New Version - October 1st (10/1)

Status
Not open for further replies.
I tried a quick game using the latest CPP 10/2

Conditions:
Rome
Pangea
8 Deity
No tech trading (hated it in civ4), RA enabled

Was fun overall but I am not able to finish my game (at least unless I force myself to).

Problems experienced:
1. I am experiencing CTDs when sometimes using planes the game will just leave when doing some manipulations with planes. It's not consistant but it happenned twice in the middle of a turn.

2. The turn times are unbearable after around T200. The AI seems to make a big carpet of units at that difficulty and it's possible some new AI routines in the DLL with units take too long with their high numbers.
I am talking 4-5 minutes turn times (calculated) with quick/quick enabled while the base game takes around 30s at the worst. None of the AI was really big but they had covered their lands of units.
1 and 2 together made me give up at around T280.

3. Sometimes when I captured a city it ended up with like 30 health, sometimes 150. Not sure if its intended or what is the reason.

The game itself:
1. Deity was rather easy. Considering it is my first game with the mod, trying to learn my way through it, and I was on my way to winning I think (2nd in tech with 4 capitals but in full conquest of the tech leader). I think the main issue the AI faces is that it remained really passive with its units. They have a tendency to camp near their cities and as long as you are not too far away in techs it shouldn't be too hard to capture the city with superior tactics. I never was threatened by a neighbour and their DoWs were really timid, sending very little. The only exception was France at the end that DoW me after saturating its land with machine guns.
Their teching was good but it's only my first game. I wasn't able to catch up to the tech leader (theodora) all game long always 8% behind her.

2. The balance of growth, science and tech cost seems spot on. It felt smooth.

3. Production of diplomat units is off after a time. With all the bonuses in my capital I was reaching some 450 production for the diplomat costing something like 250 (the girl with the phone ?). It meant that I could use my modifier bonus to produce something else while spamming these units 1 every 2 turns, using the 200 overflow to go into another building.

4. The roman arena was giving me permanent golden age with one in every city

5. I did not understand the warscore mechanic but that is possible I just need to read about it, I see it was getting up based on war results but I do not know where to go from there so I ignored it.

6. A ton of good ideas overall so goodjob despite my critics :).
 
[...]
3. Production of diplomat units is off after a time. With all the bonuses in my capital I was reaching some 450 production for the diplomat costing something like 250 (the girl with the phone ?). It meant that I could use my modifier bonus to produce something else while spamming these units 1 every 2 turns, using the 200 overflow to go into another building.
[...]

Yeah, I have the same opinion on this one. Production costs could be upped (late game is so serious spam of diplo units anyway, it could even improve performance a bit). And some of these +X% :c5production: toward diplo units are unnecessary.
 
always 8% behind her.
2nd in tech with 4 capitals

I am surprised you were that close to her. In my emperor game atm I am a similar amount behind the tech leader but with only 1 enemy capital. It seems the AI is not that good at war, in the current version. I can crush most people if I go up against them, but if I play it peacefully they out-tech me easily.
 
One oddity I've noticed with the warscore mechanic. (the actual game-version the thread was created for, btw) I have +90 warscore with my neighbor, the warscore window suggests a 'total surrender' yet all he is willing to pay is 75% of his current gold (which is a laughable sum),
He is currently sitting at about 170 GPT, and around 350 flat gold, he is offering me about 270 for peace. No cities, no capitulation, no gpt not even adding an embassy to it.
While I'm totally aware that the warscore isn't at 100, 'total surrender' should still mean 'total surrender' not 'we're willing to give you enough money for a bus-ride home'.
 
It doesn't but you need to understand one thing.If Gazebo is to continue with AI improvemnts it will be game perfomance vs AI perfomance.What I meant is that without proper means to debug AI with optimization in mind, calculation time AI needs will increase and it's quite noticable on low-mid range pc but it will perform better.
 
I have the feeling that there might smoething beeing broken with the AI in the current Version (or in the last one).
I am watching a lets play and the AI is not bad but really really horrible against the player on the naval map:

https://www.youtube.com/watch?v=8Q5ueWOSkkU&list=PLH-huzMEgGWBUU5NcRJZ7Iss4nE3jfHh4

Watch video 11-13 and see how Hiawatha declares war without having a single ship against a rather powerfull fleet.
The Same goes for England in Videos 14/15 - they have plenty of land units but almost NO navy.
I know the AI is really difficult to improve, but to me this feels like it totally neglects the fact that it is playing a Sea/Naval based map.

Also when you watch the earlier videos, is it intended that citie states are so easily conquered with out any serious repercussions?
 
I have the feeling that there might smoething beeing broken with the AI in the current Version (or in the last one).
I am watching a lets play and the AI is not bad but really really horrible against the player on the naval map:

https://www.youtube.com/watch?v=8Q5ueWOSkkU&list=PLH-huzMEgGWBUU5NcRJZ7Iss4nE3jfHh4

Watch video 11-13 and see how Hiawatha declares war without having a single ship against a rather powerfull fleet.
The Same goes for England in Videos 14/15 - they have plenty of land units but almost NO navy.
I know the AI is really difficult to improve, but to me this feels like it totally neglects the fact that it is playing a Sea/Naval based map.

