Author Topic: v1.9.4 Potential Bugs Thread  (Read 22787 times)

0 Members and 1 Guest are viewing this topic.

Offline Exultant

  • Petty Officer
  • **
  • E
  • Posts: 28
  • Thanked: 17 times
Re: v1.9.4 Potential Bugs Thread
« Reply #60 on: May 03, 2020, 08:43:26 PM »
Edit: I have screenshots embedded in my post, but the forums are adding a space after the .in the URL and they're not showing. Until I figure out how to fix this, here is the link to the imgur album.  https://imgur.com/a/nR0CElt The screenshots are uploaded in order that I discuss them, and I'll reference them in text. If a mod can fix the img code for me through a message edit please go ahead and do so. 

I have a series of catastrophic bugs in ground combat. I'll provide the requested information here, but a much more detailed walkthrough below:


The function numbers
(all prefaced with 1.9.4) 2712, 1810, 327, 1821, 2868 (This last one is elusive, but the most verbose)

The complete error texts (For 2712, 1810, 327) The object reference not set to an instance of an object.(For 1821) Attempted to divide by zero (For 2868) "Index was out of range.  Must be non-negative and less than the size of the collection.  Parameter name: Index"

The window affected Global

What you were doing at the time Invading NPR Homeworld in NN 3819

Conventional or TN start TN

Random or Real Stars Real

Is your decimal separator a comma? Natively a period

Is the bug is easy to reproduce, intermittent or a one-off? Easy to reproduce.  See below

If this is a long campaign - say 75 years or longer - let me know the length of the campaign as well 45 years, started in 1. 9. 0, migrated to . 3 and now . 4. 

Detailed Explanation

I saved the game immediately after this throwing the error.  I had (I believe) one successful round of ground combat before the errors began, but the fallout can be seen from the get-go. 

First, when loading the game, switch the view to NN 3819.  The fun is around the first planet in the system. 

You should see a ton of wrecks of the "Sheean" class around the first planet.  If you turn off the wrecks so the lifepods are visible, the first indication that something is wrong appears

Screenshot 1

Despite there being a ton of wrecks of my fighters in orbit, I've only lost 2 fighters, but they seem to have exploded multiple times, leaving identical copies of wrecks and lifepods.  I've only actually lost fighter 005 and 002.  Looking at the naval organization screen proves this:

Screenshot 2

As you can see, the only ship numbers I'm missing are 2 and 5, and you can see that, for some reason, the game has created empty fleets for them - this seems to be due to the engine being hit, the speed dropping and ship detaching, prior to the game calculating that we've reached HTK and the fighter is dead.  Below is a clip from the events window detailing the death of Sheean 012. 

Screenshot 3

Advancing time by 8 hours to the next ground combat phase will give the following errors in some random combination of times: 1810, 370.  Also, Sheean 001 is usually the next to explode multiple times. 

Additional advancements generally also include error 2712, in addition to 1810 and 370. 

Also, generally within the first or second advancement of 8 hours, one of the events will be that we've conquered a population on Mars.  This is unrelated to the Ground Combat bugs, but it is a bug that's been running for several in-game years: You can see the results on the Colony Summary window:

Screenshot 4

Several years ago the NPR tried to land troops on Mars and I defeated the fleet in orbit.  Ever since that point, every 6 months or so the NPR creates a new colony on Mars and my lone garrison immediately conquers it.  This is without the NPR having any fleet to speak of (I've been pruning them for a few years as I built an army).  No error messages are shown, so I can't give any more details as to why this occurs. 

Back to ground combat fun:

The elusive error that only shows up once or twice randomly (and which I had to reload the save many times to finally catch) is "Error 2686 Index was out of range.  Must be non-negative and less than the size of the collection.  Parameter name: Index. " I cannot consistently recreate this one, but advancing time enough will cause it to appear once or twice in the pile of error popups. 


