UE Map Editor

[Flocke]

Even if the EDO are part of a player empire and not AI, there still might be a connection with one-per-empire structures and your wonder build order type 8?

I have no clue, but it doesn't matter for when I remove a system. I just update the system id of all the wonder build orders present, and remove those with a matching id.

[Flocke]

2. did adding ships on different resolution get put on back burner?

While refactoring the code, I found a Bug with one Talarian ship image in Vanilla 1366x768

incompatible_talarian.jpg (11.33 KiB) Viewed 3364 times

I'm still working on it, so you have to wait one more week I guess.

[thunderchero]

2. did adding ships on different resolution get put on back burner?

While refactoring the code, I found a Bug with one Talarian ship image in Vanilla 1366x768


I'm still working on it, so you have to wait one more week I guess.

no problem with recent report of other files needed for UM5 waiting 1 more week or more would be a good idea

as for Talarian ship image i_m10170.tga will be fixed in update also
it took me a few to realize issue was the "improved" build images option not the original image.

just on a side note integrity check also flag that issue on that image, as well as 3 other vanilla image wrong size
Info: in edifice.bst: Image size 270x226 of Cun_b.tga deviates from default 270x225 pixels.
Info: in edifice.bst: Image size 270x226 of Fun_b.tga deviates from default 270x225 pixels.
Info: in race.rst: Image size 159x170 of antidean.tga deviates from default 160x170 pixels.

UM5 latest issue make me ask about different resolutions for structures?
should UE also check this during integrity check?

[Flocke]

but will note in "Add and remove ship models"
1. MUM uses the "TMP" slot for almost 50% of it's models. would there be any way to read shiplist (prefix of ship) instead of phaser slots? example if no dedicated slot use TMP slot for phaser data.

I didn't want to, but with the broken export reported by Benman2785 in viewtopic.php?p=59034#p59034 I had to revisit and investigate the complete ship model import / export.

And while I cleaned up the code and fixed a bunch of issues, I couldn't stop myself but also implemented the requested TMP slot support.
To get this work, I implemented a new ship model search routine, that both searches for available model HOBs and in addition also for TGA files using some pattern matching. I'm amazed how many unused hidden ships there are included with mods.

2. did adding ships on different resolution get put on back burner?

Ah well, I already mentioned it somewhere, but to give some more details, by default now all the ship images are detected by the WDF image name selectores stored in trek.exe I identified for each of the views.
Given another request, I however also implemented to optionally search and import matching images by a pattern matching algorithm. That one checks used image prefixes "i_", file name length (max 12), model name length (max 3) and non-zero leading number suffix if not "w".

Given that many mods sadly use numbers for the model names, there sadly are some potential false positives or even conflicts. E.g. ship model "m" and "m1" could both be referencing texture "i_m160.tga". Which one is correct, on bad luck both are.
Most of the time this however won't be much of an issue, you just get some redundant image copies.

UM5 latest issue make me ask about different resolutions for structures?
should UE also check this during integrity check?

I had to check on it again, but I implemented this one as well. All the edifice.bst structures are checked for missing images and wrong image sizes, with requirements loaded from WDF files and image name patterns read from trek.exe, similar to the ship models.

[Spocks-cuddly-tribble]

0x00-0x03 resultId:     incremented unique result number, starting at different offsets for the races:
                          card: 01 00 00 00 (little endian) = 1
                          fed:  AA AA AA 2A (little endian) = max_uint * 1/6
                          ferg: 55 55 55 55 (little endian) = max_uint * 2/6
                          klng: FF FF FF 7F (little endian) = max_uint * 3/6
                          rom:  AA AA AA AA (little endian) = max_uint * 4/6
                          max_uint (unsigned) = 0xFFFFFFFF = 4294967295 decimal
                          The klingon 0x8000004D here ofc flipped all F when it reached 0x80000000.

Hi Flocke, could the klingon value cause bugs with the result.list? viewtopic.php?p=60642#p60642

And if so, can we change the defaults in loc_459410 result_lst_defaults to fix this (same for ordInfo loc_457E00) ?

They use 'lea new, [old+1]' for the next number, but a simple 'inc' for the global count.

[Flocke]

Hi Flocke, could the klingon value cause bugs with the result.list? viewtopic.php?p=60642#p60642

No, already the first increment switches the F digits, so if it caused issues, it already should happen very early and after the jump over possibly stop.

