[Religion and Revolution]: Mod Development

[raystuttgart]

You have marked the whole atlas.

Can't you see my red "circle like" markings with "1." and "2." in the screenshot ?

I have made a screenshot of the whole atlas.
(Which also has red corners, but that is not what I talk about.)

I have manually marked (with red) only parts of it.

 

[Schmiddie]

Ähm, That is the whole Atlas you see? I see two! Hm, maybe something went wrong with the atlas. Ingame I only see the left part!

 

[raystuttgart]

Ähm, That is the whole Atlas you see? I see two! Hm, maybe something went wrong with the atlas. Ingame I only see the left part!

Please open the atlas with DXTBmp.
You will see what I am talking about.

 

[Schmiddie]

F......ing GIMP....

 

[Schmiddie]

I have cleaned the file and made it totally new. Now it looks and works fine.

Sorry for the confusion, ray.

I have changed some positions of resources...and sorry for that ray! Maybe you must update the xml reference...but it was worth to that work.

 

[raystuttgart]

I have cleaned the file and made it totally new. Now it looks and works fine.

Great.

Sorry for the confusion, ray.

No problem.

I have changed some positions of resources...and sorry for that ray! Maybe you must update the xml reference...but it was worth to that work.

I had almost finished all the XML references.
Oh well ok, I will check and adjust ...

 

[Schmiddie]

Furthermore I have uploaded the atlas for professions:

path: Religion_and_Revolution\Assets\Art\interface\buttons\Profession_icons\colonization_atlas_professions.dds

Do you have a preference what atlas I should do next? I would prefer to clean and arrange the whole unit graphics. With this we should make a big step forward.

Furthermore IsabelXXX teached me how to reduce the size of an atlas. I was able to reduce the size up to 3/5 without loosing quality.

 

[raystuttgart]

@Schmiddie:

Thanks for all the work you put into this.
I really believe that our mod will be a lot better, after we have finished this.
(... and created FPKs.)

@Isabelxxx:

Thanks again for all your valuable support to this mod.

@team:

I will withdraw a bit from brainstorming and discussions about new features.
It is simply, that I currently want to put all the little spare time I have to get the graphics cleaned up.

Please continue brainstorming and discussions as usual.

 

[Shoobs]

I was serious in my last post. Not getting any response is a bit of an insult.

 

[raystuttgart]

I was serious in my last post. Not getting any response is a bit of an insult.

Never meant to be impolite.
Also you already saw, that I had answered you in [Religion and Revolution]: Helping or Joining Team.
(That is the thread where we discuss offers for help.)

 

[raystuttgart]

@team and partners:

Please do currently not download the mod from SVN.
It is heavily under construction.

After Schmiddie and myself have finished our first clean up and tested it,
we will let you know.

We are progressing quite well and I really believe it will be worth the effort.
(But we will probably not get all of it done this weekend.)

 

[orlanth]

The problem I have with the current approach is that most manufactored goods (except tools production chain and hemp for ships) just feel like another kind of cash waiting to be converted at slightly different prices.

Yeah I can understand that, I agree it would be good to have at least some modest strategic gameplay differences between the yield types so they're not all basically generic units of "another kind of cash". However, it's also true as Ray mentioned the vanilla AI is fairly bad at knowing when it needs certain yields available for a unit or building. I think some AI improvements in TAC and RaR may have somewhat improved this behavior. If I recall correctly, whenever the AI gets "stuck" needing specific yields to complete a unit/building, it can rush production by paying a cash fee (perhaps the amount it pays can be adjusted by game difficulty level?). I think this is a fair method to allow some interesting yield requirements for human players, while still allowing the AI to function better with this and be a worthy opponent, compensating for the fact that its multi-step planning can never be made as flexible or sophisticated as a real human's.

In Kailric's Inventor and Medieval modcomps, technologies can require different types of yields to discover, which I think does add a lot of unique interest to the different yields, so they're not all "another kind of cash". This proved to be an interesting and fun feature because it makes the player remain alert for the opportunity to develop a variety of yields, for interesting strategic reasons other than merely converting them to some more cash. To solve issues of AI competitiveness since it can't do complicated planning, the AI does not consume these specific yield types to produce research points, but does require more research points to get each invention, and this factor is adjustable by game difficulty. In practice this is a good solution that allows the AI to play a more competitive game, while allowing players to set the game difficulty at whatever handicap they think is fair.

