Author Topic: Official v5.42 Bugs Thread  (Read 69646 times)

0 Members and 1 Guest are viewing this topic.

Offline sloanjh

  • Global Moderator
  • Admiral of the Fleet
  • *****
  • Posts: 2805
  • Thanked: 112 times
  • 2020 Supporter 2020 Supporter : Donate for 2020
    2021 Supporter 2021 Supporter : Donate for 2021
Re: Official v5.42 Bugs Thread
« Reply #360 on: June 27, 2011, 08:46:54 AM »
Looking at the code, I think the "x missiles destroyed" message may be showing a cumulative number of missiles hit, rather than hit by that particular fire control. Are there any lines showing individual weapon fire, rather than the summary total?

Steve

Aha!!  That's what I get for not reading the logs more carefully.  Yes - it looks like it's listing per-fire-control for "weapons fired" but per-enemy-salvo(?) for "missiles destroyed".  There are indeed the appropriate number of individual "hit" and "missed" messages.

Sorry about the false alarm.

John

PS - I assume you'll fix the counts to be consistent.  If so, it would be good if they were consistent with how often "Firing Summary" is printed.  If it's after every fire control then the missiles-destroyed count should be per fire control (of course).
 

Offline Charlie Beeler

  • Registered
  • Vice Admiral
  • **********
  • Posts: 1381
  • Thanked: 3 times
Re: Official v5.42 Bugs Thread
« Reply #361 on: June 27, 2011, 09:45:43 AM »
Looking at the code, I think the "x missiles destroyed" message may be showing a cumulative number of missiles hit, rather than hit by that particular fire control. Are there any lines showing individual weapon fire, rather than the summary total?

Steve

Steve I'm working from memory, but...

In my case they were for individual weapons fire, specificly gauss turrets.  I the salvo engaged started with 66 missiles.  4 ships with triple turrets (12 shots) and 4 ships with quad turrets (16 shots), all are stacked together and firing final defense fire (not self) at range 1.  Indvidual totals varied between 4 and 8 hits with a grand total of 74 hits with the last individual total being 8.  The targeted ship was hit by 58 missiles. 

It is not consistant though.  Most salvos of this size are interecepted correctly (ie all missiles destroyed) and appear to have the same type of log entries (ie individual totals vary between 4-8 intercepts with the grand total exceeding the salvo size).

I don't have a copy of the database for this game with me or I would be more specific.

Charlie
Amateurs study tactics, Professionals study logistics - paraphrase attributed to Gen Omar Bradley
 

Offline Steve Walmsley (OP)

  • Aurora Designer
  • Star Marshal
  • S
  • Posts: 11678
  • Thanked: 20471 times
Re: Official v5.42 Bugs Thread
« Reply #362 on: June 27, 2011, 09:56:57 AM »
I found that microwaves can damage non-electronic components.   I accidentally blew up a ship I was trying to disable.   It also made it monstrously hard to blind this other big ship. 

Fixed for v5.50

Steve
 

Offline Steve Walmsley (OP)

  • Aurora Designer
  • Star Marshal
  • S
  • Posts: 11678
  • Thanked: 20471 times
Re: Official v5.42 Bugs Thread
« Reply #363 on: June 27, 2011, 09:58:41 AM »
Aha!!  That's what I get for not reading the logs more carefully.  Yes - it looks like it's listing per-fire-control for "weapons fired" but per-enemy-salvo(?) for "missiles destroyed".  There are indeed the appropriate number of individual "hit" and "missed" messages.

Sorry about the false alarm.

John

PS - I assume you'll fix the counts to be consistent.  If so, it would be good if they were consistent with how often "Firing Summary" is printed.  If it's after every fire control then the missiles-destroyed count should be per fire control (of course).


Yes I have fixed the counts. The number of missiles destroyed per fire control is now tracked separately from total missiles destroyed and the former is used for the firing summary.

Steve
 

Offline Steve Walmsley (OP)

  • Aurora Designer
  • Star Marshal
  • S
  • Posts: 11678
  • Thanked: 20471 times
Re: Official v5.42 Bugs Thread
« Reply #364 on: June 27, 2011, 10:06:06 AM »
Steve I'm working from memory, but...

In my case they were for individual weapons fire, specificly gauss turrets.  I the salvo engaged started with 66 missiles.  4 ships with triple turrets (12 shots) and 4 ships with quad turrets (16 shots), all are stacked together and firing final defense fire (not self) at range 1.  Indvidual totals varied between 4 and 8 hits with a grand total of 74 hits with the last individual total being 8.  The targeted ship was hit by 58 missiles. 

It is not consistant though.  Most salvos of this size are interecepted correctly (ie all missiles destroyed) and appear to have the same type of log entries (ie individual totals vary between 4-8 intercepts with the grand total exceeding the salvo size).

I don't have a copy of the database for this game with me or I would be more specific.

