RFC Europe Files

[3Miro]

The March 10 version has been downloaded 7 times, of all the people that downloaded it, does it give you any trouble or is it just merijn_v1. I have to know if I am to help, are e talking about problem with the mod or problem with the installation.

 

[Lexad]

OK until the Burgundy and Papal scores disappeared from the corner of the screen

If they were the only states you know, should be just civ feature. Click your civ name in score table and get the full table.

 

[merijn_v1]

I've download it for the 3rd time and it still doesn't work. But when I unpack the 5th March version over it, it works fine. But I don't know if the new features are enabled then.

 

[Verily]

Agree with Lexad and merijn, the new test version is broken. When I start a new game (as Burgundy), an empty Tech Advisor screen pops up, then, once I press Enter, it goes away and I see the usual screen, but with no interface at all. Crtl-Alt-I retrieves the interface, but I have no interface when viewing cities, and the Tech Advisor screen continues to be blank. There are various other major bugs, too. Restoring to the March 5 version resolves all issues.

 

[3Miro]

And the damn thing works for me ....

OK, I need someone with the problem to do this:

-go to My Games\Beyond the Sword\ and open CivilizationIV.ini with NotePad or some other editor.
- scroll down and find where is says: LoggingEnabled = 0.
- change that to LoggingEnabled = 1.
- save the file (it will make the game produce more log files).
- start the Mar 10 mod and load the game to where you see the problem (open the tech screen and so on).
- exit the game.
- zip the Folder My Games\Beyond the Sword\Logs and post it.

In the mean time I will see what else I can possibly try.

(BTW unpacking one version over another is the wrong way to do it. Before unpacking any version, completely erase the folder of the previous one. Otherwise no one knows what bugs might appear).

 

[sedna17]

I downloaded a fresh copy of the March 10th file. It failed to launch for me, with an error on CvEventInterface. Since this file reads in RFCEBalance.py, my guess is the problem is there. I'll trying debugging it a little (since mine doesn't work it makes it much easier ). Anyhow, here are the logs.

 

[3Miro]

PythonErr gives only the failure, no line on which the error occurred.

Such erratic behavior could only be explained with a bad dll build. (That or a very bad XML file which msxml doesn't properly complain about, xml.log is very short). I will boot under windows tomorrow and rebuild the dll there. Will double check the XML files as well, I only changed the projects, the scheme and buildings text files.

 

[3Miro]

OK, sedna and the others that are having trouble with the latest build: try this .dll file, hopefully it would work. Tell me if it doesn't.

 

[jessiecat]

OK, sedna and the others that are having trouble with the latest build: try this .dll file, hopefully it would work. Tell me if it doesn't.

Where should I put it?

 

[3Miro]

Where should I put it?

It goes in RFCEurope\Assets.

Unless you have trouble with March 10 version, this file should not change anything. If everything works just fine for you, you don't even have to bother. This is a test for those that are having trouble.

 

[jessiecat]

It goes in RFCEurope\Assets.

Unless you have trouble with March 10 version, this file should not change anything. If everything works just fine for you, you don't even have to bother. This is a test for those that are having trouble.

I was having interface issues and crashes with the Mar.10 version. That's why I've gone back to the previous one (Mar.5) which works fine for me. See my report on a game up to 1760 in the playtesting thread. I think I'll wait for you guys to resolve the problem before I download the next version.

BTW Have you talked to st. lucifer lately? I'd like to do some terrian and resources fixes on the WB save,
if that's OK.

 

[3Miro]

I was having interface issues and crashes with the Mar.10 version. That's why I've gone back to the previous one (Mar.5) which works fine for me. See my report on a game up to 1760 in the playtesting thread. I think I'll wait for you guys to resolve the problem before I download the next version.

BTW Have you talked to st. lucifer lately? I'd like to do some terrian and resources fixes on the WB save,
if that's OK.

I think st. Lucifer is busy with something, coordinate whatever changes with Lexad, I think he is also changing the map.

Try the .dll file and tell me if it helps.

 

[Lexad]

I'm just thinking of tweaking naming pattern for Moscow and Kievan Rus, so feel free to go. Please, add some stone accessible for Muscowy - close to Urals or Karelia or Caucases, to make obtaining it less cheesy, but still feasible.

 

[merijn_v1]

Well, the civilclopedia still doesn't work proparly, when I added the new cvgamecoredll.

 

[sedna17]

@3Miro.

Your new DLL still crashes for me in exactly the same way, and the logs contained exactly the same information.

