Author Topic: 2.41 Bugs  (Read 9032 times)

0 Members and 1 Guest are viewing this topic.

Offline SteveAlt

  • Global Moderator
  • Rear Admiral
  • *****
  • Posts: 820
  • Thanked: 8 times
(No subject)
« Reply #45 on: December 27, 2007, 06:45:33 AM »
Quote from: "Kurt"
I installed a fresh 2.41 version of Aurora (fortunately not on top of the Quad-System Campaign <G>) and applied some damage to a ship in your Romanov campaign.  I got the same error as before.  

This either means it is a problem unique to my computer, or a general problem with 2.41 that you've since fixed.  

That was a good idea. I can't see how your computer could have a fault external to the programme that could cause this so it must be a problem with v2.41. Unfortunately I don't keep the code for past versions (altough maybe I should) so I don't think I am going to be able to fix this :(

All I can do is try and get v2.5 finished over the holidays but that isn't going to help your current situation.

Steve
« Last Edit: December 31, 1969, 06:00:00 PM by SteveAlt »
 

Offline Kurt (OP)

  • Gold Supporter
  • Vice Admiral
  • *****
  • Posts: 1766
  • Thanked: 3389 times
  • 2021 Supporter 2021 Supporter : Donate for 2021
    Gold Supporter Gold Supporter : Support the forums with a Gold subscription
    2022 Supporter 2022 Supporter : Donate for 2022
    2023 Supporter 2023 Supporter : Donate for 2023
(No subject)
« Reply #46 on: December 27, 2007, 10:01:20 AM »
Quote from: "SteveAlt"
Quote from: "Kurt"
I installed a fresh 2.41 version of Aurora (fortunately not on top of the Quad-System Campaign <G>) and applied some damage to a ship in your Romanov campaign.  I got the same error as before.  

This either means it is a problem unique to my computer, or a general problem with 2.41 that you've since fixed.  
That was a good idea. I can't see how your computer could have a fault external to the programme that could cause this so it must be a problem with v2.41. Unfortunately I don't keep the code for past versions (altough maybe I should) so I don't think I am going to be able to fix this :(

All I can do is try and get v2.5 finished over the holidays but that isn't going to help your current situation.

Steve


I can manually resolve damage that results in a kill, but this can't be good for the database.  I will have to think about this.

Kurt
« Last Edit: December 31, 1969, 06:00:00 PM by Kurt »
 

Offline Kurt (OP)

  • Gold Supporter
  • Vice Admiral
  • *****
  • Posts: 1766
  • Thanked: 3389 times
  • 2021 Supporter 2021 Supporter : Donate for 2021
    Gold Supporter Gold Supporter : Support the forums with a Gold subscription
    2022 Supporter 2022 Supporter : Donate for 2022
    2023 Supporter 2023 Supporter : Donate for 2023
(No subject)
« Reply #47 on: December 27, 2007, 10:19:30 PM »
Quote from: "Kurt"
Quote from: "SteveAlt"
Quote from: "Kurt"
I installed a fresh 2.41 version of Aurora (fortunately not on top of the Quad-System Campaign <G>) and applied some damage to a ship in your Romanov campaign.  I got the same error as before.  

This either means it is a problem unique to my computer, or a general problem with 2.41 that you've since fixed.  
That was a good idea. I can't see how your computer could have a fault external to the programme that could cause this so it must be a problem with v2.41. Unfortunately I don't keep the code for past versions (altough maybe I should) so I don't think I am going to be able to fix this :(

All I can do is try and get v2.5 finished over the holidays but that isn't going to help your current situation.

Steve

I can manually resolve damage that results in a kill, but this can't be good for the database.  I will have to think about this.

Kurt


Okay, I've been playing around with the battle that was underway to see how big a problem this is going to be.  I'm afraid it is rather larger than I thought, and will likely make this version of Aurora unplayable.  