Ground Combat fun ends approximately 14 days after the start date of this save-file, just before the life pods expire.  The game will throw error 1821 Attempted to divide by zero and all further ground combat breaks.  When this happens, I stop getting any combat reports other than estimations of enemy composition, medal awards for destroyed tonnage (which I set up) and formation breakthroughs.  The game still seems to think that I'm destroying significant amounts of enemies due to the medal awards at 5k, 10k and 25k tonnage destroyed, and the sensors in orbit detect tiny tiny changes to the ground force strength (generally 100t at a time).  I will stop losing troops, however. 

You can force the divide by zero error to happen sooner by moving supply elements from my reserve battalion (242nd Battalion HQ) into the top level of the 4 divisions (2 inf, 2 armored) so that the sub formations can resupply.  When this happens, on the next 8h time advancement divide by zero occurs, and ground combat effectively breaks. 

Screenshot 5
« Last Edit: May 04, 2020, 12:36:53 AM by Exultant »
 

Offline Ironclad Mouse

  • Petty Officer
  • **
  • I
  • Posts: 16
  • Thanked: 4 times
Re: v1.9.4 Potential Bugs Thread
« Reply #61 on: May 03, 2020, 09:41:09 PM »
Quote from: Exultant link=topic=11231. msg130501#msg130501 date=1588556606
Edit: I have screenshots embedded in my post, but the forums are adding a space after the .     in the URL and they're not showing.    Until I figure out how to fix this, here is the link to the imgur album.    https://imgur.   com/a/nR0CElt The screenshots are uploaded in order that I discuss them, and I'll reference them in text.    If a mod can fix the img code for me through a message edit please go ahead and do so. 

Just to let you know, that's because you have less than a certain amount of posts, I forget what the number is, I think it's ten posts, but once you get that many posts it'll stop doing that
« Last Edit: May 03, 2020, 09:44:21 PM by Ironclad Mouse »
"Are you a man, or a mouse"
"Squeak"
 

Offline Ironclad Mouse

  • Petty Officer
  • **
  • I
  • Posts: 16
  • Thanked: 4 times
Re: v1.9.4 Potential Bugs Thread
« Reply #62 on: May 03, 2020, 09:43:29 PM »
V. 1.9.4, checked Misc. tab to be sure
Function number:N/a
Error Text:N/a
Window Effected:Tricky to answer, but I guess the main screen
What you were doing:Surveying starting system
Start:Conventional, Spacemaster race, Spacemaster System
Stars:Random
Decimal Separator: "."
Easy, Intermittent, or one off: Easy to reproduce
Campaign Length: 21 years
I was surveying my starting system with auto-turns on and left to get a drink, when I came back I saw there had been an interrupt, so I checked what it was, it was research but under it there was a message under it saying my survey ship had insufficient crew quarters, so I checked the event logs and sure enough some turns back it had suffered a maintenance failure, but the game turns had not stopped. If that research hadn't of happened it seems I could've lost a ship just because I was thirsty
The game this happened in is named "Desert" species "Davonians" and the ship is the "Essex"
Reproduction steps: 1)Build military classed ship
2) Allow it to run out of supplies
3) Continue passing time with auto-turns on
4) Hope ship doesn't explode when you look away from it for a moment
"Are you a man, or a mouse"
"Squeak"
 

Offline Exultant

  • Petty Officer
  • **
  • E
  • Posts: 28
  • Thanked: 17 times
Re: v1.9.4 Potential Bugs Thread
« Reply #63 on: May 03, 2020, 11:14:03 PM »
Thanks Ironclad Mouse.  I'll edit the post for readability once my post count gets higher. 

In the meantime, I was only able to attach 3 screenshots with the databases, but I have 5 screenshots on the imgur album, so thread mods, please follow the link instead of opening attachments. 
 

Offline bankshot

  • Lieutenant
  • *******
  • b
  • Posts: 194
  • Thanked: 48 times
Re: v1.9.4 Potential Bugs Thread
« Reply #64 on: May 03, 2020, 11:56:39 PM »
SM mode random ruins:  when repeatedly hitting "random ruin" in an attempt to get a particular ruin type the Alien Installation type locks.  The installation remains present even when selecting "random ruin" again to delete the ruin.  And the installation does not appear to change. If you get say Ground 50% installation that is permanent even when you repeatedly re-roll the ruin.  This is different than the behavior in 1.5.1 where the installation was also deleted upon deletion of the ruin. 

