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

[Toffer90]

That would certainly be a potential issue. Was this game running toffer's overhaul? Toffer... I'm not sure about this shift. Could be a little destabilizing in some areas of the code. At least for ongoing games suddenly shifting into this or away from this... aka, might not be a safe mid-game change.

It can be a bit problematic mid-game, true, though it is no problem for the dll code.

 

[Raledon]

It's more than that. You shouldn't be able to build a hut in modern era. It's like units themselves... just shouldn't happen ever. To a point, sure, you can see a building, even an anachronism, being valid for a while, but after a margin, no longer.

That's the major issue, I think.
Units get stronger by era. An axeman has no chance against a rifleman (super-units aside). High tier settler build a city with higher base pop, and possibly more buildings. Even the property-managers (LE, healers, entertainers) get "better" with age.
On the other hand, the power level of most buildings is consistent, while certain buildings being especially good. At some point, you may build the worse buildings since they are relatively cheap. An example of this is a +1 hammers/+1 gold building at the industrial era, that cost over 1K hammers.
Personally, I'd rather remove many of these over building them to clean the build queue UI.
This is especially so for the said buildings that are the same building as the earlier era, but worse (the upgraded version of tannery is plain worse, for example). You'll build the rifleman since it can kill the axeman- if an axeman kills the rifleman, why bother with the rifleman?

On another note, it turns out that gatherers are "better" than workers. My gatherers can build Nomadic Herd on poultry, while the worker cannot. The worker can only build Pasture, which requires canal systems (not researched). This is strange, since I cannot even build gatherers anymore, so using them on this is my only option to get poultry in a timely manner.

 

[JosEPh_II]

But if the original game is on Toffer's overhaul in the UI and more, then it's possible something is causing the Town Hall issue from that if we're testing the game without it. Maybe someone needs to test the game WITH it, trading for furniture before the end of turn of course.

In his game he can build Furniture factory and Brick factory. Those will be necessary (well at least the Furniture factory) for the Luxury Furniture Workshop iirc. Still not clear on how bricks got tied into furniture though. Can see bricks as maybe a prereq for town hall. But wood would've worked as well there.

 

[Dancing Hoskuld]

On the issue of upgrading improvements. The base time it takes is defined in the XML. That is modified by game speed multipliers also in the XML. Obviously someone needs to go through and see exactly which are used and then find out where the problem is.

It's more than that. You shouldn't be able to build a hut in modern era. It's like units themselves... just shouldn't happen ever. To a point, sure, you can see a building, even an anachronism, being valid for a while, but after a margin, no longer. However, we've now discovered the HUGE problem with why Police Stations and such are breaking down and it was that furniture prereq being lost. It was never an obsoletion issue.

And yet in reality they still do. Most are called small businesses or start-ups but they are basically "huts".

Reledon expresses it well. If it were like units then of course you would build the later ones, but it isn't. Not only that it is way too expensive to build the new versions in new cities.

On another note, it turns out that gatherers are "better" than workers. My gatherers can build Nomadic Herd on poultry, while the worker cannot. The worker can only build Pasture, which requires canal systems (not researched). This is strange, since I cannot even build gatherers anymore, so using them on this is my only option to get poultry in a timely manner.

This is a bug.

 

[JosEPh_II]

It can be a bit problematic mid-game, true, though it is no problem for the dll code.

Curious about why 4% is better than 5%?

 

[Dancing Hoskuld]

Curious about why 4% is better than 5%?

Go into GlobalDefines.XML or what ever and set it to what you want. Use notepad. It is that simple. I do it in mods which leave it at 10%. Mostly people use 5% or 10% but any number can be used, though some will cause problems eg -10 and 10000.

 

[JosEPh_II]

On the issue of upgrading improvements. The base time it takes is defined in the XML. That is modified by game speed multipliers also in the XML. Obviously someone needs to go through and see exactly which are used and then find out where the problem is.

