Single Player bugs and crashes v36 plus (SVN) - After the 24th of October 2015

Status
Not open for further replies.
I know that it is not a 24-hour-long case, but few MONTHS? You are kidding, right? I hope so, but if you are not, the only reasonable solution to my crash is getting rid of the buggy mod and/or uninstall the game completely. Nobody waits months for troubleshooting/support for anything, guys :(.

Not joking. Life is a bit busy for many of the modders at the moment. Remember this is a hobby for us not a job.
 
need to fix this.


No but a known problem that's been narrowed down to a cause somewhere in the civic changes that have been made.


No this is a huge problem since the civic changes introduced it. Revolt starts happening before any options to mitigate revolt are even introduced to the player.


<snip>.

Question: why are players that use the longer GS say it's not happening with them.
Parsec wrote:but only if the use a fast game-speed (regarding your turn-counter it is a fast one)... in slow games there are only sometimes a problem, but never in the capital city...

So basically the Civic changes need compared to the Civics before SO changed them all. Has anyone asked SO what he did to the Rev section of the Civics? Probably not huh.

JosEPh
 
Yup. Real life before everything else. And yes, I know that all of this is your hobby. Good luck with all things (in RL and in modding)!
 
If I remember correctly the % is a special character in the text. If you want a % you have to enter %% especially if you are using the text with a variable ie where the 50 comes from.
Correct.

edit I can't find "Researches the tech" in the text files or the python so I don't know how it is added to the pedia or how it is translated into non-English languages.
I did find one on the TRAITS but not one which looks like the error.​
I think that may be it but I can't recall. I can also find out by tracking the tag itself to the text call in the code. This task was as much a note to self to address but thanks for trying.

edit2 The Observation Post etc going obsolete is Hydro's design. He has this happening with a number of building lines to force you to "upgrade". It has always been a pain especially for cities that need something but can't afford the most recent. A better way with these security buildings is needed but has not been thought out yet.
I want to get them to at least hold off on obsoleting until the building 2 steps past them has been unlocked, giving the full extent of the next upgrade's 'time' to upgrade before being faced with absolute obsoletion. I THOUGHT I had included that consideration when I audited those buildings/units. I probably didn't audit the building itself closely enough though.

I know that it is not a 24-hour-long case, but few MONTHS? You are kidding, right? I hope so, but if you are not, the only reasonable solution to my crash is getting rid of the buggy mod and/or uninstall the game completely. Nobody waits months for troubleshooting/support for anything, guys :(.
Most crashes or hangs don't affect MOST games. Therefore, rather than removing the mod, you can probably simply start a new game. Anyhow, I know it's a pain but no previous bug report that has come in has any less validity to get addressed than the ones that have just recently been made. I fell pretty far behind here because of a bunch of other things going on but I'm putting more time in lately so it's probably not going to take months, more like weeks perhaps. Or even possibly much less. I'm not sure what all lies between the bugs reported a few months ago and now.

Question: why are players that use the longer GS say it's not happening with them.
Happened to Whisperr and she plays Snail. How much longer are we talking? I would think it would be worse for them. But we've HAD loudly complaining voices on this subject since around the time the civics were adjusted. I was figuring that those who weren't complaining weren't playing Rev.

So basically the Civic changes need compared to the Civics before SO changed them all. Has anyone asked SO what he did to the Rev section of the Civics? Probably not huh.
Exactly. And yeah I haven't had time to address him personally. I figured he'd be keeping up on the forum posts. Plus, as I theorized earlier, it may not be anything directly about those civic changes except for whatever the change was that made it possible for the capital to be unhappy in the first stage of the game. I'm not SURE that's it but I wonder if it is.
 
the issue with the changing of the BUG screen and the automation table I could isolate, however, the F1 problem is not yet found, it is independent from this:
I upload here two sav, the first is before I finish "road building" by using a Great Trader, the second after that. I made no other between, just only got the "road building" tech.

in "before" I have the full table in BUG as well as in /automatic, in "after" I have the reduced BUG table screen and the reduced table in automatics.

I believe, that a similar fairure also is causing the F1 problem.
(maybe a change or typing mistake in a string, i.e. a , instaed a ; or something else, but there must be something just with that research in BUG or somewhere, that gives a different BUG version/table.)

I hope with this information you can find what it is.
 