Minor display bug: the ruin text does not update after deletion, you have to close and re-open the window to see that is deleted.  The text will update with the new ruin type if you reroll the ruin again.


Version: 1.9.4

The function number: no error
The complete error text: N/A
The window affected: System Generation and display
What you were doing at the time: Repeatedly pressing "random ruin" in SM mode
Conventional or TN start: TN
Random or Real Stars: read
Is your decimal separator a comma? no
Is the bug is easy to reproduce, intermittent or a one-off? easy
 

Offline TMaekler

  • Vice Admiral
  • **********
  • Posts: 1112
  • Thanked: 298 times
Re: v1.9.4 Potential Bugs Thread
« Reply #65 on: May 04, 2020, 12:51:50 AM »
Reported in 1.9.3, still there in 1.9.4:

Some DSTS circles are displayed twice around a system like this:
https://drive.google.com/open?id=1eyTh6bn5dYd55jOulg-I5SiwnKUAAtX6


Here is the DB:
https://drive.google.com/open?id=1ZGu9d3CmE4behsnqwmX7kQQk38RTsdHI

There are two bodies cery close to each other, Pandora & Daphne. But only one of them has DSTS (Daphne).
 

Offline Demonius

  • Warrant Officer, Class 1
  • *****
  • Posts: 83
  • Thanked: 25 times
Re: v1.9.4 Potential Bugs Thread
« Reply #66 on: May 04, 2020, 01:59:20 AM »
1.9.4, started in 1.9.0, 21 years into the Campaign.

Possible civilian pathfinding bug, no error message.

Looking at the JP1 from Sol to Alpha Centauri, theres 3 civilian SLs doing a merry go round around the jumppoint.

Advancing time by below 3hrs shows the currently active Liner just sitting at the Jumppoint indefinitely doing nothing.
Advancing time by 3 hrs or more makes the ships continually exchange position on this or the far side of the JP.
Ships affected: SL Dong L4 002, SL Briseno L5 002, SL Winborn L4 011

All other FT/CS of all the shipping lines ships seem to work normally.
 

Offline serger

  • Commodore
  • **********
  • Posts: 639
  • Thanked: 120 times
  • Silver Supporter Silver Supporter : Support the forums with a Silver subscription
    2021 Supporter 2021 Supporter : Donate for 2021
    2022 Supporter 2022 Supporter : Donate for 2022
Re: v1.9.4 Potential Bugs Thread
« Reply #67 on: May 04, 2020, 02:32:34 AM »
You have to press "default fleet" button to make it remember which fleet to use by default:



I just tested and the button works as it should in 1.9.4

Yep, I missed this button somehow, thank you!
 

Offline SpaceMarine

  • Bug Moderators
  • Rear Admiral
  • ***
  • Posts: 904
  • Thanked: 877 times
Re: v1.9.4 Potential Bugs Thread
« Reply #68 on: May 04, 2020, 02:44:17 AM »
Reported in 1.9.3, still there in 1.9.4:

Some DSTS circles are displayed twice around a system like this:
https://drive.google.com/open?id=1eyTh6bn5dYd55jOulg-I5SiwnKUAAtX6


Here is the DB:
https://drive.google.com/open?id=1ZGu9d3CmE4behsnqwmX7kQQk38RTsdHI

There are two bodies cery close to each other, Pandora & Daphne. But only one of them has DSTS (Daphne).

Thank you for the report, I have entered the DB and determined that the bug does indeed exist and is confirmed, I will be moving this over to confirmed bugs thank you.
 

