Slitherine

If you in 1.10 save and reload a scenario made with 1.05, strings for citys .. are corrupted after reloading.

orlinos found out the reason
viewtopic.php?f=121&t=36329&p=342075#p342184



Hallo!

I tested a 1.05 scenario from a campaign in 1.10.
The order of strings in the scenario.pzloc will destroyed in 1.10. after saving the game and reloading.

Example:
-------------------------- 1.05-----------1.10
IDS_STRING16 Krasnyy Dorogobuzh
IDS_STRING17 Yartsevo Demidov
IDS_STRING18 Elnya Dukhovshchina
IDS_STRING19 Pochinck Talashkino
IDS_STRING20 Smolensk Weiher
IDS_STRING21 Safonovo Dnjepr
IDS_STRING22 Dorogobuzh Dorogobuzh
IDS_STRING23 Demidov Demidov
IDS_STRING24 Dukhovshchina Dukhovshchina
IDS_STRING25 Talashkino Talashkino
IDS_STRING26 Weiher Weiher
IDS_STRING27 Dnjepr Dnjepr

could it be that the strings foer 16 to 21 in 1.10 used for the new terrain types. New terrain types: 6, changed strings: 6
Should I add the 6 first strings after string 27?
Or will Slitherine fix this bug?

wrong information - deleted

deleted

It seems the same issue scrambles the german localization installed by PzC 1.10.

I did not have time yet to analyze this issue in detail, but it will definitely be fixed.

deleted

What have I done?

I started 1.10 and cheated to the scenario.
After that, I started 1.05 and load a savegame.
I compared the units and cheated the new unit with exp and kills from the 1.05 savegame, because I would not start at the beginnig of the campaign.
I saved the new 1.10 situation.

After loading this savegame, the strings are corrupted.
Dont use two different Versions of PC simultaniusly!

Generally, all saves and mod campaigns are stored in the same location - My Games\Panzer Corps\Mods (or Scenario) - so a 1.05 and a 1.10 version will access the same location. I'm not sure what changes each does to the pzloc files in which the strings are stored, though.

I had the same problem and I think I also had it during my last 1.05 playthrough (so maybe it's not patch related). I couldn't find Arras on the map (the name, the city itself stood there, of course, but had a wrong name). I started the whole Grand Campaign yet again after installing 1.10 and found the names mixed up.

I checked the previous saves from current playthrough - all scenario maps had incorrect city names - I just didn't notice it. Here is an example of a screenshot (from Eben-Emael).



I just finished playing Arras scenario and the city names on the next map, Calais, were all wrong. I reloaded the save with the last turn from Arras (a save with wrong names), finished the scenario - and this time names in Calais are correct. I had to resort to a save to achieve it, just reloading the scenario won't work.

I only play the English version of the game but I did switch the languages a few times, just to check how they look like.

EDIT: I do keep a backup of the PzC 1.05 in another folder. I run it a few times after installing 1.10 (to compare the differences). So maybe this was the thing that did the damage (I do not know, if PzC reads the game data from the subfolders the executable is in, or from the game folder recognized in the registry).

More news from the name string bug front.

1. Advancing to a new scenario in the Grand Campaign (by a cheat or from a save, even a save with wrong names) - the name strings are correct.

2. Saving and reloading the scenario - corrupts the names. Advancing to the next scenario - again means proper strings in the next scenario - until saved and reloaded. So I guess the problem is created during saving or loading the game?

3. I tried my saves from the 1.05 playthrough in a backup copy of Pzc 1.05. No name string errors. I loaded the saves into 1.10, saved the game and reloaded - names became corrupted.

fsx wrote:
If you in 1.10 save and reload a scenario made with 1.05, strings for citys .. are corrupted after reloading.
[...]

could it be that the strings foer 16 to 21 in 1.10 used for the new terrain types. New terrain types: 6, changed strings: 6
Should I add the 6 first strings after string 27?
Or will Slitherine fix this bug?

I checked the .pzloc of of backup versions of the DLC's. Some of them do feature (I think) new terrain in string numbers you mention, for example Arras:
The Afrika Corps scenarios in the main data folder (that were installed with the update but are yet unplayable) have the same sequence. The vanilla main campaign pzloc files were also updated accordingly (I checked against my backup).

With DLC's it seems strange - most of the '40 scenarios have those new strings, other DLC's do not - they start straight with city names. But I think both "types" of scenarios get the name corruption problem when reloaded.

I checked if I could "borrow" some file from my 1.05 backup - but the problem stems from the 1.10 exe. I copied the 1.05 executable into the main folder (1.10) and and the mistake disappeared. Of course, the patch improvements did disappear, too.

I tried to "transform" a 1.05 scenario to 1.10.

Loaded the 1.05 scenario in the 1.10 scenario-editor all strings are correct. Looking at the strings-list in the scenario editor: the new terrain strings added automatically.
Saved. If I started the scenario (not from a savegame), the strings are corrupted (like desribed with an offset of 6). Saved, reloaded the savegame - now again the strings changed. An other wrong string for the city I observed. (Smolensk -> Weiher -> Elnya)

Configuation:
* DMP-Mod with "old" terrain- and tile-files (without the new terrains in terrain.pzloc and without new terrain in the layers and in the terrain.png)
* scenario version 1.05
* german version

strings in pzloc (saved with editor-version 1.10):
Compare with my first post!