Attachments

  • Friedrich v. Chr.-5287-before.7z
    Friedrich v. Chr.-5287-before.7z
    4.1 MB · Views: 47
  • Friedrich v. Chr.-5287-after.7z
    Friedrich v. Chr.-5287-after.7z
    4.1 MB · Views: 30
  • BUG_before.jpg
    BUG_before.jpg
    91.3 KB · Views: 55
  • BUG_after.jpg
    BUG_after.jpg
    84.8 KB · Views: 78
  • automations_before.jpg
    automations_before.jpg
    102 KB · Views: 53
  • automations_after.jpg
    automations_after.jpg
    90 KB · Views: 32
which can travel across Sea tiles at a much earlier stage in the tech tree than other ships. Are you sure they didn't gain contact by means of an early water crossing?
pretty sure. I checked with world builder before and after, and both don't have tech or units to do a contact.
 
I think that Neanderthal cities should always be razed when conquered by homo sapiens.
but that really would not be according the historical things. Neandertaler and modern human did live together for mire than 50.000 years and had children together over generations, again and again, and our gene still has more than 15% from Neandertaler.

Neandertsaler did not die when they meet the modern human. They lived side at side for long time.
 
Question: why are players that use the longer GS say it's not happening with them.
don't be offended. One who is not a modder but looks for a interesting game, has expectations and hopes. And he may hope again and again, that the problem(s) got fixed. If not , he is disappointed, but this does not mean, that he does not like the game in general.

If you have a racing car, you like it, even if it makes problems. But just because that you alwys press the engineer to fix the problems faster.
Take it as stimulus, not as bad critic ;)
 
ThunderBrd wrote:

Originally Posted by JosEPh_II
Question: why are players that use the longer GS say it's not happening with them?

Happened to Whisperr and she plays Snail. How much longer are we talking? I would think it would be worse for them. But we've HAD loudly complaining voices on this subject since around the time the civics were adjusted. I was figuring that those who weren't complaining weren't playing Rev.

So basically the Civic changes need compared to the Civics before SO changed them all. Has anyone asked SO what he did to the Rev section of the Civics? Probably not huh.

Exactly. And yeah I haven't had time to address him personally. I figured he'd be keeping up on the forum posts. Plus, as I theorized earlier, it may not be anything directly about those civic changes except for whatever the change was that made it possible for the capital to be unhappy in the first stage of the game. I'm not SURE that's it but I wonder if it is.

Done some digging this morning, started comparing v36 starting Civics (especially the Rev sections of these) and yes, Rev modifiers have been changed several times since then all from SO.

Want me to revert all these Rev changes back to v36 levels?

(And here I am volunteering to work on a section of the Mod I loathe! :crazyeye::p :hammer2: )

EDIT:
@T-brd,
Finding lots of instances of this type changes in the Rev section:
<iRevIdxDistanceModifier>0.4</iRevIdxDistanceModifier> Current SVN
Instead of this from V36:
<iRevIdxDistanceModifier>35</iRevIdxDistanceModifier>

The modifier has been changed from a whole integer to a decimal form. This looks very wrong. Is it?

Also Found 2 sets of Rev Modifiers In one of the early Military Civics.
JosEPh
 
Done some digging this morning, started comparing v36 starting Civics (especially the Rev sections of these) and yes, Rev modifiers have been changed several times since then all from SO.

Want me to revert all these Rev changes back to v36 levels?

(And here I am volunteering to work on a section of the Mod I loathe! :crazyeye::p :hammer2: )

JosEPh

Look at the civics thread i necro'ed?
 
SO you had 2 sets of Rev Modifiers in one of the Warfare Civics alone. This and several other changes in the Rev section of the starting Civics in all categories is causing reported problems that T-brd and I were discussing here and that Winter Warrior and several others like whisperr are saying breaks the early game.

As for the rest of your Civic changes, yes they should be discussed. I have toned done several already like Banditry and Tribal Warfare.

Once I get an answer from T-brd about the Rev question I have (see above post) then I will be committing the Rev corrections (ie changing back to Official v36 values). We can then go from there.

JosEPh
 
I'm not quite sure what the question is except if your asking if we should revert to the previous rev modifiers.

Since I have never fully analyzed the mathematics underlying rev, I think it's safe to say that we probably should revert to when we last knew it to be working in a more stable manner. That's a start.

The problem, imo, is we're all wandering around in a dark room taking guesses as to what the numbers mean exactly. What you pointed out about the rev distance modifier is a perfect example. I'd have to look deeply into the code to see but it looks like that would be setup as an integer rather than an float and thus to put an decimal into play there would probably be far outside the expected range by the original programmer and would likely throw things way off in some way (probably rounding down to 0).

