v2.1.1 Bugs Thread

[nuclearslurpee]

Firstly designing a prototype using 'next tech' seems to result in the prototype not detecting that the missing research to allow it has been completed leaving it forever as a 'Future Prototype' even when you can recreate it without any extra theoretical tech. It can be worked around by recreating the exact settings for a new research project/prototype, but it means that you can't tell the previous itteration to be made into a research project for your queue.

This one is a known behavior, technically is is WAD and should be a suggestion but either way it would be nice to have this work more smoothly.

The second one I believe has been spotted, or at least there's a couple of threads on something similar. UK officer ranks, some organisation set up with the various Admin Commands having their lowest rank as Commodore (rank 4). Currently no ships requiring above Rank 1 resulting in the admin commands not having anybody taking control because nobody is reaching Commander (Rank 2), much less Captain or Commodore. Realisitic promotions on and it's around 50 years in.

Not a bug, this is 100% WAD and WAI. You need to have roles at the intermediate ranks for the promotions-on-demand system to work.

Found an odd bug.

I've also recently been struggling with some DB corruption problems. I have assumed these were due to either an ill-advised DB modification or a problem with my hard drive or other aging hardware (neither of which is reportable, of course), but if there is actually some weirdness about copying and renaming the DB file that would explain a lot.

[Ghrathryn]

This one is a known behavior, technically is is WAD and should be a suggestion but either way it would be nice to have this work more smoothly.

Fair enough, though it'd be nice to be able to do a prototype with the next tech level active and be able to research the prototype once you've researched whichever techs were missing from it. I'm not sure how you'd code things to recognise when a project can be researched if you make a prototype without all the required tech, but it'd certainly be useful to have it recognise, particularly for people part way through researching a tech that they need for a future tech prototype when they make said prototype.

Not a bug, this is 100% WAD and WAI. You need to have roles at the intermediate ranks for the promotions-on-demand system to work.

So probably worth mentioning in the suggestions thread about having the on demand promotions allowing promotions through ranks that don't have things to do in order to fill upper ranks if there's a requirement for those?

[Blacklight]

Im not dead certain, it is hard to be, but ive heard from multiple people plus my current game, it doesnt seem to be generating JP's to "local systems" In that i cannot find the 5 other NPR's on a 250 star map and in 2 games where ive explored 50-60 stars, havent had a single system loop back to one id already explored.  not one.  I have seen 2-3 other people mention that they havent had any either in their games on the discord.

SJW: This is working fine in my own games - looks like RNG rather than a bug.

[nuclearslurpee]

Im not dead certain, it is hard to be, but ive heard from multiple people plus my current game, it doesnt seem to be generating JP's to "local systems" In that i cannot find the 5 other NPR's on a 250 star map and in 2 games where ive explored 50-60 stars, havent had a single system loop back to one id already explored.  not one.  I have seen 2-3 other people mention that they havent had any either in their games on the discord.

This happens a lot on the default settings. The JP network is generated completely randomly, so it is not uncommon to get maps that are different than expected or hoped for, indeed every so often we see someone who explores a handful of systems and finds no further JPs leaving them basically trapped in known space. On random stars, this is compounded because there is no rule about how "near" or "far" the starting NPRs might generate, since the distance settings only affect Real Stars games, but even in Real Stars some weirdness could happen. Basically, if there is a bug it is difficult to say for sure because it is much more probable to be a quirk of randomness. If this were a bug and not just a rare random occurrence, we would expect to see a lot more players reporting it.

What are your system generation settings? In Real Stars it is not terribly uncommon to not see loops generating for even a large number of systems, but in Random Stars there are some settings you can tweak.

[Blacklight]

These are the settings for this game, so far ive found 1 NPR and it was a generated one.   Im playing with known stars cause the game crashes like crazy when you try start a game with it off.   39 stars explored, my other game is similar settings tho a few more stars total, it has 48 systems explored with no NPR's found at all. 
I have heard of the whole people getting locked in with no JP's etc, but i felt this was different cause its happened twice to games on  the new versions, there was also 3 people on the discord who also commented that they had zero loopbacks on the discord and i suspect theyre hitting a similar issue.

[nuclearslurpee]

Im playing with known stars cause the game crashes like crazy when you try start a game with it off.

@Steve plz

But yeah, Known Stars is usually pretty bad for jump loops, especially if you don't get one or two in the first 15-20 systems which is far from guaranteed.

[Blacklight]

yeah i can live without jump loops, but my concern is that its cause the local system generation is borked, and if it is then that would be why im not running into NPR's, as the systems we discover arent being looped to systems they discover.  So you practically have to brute force find their systems as theres no chance for it to link it instead of generating a new system.

SJW: Jump loops are fine in Known Stars

[nanomage]

yeah i can live without jump loops, but my concern is that its cause the local system generation is borked, and if it is then that would be why im not running into NPR's, as the systems we discover arent being looped to systems they discover.   So you practically have to brute force find their systems as theres no chance for it to link it instead of generating a new system.

i've run a test in Known Stars:
1.  create several close star systems - in my case Sol, α Cen, Proxima Cen, and WISE-1639-6847
2.  in each of them, create several dozen JP's (i was using 20), and explore them all
in my test, this created 4 non-connected star graphs.  All these systems are quite close to Sol, and so to each other, yet they never interconnect
It looks like the probability for the game to link a newly explored JP to an already generated system is low enough that it's not observed, even for very close systems. 
This behaviour seems to make for boring loopless maps in Known Stars, and makes ever encountering starting NPR's very unlikely.

