Single Player bugs and crashes v37 plus (SVN) - After the 24th of December 2016

[Thunderbrd]

@Team: Should I go ahead and set the mod to start at the year 200 000 BC, I'm going to have to change the callendar regardless of the starting year.

Yes please

 

[Noriad2]

While I understand I said you no longer exist prior, I do need to address this post. I did not know that Deity had 2 starting settlers... and I do not know why they aren't using both of them most of the time ( IE, they -usually- only settle one city ). This said, how in the blazes am I meant to close the gap with these AI if they are able to blossom out into massive sprawling empires before I can even get a second city myself??

If you start close to another civ you can rush him. Although it is hard to financially hold on to a 2nd city until you have Barter. Third cities and more that you conquer will probably have to be disbanded if before Chiefdom. A good guide for rushing in Civ 4 BTS can be found here:

https://forums.civfanatics.com/threads/the-early-rush.244075/

Also, the game is 100,000 years. There will be plenty of opportunities to catch up later. The strong point of Civ 4 is that there are several viable strategies towards victory. It is best to climb the ladder of difficulty levels, trying out different strategies as you go up each level.

Civ 4 BTS deity was so hard that you needed to know several strategies cold, continually analyze your situation and be able to quickly switch to the best strategy for your specific current situation. It was said that only 1% of players ever won a civ 4 BTS game on deity level. It took me months of playing and studying several strategy guides to get my first wins on Civ 4 BTS deity.

C2C indeed tended to be a lot easier than BTS: there are tons of let's plays on youtube of mediocre players crushing everything, be number one in score , and have a higher score than the next three civs combined. However, *after* V37 (i.e. SVN version) deity/nightmare was made a lot more difficult, and is a decent challenge now as far as I can see (still in the middle of a game). If the AI is improved and certain exploits fixed (like buying techs with workers, or power stacking research bonuses from religions) then Deity/Nightmare might actually come a lot closer to BTS in difficulty.

 

[Toffer90]

Glad to be of help.


Another question - are Microsoft Visual C++ Runtime Error crashes unavoidable with this mod? Because I get them every couple of hours, almost like on a timer... I am running a ridiculously sized map with a good chunk of AI's, so maybe there isn't much to do about it, considering it's an old game.

Do any of you know of discussions out there about fixing or reducing this problem? Like if there's something you can do yourself...

There are some memory leaks here and there that will lead to crashes to desktop (CTD),
I've noticed that loading saves mid-game will take you to one of those CTD faster. I therefore avoid the "go back a couple of turns to see if I can avert disaster" loading in my games.

Giant and gigantic maps are so big that you should use view-ports when they start to crash regularly; when that is, depends entirely on the number of civs/units/cities on the map.

There isn't really that much you can do to avoid CTD due to memory allocation failure (MAF).

 

[Dancing Hoskuld]

There isn't really that much you can do to avoid CTD due to memory allocation failure (MAF).

Save the game, exit Civ IV and then start it and load the save works up to a point. However in late game you may need to do this every turn to avoid MAFs.

 

[Septimius]

Save the game, exit Civ IV and then start it and load the save works up to a point. However in late game you may need to do this every turn to avoid MAFs.

I have autosaving on for evey turn.

But what I really wanted to know, is whether the Runtime error crash after a certain period is normal for Civ IV? I mean is there nothing really that can be done about it? It is kind of tiresome to have to restart the game every couple of hours.

 

[Dancing Hoskuld]

You need to exit Civ IV to get rid of the memory leaks that cause the MAF. Loading a save without exiting can actually increase the problem.

Yes the runtime error occurs for Civ IV for all mods and even vanilla. It is due to memory leaks. Memory leaks occur in the dll, Python and exe. Attempts to get rid of them in the dll and Python can be made but there is no way to address the ones in the exe.

When it occurs varies but it happens sooner the more stuff is in a mod.

 

[Septimius]

You need to exit Civ IV to get rid of the memory leaks that cause the MAF. Loading a save without exiting can actually increase the problem.

