Version 4.1 Discussion Thread

I finished downloading MM 4.1. Nice new terrain and features for my Africa map, and for my planned Oceania map.:)
I can't say more about it yet, because I am quite busy for my job and I first want to finish a running game in version 4.0.1.

This also means that I won't be able to find any bugs or imbalancies for the coming weeks/months. Lunarmongoose, I thus recommend you to play a couple of games by yourself before you start your work to improve the AI.
 
AAAAAAAAAAAND... We're live.

At last.

Been working on this version on-and-off for 18 months now hehe.

There are several ways to look at size - version 4.0 probably had more SDK work for example - but if you go by the number of lines in the Change Log and Release Notes, then version 4.1 is my NEW single biggest patch ever. TAKE THAT, you stupid pathetic version 4.0... Mwa ha ha.

To summarize, this has: MASSIVE graphics updates, a MASSIVE PerfectMongoose mapscript update, substantial updates in the usual mod content departments, and a STILL-RATHER-IMPRESSIVE list of SDK changes.

(I suppose at this juncture I should also mention a few remaining icon selections that didn't make it into the last Historical Document post: Atheism black Medieval crest lion on light gold background, Witch Doctor skeletal hand on black background, Shaman yellow energy orb on red background, Atlatl two darts flying left on bright red background, Air Bomb two B52s on tarmac, Air Explore old map, Air Intercept green fighter shooting on sunset background, Air Rebase purple and red suitcase on gray background, Air Recon white drone with sharp angles over dark blue water, Air Sentinel rotated Harrier over light blue water. As always I didn't MAKE these icons, but I found them and chose them for these roles and that's a skill too.)

Just happened to check here today and so Glad you're back to release 4.1. Thank you.

JosEPh :)
 
Doesn't your mapscript work with other mods?? ie C2C, I keep getting a CTD when loading in C2C?

No, it doesn't, and neither would any other mapscript that uses any non-default behavior for placing terrain or features or resources, because it has to reference those directly, and they are different in every mod.

As was the case in the past as well, the C2C people will have to modify the mapscript to work in C2C before you can use it. It's not hard to do, but it takes a little time. The next thing up for me personally is to make a vanilla-compatible version of PM 3.3, for the same reason.

But please give it a try in MM! I know everyone loves C2C, but my mod is really freakin awesome at this point. :p
 
Started a game last night, selected Ronald Reagan as leader but got Geo. Washington instead.

Set up 12 AI, PM map, Prince, on the 1st Mongoose (extra large maps); by turn 70 started seeing 45 sec to 1 min End of Turn processing time (i7 2600k 3.0 GHz, 8GB ddr3 1660Mhz ram, Geforce GT 550i vid card 2GB DDR5 ram), was kind of surprised at EoT time being that long and seeming to get longer by turn 85.

Very interesting features in this mod.

Just a short report for now.

JosEPh
 
Started a game last night, selected Ronald Reagan as leader but got Geo. Washington instead.

If you're new to the mod, it's very helpful to read the Stuff to Know document, which is in the mod folder and also in the sticky thread here! :)

I've used chronological leader cycling for a while now. (I THOUGHT it was a completely original idea, but apparently someone else did it independently around the same time, heh.) You only pick your civ, not your leader. You start with their earliest leader historically, and it switches automatically in later eras.

Set up 12 AI, PM map, Prince, on the 1st Mongoose (extra large maps); by turn 70 started seeing 45 sec to 1 min End of Turn processing time (i7 2600k 3.0 GHz, 8GB ddr3 1660Mhz ram, Geforce GT 550i vid card 2GB DDR5 ram), was kind of surprised at EoT time being that long and seeming to get longer by turn 85.

It's generally much worse on the large map sizes, but yes, turnover time can be bad. I'm still using BBAI for my AI code, with only minor additions to accommodate specific features in the mod that shouldn't affect turn time much. And BBAI is supposed to be quite a bit faster than the vanilla AI code. So I'm fairly sure it'd be the same in other mods at that map size.

BUT, I could always be wrong, heh. Regardless, in the next few weeks or months I'm looking to merge in the K-Mod and/or C2C AI code, which should help further. I do wish Karadoc weren't gone at the moment though.