Also when you watch the earlier videos, is it intended that citie states are so easily conquered with out any serious repercussions?

Naval operations have always been on our list, and trust me, it's doing better now than it was before this all started (the AI was actually unable to move naval units offensively prior to the CP). That said, ilteroi and I have been working a bit more on the naval operations lately, so expect some improvements soon.

G
 
Watch video 11-13 and see how Hiawatha declares war without having a single ship against a rather powerfull fleet.
The Same goes for England in Videos 14/15 - they have plenty of land units but almost NO navy.
I know the AI is really difficult to improve, but to me this feels like it totally neglects the fact that it is playing a Sea/Naval based map.

The random DoWs are from my experience usually other AI bribing them to attack, that's probably the reason why AI cities keep switching owners without actually going to war.
I was thinking that there there should be be some kind of block that stops the AI from declaring a losing war, even if they were paid for it. But then I realized, that the main problem with the entire game is that the AI just keep underestimating the player. They probably still use that non-sense vanilla calculation for military might, meaning that even if they have nothing and are sure to lose the war, they will still accept the bribe. This is actually a huge game-play issue, if the AI can't properly assess your military power or realize the fact that they don't stand a chance in a war, that is going to lead them into making extremely stupid decisions.




Also when you watch the earlier videos, is it intended that citie states are so easily conquered with out any serious repercussions?
Is the guy playing an outdated version? Because in my current game, mid-renaissance era city-states have 90 combatstrength, making them sturdy as all hell. I mean it took me like 15 turns with 7 cannons to capture one.
 
no capitulation
Yes I sort of expected a 90+ warscore to result in capitulation when they are down to remaining cities.
Perhaps they don't capitulated to me because they hate me so much, but then I at least expect them to look for a vassal state agreement with someone with a large army when I sue for peace. The result is instead, of course, that some other powerful army DoWs them and takes their last city(ies) instead.
 
It doesn't but you need to understand one thing.If Gazebo is to continue with AI improvemnts it will be game perfomance vs AI perfomance.What I meant is that without proper means to debug AI with optimization in mind, calculation time AI needs will increase and it's quite noticable on low-mid range pc but it will perform better.

I7-4790 + 16 Gb DDR3 1600 + GTX 760 1 GB DDR5, mid end??? Nahhh.

Something is looping badly for some AI turns, which was not happening before... and I'm sure G will find it, if not already. But it is bug for sure, and a bad one.
 
I7-4790 + 16 Gb DDR3 1600 + GTX 760 1 GB DDR5, mid end??? Nahhh.

Something is looping badly for some AI turns, which was not happening before... and I'm sure G will find it, if not already. But it is bug for sure, and a bad one.

Seems to be one particular AI in my game right now. Seemed to start when I put a spy in their capital. Perhaps it is moving units past there a lot?
 
Trying out the new beta version now, but haven't put enough time into it to comment. But I just wanted to say, that while I have gotten the Diplomacy Screen CTD (non-beta version), I haven't had any issues with turn time. My current non-beta game, it is turn 305 (epic speed), every civ has a navy, the aztecs have about 10 trillion units, and my turn times are still under 30 seconds.

My specs are poo by today's standards, more or less. i5 2500k, 8gb RAM, game sits on a 1TB 7200rpm HD.
 
Trying out the new beta version now, but haven't put enough time into it to comment. But I just wanted to say, that while I have gotten the Diplomacy Screen CTD (non-beta version), I haven't had any issues with turn time. My current non-beta game, it is turn 305 (epic speed), every civ has a navy, the aztecs have about 10 trillion units, and my turn times are still under 30 seconds.

My specs are poo by today's standards, more or less. i5 2500k, 8gb RAM, game sits on a 1TB 7200rpm HD.

Thanks. The turn time issue is most likely a rare break in a loop somewhere. We'll find it, hopefully.


G
 
A small thing - for some reason, when I am using the beta patch, all yields show up as food (hammers/gold/etc). They are the real thing (so I do get hammers from forests) but just all food. I simply cut the non-beta folder and moved it, and dropped in the beta folder. I deleted the cache too. It's a small thing, I just wondered if I just missed a step (I even followed the PDF how-to guide). If not it's okay, because I'm just trying to test out the beta and a UI thing isn't going to stop me. :-)
 
A small thing - for some reason, when I am using the beta patch, all yields show up as food (hammers/gold/etc). They are the real thing (so I do get hammers from forests) but just all food. I simply cut the non-beta folder and moved it, and dropped in the beta folder. I deleted the cache too. It's a small thing, I just wondered if I just missed a step (I even followed the PDF how-to guide). If not it's okay, because I'm just trying to test out the beta and a UI thing isn't going to stop me. :-)

Did you install the old beta version as well? The one that changed tradition to reduce the foodcost of specialists by 50%?
 
Did you install the old beta version as well? The one that changed tradition to reduce the foodcost of specialists by 50%?

I don't think so? I got it from G's post #104. And the tradition policy still says 5 specialists for free. So I'm not sure. :-/
 
Status
Not open for further replies.
Back
Top Bottom