The difference between my result (no launch) and others (interface crashing) could be due to operating system. I recently upgraded to Vista, which closed Civ down for me saying it had crashed. Normally when Civ crashes on load I have to quit it myself from the Task Manager. This theory doesn't explain why the mod works on your system.

I conducted a few experiments:

1) Commented out tReligionSpreadFactor in Consts.py. This caused errors on load (in CvEventInterface et al) of course, but interestingly enough the game did still launch (I lost immediately upon starting a new game as Burgundy).

2) I did #1 and also commented out where gc.setReligionSpread is called in RFCEBalance.py. The game loaded and played properly, and I was unable to find any interface problems. I assume religion-spread would be funny, but I didn't test that.

Attempting to pursue this further, I tried to find setReligionSpread and make sure you were calling it with the correct arguments, but I couldn't find this function in the latest CvGameCoreDLL folder you included, which is odd. Where is this function defined?

EDIT: Ignore that last paragraph -- I found it. Lazy file search. Everything looks okay though, so...

 

[3Miro]

@3Miro.

Your new DLL still crashes for me in exactly the same way, and the logs contained exactly the same information.

The difference between my result (no launch) and others (interface crashing) could be due to operating system. I recently upgraded to Vista, which closed Civ down for me saying it had crashed. Normally when Civ crashes on load I have to quit it myself from the Task Manager. This theory doesn't explain why the mod works on your system.

I conducted a few experiments:

1) Commented out tReligionSpreadFactor in Consts.py. This caused errors on load (in CvEventInterface et al) of course, but interestingly enough the game did still launch (I lost immediately upon starting a new game as Burgundy).

2) I did #1 and also commented out where gc.setReligionSpread is called in RFCEBalance.py. The game loaded and played properly, and I was unable to find any interface problems. I assume religion-spread would be funny, but I didn't test that.

Attempting to pursue this further, I tried to find setReligionSpread and make sure you were calling it with the correct arguments, but I couldn't find this function in the latest CvGameCoreDLL folder you included, which is odd. Where is this function defined?

EDIT: Ignore that last paragraph -- I found it. Lazy file search. Everything looks okay though, so...

Thanks sedna, you have pinpointed the problem. The arguments that apssed to the function about religion spread are fine, it in itself does something funny. Now that I know what I am looking for, I should be able to fix the issue today and post a new .dll or complete version (if you are planning on sending a new version, I can just give you the .dll so that you can include it)

 

[sedna17]

Go ahead and post a complete new version. It'll be a bit more work before I have something worthwhile to release.

 

[3Miro]

Thanks to sedna and the others the bug has been found and successfully swatted. Expect the next version is couple of minutes.

I was trying to decrease the need for Python - XML synchronization that cause many problems in the past. I am also trying to make the mod completely Python/XML controllable (reduce the need to do C++). Thus I moved things like the map sizes and such from static to dynamic memory, meaning arrays sizes are declared on game start (out .dll is much smalled than Rhye's). To do the religion spread properly, I need to know how many religions there are, however, when I decide on the size of the array (in one of the functions called in RFCEBalance) XML has not been read yet. With slight change in Python and C++ that is fixed now.

On Archais OS and such:
sedna uses MacOSX and I use Linux, interestingly modern OSX (since 2001) and all of Linux versions share virtually the same Unix core (kernel). Unix was first build in the late 60's early 70's and DOS and Windows did not exist until mid to late 80's. However, Unix is vastly superior to Windows.

Having said that, I should address the possible issues for working on different systems. When I compile the c++ code, I am actually using a virtual machine (the program is called Virtual Box). I have a full copy of windows running within my Linux, windows just sees somewhat different hardware (it even sees the right CPU, just the Hard Drive and some other devices are different). You can see the screen-shot. One of the other devices that is different is the Video Card, as such I cannot play games with Virtual Box. To play Civilization I am using wine, which is a program that runs windows programs, but is not windows in itself. Wine emulates the API, so Civilization thinks it is working with windows and yet it is working with Linux and wine is the interpreter that lets them talk to each other. The bug should have appeared under Linux (wine), but it did not for some reason, which made it hard for me to capture.

Python and XML, however, are simply text files which are platform independent. This means that it makes no difference or whatsoever which system one uses to edit them. The only peculiarity might be in a __MacOSX or .something or something~ files appearing depending on the text editor/zip program that senda and I use. Those should make no difference for the game itself.

 

[BurnEmDown]

Thanks 3Miro, been waiting for a working, up-to-date version for some time now .

 

[sedna17]

I confirm that the new version works for me.