The default value should only be used for initialization. Furthermore the number is only used as an identifier, and therefore just needs to be unique.
If there was any issue, it would equally affect any result message and be much more noticable. Once it goes out of sync, any event of same race should be affected.

[Spocks-cuddly-tribble]

Thanks, I see, I misread the redundant comment about flipping all F -> 0x7FFFFFFF + 1 = 0x80000000

[Spocks-cuddly-tribble]

Hi Flocke, I hope you're doing well. Some additions/corrections:

--- treaty file ---
0x30-0x33: treatyType (1:peace 2:alliance 3:war pact 4:accepted demand 5:friendship 6:affiliation 7:membership 8:gift)
0x70-0x71: targetRaceId (for war pact, else -1)

treatyType 9 = Bribe aka request to minor race to terminate treaty with empire targetRaceId X (can be combined with credit offer)
treatyType 4 is for every demand (credits & contested territory claims), not only accepted ones

AI_Diplomacy.png (44.77 KiB) Viewed 2847 times

--- GTForceList/GWTForce ---
ability mask 0x0008 (08 00) = ?? never seen used

sub_4485D8 SetTaskForceAbilities -> see IDA: https://workupload.com/file/4yVDdUwcf7T

Indicates at least one ship has zero hull points in GShipList (possible due to vanilla bugs e.g. orbital batteries and neutron stars).
For an unused feature I guess (TF features crippled, repair, forced scrap?)

Also @ +50 -> greatest colony ship production value, so your 'contains colony ships' = 1 doesn't work for patches (but greater null will).
+44 = colony ship ID with greatest production value (vanilla different values possible due to missed upgrade bug) Both values should be unused by default.

Using the info I implemented a workaround for a TF NULL crash problem due an unknown AI colony ship task issue in one of my patches.

00448D7E     8078 50 01     CMP BYTE PTR DS:[EAX+50],1  // greatest colony ship production value
00448D82     7D 5B          JGE SHORT 448DDF     //  contains colony ships

It won't relocate stranded fleets containing colony ships anymore, but better than sporadic crashes or no patch: viewtopic.php?p=57729#p57729


More than that I fixed an annoying bug with AI fleets keeping old/bad stats after re-deploying. Limping with speed 1 (and/or short range) to the next staging point after some tasks can cause massive delays between the AI fleet tasks (worse in larger maps).

Example for the vanilla AI Fleets glitch (note wrong speed 1, TT with scout scan value and scout with med range):

AI_Fleets_Glitch.png (52.59 KiB) Viewed 2952 times

► Show Spoiler

NAME: AI Fleets with wrong Speed/Range after redeploy (Fix)

>> 0x000086ce 8b 40 4e ba 01 00 00 00 c1 f8 10 e8 42 05 04 00 8b 50 04 89 c1
>> 0x00008727 15 03 04
>> 0x0000872c 44 24 fe ba 01
>> 0x00008732 00 00 c1 f8 10
>> 0x0000c4db 8d 94 24 20 01 00 00
>> 0x0000c51d 84 24 20 01 00 00
>> 0x0000c525 0f bf 16 e8 73 af 03 00 8d 84 24 20 01 00 00

<< 0x000086ce 0f b7 40 50 42 e8 48 05 04 00 8b c8 e8 f9 f2 03 00 8b 51 04 90
<< 0x00008727 ad f2 03
<< 0x0000872c c4 e8 0e 03 04
<< 0x00008732 0f b7 04 24 42
<< 0x0000c4db 8b d0 8b d8 90 90 90
<< 0x0000c51d 03 0f bf 16 90 90
<< 0x0000c525 e8 76 af 03 00 8b c1 e8 a7 b4 03 00 8b c3 90

004092CE     0FB740 50        MOVZX EAX,WORD [EAX+50]  // fleet ID from GShipList
004092D2     42               INC EDX
004092D3     E8 48050400      CALL 449820     // GetTaskForceEntry
004092D8     8BC8             MOV ECX,EAX
004092DA     E8 F9F20300      CALL 4485D8     // update TaskForceAbilities
004092DF     8B51 04          MOV EDX,[ECX+4]  // ship count
004092E2     90               NOP

00409326     E8 ADF20300      CALL 4485D8   // update TaskForceAbilities
0040932B     8BC4             MOV EAX,ESP
0040932D     E8 0E030400      CALL 449640   // AI_add_to_TF_sector_list_3C
00409332     0FB70424         MOVZX EAX,WORD [ESP]
00409336     42               INC EDX