Thanks for posting! Appreciate the feedback. :)
 
As was the case in the past as well, the C2C people will have to modify the mapscript to work in C2C before you can use it. It's not hard to do, but it takes a little time. The next thing up for me personally is to make a vanilla-compatible version of PM 3.3, for the same reason.

But please give it a try in MM! I know everyone loves C2C, but my mod is really freakin awesome at this point. :p

Sorry all of C2C's python personal are now gone :blush:
 
Sorry all of C2C's python personal are now gone :blush:

Lol... Never expected to hear THAT response. Really? ... Seriously? Wow. Erm... Well Rakete mentioned he was gonna provide a version of his maps for C2C people to set up for their mod, but I guess that might not happen either.

Sorry, but it'd take several hours to several days of work to convert it depending on how fancy you want it to be with handling the C2C terrain types, and helping support rival mods isn't high on my priority list. :(
 
Lol... Never expected to hear THAT response. Really? ... Seriously? Wow. Erm... Well Rakete mentioned he was gonna provide a version of his maps for C2C people to set up for their mod, but I guess that might not happen either.

Sorry, but it'd take several hours to several days of work to convert it depending on how fancy you want it to be with handling the C2C terrain types, and helping support rival mods isn't high on my priority list. :(

Not to worried about terrain and stuff, just so it doesn't crash is all. :p
 
Okay, so, you might have noticed by now, there's a fairly serious bug where all sea monster types (Turtle, Serpent, and unconfirmed but I assume Kraken too) spawn immediately when the Barbs start, regardless of their tech requirements. That's almost certainly b/c I didn't update them for the new water terrain types, but I'll know for sure when I look at the code in a minute.

I've made a few other tweaks since launch as well, so I'm gonna upload a 4.1a patch soon. It WILL break 4.1 save files b/c of a couple XML additions... Sorry heh. (I would save those additions for 4.2, but I have to patch the bug, and Rakete can use the new animal type in his maps, so what the heck. This SHOULD mean 4.2 will be save-compatible with 4.1a, though.)

Not to worried about terrain and stuff, just so it doesn't crash is all. :p

The 2 things are related. The terrain, features, and resources are WHY it's crashing.
 
Well, THAT's fixed at least... Had nothing to do with the new water types of course. :p When I made the fix in 4.1 for Barb cities being able to build monsters normally, and for Sandworms spawning as cargo in Barb transports, I inadvertently bypassed all the normal unit training checks for the special case of Barbs spawning on their own every turn. Soooo not only did you have Krakens spawning immediately, but there was probably no limit on how many there could be, either. Oops. :)

Unfortunately I just found a second serious bug: my new river pathing code doesn't handle river cities built directly on resources correctly. As soon as I get THIS fixed I should be ready to upload the patch.
 
Thanks for the clarifications. And no I didn't read the Stuff to Know document. :p Just wanted to play right away. ;)

Thank you for continuing to work on mod.

JosEPh
 
Thank you for continuing to work on mod.

Well, since I'm now actively playing a game myself again (for the first time in about a year), I don't have much choice... I have to AT LEAST fix the serious bugs, lol.

Although 4.1 has now emerged from the play-testing crucible in much better shape than my previous mega-update, 4.0, did. THAT one required a fairly substantial emergency patch, 4.0.1, which took me a month to finish at the time. By comparison, the current emergency patch has only taken a week. ;)

Speaking of which, I DID finally fix the city resource problem Thursday night, after a whole day poring over my custom river pathing code from last Spring which I honestly hoped I'd never have to see again due to how complicated it is... Sigh. Turns out it had nothing to do with rivers directly; ALL cities located on resource tiles were getting permanently bugged, though only for revealed and enabled resources. Sorry about that, guys!

In my defense though, it was a REALLY subtle error, that took me 8 HOURS to find even when I knew exactly what I was looking for, that I didn't see til I was literally staring right at it, and that seemed quite harmless at first glance due to every other use of this type of code block being done quite correctly this way in reliable vanilla code.

Well, at least I don't have to worry about anyone duplicating my river link fix, lol.

I'm feeling pretty confident at this point that I've found all the bugs in 4.1, so I'm about ready to post 4.1a. Right now I'm going to sleep though.
 