Yes the runtime error occurs for Civ IV for all mods and even vanilla. It is due to memory leaks. Memory leaks occur in the dll, Python and exe. Attempts to get rid of them in the dll and Python can be made but there is no way to address the ones in the exe.

When it occurs varies but it happens sooner the more stuff is in a mod.

Ok, figured as much.

Well, the runtime errors ctd's the game completely. I've never have it crash me to the menu or give me the option to load a game

 

[JosEPh_II]

Ok, figured as much.

Well, the runtime errors ctd's the game completely. I've never have it crash me to the menu or give me the option to load a game

Of course it does (runtime errors ctd's the game completely). And it will never take you back "to menu" or "option to load" ever. Not how a Memory Allocation Error works.

Save on a periodic time frame say every 5 turns on anything less than Eternity GS and Gigantic map. On Eternity and Gigantic every other turn after the 1st 500 turns. Buy playing what you play and not saving for hours you're overflowing what the game can hold in data. Give it a break and save more often, in your case a lot more often.

You will have experiment to find your "Happy medium".

 

[Septimius]

Of course it does (runtime errors ctd's the game completely). And it will never take you back "to menu" or "option to load" ever. Not how a Memory Allocation Error works.

Save on a periodic time frame say every 5 turns on anything less than Eternity GS and Gigantic map. On Eternity and Gigantic every other turn after the 1st 500 turns. Buy playing what you play and not saving for hours you're overflowing what the game can hold in data. Give it a break and save more often, in your case a lot more often.

You will have experiment to find your "Happy medium".

Are you saying that manual saves help against runtime errors? Because I am autosaving every turn.

 

[Noriad2]

I see, then I have a lot of work ahead of me.

@Team: Should I go ahead and set the mod to start at the year 200 000 BC, I'm going to have to change the callendar regardless of the starting year.
I'll of course adjust animal spawning to the new dates.

I don't think that it is a good idea to set the start date to 200K BC.

Anatomically correct humans evolved 200K BC yes, but according to current scientific knowledge they only evolved in East Africa. Homo Sapiens didn't leave Africa until ~50K BC.
Homo Erectus moved out of Africa much earlier, but evolved to other human species, like Neanderthals and Denisovans, in the rest of the world.

So a 200K BC start means all the homo sapiens civs must start close very together (in Ethiopia) then start migrating to the rest of the world after 150K years. While the game starts with different civs spread out over the (old) world (i.e. everywhere except the Americas).

There is no way to make the tech tree historically correct, as firemaking and stone tools probably predate Homo Sapiens by a few million years. As Homo Sapiens started migrating out of Africa only after 50K BC the game should not start earlier than 50K BC and perhaps even a bit later.

 

[Toffer90]

I don't think that it is a good idea to set the start date to 200K BC.

Anatomically correct humans evolved 200K BC yes, but according to current scientific knowledge they only evolved in East Africa. Homo Sapiens didn't leave Africa until ~50K BC.
Homo Erectus moved out of Africa much earlier, but evolved to other human species, like Neanderthals and Denisovans, in the rest of the world.

So a 200K BC start means all the homo sapiens civs must start close very together (in Ethiopia) then start migrating to the rest of the world after 150K years. While the game starts with different civs spread out over the (old) world (i.e. everywhere except the Americas).

There is no way to make the tech tree historically correct, as firemaking and stone tools probably predate Homo Sapiens by a few million years. As Homo Sapiens started migrating out of Africa after 50K BC the game should not start earlier than 50K BC and perhaps even a bit later.

Remember that we are planning a nomadic start feature, where you won't be able to settle a city before some tech. Your first "city" would be a unit that can produce units.

↓↓ Subject to change ↓↓
On Normal gamespeed the first 150 000 years take atm. (latest SVN) 47 turns.
On Eternity gamespeed the first 150 000 years take atm. (latest SVN) 469 turns.
↑↑ Subject to change ↑↑

Anyhow, why do we have to accurately simulate history in every possible way?