I don't think you or I want to mess with rev too much yet, at least not until we have time to really look at what the numbers mean and at what tolerances they begin triggering events and so on. It's going to take analysis and we'll probably find a few hundred places we'd want to improve it. So we can't underestimate the depth of the project, particularly since it spans C++ and Python and there's a known misprogrammed tag and likely room for a lot more detailed tags for further interaction that would be inspired by the investigation. So for now, I think putting it to where it used to work fine would be wise.

I don't blame SO for wanting to try a few tweaks, but it needs careful changes and monitoring to get away with that when we aren't fully clear what the changes would mean. A scientific method approach to making adjustments would need to be employed if you can't read the code and put it all together into a cohesive picture (which I CAN do someday.) Somehow the previous civic designer didn't poo the scrooch so he had a decent idea of what the numbers would indicate. I'm not yet so blessed as to have an educated enough opinion of what they should be and my modding side of the computer is wrapped up in a deeper issue at the moment.
 
The v36 set had the <iRevIdxDistanceModifier>35</iRevIdxDistanceModifier> type.

Only the recent SVN had the <iRevIdxDistanceModifier>0.4</iRevIdxDistanceModifier> type. I did briefly look at the 9089 set SO put in back in Feb/march.

I have spent the last 4+ hours combing thru the Rev Mod sections for ALL Civics. Most of the Changes like the above example are in the Gov't and Military/Warfare Civics.

But there are others I did not post that were polar opposites of what v36 had.

Couple this with the Double Rev set found in one of the early Military with 2 of the modifiers from each set differing in the value, and this is most likely Winter warrior and Whisperr's source of their particular early game rev problem.

I have just completed the reversion of the Rev mod set back to v36 values. But have not committed them yet.

EDIT: SO has posted a New CivicInfos file in the Civic Discussion thread that he wants ppl to look at and discuss.

JosEPh
 
The v36 set had the <iRevIdxDistanceModifier>35</iRevIdxDistanceModifier> type.

Only the recent SVN had the <iRevIdxDistanceModifier>0.4</iRevIdxDistanceModifier> type. I did briefly look at the 9089 set SO put in back in Feb/march.

I have spent the last 4+ hours combing thru the Rev Mod sections for ALL Civics. Most of the Changes like the above example are in the Gov't and Military/Warfare Civics.

But there are others I did not post that were polar opposites of what v36 had.

Couple this with the Double Rev set found in one of the early Military with 2 of the modifiers from each set differing in the value, and this is most likely Winter warrior and Whisperr's source of their particular early game rev problem.

I have just completed the reversion of the Rev mod set back to v36 values. But have not committed them yet.

EDIT: SO has posted a New CivicInfos file in the Civic Discussion thread that he wants ppl to look at and discuss.

JosEPh

I reread the first part in your last post and edited my comments after - they still apply. The problem is not knowing what different numbers will really mean either way.

I suppose with that file, we'd just have to run compares perhaps. You've been working with some other factors there so I should really stay out of it myself.
 
So agreement on letting me commit the Rev revert and clean up set of CivicsInfos?

I think you are saying yes but then I'm not sure if you really are.

JosEPh
 
You shouldn't be able to attack him in the city without using an arrest or assassination to do so. That's a problem. Are you on the latest SVN or something a little earlier? I've had to tinker with this stuff quite a bit to get it right and I'm not sure it is quite right still. As an extension of that 'rule', he also shouldn't be stopping units from entering the city.

That criminals are getting city defense modifiers is also agreeably not optimal and I see I'll need to fix that.

I'm on SVN9271, yes.

Note that he's only stopping units from entering the city, and attackable, once he becomes Wanted. Also, I'm playing without any of the Combat Mod options (No Hide and Seek, for example), not sure if that makes a difference.
Note to self to fix a few issues here.


So agreement on letting me commit the Rev revert and clean up set of CivicsInfos?

I think you are saying yes but then I'm not sure if you really are.
Yeah, but I'm not sure how SO feels. I'm putting in my opinion which is to say, yes, commit your cleanup!
 
Revert done.

JosEPh
 
Note to self to fix a few issues here.



Yeah, but I'm not sure how SO feels. I'm putting in my opinion which is to say, yes, commit your cleanup!

OK u dont have to ask me on that stuff, u'll have agreed that TB is the CEO of the current C2C as DH said he did not want it . .
 
SVN 9271 Save and Minidump.

A game I started a few months back to try new features. Managed to get to almost to the end of Modern Age.

Press enter to end the turn, after a few seconds CTD.

Now works - 9280.
 
Musketeers, the French culture unit, are either losing their free Speed promotion immediately or failing to get it. Probably related to the promotion validity changes in SVN9270. Might as well just give them 2 base movement speed instead of messing with that, I'd say.
 
Status
Not open for further replies.
Back
Top Bottom