Offline SpaceMarine

  • Bug Moderators
  • Rear Admiral
  • ***
  • Posts: 904
  • Thanked: 877 times
Re: v1.9.4 Potential Bugs Thread
« Reply #69 on: May 04, 2020, 02:50:06 AM »
SM mode random ruins:  when repeatedly hitting "random ruin" in an attempt to get a particular ruin type the Alien Installation type locks.  The installation remains present even when selecting "random ruin" again to delete the ruin.  And the installation does not appear to change. If you get say Ground 50% installation that is permanent even when you repeatedly re-roll the ruin.  This is different than the behavior in 1.5.1 where the installation was also deleted upon deletion of the ruin. 

Minor display bug: the ruin text does not update after deletion, you have to close and re-open the window to see that is deleted.  The text will update with the new ruin type if you reroll the ruin again.


Version: 1.9.4

The function number: no error
The complete error text: N/A
The window affected: System Generation and display
What you were doing at the time: Repeatedly pressing "random ruin" in SM mode
Conventional or TN start: TN
Random or Real Stars: read
Is your decimal separator a comma? no
Is the bug is easy to reproduce, intermittent or a one-off? easy

Thank you for following the formatting, I have confirmed that you are not able to delete the ruins and also the bonus stays the same, though the installation does appear to change, either way this is a bug of somekind and will be logged in confirmed, thank you for the report.
 

Offline SpaceMarine

  • Bug Moderators
  • Rear Admiral
  • ***
  • Posts: 904
  • Thanked: 877 times
Re: v1.9.4 Potential Bugs Thread
« Reply #70 on: May 04, 2020, 02:56:34 AM »
1.9.4, started in 1.9.0, 21 years into the Campaign.

Possible civilian pathfinding bug, no error message.

Looking at the JP1 from Sol to Alpha Centauri, theres 3 civilian SLs doing a merry go round around the jumppoint.

Advancing time by below 3hrs shows the currently active Liner just sitting at the Jumppoint indefinitely doing nothing.
Advancing time by 3 hrs or more makes the ships continually exchange position on this or the far side of the JP.
Ships affected: SL Dong L4 002, SL Briseno L5 002, SL Winborn L4 011

All other FT/CS of all the shipping lines ships seem to work normally.

Thank you for the report, I have been able to confirm this and I will be moving it to the confirmed thread, let me know if you get any more bugs to report :)
 

Offline SpaceMarine

  • Bug Moderators
  • Rear Admiral
  • ***
  • Posts: 904
  • Thanked: 877 times
Re: v1.9.4 Potential Bugs Thread
« Reply #71 on: May 04, 2020, 03:09:41 AM »
V. 1.9.4, checked Misc. tab to be sure
Function number:N/a
Error Text:N/a
Window Effected:Tricky to answer, but I guess the main screen
What you were doing:Surveying starting system
Start:Conventional, Spacemaster race, Spacemaster System
Stars:Random
Decimal Separator: "."
Easy, Intermittent, or one off: Easy to reproduce
Campaign Length: 21 years
I was surveying my starting system with auto-turns on and left to get a drink, when I came back I saw there had been an interrupt, so I checked what it was, it was research but under it there was a message under it saying my survey ship had insufficient crew quarters, so I checked the event logs and sure enough some turns back it had suffered a maintenance failure, but the game turns had not stopped. If that research hadn't of happened it seems I could've lost a ship just because I was thirsty
The game this happened in is named "Desert" species "Davonians" and the ship is the "Essex"
Reproduction steps: 1)Build military classed ship
2) Allow it to run out of supplies
3) Continue passing time with auto-turns on
4) Hope ship doesn't explode when you look away from it for a moment

First thank you for providing a detailed report, second I have looked in the DB and can confirm that for some reason you are not getting an interrupt when losing components onboard the ship or suffering any kind of maintenance failure, while suffering a maintenance failure that gets repaired and that not interrupting is fine, when you cant repair the maintenance failure it should give some kind of an interrupt, either way I will be sending this to confirmed bug, we will see what steve says.

P.S your deployment time on your ship design was 3 months, might wanna fix that for a survey craft.
 