I originally did that with the game speeds Before the New Math formula came into being. Afterwards with the games I've played as test I've not seen a big problem with improvement upgrades. Especially since the slower GS players don't Want them to be built faster. At least that was the feedback i was getting.

Thunderbrd said: ↑
It's more than that. You shouldn't be able to build a hut in modern era. It's like units themselves... just shouldn't happen ever. To a point, sure, you can see a building, even an anachronism, being valid for a while, but after a margin, no longer. However, we've now discovered the HUGE problem with why Police Stations and such are breaking down and it was that furniture prereq being lost. It was never an obsoletion issue.

And yet in reality they still do. Most are called small businesses or start-ups but they are basically "huts".

Agree with DH on this one.

 

[JosEPh_II]

Go into GlobalDefines.XML or what ever and set it to what you want. Use notepad. It is that simple. I do it in mods which leave it at 10%. Mostly people use 5% or 10% but any number can be used, though some will cause problems eg -10 and 10000.

Not the point. Just wanted to know why Toffermade the change as it stuck out when examining KaTioN_PT's save game.

 

[Dancing Hoskuld]

I originally did that with the game speeds Before the New Math formula came into being. Afterwards with the games I've played as test I've not seen a big problem with improvement upgrades. Especially since the slower GS players don't Want them to be built faster. At least that was the feedback i was getting.

The problem is (just to be clear about what we are talking about)

1. a worked improvement will upgrade in 100 turns
2. a worker can upgrade the improvement in 15 turns

This is the disparity being talked about. However there is another similar

1. a fortification with a unit fortified in it will upgrade in 50 turns.
2. a worker can upgrade the fortification in 10 turns.

As far as I can see you do want the "natural" upgrade time to be slower that the "active" upgrade time but do we want it to be such a big difference? Personally I think 5x is OK but since it is based on the cost to build the improvement this may get out of hand over the time of the game.

 

[JosEPh_II]

The problem is (just to be clear about what we are talking about)

1. a worked improvement will upgrade in 100 turns
2. a worker can upgrade the improvement in 15 turns

This is the disparity being talked about. However there is another similar

1. a fortification with a unit fortified in it will upgrade in 50 turns.
2. a worker can upgrade the fortification in 10 turns.

As far as I can see you do want the "natural" upgrade time to be slower that the "active" upgrade time but do we want it to be such a big difference? Personally I think 5x is OK but since it is based on the cost to build the improvement this may get out of hand over the time of the game.

And you are saying #1 is attributed to the GS settings; iImprovementPercent = How fast tile improvements upgrade to the next level (eg cottages to hamlets). Normal is 100

While #2 is dependent upon: "iBuildPercent = How fast tile improvements are built. Normal is 100".

So you want 2 to be smaller/faster than 1. Right?

Because currently they are the same for most GS. Just like the "Readme section" suggests.

 

[Dancing Hoskuld]

And you are saying #1 is attributed to the GS settings; iImprovementPercent = How fast tile improvements upgrade to the next level (eg cottages to hamlets). Normal is 100

While #2 is dependent upon: "iBuildPercent = How fast tile improvements are built. Normal is 100".

So you want 2 to be smaller/faster than 1. Right?

Because currently they are the same for most GS. Just like the "Readme section" suggests.

It is all more complicated that that and someone needs to take the time to review it. I suspect we have far too many adjustments made to the work rate of the worker and that that is what is actually causing the big disparity shown.

Upgrade time is defined by iUpgradeTime in ImprovementInfos.

Build time is defined by iTime in BuildInfos but modified by adjustments to the units Build/Work Rate which defined on the UnitInfos in iWorkRate but is modified by buildings, promotions, civics and technologies.​

Does anyone know if the building adjustment to work rate on buildings is universal or just in the vicinity of the city with the buildin? I think we have some normal buildings that are adjusting work rate so multiple of them can cause problems.