Charlie

Another issue with the point defence code is that overkill is possible, so more missiles appear to be shot down than are actually destroyed. For example, in v5.42 assume you have two fire controls, each with two triple laser turrets each firing at separate incoming salvos of four missiles. It is possible fire control A might show a summary of six hits, even though there are only four missiles to hit, and fire control B might show a summary of two hits. With eight missiles incoming and eight missile hit messages, you may think you have killed all eight when in fact two incoming missiles remain. This situation only occurs for multiple shot weapons such as turrets, railguns, CIWS, etc. In v5.50 I have added some code to stop firing as soon as all targeted missiles have been killed, even if this means a multi-shot weapon doesn't fire all available shots. This should mean that the number of successful shots matches the number of missiles destroyed.

Steve
 

Offline Charlie Beeler

  • Registered
  • Vice Admiral
  • **********
  • Posts: 1381
  • Thanked: 3 times
Re: Official v5.42 Bugs Thread
« Reply #365 on: June 27, 2011, 11:01:14 AM »
Another issue with the point defence code is that overkill is possible, so more missiles appear to be shot down than are actually destroyed. For example, in v5.42 assume you have two fire controls, each with two triple laser turrets each firing at separate incoming salvos of four missiles. It is possible fire control A might show a summary of six hits, even though there are only four missiles to hit, and fire control B might show a summary of two hits. With eight missiles incoming and eight missile hit messages, you may think you have killed all eight when in fact two incoming missiles remain. This situation only occurs for multiple shot weapons such as turrets, railguns, CIWS, etc. In v5.50 I have added some code to stop firing as soon as all targeted missiles have been killed, even if this means a multi-shot weapon doesn't fire all available shots. This should mean that the number of successful shots matches the number of missiles destroyed.

Steve

Let's hope that solves the issue I've had.  I have a feeling it will.

In my example there is only 1 salvo of 66 missiles.  There are a total of 8 turret/fire controls suites (1 per ship).  The log shows that each suite destroyed between 4 and 8 missiles with the last suite destroying 8.  I think the issue is that fire control #7 exactly destroyed the last of the salvo.  It would appear that since there was another fire control targeting the salvo in final defense the salvo size was reset to the original 66 and then the final fire control destroyed 8 leaving the 58 missiles that hit the targeted ship. 
Amateurs study tactics, Professionals study logistics - paraphrase attributed to Gen Omar Bradley
 

Offline Steve Walmsley (OP)

  • Aurora Designer
  • Star Marshal
  • S
  • Posts: 11678
  • Thanked: 20471 times
Re: Official v5.42 Bugs Thread
« Reply #366 on: June 27, 2011, 12:23:28 PM »
I can't see the stats for an NPR race on the "View Race Details" (ctrl-F2) screen (including in SM mode) even if I've captured thousands of them from lifepods.

I understand that you might not want to give players access to the edit functions on the race details screen, 'cuz they might manage to screw up the computer-controlled NPR, but that could be fixed by disabling the edit mode if it's a computer-controlled race.  I think ctrl-F2 information should be available for any race with which you've got communications....

For v5.50, an Empire can conduct "Alien Autopsies" (using a 5000 RP research project) on each species that it recovers from lifepods. These species will then show up for the appropriate alien empire on the Alien Race window along with their environmental tolerances. Tracking known alien species has been running for a while in the background so this was fairly easy to introduce. When I get around to biowarfare, this will be the first step on the road to building bioweapons targeted on a particular species.

Steve

 

Offline Steve Walmsley (OP)

  • Aurora Designer
  • Star Marshal
  • S
  • Posts: 11678
  • Thanked: 20471 times
Re: Official v5.42 Bugs Thread
« Reply #367 on: June 27, 2011, 12:31:49 PM »
After shutting down an industry, reactivating it usually requires a 180 day wait.
When you reactivate a shutdown industry, the time in the box shows how much longer before the industry is active.
If you click the cancel button you would assume that this would cancel the reactivation, but instead it makes the industry active immediately.

Fixed for v5.50

Steve
 

Offline Steve Walmsley (OP)

  • Aurora Designer
  • Star Marshal
  • S
  • Posts: 11678
  • Thanked: 20471 times
Re: Official v5.42 Bugs Thread
« Reply #368 on: June 27, 2011, 12:32:46 PM »
YEAH!!!!  So does that mean there's an easy workaround - SM your relation with them up to 100 and grant them trade access?
(And remember how many points you added so they can be subtracted later when the relationship is strong enough.)

John

Yes, that workaround will fix the problem.

Steve
 

Offline sloanjh

  • Global Moderator
  • Admiral of the Fleet
  • *****
  • Posts: 2805
  • Thanked: 112 times
  • 2020 Supporter 2020 Supporter : Donate for 2020
    2021 Supporter 2021 Supporter : Donate for 2021
Re: Official v5.42 Bugs Thread
« Reply #369 on: June 27, 2011, 11:58:37 PM »
Instant OOB ignores the "conscript" flag.