The following users thanked this post: Ironclad Mouse

Offline Demakustus

  • Chief Petty Officer
  • ***
  • D
  • Posts: 30
  • Thanked: 19 times
Re: v1.9.4 Potential Bugs Thread
« Reply #72 on: May 04, 2020, 03:15:38 AM »
The function number - (both prefaced with 1. 9. 4) 2608, 1838
The complete error text - A reference to an object was not set to an instance of an object (both functions)
The window affected - Ground Forces -> Unit Class Design tab
What you were doing at the time - Trying to design an STO, without any Active Grav Sensor Strength tech
Conventional or TN start - Conventional only
Random or Real Stars - Real Stars, but it shouldn't matter
Is your decimal separator a comma? - No
Is the bug is easy to reproduce, intermittent or a one-off? - Easy
If this is a long campaign - say 75 years or longer - let me know the length of the campaign as well - No length at all, can happen immediately

Steps to reproduce - Begin a new conventional start game, SM in a tech for a direct fire weapon (tested with particle beam and laser), try to design a ground STO unit, then when switching the component type for a static unit I get a regular popup that an STO requires sensor tech followed by the 2608 and 1838 error, and this happens 4 times in a row.
It seems that the error situation happens for every UI element that shows when selecting the STO component (like the panel, the PDW checkbox, the ECCM checkbox and the weapon itself).
I don't think this causes any other issues, it's quite situational and hardly game breaking.
 

Offline SpaceMarine

  • Bug Moderators
  • Rear Admiral
  • ***
  • Posts: 904
  • Thanked: 877 times
Re: v1.9.4 Potential Bugs Thread
« Reply #73 on: May 04, 2020, 03:18:08 AM »
Edit: I have screenshots embedded in my post, but the forums are adding a space after the .in the URL and they're not showing. Until I figure out how to fix this, here is the link to the imgur album.  https://imgur.com/a/nR0CElt The screenshots are uploaded in order that I discuss them, and I'll reference them in text. If a mod can fix the img code for me through a message edit please go ahead and do so. 

I have a series of catastrophic bugs in ground combat. I'll provide the requested information here, but a much more detailed walkthrough below:


The function numbers
(all prefaced with 1.9.4) 2712, 1810, 327, 1821, 2868 (This last one is elusive, but the most verbose)

The complete error texts (For 2712, 1810, 327) The object reference not set to an instance of an object.(For 1821) Attempted to divide by zero (For 2868) "Index was out of range.  Must be non-negative and less than the size of the collection.  Parameter name: Index"

The window affected Global

What you were doing at the time Invading NPR Homeworld in NN 3819

Conventional or TN start TN

Random or Real Stars Real

Is your decimal separator a comma? Natively a period

Is the bug is easy to reproduce, intermittent or a one-off? Easy to reproduce.  See below

If this is a long campaign - say 75 years or longer - let me know the length of the campaign as well 45 years, started in 1. 9. 0, migrated to . 3 and now . 4. 

Detailed Explanation

I saved the game immediately after this throwing the error.  I had (I believe) one successful round of ground combat before the errors began, but the fallout can be seen from the get-go. 

First, when loading the game, switch the view to NN 3819.  The fun is around the first planet in the system. 

You should see a ton of wrecks of the "Sheean" class around the first planet.  If you turn off the wrecks so the lifepods are visible, the first indication that something is wrong appears

Screenshot 1

Despite there being a ton of wrecks of my fighters in orbit, I've only lost 2 fighters, but they seem to have exploded multiple times, leaving identical copies of wrecks and lifepods.  I've only actually lost fighter 005 and 002.  Looking at the naval organization screen proves this:

Screenshot 2

As you can see, the only ship numbers I'm missing are 2 and 5, and you can see that, for some reason, the game has created empty fleets for them - this seems to be due to the engine being hit, the speed dropping and ship detaching, prior to the game calculating that we've reached HTK and the fighter is dead.  Below is a clip from the events window detailing the death of Sheean 012. 

Screenshot 3

Advancing time by 8 hours to the next ground combat phase will give the following errors in some random combination of times: 1810, 370.  Also, Sheean 001 is usually the next to explode multiple times. 