 

[Thunderbrd]

In his game he can build Furniture factory and Brick factory. Those will be necessary (well at least the Furniture factory) for the Luxury Furniture Workshop iirc. Still not clear on how bricks got tied into furniture though. Can see bricks as maybe a prereq for town hall. But wood would've worked as well there.

He had access to lots of Bricks already. I didn't look further than that to look at concrete because you only need either bricks or concrete (I agree wood should be added to that list for the Town Hall). So once he gets furniture from trading, he should be able to build them. None of these are vicinity prerequisites. Something is wrong here and I don't know if its the save itself, caching issues (building stuff is tied into caching of the most horrific magnitude of complexity to follow) or our index changes. The answer to not obsolete the furniture shop is sufficient for now.

And yet in reality they still do. Most are called small businesses or start-ups but they are basically "huts".

Call me a stickler but people don't build huts anymore and the name and button art matter to maintain the feel of the era. It disrupts the suspension of disbelief as a player far too much to still be constructing buildings of bygone eras.

However, I agree with your premise from a game design perspective. We do have need for some lesser version buildings that can help in building the infrastructure of a city up so it can ramp up to speed without being stuck on huge buildings to generate its production forever. But these can't be the same buildings as those being replaced and they also need to be replaced when the more advanced forms are constructed after opening up due to larger populations (or admin center building prereq which is just a population and culture prereq template ultimately if you use it as a prereq.) As we examine building chains soon, let's consider how we can make some ramp-in buildings to support smaller communities and captured cities in growing up.

So there's 2 dimensional axises for buildings to evolve along... interesting. More complex but better for gameplay.

The problem is (just to be clear about what we are talking about)

1. a worked improvement will upgrade in 100 turns
2. a worker can upgrade the improvement in 15 turns

This is the disparity being talked about. However there is another similar

This is not disparity. The worked improvement is not supposed to upgrade anywhere near as fast as you can get it to if you work over the improvement with a worker, showing both intent to solve the upgrade faster and the application of your national resources to do so without just sitting back and waiting for it to happen organically. Traits in my design will commonly manipulate one or the other or both - they mean different things. Thing is, normal organic upgrade times have to be significant enough to matter to want to get a bonus to that upgrade time. If we shrink the difference much, there won't be any motive to care to get a boost to that. Same must be said for work speeds. Who cares your worker just promoted when it doesn't make any difference. Even on faster gamespeeds it can make a difference with the current values so that's good.

And yes, to whoever said it earlier, gatherers ARE faster workers. But they also sacrifice themselves to complete most builds.

1. a fortification with a unit fortified in it will upgrade in 50 turns.
2. a worker can upgrade the fortification in 10 turns.

A fortification without a military unit in it will not count down to upgrade. A worker can go in and make improvements much faster than military staffing the gates.

As far as I can see you do want the "natural" upgrade time to be slower that the "active" upgrade time but do we want it to be such a big difference?

Yes. This rewards paying attention and looking for ways to improve how quickly you grow and thrive. And like I said, the intent is soon to have some leader traits be better at letting you ignore upgrades and just letting them take place in their own time. I think some civics can manipulate this rate of speed too but I don't think it's been a much used tag.

And you are saying #1 is attributed to the GS settings; iImprovementPercent = How fast tile improvements upgrade to the next level (eg cottages to hamlets). Normal is 100

Both are modified by gamespeed. The base time to upgrade on its own does factor off the same base construction costs though.

Does anyone know if the building adjustment to work rate on buildings is universal or just in the vicinity of the city with the buildin? I think we have some normal buildings that are adjusting work rate so multiple of them can cause problems.

I believe there's only a player wide global effect with that and it doesn't factor in to upgrade times, just worker work rates. Common buildings with this effect can compile far too dramatically so it should really only be a minor wonder ability and shouldn't be overused toooo much.

 

[JosEPh_II]

Both are modified by gamespeed. The base time to upgrade on its own does factor off the same base construction costs though.

Uh duh, did I not say so?

 

[Thunderbrd]

Uh duh, did I not say so?

