• We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.
Yes, it was a link to a post dealing with the issue. By following the instructions the game will speed up considerably. But here it is copypasted:

Open the mod-CORE2\ai\usa_1936.ai file in notepad (same goes for the other USA AI files), and change the following line:
683 # Hawaii into 663 # Pearl Harbor
 
It does work, but CORE is slower than vanilla - if that's what you're comparing too. I tried TRP the other day. Load that up and compare - CORE plays like a sports car in comparison...

This problem exists in regular Doomsday as well, but since that game plays faster from the start its less noticeable. Even if you don't like CORE, make the change and play the regular game - it should move slightly faster still than what you're used to.
 
How does such a small thing make a difference?
 
GeneralHannibal said:
How does such a small thing make a difference?
The AI tries to send either naval units or convoys to a non-existing port, and keeps retrying it apparently. As CORE has more naval units than DD it affects CORE rather badly. I really have no clue why DD itself doesn't appear to be suffering from it - it's a Vanilla bug after all.

After applying the fix CORE was about 25% slower than regular DD IIRC.
 
It does affect vanilla, I tested this the other day - but even with the bug in place Doomsday is usually 'fast enough'. I can only say that on my machine, the vanilla Doomsday is lightning quick with the bug fixed - noticeably (but not much) with it fixed.
 
kel varnsen said:
ok, was a lot better when i started a new game. guess it works then...
That's good to hear. :)
 
Hagar said:
The AI tries to send either naval units or convoys to a non-existing port, and keeps retrying it apparently. As CORE has more naval units than DD it affects CORE rather badly. I really have no clue why DD itself doesn't appear to be suffering from it - it's a Vanilla bug after all.

After applying the fix CORE was about 25% slower than regular DD IIRC.

Ran through a log I created for it.....

The problem is with the US convoys in my game.

Naval units automatically went to Pearl, but as soon as they put a unit on Hawaii, oh dear lord. The convoys referenced it a lot.

One quick fix I could think of would be to change the error threshold, so if it doesn't work on the first try, give up, but that requires messing with the .exe most likely, which I'm not about to do.

The situation might vary depending on the current game, so I am gonna test it out some more.
 
Nomble said:
Still getting lots of crash to desktop with CORE. It's a real shame :(
DD itself isn't too stable, so that might be causing it. I do suspect a memory leak is the problem, probably goes for all Paradox games really. Just set your autosave to montly, and you won't suffer too much from it.

What's your internal memory like?
 
I once played a whole game with that lag game that normaly takes some houres took some days
 
Hagar said:
DD itself isn't too stable, so that might be causing it. I do suspect a memory leak is the problem, probably goes for all Paradox games really. Just set your autosave to montly, and you won't suffer too much from it.

What's your internal memory like?

It's not a constant memory leak, that much I have noticed.

Just from what I have noticed while playing, it has a greater chance of happening in two situations:

1) number of units
2) how far into the game you are

Case in point:

Playing as America, I wait until 1943 before actually really doing anything, and by that time I have usually built up a good deal of units (marines, infantry, armor+motorized), and the instant I kick off say invading North Africa? BAM! CTD....

I did actually manually track one memory leak related to actual country tags. It referenced Belgium, and it received my favorite problem ever from coding: segmentation fault.... It went out of bounds somewhere in that call, I don't know specifically where though.

Can I just get a buffer underrun please???????
 
DarkSoul1984 said:
Ran through a log I created for it.....

The problem is with the US convoys in my game.

Naval units automatically went to Pearl, but as soon as they put a unit on Hawaii, oh dear lord. The convoys referenced it a lot.

One quick fix I could think of would be to change the error threshold, so if it doesn't work on the first try, give up, but that requires messing with the .exe most likely, which I'm not about to do.

The situation might vary depending on the current game, so I am gonna test it out some more.

Hi,

Obviouslly you know more about technical programing than I do. Could you be a bit more specific about what exactly you saw during your checks that was causing the problems. I do write code for CORE but it's just a self taught hobby so when you start getting into the workings of the code I need very specific explanations.

mm
 
dec152000 said:
Hi,

Obviouslly you know more about technical programing than I do. Could you be a bit more specific about what exactly you saw during your checks that was causing the problems. I do write code for CORE but it's just a self taught hobby so when you start getting into the workings of the code I need very specific explanations.

mm

Ok, let me see if I can explain what's wrong...

It's in the .exe file.

I haven't de-compiled it, so I don't know exactly what's written there or how it's referencing it, but....

It would appear that it loops the supply convoy and/or troop movements for the ai. Unfortunately, if you give it the wrong information it runs repeatedly until it hits a break statement to get it out of the loop. (Computers really can't deal with infinity, unless we're talking skynet, in which case an erroneous line of code in an information file doesn't really matter) Basically, it keeps returning false when the computer sends the convoy or troops and so it creates it again, and again, and again...

Changing the line to read Pearl Harbor works, or you can change the set-up of the bases. I honestly don't know which one would be better as I don't know if performing the latter would fix issues with subsequent scenarios in Vanilla (I think everyone who's played the '44 scenario knows it runs really slow), or if that particular line is a mere oversight in only the '36 USA ai file.

As for other things, I haven't really tested anything in particular with CORE past that and the intelligence tree (which is horribly messed up). I've been trying other mods to see if I can possibly make a combination of different changes that might work better, but after messing with HSR last night, I gave that up... (even though I love the heavy division idea it take WAY too much just to implement that.....)
 
If its the .exe, then you should definately take your findings to Paradox themselves. If this situation remains for Armageddon, perhaps they can look at it for a patch. Even if this particular situation can be resolved by moving the base, an actual fix for the behavior would perhaps deal with any similar things that haven't been found yet.