 

[Septimius]

I don't think that that is a good idea. Anatomically correct humans evolved 200K BC yes, but according to current scientific knowledge they only evolved in East Africa. Homo Sapiens didn't leave Africa until ~50K BC.
Homo Erectus moved out of Africa much earlier, but evolved to other human species, like Neanderthals, in the rest of the world.

So a 200K BC start means all the homo sapiens civs must start close very together then start migrating to the rest of the world after 150K years.

There is no way to make the tech tree historically correct, as firemaking and stone tools probably predate homo sapiens by a few million years. As Homo Sapiens started migrating out of Africa after 50K BC the game should not start earlier than 50K BC and perhaps even a bit later.

A few corrections, latest research shows sapiens spreading out from East Africa around 70.000 B.C. This is known as the cognitive revolution. The point in time when Sapiens begin to rule the planet and begin to take out any competition. Be they other Homo species or animals for that matter.

Already 100.000 B.C. you have archaeological discoveries of Sapiens and Neanderthal living in the same area of southern Europe, but Sapiens disappear from the area.

What an early 200.000 BC era should portray, is a comparatively weak creature, primarily a gatherer and scavenger, that run from predators and can't go much more than toe to toe with neanderthals. The cognitive revolution is the point when Sapiens apparently develop the necessary abilities to work together in much more complex systems than f.ex. neanderthals. Our abilities of myth making, stories, complicated descriptions, are considered the real human revolution, not so much fire or stone tools, which you correctly point out, is much much older and used by several of the Homo species.

 

[Toffer90]

Note to all:

Games started before SVN rev. 9519, will have a callendar date that is exactly 150 000 years wrong compared to if the game had been started at or after rev. 9519.

 

[JosEPh_II]

Are you saying that manual saves help against runtime errors? Because I am autosaving every turn.

Yes exactly. Autosaves only really help when you have a CTD to get back to close to what turn you were on when it goes. It's more of a placeholder save point. Manual is a hard point save and clears out the build up of data from playing. That way all the lists and arrays don't have to keep adding more data. They have a fresh point to work from. This is all a laymans description. Best I can do on my memories from 42 years ago when I had my last programming course.

Just go into the Save folder every couple of days and delete the build up of saves. Once the game gets more than 80 iirc it too can cause problems. Civ IV is after all only a 32 bit engine single core.

 

[Toffer90]

Yes exactly. Manual is a hard point save and clears out the build up of data from playing. That way all the lists and arrays don't have to keep adding more data. They have a fresh point to work from. This is all a laymans description. Best I can do on my memories from 42 years ago when I had my last programming course.

@Septimius: Though it may sound strange, I believe there is truth to this one.

 

[JosEPh_II]

@Septimius: Though it may sound strange, I believe there is truth to this one.

Knock me over with a feather!

 

[Toffer90]

Knock me over with a feather!

Not sure what you mean by that, the expression is unknown to me.

I was not in any way trying to say that you tell untruths.
It sounds strange, not because you said it, but because it can remind one of typical myths that often gets a life on its own in forums such as this.
Like the notion that the steam version of BtS causes lots of problems for C2C. That steam installs the wrong v3.19 patch and so on.
Or that it helps to use a large adress aware patch on bts even though BtS v3.19 already was made large address aware by fireaxis.

 

[Septimius]

I believe that is what is commonly known as a "heureka" moment - at least for me.

 

[Harrier]

Not sure what you mean by that, the expression is unknown to me.

It means an unexpected reply.

I can knock you over with an axe - it is hard and expected to knock you over.
But a feather is soft so should not knock you over.

Therefore the unexpected response - he was expecting an axe (against) but got a feather (agree).

It is an English saying, the Americans adopted. (then probably changed the meaning a bit!!)

 

[Toffer90]

It means an unexpected reply.

I can knock you over with an axe - it is hard and expected to knock you over.
But a feather is soft so should not knock you over.

Therefore the unexpected response - he was expecting an axe (against) but got a feather (agree).

It is an English saying, the Americans adopted.

I see, thanks. ^^