I had a fleet of 4 cargo ships in a weird state - every time they offloaded (or loaded) something I would get a "failed to unload" message.  I instant-OOB'd another 4 replacement ships and that seemed to solve the problem, but the instant ships came in with my racial training level even though they're supposed to be conscript-only.

As for the weird state, I can think of 3 things that might have triggered it:
1)  The TG they were in was dropping components off after salvaging an Invader.
2)  I had just gone through a big commander auto-update cycle.
3)  The TG had both a tug and a salvage ship (with the tug towing the salvage ship) when I tried to offload.  I later pulled them out into a different TG and was seeing the same problems, though.

None of the above seem to me like it should have triggered this state, so I don't know what else is going on. 

John
I tracked this (the weird state) issue down: if a TF Logistics bonus is too good, you'll consistently get failed to load/unload messages for TG.

In my last bout of officer reassignments, I assigned a captain with a 100% logistics bonus to Logistics Command.  The commander of Logistics Command also has a bonus, so the total bonus (from the TF screen) is 110%.  It just occured to me while typing this that this probably means it takes negative time to unload a TG in that TF. :)

In any event, this was causing the weird behavior - when I swapped out the logistics officer for one with a lower bonus the problem went away.

John

PS - I'm not just guessing here - I actually did a lot of experimenting in SM mode and looking in the DB - the only difference between a TG that worked and one that didn't was the TF ID.

 

Offline Napoleon XIX

  • Petty Officer
  • **
  • N
  • Posts: 26
Re: Official v5.42 Bugs Thread
« Reply #370 on: June 29, 2011, 08:27:20 AM »
Whenever a try to advance the time, I get flooded with a bunch of these messages:

"Error in ExecuteOrders

Error 6 was generated by Aurora
Overflow
Please report to .  .  .  "

What could be causing it?

EDIT: the above (but not the below) problem was solved by deleting the training TG that had no units in it.

I have also been getting the following message ever since I discovered a nebula system:

"Error in OrbitalMovement

Error 11 was generated by Aurora
Division by 0
Pl.  .  .  "
« Last Edit: June 29, 2011, 08:54:37 AM by Napoleon XIX »
 

Offline enkril

  • Able Ordinary Rate
  • e
  • Posts: 2
Re: Official v5.42 Bugs Thread
« Reply #371 on: June 30, 2011, 10:23:30 AM »
Hey for my first post 2 small bugs I founded :
- I created a crappy design just to make some test.  The problem is that it was based on my Colony Ship so some shipping lines chose to build it.  It wasn't that much of a problem because the speed was 1Km/s so it was just a waste of money for them.
But a while latter I did some more experimenting with my test design.  So I removed the Cryo Transport and put only engines.  As a result, I keep getting the following message:
12th August 2067 05:04:46,Sol,CS Gaudin Project 004 failed to Unload Colonists at Mars.  Please reschedule the orders for this fleet

As a result I changed again the design so it is again able to hold colonists and the message stopped.
I think design used by civilian should be locked or cloned to prevent such problem.
I haven't tried it but I think an exploit would be to oversized the design of an already built civilian ship

- This one is fairly easy to reproduce :
Go to the missile design screen and select the drone button
Close the screen and open it later.

As the result, the missile button is selected but the name and calculation are made according to the drone rules.  On the other side, engine and agility aren't read-only so it is probably possible to overrule the limitations of drones. . .
As I founded out, the biggest issue is that if you select a missile design it sticks with the drone calculation so you can end with something really different from what you want. . .


Since this is my first post, I must say thank you Steve, in many ways this is my dreams game!
 

Offline Cavgunner

  • Lt. Commander
  • ********
  • Posts: 283
  • Thanked: 129 times
Re: Official v5.42 Bugs Thread
« Reply #372 on: July 02, 2011, 09:58:11 AM »
I don't think I have ever seen a Cannot Open Database error before. It sounds like the database file itself might be corrupted somehow. Have you opened the Stevefire.mdb file in Access or used designer mode in some way?

Steve

I appreciate the response.  Imagine someone who has the least possible knowledge about programming.  That would be me.  So to answer the question, I don't recall opening any such file, and I wouldn't know what to do there if I had.  Also I have not used designer mode, though I do usually keep SM mode on.
 

Offline Jacob/Lee

  • Lt. Commander
  • ********
  • Posts: 203
  • Thanked: 1 times
  • (Where I got this .gif beats me!)
Re: Official v5.42 Bugs Thread
« Reply #373 on: July 04, 2011, 11:17:33 PM »
For some reason, just out of the blue, the system map will chop in half. Something might be causing it but I haven't caught on to what it may be.
 

Offline welchbloke

  • Vice Admiral
  • **********
  • Posts: 1044
  • Thanked: 9 times
Re: Official v5.42 Bugs Thread
« Reply #374 on: July 05, 2011, 01:01:44 AM »
For some reason, just out of the blue, the system map will chop in half. Something might be causing it but I haven't caught on to what it may be.

I've had this as well  it seems to be related to when another program uses the graphics card/chip. If I play  another game whilst Aurora is on I get this effect.
Welchbloke