IDS_STRING16 Desert
IDS_STRING17 Dunes
IDS_STRING18 Rough Desert
IDS_STRING19 Escarpment
IDS_STRING20 Strait
IDS_STRING21 Jungle
IDS_STRING22 Krasnyy
IDS_STRING23 Yartsevo
IDS_STRING24 Elnya
IDS_STRING25 Pochinck
IDS_STRING26 Smolensk
IDS_STRING27 Safonovo
IDS_STRING28 Dorogobuzh
IDS_STRING29 Demidov
IDS_STRING30 Dukhovshchina
IDS_STRING31 Talashkino
IDS_STRING32 Weiher
IDS_STRING33 Dnjepr

AND: If I load the scenario again in the 1.10 editor, the result is:

IDS_STRING16 Desert
IDS_STRING17 Dunes
IDS_STRING18 Rough Desert
IDS_STRING19 Escarpment
IDS_STRING20 Strait
IDS_STRING21 Jungle
IDS_STRING22 Rough Desert
IDS_STRING23 Escarpment
IDS_STRING24 Strait
IDS_STRING25 Jungle
IDS_STRING26 Krasnyy
IDS_STRING27 Yartsevo
IDS_STRING28 Elnya
IDS_STRING29 Pochinck
IDS_STRING30 Smolensk
IDS_STRING31 Safonovo
IDS_STRING32 Dorogobuzh
IDS_STRING33 Demidov
IDS_STRING34 Dukhovshchina
IDS_STRING35 Talashkino
IDS_STRING36 Weiher
IDS_STRING37 Dnjepr

The 4 red lines move the city-name again! If is the sane number as the offset from the post below (weiher -> Elnya)

Perhaps, I found a first solution.

Configuration:
DMP-Mod with EXE-Files from 1.10 + hero1,2,3.png
Scenario version 1.05

I copied the new terrains.pzdat from original 1.10.
I added the 6 new strings into the pzloc file of the scenario manually.
If I reached the scenario first time, all cities are okay.
Save, reload -> the city I observe is okay.

And now for the "update" of a 1.05 scenario to 1.10.

Same configuration.

Delete the 4 twice included scenario-strings in the editor, and save.
Starting scenario ok, save & reload okay.

BUT..
The pzloc is in german and the string in the scenario (single scenario) are in english.
BUT...
This scenario included in a campaign has an english (terrain-strings only) pzloc but I see german terrain-descriptions.

In the old terrain.pzloc are used ID for the name, in the new not.

old:
IDS_TRN_SHALLOW_SEA 100 0

new:
Shallow Sea 100 0

fsx wrote:And now for the "update" of a 1.05 scenario to 1.10.

Same configuration.

Delete the 4 twice included scenario-strings in the editor, and save.
Starting scenario ok, save & reload okay.

BUT..
The pzloc is in german and the string in the scenario (single scenario) are in english.
BUT...
This scenario included in a campaign has an english (terrain-strings only) pzloc but I see german terrain-descriptions.

Could it be, that the 4 additional strings only appear in german version of the scenario editor?
I now used for update an english 1.10 scenario editor with the new terrain-file. And I tested the result as a stand-alone-scenario.

I hope you join us in the next beta testing.

VPaulus wrote:I hope you join us in the next beta testing.

I also thought about it...

But you have to know:
1. It will be a part time job for me (because DMP-Mods)
2. I woud say, I am not "playing" the scenarios. If I test a scenario -for example- I dont use FoW, looking for possible problems, that triggers works...
Many other testers can better say if balance is right or not and play much better than me. Sorry, but I am a nitpicker (Erbenszähler).
About the tests for the string-issue: If I saw the string issue, I remembered, at my try to add a terrain in 1.05. In the editor it was possible, but I had to add the new terrain string.
Thats the reason, because I try to test with(out) the terrain-file.

These followed issues are not bugs. My feedback for the first AK-scenarios:
* many new features are used, but perhaps less could be more. Perhaps AK should introduce the features step by step, not so massive (messages, changed objectives MORE than once)
* I remember that an airfield had a city-name without "airfield". It is not so easy to see this hex as an airfield, if an (big) unit on it.
* In Brevity some fighters arrives from nowhere. It coud be more "logical"?, if new enemy units arrives - they should appear at the start of the own turn and a message box opens (message box is in AK the normal way, but not in this scenario - if I remember right). Why dont use in Brevity the new feature to give each fighters a target zone and place these units "normal" at the scenario?

In DLC 40, scenario 2 (Albert Canal) i found 2 wrong town names in the south eastern part of the map

a) one town named "Scheldt River" (no river nearby)

b) one town called "Albert Canal Bridge" ( canal is a good distance up north)

Don't know if that's related to patch 1.10 or if it was wrong from the start (i started DLC40 before patching to v1.10)

PS: There's a town on the map called Wsterlo, should have been Westerlo instead, right ?

PPS: I opened the map in the scenario editor and actually not only 2 but mostly all town names are mixed around.