As I stated before, I had to exit out of Aurora to escape the endless error messages.  Once I re-entered, I determined which ships were hit and for how much damage, and which damage got applied before Aurora entered the loop.  I then manually applied the remaining damage.  I figured that this might be an acceptable stop-gap measure, but I wasn't sure that this would be workable in the long-term.  Whether or not it is workable in the long-term now appears moot.  

When I opened the system map, I noticed that neither race now showed any alien contacts at the battle-site, only their own units and missiles.  The Battle Control window also showed no target contacts what-so-ever.  I figured that this would clear up after I advanced the time, but it was troubling, especially if this was going to happen every time a ship was destroyed.  

At this point I advanced the time by five seconds, and Aurora immediately went into an endless error loop over allocating damage, as before, even though I was pretty sure that all weapons were recharging.  I exited out and restarted, and checked the error log, and it appeared that Aurora was trying to apply the damage that happened during the last time advance, after the endless loop.  To check this, I restored the database from the backup, ordered all ships to cease fire, and advanced the time again.  Yet again Aurora went into an endless damage allocation loop, and checking the event log verified that it was allocating damage that occured during the previous turn, but which went unallocated because of the loop.  

It appears that Aurora keeps a tally of damage that needs to be allocated, so even though it fails out on one, all of the other damage is still listed someplace, waiting for a new time advance so that it can be allocated.  

At this point this appears to be a fatal error for the campaign.  While manual damage allocation is possible, it won't work if Aurora keeps trying to allocate damage from past turns.  I don't see any way to rectify this, short of some sort of fix, but that doesn't appear to be possible.  

Kurt
« Last Edit: December 31, 1969, 06:00:00 PM by Kurt »
 

Offline SteveAlt

  • Global Moderator
  • Rear Admiral
  • *****
  • Posts: 820
  • Thanked: 8 times
(No subject)
« Reply #48 on: December 28, 2007, 08:20:24 AM »
Quote from: "Kurt"
It appears that Aurora keeps a tally of damage that needs to be allocated, so even though it fails out on one, all of the other damage is still listed someplace, waiting for a new time advance so that it can be allocated.  
That's correct. This is because all firing is simultaneous so I can't apply damage until all ships have fired. The damage records are in tthe FireResult table so I guess you could delete them before applying damage manually but that is hardly ideal.

Quote
At this point this appears to be a fatal error for the campaign.  While manual damage allocation is possible, it won't work if Aurora keeps trying to allocate damage from past turns.  I don't see any way to rectify this, short of some sort of fix, but that doesn't appear to be possible.  

As a final test I have downloaded Aurora from this site on to a clean computer and destroyed a ship. I got the same error so it is definitely a problem with the program and not your PC. I can't believe I released it with so fundamental an error or that its taken so long to encounter it. I was playing a campaigns while coding it so it must have been something I introduced between the last time a ship was destroyed and the release of the code. It also must have been something not directly related to ship destruction or I would have tested it. I also started a new campaign after v2.41 was released because I made some significant code changes for v2.5 so I didn't destroy any ships in the short term and before I did destroy a ship I obviously fixed whatever the problem was without realising it.

I'll try and get v2.5 released as soon as I can, probably early in the new year. On the bright side, there is a lot of really cool stuff in v2.5 :)

Steve
« Last Edit: December 31, 1969, 06:00:00 PM by SteveAlt »
 

Offline Kurt (OP)

  • Gold Supporter
  • Vice Admiral
  • *****
  • Posts: 1766
  • Thanked: 3389 times
  • 2021 Supporter 2021 Supporter : Donate for 2021
    Gold Supporter Gold Supporter : Support the forums with a Gold subscription
    2022 Supporter 2022 Supporter : Donate for 2022
    2023 Supporter 2023 Supporter : Donate for 2023
(No subject)
« Reply #49 on: December 28, 2007, 09:13:35 AM »
Quote from: "SteveAlt"
Quote from: "Kurt"
It appears that Aurora keeps a tally of damage that needs to be allocated, so even though it fails out on one, all of the other damage is still listed someplace, waiting for a new time advance so that it can be allocated.  
That's correct. This is because all firing is simultaneous so I can't apply damage until all ships have fired. The damage records are in tthe FireResult table so I guess you could delete them before applying damage manually but that is hardly ideal.