Well, since I'm now actively playing a game myself again (for the first time in about a year), I don't have much choice... I have to AT LEAST fix the serious bugs, lol.

Just continue to play yourself, and I bet you will find more bugs ;)

- Not kidding on you, it's pretty normal in humane programming that these things happen.

I experienced that my myself when I was programming with Turbo Pascal in the late 1990s, and the company that I currently work for has a database software of its own that gets patched every couple of weeks since ever.
 
where is my stone?
I m playig 4.1 as the ottomans, and I can't see stone in my capital city when i am supposed to have two stones (one from Skara Brae and one from a quarry)?

As I discussed in the last few posts, there's a bug with cities built directly on revealed resources. It also causes them to get messed up with wonders that provide resources, like Skara Brae. I already fixed it in 4.1a, which will be up soon, but it will require that you start a new game, since the cities are permanently messed up once this happens. Sorry, this sort of thing can happen when you completely rewrite a major chunk of vanilla code. :(

I also found another bug, this one with the black crosshatch tile overlay not working right with blockades. It appears to only be graphical; the blockade effect itself seems to be working fine. But I wanted to take a look at this before posting the update, which I'm just getting to now.

Also, the turns DO seem to be taking longer than before, though it's been so long for me personally that it's hard to tell. This may have to do with my new PlotGroup code, but I really hope it's something else b/c I love that feature, sigh. Anyway, wanted to mention I'm aware of the issue. Version 4.2 will pretty much just be a massive DLL update, so I'm going to try to get to the bottom of it as part of that.
 
Thanks for the updates and the coming patch(s).

Since I almost never build a city on a resource I haven't seen this problem in my empire. But if it does it for a player then I would expect it to do that to the AI as well. Since the AI Loves to build cities on resources. :p

JosEPh :)
JosEPh
 
Since I almost never build a city on a resource I haven't seen this problem in my empire.

That's also why _I_ was a bit slow noticing it, heh.

But if it does it for a player then I would expect it to do that to the AI as well. Since the AI Loves to build cities on resources. :p

Yes it does, since it has nothing to do with who the players are. As I said above, it was the result of a very subtle bug in my new PlotGroup code, which was written solely for fixing river trade links.

After repeatedly experiencing extra lag when founding and especially capturing cities in my current game, I am also warming up to the idea that it must also be the source of the extra turnover lag. I may have missed some ways to optimize things, so I will closely compare it to the vanilla code when I get a chance. But if that fails I'll just go back to using the vanilla code, and let trade links jump between rivers again. :(

Either way I'll fix the lag in 4.2, but if I'm right about the cause then it ISN'T something that should get any worse over the course of a game. In fact, it should get LESS severe as more of the world gets covered in roads, since those simplify the PlotGroups compared to rivers.

Sorry for the continued delay getting 4.1a posted. I ran into a Python error with one of the events, and 2 minor string errors in the DLL, last night. Those won't take me long to fix, but it's hard to get things prepped for public release when your best friend is playing a Civ4 game with you 6-8 hours every night, and when you're fairly busy with other stuff on top of that. (I just spent several hours looking for a new pair of winter gloves online, for example.)

I am HOPING to get it up in the next few hours, but at the moment I have to go get a 2-weeks-overdue allergy shot, and there's a new episode of my can't-miss tv show on in 6 hours (Arrow), so we'll see.
 
Forgot to mention in my last post that I had successfully fixed the blockade tile graphics bug I mentioned in my post before that, heh. (And the fix also fixes a longstanding vanilla glitch with the black crosshatch overlays.)

Anyway, 4.1a is finally up. I don't have time to fix the PlotGroup recalculation lag just yet, but even with it, 4.1 should be fully playable now. I HAD been aiming for a perfect result with the main release; sorry that didn't QUITE go according to plan, lol.

(Historical Document note: Use of the "Antigravity Boots" icon as a widget in Techs that allow Terrain/Feature movement restrictions to be lifted for certain units.)
 
Don't everyone be happy all at once, or anything. :p

Been very busy again here lately, but I'm hoping to work on the PlotGroup lag issue soon, if nothing else. Aside from that though, 4.1a looks pretty good so far.
 
Top Bottom