0040D0DB     8BD0             MOV EDX,EAX
0040D0DD     8BD8             MOV EBX,EAX
0040D0DF     909090           NOP

0040D11B     66:8903          MOV [EBX],AX
0040D11E     0FBF16           MOVSX EDX,WORD [ESI]
0040D121     9090             NOP

0040D125     E8 76AF0300      CALL 4480A0     // AI_remove_shipID_EDX_fleet_EAX
0040D12A     8BC1             MOV EAX,ECX     // old fleet entry
0040D12C     E8 A7B40300      CALL 4485D8     // update TaskForceAbilities
0040D131     8BC3             MOV EAX,EBX
0040D133     90               NOP

Also upcoming:

- Disable AI Fleet Buildup Support Limit in Peacetime for Affluent AIs
-> Fed AI with 1.5+ Mio. credits but -3k income (without AI cheats) stopped after only 3 Defiants in one of my test games

► Show Spoiler

NAME: Disable AI Fleet Buildup Support Limit in Peacetime for Affluent AIs
DESC: AIs with more than 400,000 credits use the increased war NETCRALLOC limit even in peacetime.
AUTHOR: Spocks-cuddly-tribble
URL: 

>> 0x00006294 8b 6a 04 83 fd 07
>> 0x0000629b 85
>> 0x00006390 83 fd 06 0f 84 07 ff ff ff
>> 0x0000639d 84

<< 0x00006294 81 fd 80 1a 06 00
<< 0x0000629b 8c
<< 0x00006390 8b 6a 04 83 fd 06 7c 0a 90
<< 0x0000639d 8e


#00406E94     81FD 801A0600     CMP EBP,61A80  // 400,000
#00406E9A     0F8C F0000000     JL 406F90      // check war agendas

#00406F90     8B6A 04           MOV EBP,[EDX+4]  // AI agenda
#00406F93     83FD 06           CMP EBP,6
#00406F96     7C 0A             JL SHORT 406FA2  // not war
#00406F98     90                NOP

#00406F9C    ^0F8E FEFEFFFF     JLE 406EA0       // is war

- AI Rejecting Alliance Fix to allow Domination team Victory -> Example for the issue: viewtopic.php?p=57051#p57051
-> this one was tricky, default requires 20+ 'Good' special events (my klingons have only 6 with the feds in below pic):

map_ai_data_types_E_F_12.png (11.58 KiB) Viewed 2952 times

► Show Spoiler

NAME: AI Rejecting Alliance Fix to allow Domination team Victory
DESC: 
AUTHOR: Spocks-cuddly-tribble
URL: 

>> 0x00030969 31 d2 88 fa 31 c9
>> 0x00030970 6c 49 fd ff 66 8b 4e 08 89 c5 8d 04 cd 00 00 00 00 01 c8 c1 e0 02 01 c8 c1 e0 02 89 c1 c1 e0 02 bf 98 1a 5b 00 29 c8 30 db 01 c7 31 c9 88 d9 89 f0 89 ca e8 38 49 fd ff 85 c0 75 2a fe c3 80 fb 05 72 e8 31 d2 88 fa 89 d0 e8 02 37 fd ff 85 c0 74 29 7c 0a 66 83 bd d6 01 00 00 03 7e 40
>> 0x000309d7 bc 8f 10 01 00 00 02
>> 0x000309e8 c3
>> 0x000309ec bf 06 e8 ad 43 fd ff 83 f8 07 7c d6 66 83 bd d4 01 00 00 14 7c cc 66 83 bd d6 01 00 00 03 7e 17

<< 0x00030969 e8 72 49 fd ff 8b
<< 0x00030970 b9 bc 01 00 00 0f b6 7e 08 0f af f9 81 c7 a8 1b 5b 00 32 db 33 d2 8a d3 8b c6 52 e8 50 49 fd ff 5a 85 c0 75 41 fe c3 80 fb 05 72 e8 80 7c 29 1a 03 7f 2b 8a d7 8b c2 e8 14 37 fd ff 85 c0 7c 1e 75 5c 66 81 7d 02 ee 02 7c 14 80 7c 29 18 14 7d 62 8b 77 a0 ba 48 1b 5b 00 0f b6 c7 eb 1d
<< 0x000309d7 3c 97 02 90 90 90 90
<< 0x000309e8 ac
<< 0x000309ec af c1 03 34 10 33 db 8d 04 8b 03 1c 10 2b c1 7d f9 8d 04 5b c1 e8 02 3b f0 7f 1c 90 90 90 90 90