Quote
At this point this appears to be a fatal error for the campaign.  While manual damage allocation is possible, it won't work if Aurora keeps trying to allocate damage from past turns.  I don't see any way to rectify this, short of some sort of fix, but that doesn't appear to be possible.  
As a final test I have downloaded Aurora from this site on to a clean computer and destroyed a ship. I got the same error so it is definitely a problem with the program and not your PC. I can't believe I released it with so fundamental an error or that its taken so long to encounter it. I was playing a campaigns while coding it so it must have been something I introduced between the last time a ship was destroyed and the release of the code. It also must have been something not directly related to ship destruction or I would have tested it. I also started a new campaign after v2.41 was released because I made some significant code changes for v2.5 so I didn't destroy any ships in the short term and before I did destroy a ship I obviously fixed whatever the problem was without realising it.

I'll try and get v2.5 released as soon as I can, probably early in the new year. On the bright side, there is a lot of really cool stuff in v2.5 :)

Steve


That is pretty much what I thought.  I have been thinking about this since last night, and I think I can recreate the Quad-System campaign in a fresh 2.5 version of Aurora without too much trouble.  After all, by the year 42, where the first battle occurred and Aurora freaked out, New Iowa had only explored three systems.  All I really need to do is create a new quaternary system with one or more habitable planets, and several that can be terraformed, and go from there.  I'll need to "modify" what I've already posted to conform to the new reality, but that shouldn't be too difficult.  

Of course, I'll need 2.5 to do all of that.  No pressure <G>.

Kurt
« Last Edit: December 31, 1969, 06:00:00 PM by Kurt »
 

Offline Charlie Beeler

  • Registered
  • Vice Admiral
  • **********
  • Posts: 1381
  • Thanked: 3 times
(No subject)
« Reply #50 on: December 28, 2007, 09:13:42 AM »
I did make a vague reference in Nov 11 in the 2.40 bug list

Quote
Posted: Sun Nov 11, 2007 9:21 am    Post subject:    

--------------------------------------------------------------------------------
 
Has anyone been having problems with movement orders and damage allocation after a ship is destroyed in combat?

I've been doing some quick and dirty games to play with beam v missile and beam v fighter/missile combat. So my problems may be of my own making.


When no one responded, I just figured it was something I was doing wrong and it would shake out as I discovered my many functional mistakes.  

I'll endeavor to be more detailed in reporting issues I'm having moving forward.

Charlie
« Last Edit: December 31, 1969, 06:00:00 PM by Charlie Beeler »
Amateurs study tactics, Professionals study logistics - paraphrase attributed to Gen Omar Bradley
 

Offline SteveAlt

  • Global Moderator
  • Rear Admiral
  • *****
  • Posts: 820
  • Thanked: 8 times
(No subject)
« Reply #51 on: December 29, 2007, 07:19:13 PM »
Quote from: "Charlie Beeler"
When no one responded, I just figured it was something I was doing wrong and it would shake out as I discovered my many functional mistakes.  

I'll endeavor to be more detailed in reporting issues I'm having moving forward.

I still haven't got through all the bugs reports for v2.41 so I would have got to them eventually :)

More detail always helps though as I tend to tackle the bug reports with the most information first as they are usually the easiest to replicate and track down. Error numbers and messages are very useful, as is exactly what you were doing when the error occured.

Steve
« Last Edit: December 31, 1969, 06:00:00 PM by SteveAlt »
 

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
(No subject)
« Reply #52 on: December 31, 2007, 09:44:22 AM »
Hi Steve,

  This one's a bit witchy, so you might have trouble tracking it down.

  Every now and then, my construction queue gets messed up - it's not building the thing it says it's building.  In the most recent example, it's supposedly 80% done building a mine, but during the update cycle it's eating tritanium and when I look at the installations in SM mode, I see 80% of an ordnance factory.  My recollection is that if I delete out the corrupted job(s) and re-queue them, the problem goes away.  The only two clues I can add are that in the most recent case I had just finished building a spaceport and that I think I've seen it last through more than one completion of a queued item (automated mines) in another case.  Could spaceports be having some sort of long-term corruption effect?