 

[raystuttgart]

If I recall correctly, whenever the AI gets "stuck" needing specific yields to complete a unit/building, it can rush production by paying a cash fee (perhaps the amount it pays can be adjusted by game difficulty level?). I think this is a fair method to allow some interesting yield requirements for human players, while still allowing the AI to function better with this and be a worthy opponent, compensating for the fact that its multi-step planning can never be made as flexible or sophisticated as a real human's.

Guys seriously, let us not mess with this.

If AI needs to buy yields too often to "rush" production of a building, this will get problematic for AI as well,
because it will need to much money which it then cannot spend otherwise !

And of course AI can only rush, if it has the money.
This would lead to many unfinished build projects in AI cities.

Also, we might get in conflict with other features like "Domestic Market".

A definite veto from my side to stronger mix strategic and economical aspects of yields.

Implementing things that might be cool for Human Players is not worth totally ruining AI.
And I do not want to program massive AI cheats.

Also, please let us have discussions about adding features to our mod in the Mod-Development thread again.

 

[Robert Surcouf]

Hi all,

I've been playing CivCol from Vanilla on, switched to TAC later and have now played games with both 1.0 and 1.1 of R&R: I have to say you guys did an awesome job here! Thanks a lot to you all (including the TAC-team) for making CivCol a great game!

Thank you Heartsbane!

We explicitly do not do this because AI cannot handle it very well.

I would rather avoid this too... This might even cost us some effort too. I mean this could increase micro-management. And we could have some problems when we try to rush some productions...

 

[raystuttgart]

@team:

Small update from the cleanup.
More than 5300 files have been deleted so far !!!!
(Just to give you an impression how much potential for improvement we had.)

The mod already looks so much more cleaned up !

Tests are very very positive so far.
I cannot give you numbers but it feels a lot faster.

This clean up is a gigantic investion in the future of our mod !

Most important benefits we will get from this:

• Better performance
• Higher stability
• Easier to do further development

We are not done yet but we are really making good progress.

Huge thanks to Schmiddie.
He is really working on this like a graphical modding machine.

 

[Shoobs]

Never meant to be impolite.
Also you already saw, that I had answered you in [Religion and Revolution]: Helping or Joining Team.
(That is the thread where we discuss offers for help.)

I tried to edit and or remove that post but it seems to have FAILED. My apologies.

 

[raystuttgart]

@team and partners:

The clean up is more or less finished.
(There are only minor things we might still do within the next days.)

So everybody please get the newest revision.

Please help testing and let Schmiddie or myself know, if you find any bugs considering graphics.

@Schmiddie:

This really was a fantastic team work.

 

[Robert Surcouf]

I've fixed the domestic Advisor screen bug.
There was a missing condition:

 if pCitizen.getProfession() != ProfessionTypes.NO_PROFESSION:

I've added an enhancement to see the ANGRY POPULATION ICON during city occupation.
What do you think?

 

[raystuttgart]

What do you think ?

Great.

 

[raystuttgart]

@team:

I have got some good answers about FPKs.

1. All that FPKs are doing is to decrease initial loading of the mod.
2. On newer file systems it does not really make a real difference at all.
3. They have no positive impact to performance or caching ingame.

Finally this does make sense to me.
Everything else I had heard about performance or caching simply never had any plausible technical explanation.
(That is why I was asking several experienced modders.)

To Summarize:

I think we can forget about FPKs.
(At least for now. Maybe we can reconsider if we ever end the mod and create a "final" release.)

1. We have already cleaned up the mod and thus it is loading faster.
2. It would only be additional work and make changes to the graphics more complicated.
3. For newer file systems it does not really matter anyways.

Comment:

Improved performance ingame would have been a real reason for me to introduce one or more FPKs.
Faster initial loading of the mod on old machines alone is not reason enough to my opinion.
It simply does not justify the additional efforts we would have during our modding work.
Also I have gotten many unverified claims about negative impacts on ingame performance of FPKs.

Edit:

Made some tests considering initial loading times of the mod.
(With our nicely cleaned up current revision.)

A) My old development machine

Empty caches: <45s
Filled caches: <15s

B) My strong business machine

Empty caches: <15s
Filled caches: <5s