Additional advancements generally also include error 2712, in addition to 1810 and 370. 

Also, generally within the first or second advancement of 8 hours, one of the events will be that we've conquered a population on Mars.  This is unrelated to the Ground Combat bugs, but it is a bug that's been running for several in-game years: You can see the results on the Colony Summary window:

Screenshot 4

Several years ago the NPR tried to land troops on Mars and I defeated the fleet in orbit.  Ever since that point, every 6 months or so the NPR creates a new colony on Mars and my lone garrison immediately conquers it.  This is without the NPR having any fleet to speak of (I've been pruning them for a few years as I built an army).  No error messages are shown, so I can't give any more details as to why this occurs. 

Back to ground combat fun:

The elusive error that only shows up once or twice randomly (and which I had to reload the save many times to finally catch) is "Error 2686 Index was out of range.  Must be non-negative and less than the size of the collection.  Parameter name: Index. " I cannot consistently recreate this one, but advancing time enough will cause it to appear once or twice in the pile of error popups. 


Ground Combat fun ends approximately 14 days after the start date of this save-file, just before the life pods expire.  The game will throw error 1821 Attempted to divide by zero and all further ground combat breaks.  When this happens, I stop getting any combat reports other than estimations of enemy composition, medal awards for destroyed tonnage (which I set up) and formation breakthroughs.  The game still seems to think that I'm destroying significant amounts of enemies due to the medal awards at 5k, 10k and 25k tonnage destroyed, and the sensors in orbit detect tiny tiny changes to the ground force strength (generally 100t at a time).  I will stop losing troops, however. 

You can force the divide by zero error to happen sooner by moving supply elements from my reserve battalion (242nd Battalion HQ) into the top level of the 4 divisions (2 inf, 2 armored) so that the sub formations can resupply.  When this happens, on the next 8h time advancement divide by zero occurs, and ground combat effectively breaks. 

Screenshot 5

Thank you for the extremely detailed post, I have had a quick glance into the DB and thats a pretty screwed up game, the Mars issue alone is weird, so I will be moving this to confirmed, I was also able to see the lifepod issue, Thank you for the report.
 

Offline SpaceMarine

  • Bug Moderators
  • Rear Admiral
  • ***
  • Posts: 904
  • Thanked: 877 times
Re: v1.9.4 Potential Bugs Thread
« Reply #74 on: May 04, 2020, 03:27:06 AM »
The function number - (both prefaced with 1. 9. 4) 2608, 1838
The complete error text - A reference to an object was not set to an instance of an object (both functions)
The window affected - Ground Forces -> Unit Class Design tab
What you were doing at the time - Trying to design an STO, without any Active Grav Sensor Strength tech
Conventional or TN start - Conventional only
Random or Real Stars - Real Stars, but it shouldn't matter
Is your decimal separator a comma? - No
Is the bug is easy to reproduce, intermittent or a one-off? - Easy
If this is a long campaign - say 75 years or longer - let me know the length of the campaign as well - No length at all, can happen immediately

Steps to reproduce - Begin a new conventional start game, SM in a tech for a direct fire weapon (tested with particle beam and laser), try to design a ground STO unit, then when switching the component type for a static unit I get a regular popup that an STO requires sensor tech followed by the 2608 and 1838 error, and this happens 4 times in a row.
It seems that the error situation happens for every UI element that shows when selecting the STO component (like the panel, the PDW checkbox, the ECCM checkbox and the weapon itself).
I don't think this causes any other issues, it's quite situational and hardly game breaking.

Thank you for the well formatted bug report, i have been able to reproduce this, it seems more like the errors its spitting out are the issue not the fact that its refusing to build without sensors, steve has dealt with bugs like this in the past, as such I will be sending it to confirmed.

P.S Am clocking off for now so wont be answering new bug reports as I have to work on my AAR and more tutorials.
« Last Edit: May 04, 2020, 03:32:48 AM by SpaceMarine »
 
The following users thanked this post: Demakustus