  If you can't track the problem down, you might see if the same query is being used during the update cycle to decide what to build and during display of the construction queue - that way the corruption would be obvious when it happens.

Thanks,
John
« Last Edit: December 31, 1969, 06:00:00 PM by sloanjh »
 

Offline Charlie Beeler

  • Registered
  • Vice Admiral
  • **********
  • Posts: 1381
  • Thanked: 3 times
Cross posted from 2.4 bugs
« Reply #53 on: December 31, 2007, 09:59:09 AM »
Quote
Posted: Mon Nov 19, 2007 2:34 pm    Post subject:    

--------------------------------------------------------------------------------
 
1)
Error in MoveFleets
Error 3265 was generated by DAO.fields
Item not found in collection

3 times per time advance for each fighter group in flight.

2) Fighters will launch missiles while on board ship if ceasefire isn't initiated.

3) If I don't have sub-pulses activated, slower ships will over take faster ships even though the faster ships have been given greater standoff ranges. (ie 1800kps with min of 50 will over take a 3000kps ship with min of 1000)
« Last Edit: December 31, 1969, 06:00:00 PM by Charlie Beeler »
Amateurs study tactics, Professionals study logistics - paraphrase attributed to Gen Omar Bradley
 

Offline Erik L

  • Administrator
  • Admiral of the Fleet
  • *****
  • Posts: 5658
  • Thanked: 374 times
  • Forum Admin
  • Discord Username: icehawke
  • 2020 Supporter 2020 Supporter : Donate for 2020
    2022 Supporter 2022 Supporter : Donate for 2022
    Gold Supporter Gold Supporter : Support the forums with a Gold subscription
    2021 Supporter 2021 Supporter : Donate for 2021
(No subject)
« Reply #54 on: December 31, 2007, 10:27:06 AM »
Quote from: "SteveAlt"
Quote from: "Kurt"
It appears that Aurora keeps a tally of damage that needs to be allocated, so even though it fails out on one, all of the other damage is still listed someplace, waiting for a new time advance so that it can be allocated.  
That's correct. This is because all firing is simultaneous so I can't apply damage until all ships have fired. The damage records are in tthe FireResult table so I guess you could delete them before applying damage manually but that is hardly ideal.

Quote
At this point this appears to be a fatal error for the campaign.  While manual damage allocation is possible, it won't work if Aurora keeps trying to allocate damage from past turns.  I don't see any way to rectify this, short of some sort of fix, but that doesn't appear to be possible.  
As a final test I have downloaded Aurora from this site on to a clean computer and destroyed a ship. I got the same error so it is definitely a problem with the program and not your PC. I can't believe I released it with so fundamental an error or that its taken so long to encounter it. I was playing a campaigns while coding it so it must have been something I introduced between the last time a ship was destroyed and the release of the code. It also must have been something not directly related to ship destruction or I would have tested it. I also started a new campaign after v2.41 was released because I made some significant code changes for v2.5 so I didn't destroy any ships in the short term and before I did destroy a ship I obviously fixed whatever the problem was without realising it.

I'll try and get v2.5 released as soon as I can, probably early in the new year. On the bright side, there is a lot of really cool stuff in v2.5 :)

Steve


Just so you know, this affects the db as a whole. A new game started on the same db/version will be affected also...
« Last Edit: December 31, 1969, 06:00:00 PM by Erik Luken »
 

Offline Erik L

  • Administrator
  • Admiral of the Fleet
  • *****
  • Posts: 5658
  • Thanked: 374 times
  • Forum Admin
  • Discord Username: icehawke
  • 2020 Supporter 2020 Supporter : Donate for 2020
    2022 Supporter 2022 Supporter : Donate for 2022
    Gold Supporter Gold Supporter : Support the forums with a Gold subscription
    2021 Supporter 2021 Supporter : Donate for 2021
(No subject)
« Reply #55 on: December 31, 2007, 11:51:00 AM »
Quote from: "Erik Luken"
Just so you know, this affects the db as a whole. A new game started on the same db/version will be affected also...