Yeah so take it as confirmation, that's all. I got the impression you were saying only one was modified but I probably just misread your intended comment.

 

[JosEPh_II]

@Thunderbrd
Heads up,
Chaotic's save game played just fine. I uploaded the next turn for him in his thread about it.

 

[Thunderbrd]

@Thunderbrd
Heads up,
Chaotic's save game played just fine. I uploaded the next turn for him in his thread about it.

Thanks for checking on that. That's a huge help. I might still confirm on my end but it means I need to try it on the faster dll before worrying about a debug run. So that can save me tremendous amounts of time.

 

[Thunderbrd]

He had access to lots of Bricks already. I didn't look further than that to look at concrete because you only need either bricks or concrete (I agree wood should be added to that list for the Town Hall). So once he gets furniture from trading, he should be able to build them. None of these are vicinity prerequisites. Something is wrong here and I don't know if its the save itself, caching issues (building stuff is tied into caching of the most horrific magnitude of complexity to follow) or our index changes. The answer to not obsolete the furniture shop is sufficient for now.

Ok, through the magic of debug dll evaluations I have determined why the Town Hall stops being buildable in those few cities. It is because of the furniture at first but then it's because he's already constructed a Metropolitan administration in those cities and it replaces the Town Hall, which was constructable only when the Metro Admin was shut down. How it was constructable and yet the Metro admin had been disabled is something I haven't figured out - THAT might have been a matter of caching not yet catching that the Town Hall SHOULDN'T have been constructable in the round he set them to build (post the discovery of Industrialization when the Metro's had already been disabled - disable vs buildable might have varied in an odd clash due to one being cached and the other not.)

Anyhow, from this point, there is NO bug of note in this save. Case Closed.

 

[Thunderbrd]

SVN 9849

Got into a curious situation, usually when you have units waiting for orders the game toggles to them, this is the same with buildings. In the end when all units/buildings are occupied you are given the warning to end your turn.

Well, what happened was that I pressed next turn, but it no longer toggled to inactive units or buildings, even after checking them out by moving them. So I saved the game, lucky I auto save every turn.

Now the curious thing is that when loading that weird save I'm getting a hard crash related to Civ4 runtime.

Please see both attached files, "Non buggy" is the good turn where no problems occurred and "Buggy" is the next turn, the one with all the weirdness and the one that crashes when loading. In any case going into the next turn on the "Non buggy" save turns on the weirdness (so at least its repeatable!)

EDIT: Probably related to the whole "World War" i'm doing against the AI?

Sorry... not having a problem with this game now. I didn't verify the buggy one but usually you have to catch a bad gamestate when it becomes one and I didn't have it happen... often this can be because I'm running a more updated DLL and small changes make a different processing path get taken. If you see it happen again I'll try to fix it with a higher priority to catch it before changes might invalidate the situation.

 

[Toffer90]

Not the point. Just wanted to know why Toffermade the change as it stuck out when examining KaTioN_PT's save game.

No big reason really, I may prefer to have 25 steps instead of 20 steps in the range 0-100%, then again I might not prefer it... Sometimes I just mess around with stuff to see what works and what doesn't...

I can't guarantee that using an unfinished and experimental modmod will be safe for your saves, but I think that anyone who tries out WIP modmods understand that there is a risk involved.

Curious about why 4% is better than 5%?

The only important thing is that the number X gives an integer result from this calculation: 100 / X
So the number 3 would make it strange as it e.g. would hit the value 99%. It may be, I seem to recall, that I in one version of the modmod sat the number to 3 and that I in the next update changed it to 4.

 

[KaTiON_PT]

Sorry... not having a problem with this game now. I didn't verify the buggy one but usually you have to catch a bad gamestate when it becomes one and I didn't have it happen... often this can be because I'm running a more updated DLL and small changes make a different processing path get taken. If you see it happen again I'll try to fix it with a higher priority to catch it before changes might invalidate the situation.

Thank you Thunderbrd, for the lengths you went to help my situation.