► Show Spoiler

00431569     E8 7249FDFF       CALL 405EE0
0043156E     8BE8              MOV EBP,EAX
00431570     B9 BC010000       MOV ECX,1BC     // entry size empsInfo
00431575     0FB67E 08         MOVZX EDI,BYTE [ESI+8]
00431579     0FAFF9            IMUL EDI,ECX
0043157C     81C7 A81B5B00     ADD EDI,5B1BA8  // empsInfo 0x5B1A98+0x110=0x5B1BA8
00431582     32DB              XOR BL,BL
00431584     33D2              XOR EDX,EDX
00431586     8AD3              MOV DL,BL
00431588     8BC6              MOV EAX,ESI
0043158A     52                PUSH EDX
0043158B     E8 5049FDFF       CALL 405EE0
00431590     5A                POP EDX
00431591     85C0              TEST EAX,EAX
00431593     75 41             JNZ SHORT 4315D6
00431595     FEC3              INC BL
00431597     80FB 05           CMP BL,5
0043159A    ^72 E8             JB SHORT 431584
0043159C     807C29 1A 03      CMP BYTE [ECX+EBP+1A],3
004315A1     7F 2B             JG SHORT 4315CE
004315A3     8AD7              MOV DL,BH
004315A5     8BC2              MOV EAX,EDX
004315A7     E8 1437FDFF       CALL 404CC0
004315AC     85C0              TEST EAX,EAX
004315AE     7C 1E             JL SHORT 4315CE
004315B0     75 5C             JNZ SHORT 43160E
004315B2     66:817D 02 EE02   CMP WORD [EBP+2],2EE  // 750
004315B8     7C 14             JL SHORT 4315CE
004315BA     807C29 18 14      CMP BYTE [ECX+EBP+18],14
004315BF     7D 62             JGE SHORT 431623
004315C1     8B77 A0           MOV ESI,DWORD [EDI-60]  // 0x110-0x60=0xB0 // cumulated total population AI empire
004315C4     BA 481B5B00       MOV EDX, 5B1B48   // empsInfo 0x5B1A98+0xB0=0x5B1B48 // cumulated total population 
004315C9     0FB6C7            MOVZX EAX,BH
004315CC     EB 1D             JMP SHORT 4315EB

004315D6     833C97 02         CMP DWORD [EDI+EDX*4],2
004315DA     90909090          NOP

004315E7    ^74 AC             JE SHORT 431595

004315EB     0FAFC1            IMUL EAX,ECX
004315EE     033410            ADD ESI,DWORD [EAX+EDX]   // + cumulated total population player empire
004315F1     33DB              XOR EBX,EBX
004315F3     8D048B            LEA EAX,DWORD [EBX+ECX*4]
004315F6     031C10            ADD EBX,DWORD [EAX+EDX]  // cumulated total population of all empires
004315F9     2BC1              SUB EAX,ECX
004315FB    ^7D F9             JGE SHORT 4315F6
004315FD     8D045B            LEA EAX,DWORD [EBX+EBX*2]  // *3
00431600     C1E8 02           SHR EAX,2    // *3/4 = 75 %
00431603     3BF0              CMP ESI,EAX
00431605     7F 1C             JG SHORT 431623 // Alliance can trigger domination Victory
00431607     9090909090        NOP

[Flocke]

I totally missed to answer above posting. Been a stressful time when last year ended.

treatyType 9 = Bribe aka request to minor race to terminate treaty with empire targetRaceId X (can be combined with credit offer)
treatyType 4 is for every demand (credits & contested territory claims), not only accepted ones

updated

sub_4485D8 SetTaskForceAbilities -> see IDA: https://workupload.com/file/4yVDdUwcf7T

Indicates at least one ship has zero hull points in GShipList (possible due to vanilla bugs e.g. orbital batteries and neutron stars).
For an unused feature I guess (TF features crippled, repair, forced scrap?)

added

Also @ +50 -> greatest colony ship production value, so your 'contains colony ships' = 1 doesn't work for patches (but greater null will).

That is long fixed, what location did I miss?