Just for grins, I made a backup of the database, and then deleted all games other than the Remnant game. I then incremented 5 seconds and got some combat spam on 6th Jan 3000 for the previous game (id'd by ship names). These showed up only on SM View and got listed under the SM Race.

I followed this with a couple more time increments, and no errors.

So if you don't have combat, this will work to get things moving.
« Last Edit: December 31, 1969, 06:00:00 PM by Erik Luken »
 

Offline Father Tim

  • Vice Admiral
  • **********
  • Posts: 2162
  • Thanked: 531 times
Re: Cross posted from 2.4 bugs
« Reply #56 on: December 31, 2007, 04:35:34 PM »
Quote from: "Charlie Beeler"
3) If I don't have sub-pulses activated, slower ships will over take faster ships even though the faster ships have been given greater standoff ranges. (ie 1800kps with min of 50 will over take a 3000kps ship with min of 1000)


I expect the problem is arising from commander initiative.  That is, the faster ships are moving first to their longer stand-off range, and then the slower ships are closing up the rest of the distance.

The solution, as you have noted, is to increment time by smaller amounts.  Keep in mind, however, that the best you can ever do against a slower opponent with superior initiative is (your chosen range + 5*his speed in the direction of his choice).  If you're trying for 300,000 km for missile launches i'ts not a big deal, but if you're trying for point-blank range for your energy weapons, you could easily end up 15,000 km away in the wrong direction.  Curse those wily Federats and their Picard Maneuver!
« Last Edit: December 31, 1969, 06:00:00 PM by Father Tim »
 

Offline Charlie Beeler

  • Registered
  • Vice Admiral
  • **********
  • Posts: 1381
  • Thanked: 3 times
(No subject)
« Reply #57 on: January 01, 2008, 03:46:55 PM »
I've long since killed the game that the example came from.  But,  the intent was to have a carrier standoff around 1m km and launch fighters.  Since the carrier's speed is 3000kps and the targets is 1800kps this should be doeable.  I'd turned off the sub-pulses do the the movement errors I'm getting when fighters move.  I was more than a little surprised to find (think after 5 minutes) that the slower ship was something like 15k away from the carrier and pounding it into scrap.  Oh and the fighters speed was something like 12k kps and about 500k km away from the fight.

In the interim, I've gone to way points, micro-managing the speeds and living with thumping enter (a lot) to keep fast carriers at loyter ranges while fighters strikes overrun targets.
« Last Edit: December 31, 1969, 06:00:00 PM by Charlie Beeler »
Amateurs study tactics, Professionals study logistics - paraphrase attributed to Gen Omar Bradley
 

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
(No subject)
« Reply #58 on: January 02, 2008, 12:15:20 AM »
Not sure if this is a bug or feature, but it looks like trade convoys aren't taking distance into account when calculating payoff.  I've been running a 1-jump convoy and getting 10% of the smaller population's wealth as a payoff.  When I switched to a 3-jump convoy, it was still just 10% (rather than the 30% I expected).

John
« Last Edit: December 31, 1969, 06:00:00 PM by sloanjh »
 

Offline Kurt (OP)

  • Gold Supporter
  • Vice Admiral
  • *****
  • Posts: 1766
  • Thanked: 3389 times
  • 2021 Supporter 2021 Supporter : Donate for 2021
    Gold Supporter Gold Supporter : Support the forums with a Gold subscription
    2022 Supporter 2022 Supporter : Donate for 2022
    2023 Supporter 2023 Supporter : Donate for 2023
(No subject)
« Reply #59 on: January 09, 2008, 08:09:31 PM »
Steve -

I posted this before, but it disappeared.  Not to rush you, but do you have an ETA for 2.5?  

Kurt
« Last Edit: December 31, 1969, 06:00:00 PM by Kurt »