SJW: Jump loops are fine in Known Stars. Chance for connection is 6.7%.

[Ragnarsson]

In reference to the question of whether or not Jump Point loops are forming in the current version, I can definitively say they are (at least in games that do NOT utilize Known Stars). That said, they are not forming at the proper rate (see below).

The test I conducted:
- Created a new game
- Deselected Known Stars (finally got the game to properly generate after the 3rd attempt, but that's a different bug...)
- Set Local System Generation % to 100 and Spread to 5

These settings should limit any given system to connecting to the 10 closest systems. As I was not using Known Stars, this is the 10 closest systems by System Number. To see what system numbers were generating I used no system naming theme, thus making it apparent which system numbers were generating.

In the attached image, System #1000 connected to System #2.

I suspected loops might be forming only when they had no other choice; for example, if all other systems within 5 had already been generated. As can be seen from the attached image, that is not, however, the case. Valid options for System #1000 to connect to should be Systems #0-4 and #995-999 (I'm fairly certain Sol is counted as System #0, though if that's the case it shouldn't have been able to generate System #995 given the range of System Spread of 5... honestly, I don't know what the hell is happening there). System #997 has not yet been generated, and would have been a valid choice.

I continued my exploration for a while past what's shown in the screenshot. And I've come to the inescapable conclusion that there IS something wrong with system generation / loop generation logic. The number of times an unexplored jump point had (for example) an 80% chance of connecting to an existing system and a 20% chance of connecting to a new system, and yet generated a new system defies random chance.

Time after time, no matter the odds against it, a new system was generated rather than a connection to an existing one.

I strongly suspect this problem began due to the "Avoidance of Closed Universe" changes in 2.1.1, as I haven't observed it previously.

Hopefully this is somewhat helpful.

[nuclearslurpee]

I've come to the inescapable conclusion that there IS something wrong with system generation / loop generation logic.

While this conclusion may be correct for Known Stars, per the preceding discussion, what you've shown here is exactly what I would expect if the system ID# for Random Stars generation was treated as periodic (i.e., counting as ..., 998, 999, 1000, 1, 2, 3, ...).

However I do share your suspicion about the avoidance-of-closed-universe changes as this does seem like it could be related, at least for Real Stars (would be unsurprised if it was also causing a different bug leading to the generation problems for Random Stars).

[Ragnarsson]

While this conclusion may be correct for Known Stars, per the preceding discussion, what you've shown here is exactly what I would expect if the system ID# for Random Stars generation was treated as periodic (i.e., counting as ..., 998, 999, 1000, 1, 2, 3, ...).

I didn't mean to suggest the sequencing of system ID's was unusual; it's precisely as I'd expect, too. System 1000 connecting to System 2 was unremarkable, as they'd be within the 5 system threshold. I was only attempting to show loops are not impossible which was really my point, but with the important observation of the probability-defying dearth of connections to pre-existing systems (which I observed later, but is not shown in the image above), even when the odds heavily favor such connections over a large number of iterations - and to demonstrate that the lack of loops is affecting games which do not use Known Stars, whereas most reports above were concerning games in which Known Stars were used. It doesn't seem to matter which is used, the same issue seems to be occurring regardless.

[nanomage]

Yes, there is no doubt that loops are possible in Random Stars, at least if Local generation chance is cranked up to 95+.  In Known Stars however I strongly suspect they are not

SJW: Jump loops are fine in Known Stars

[Vandermeer]

Yes, there is no doubt that loops are possible in Random Stars, at least if Local generation chance is cranked up to 95+.  In Known Stars however I strongly suspect they are not

I can lend weight to the observations discussed here. I had it at settings of 50/15 in random stars and got 2 loops within the first 20 systems in 2.1. My current game of 75/30 known stars has expanded nearly as far and seen none, where I also had two loops in a previous 2.0 known stars game at that point.

[vesisko]

I've been unable to generate a new game with random stars.  Game freezes after player race creation, and when I reload I just get a blank screen where clicking on anything throws an error from Function $155.  With real stars turned on it works okay.  DB attached, I don't believe I've done any modding beyond tweaking a few tech or component names/values.  I recall there was a bug about this in v2. 0 - maybe something in that wasn't completely fixed?

I think this is an intermittent problem with all system generation in random stars games.  I reported a similar bug in v2. 0. 2 that I believe to be a different manifestation of the exact same issue, and I don't think it was ever addressed.

hxxp: aurora2. pentarch. org/index. php?topic=13028. msg161111#msg161111

I also have the problem with game freezing.  It also sometimes happens when a ship jupms to a new system.

SJW: Fixed for v2.2

[James Patten]

I am running what I think is 2.1.1 (Aurora.exe dated 8/6/22, but can't figure out how to find out in the program).  I'm running a normal (not conventional) start.  Currently I'm only in Sol.  When I start Aurora I get this error:

2.0.1 Function #3040: Value was either too large or too small for a Decimal.

I hit OK and it's fine.  However I get this error whenever I try to open the System View (LOTS of times - I hold down the Enter key for a minute or so) and Mining Survey Window and search (about a dozen times).