+44 = colony ship ID with greatest production value (vanilla different values possible due to missed upgrade bug) Both values should be unused by default.

updated

AI Fleets with wrong Speed/Range after redeploy (Fix)

for UE I only accept valid and reproducable save uploads

Nice proggress on the patches and thx for the updates!

[Spocks-cuddly-tribble]

Also @ +50 -> greatest colony ship production value, so your 'contains colony ships' = 1 doesn't work for patches (but greater null will).

That is long fixed, what location did I miss?

Just from the wiki, if UE already sets 'greatest colony ship production value in the fleet' (not always 1 if CS), then you're fine.

[Flocke]

Just from the wiki, if UE already sets 'greatest colony ship production value in the fleet' (not always 1 if CS), then you're fine.

The wiki on the other hand I don't care on for now. There is too much waiting for update and given the little usage I rather move the info to AFC at some point, or I set up a new wiki..

I see, at some point I need to update it again. Maintaining a wiki for so little use however is not much reasonable.
For now I have better to do.

[Flocke]

with latest version of UE I attempted to switch to single player game and switch race and got crash on game load.
I duplicated this issue for testing, using latest download "modded_save_detection"

saved game feds (host) cards (join)
save 2 feds MP game save
save 3 sp feds (UE saved) no issues
save 4 sp fer (UE saved) crash turn 2 with error log
File: ..\..\source\game\ship.c, Line: 968, taskForce != NULL

MUM 1366 beta version 4 (not released yet)

That save games tempt to crash when you switch AI with player race I think is nothing new, given that the AI is not removed but keeps messing with the player race.

I however went into analysing this turn 11 game some little and managed to fix 2nd turn crash. Now it crashes when ending turn 53 instead with:
Ausnahmecode: 0xc0000005
Fehleroffset: 0x00005f60

But it is progress and with further analysis and testing might be possible to fix.

The cause to 2nd turn crash was that the Ferengi AI in some routine fails to lookup task force 00 01 00 00 with the two transports 41 00 and 42 00 that build a starbase in Ferenginar. When resetting the AIShpUnt task and TskSh60, the AI skips to mess with the task force and you get to turn 53.

resetAITask.jpg (163.46 KiB) Viewed 221 times

I added two new task force map menu options to either reset or completely remove AI tasks.
The cleanup options also allow to remove all ship AI tasks, but that only delayed above 0x00005f60 crash a few more turns.

Removing the task force instead, it crashes at turn 51 with same old taskForce != NULL error.
This however misses to remove AI task 3C 00 (tsksh60, aitask98, tsksy60.cnt, tsksy60, tsksh60.cnt) and update AgtTk1 for the removed AI task.
An optimization implemented for the player race in UE, so no surprise.

Having first reset AI ship tasks and then removed the task force, it now crashed ending turn 54. Again with:
Ausnahmecode: 0xc0000005
Fehleroffset: 0x00005f60

Think I will try some integrity check to remove all player AI tasks next. Including non task force related ones.

Edit: Wait, we have the map cleanup options. When I remove all AI ship tasks it fixes above turn 53 crash and I managed to play up to turn 74 before it crashed with same offset error again.
Same like turn 53 crash it can't be continued without removing the AI ahip tasks again. Next it crashed ending turn 81, and guess what, same error again.

Edit2: Ok, the second crash is caused either by the Romulans meeting the Terran empire, or by the Klingons meeting the Cardassians. This is a bug I can't fix in UE in reasonable time. I'd first have to figure how to add a whole AI agent back.
But I'll improve integrity checks and task force removal to at least address the taskForce != NULL crash.

[Spocks-cuddly-tribble]

Just tell them to use one of the two methods to switch the player empire: viewtopic.php?p=57851#p57851 (instead of UE)

[Flocke]

Just tell them to use one of the two methods to switch the player empire: viewtopic.php?p=57851#p57851 (instead of UE)

Yes will do.

Same issue however bugged me before when I tried to investigate some MP savegames. And I refuse to install any patches for this.
Given I already had AI cleanup routines in place for ship removal and stuff, I at least needed to check on what can be done.

[Spocks-cuddly-tribble]

It might help to look at the BotF player/AI takeover codes:

sub_43A200 player_takeover_empire_AL (ex-AI)
sub_43A160 AI_takeover_empire_AL (ex-player)

Related to our other conversation:

43A1D2 mov dword [eax+54h], 0 ; ?(mislabeled as numWayPoints) zero for all ex-player fleets now AI (ex-AI to player keep entry)