content, leaving existing features to bite the dust.

 

[sir Eyeball]

Hey all,

Given the option of me and my devs either a. hanging out on the forums filled with questions we can't answer until official announcements, or b. working actively on the game, I think in the long run most people would prefer "b."

So, please don't interpret lack of voluminous and immediate replies as indifference. We do follow the threads but don't always reply. Our priority is on the game.

I just want to reaffirm that 1.7.0/Bradbury, the new multiplayer beta, is done by a separate team and doesn't take any resources away from my game team for bug fixing or new development. It will go live when we've had a full beta and testing cycle and are confident that it works well.

Because I don't want to have two betas out at the same time, confusing people, gameplay fixes will happen separately from that. So, the plan right now (subject to change, of course) is to release an optional beta with gameplay fixes separate from the full Bradbury MP release. I'm not going to rush to put something half-baked in your hands too soon. When? "When it's ready."

Your concerns are heard and are being acted upon. If you are upset, please direct it at me. My developers are incredibly hard working and passionate people doing a complex job with a lot of moving parts. Any failure of planning is mine.

While I do agree that b is the most preferable option I also think that in the case of bug reports a simple acknowledgement of it being received and worked on would be nice and prudent of you. I have a bug report with no acknowledgement and I do consider making a new because it is so long down the list now that it might have slipped through. If you have read it and reacted to it a new one is just more clutter and if it slipped a new one is prudent.

 

[Aotrs Commander]

This goes double for jobs were creativity plays a role. And despite all efforts, programming still is one of those.
For my Bachelors' thesis, I compared several studies on avoiding projects and programs being late and/or error-ridden. Turns out, not overworking your project team is one of the key factors. From that data and asking around, I have the impression that the average usable "programming time" averages at 4-6 hours, with the rest of the workday spent on preparation and coordination.

That's interesting, because (as a self-employed CAD modeller), 4-6 hours per day of actual CAD design is all I find I can manage, before I reach the point where I start to get frustrated and make loads of mistakes. (And I'm told (by the engineers of my acquiantance) that that's is about all actual CAD time you would spend in an office per day, with the rest, as you say, on the other general work-gubbins/meetings etc.)

 

[Cat_Fuzz]

To be fair to Paradox, they are doing a better job of fixing this than Stardock did with GC III. Paradox have actually tried (and admittedly failed) to release a bug fix patch a year after release and after it made things worse, are now in the process of trying fix things.

GC III had a lot of problems out the door, and after a year of release, with many complaints of needing too many constructors for star bases, a borked trading AI and numerous balance issues, we got none of that fixed and instead had three paid DLC's that added mostly fluff to the game.

I understand Stardock have tried to (finally) address some these issues with their latest expansion, but unlike Paradox, there's a price tag attached to their bug fixes. At least Paradox knows to keep fixes free, rather than try to resell as an expansion

 

[Jularbo Kizn]

To be fair to Paradox, they are doing a better job of fixing this than Stardock did with GC III. Paradox have actually tried (and admittedly failed) to release a bug fix patch a year after release and after it made things worse, are now in the process of trying fix things.

GC III had a lot of problems out the door, and after a year of release, with many complaints of needing too many constructors for star bases, a borked trading AI and numerous balance issues, we got none of that fixed and instead had three paid DLC's that added mostly fluff to the game.

I understand Stardock have tried to (finally) address some these issues with their latest expansion, but unlike Paradox, there's a price tag attached to their bug fixes. At least Paradox knows to keep fixes free, rather than try to resell as an expansion

Also to give some head here I don't know of any other big studio that makes strategy games that helps out it online community as much and continue working on the game so long after release, the only thing that comes close is Civ4 back in the day. I think the problem is that EU series and Hoi4 is the favourites who get so much love that Stellaris is the black sheep of the family that is desperate for attention but seems to receive next to nothing comparatively (while Victoria is the ugly unmentionable secret locked away in the loft or basement).

 

[Milten]

(while Victoria is the ugly unmentionable secret locked away in the loft or basement)

There there, it's a loved child, just in another way.

 

[permanently_afk]

That's interesting, because (as a self-employed CAD modeller), 4-6 hours per day of actual CAD design is all I find I can manage, before I reach the point where I start to get frustrated and make loads of mistakes. (And I'm told (by the engineers of my acquiantance) that that's is about all actual CAD time you would spend in an office per day, with the rest, as you say, on the other general work-gubbins/meetings etc.)

Thats interesting. If I surmise both high complexity programming (i.e. modeling, architecture, algorithms, multithread, etc.) and CADing work under "engineering" this would lead to the statement that an engineer has about 4-6 hours of work time, with the rest spent on auxillary work (e.g. coordination, checking back, simple tests, light documentation). And that it is not possible to increase this amount permanently or extend working time. Which means the work hours you get out of an average year are fixed. Together with the mythical man-month, thats a solid basis to argue on.

This would also nicely dovetail with the trouble Paradox is in: Their work amount increased (HOI4, Stellaris, new projects and their own compexity), while their manpower remained constant. If I speculate, I see a solution in two parts to this, which both require effort (i.e. takes away quite a bit of Dev time on the games):

• Re-Organise the workload distribution in the teams (target: @Jamor != Ramirez). Automate testing. Improve infrastructure (i.e. tooling, procedures). Create internal documentation (for reference). Review high-level architecture. Thats all an upfront cost in hours. Furthermore, all of those items tend to be incremental - meaning that more than one wave of effort will be required. Also, the last one will touch the Clausewitz engine, with all the problems that entails.
• Hire more people. For why this can drastically slow development, see The mythical man-month

The consequences from those two points leave me quite content with the state of affairs, since statements by the devs indicate that they know this and have no intention of obscuring this (evident by the lack of corp-speak). Problem is, these things take time - and patience gets rarer these days.