Caveman2Cosmos Let's Play - AstralPhaser/12padams

40 sec for the AI isn't long. Unless you are doing a video Let's Play that is. ;)

Are you feeling better?

JosEPh :)
 
I'm no longer vomiting... Now I have other symtoms which I would prefer not to discuss. I guess I'm out off editing even more now that I'm sick...

I should be better soon though... Ready to make more epsiodes of my lets plays and a new one that's coming up...
 
I hope all your childhood vaccinations are up to date. Considering your line of work.

JosEPh
 
Boy its been a long time... here is episode 25

Episode 25: http://youtu.be/7Nz2ol_lakY
I may have just wiped out the Scandinavians but I am no where near done yet! Some of my cities are still yearning to rejoin the Maori Civilization so what should I do? Destroy the last two Maori cites so that my people no longer complain about wanting to join that puny civilization.


Please note however that I have lost the mood for caveman2cosmos right now so I will work on my other lets plays to get back into the mood again a little later.

I am doing a multiplayer caveman2cosmos lets play though so the first epsiode of that will be coming soon...
 
12padams said:
Boy its been a long time... here is episode 25

Episode 25: http://youtu.be/7Nz2ol_lakY
I may have just wiped out the Scandinavians but I am no where near done yet! Some of my cities are still yearning to rejoin the Maori Civilization so what should I do? Destroy the last two Maori cites so that my people no longer complain about wanting to join that puny civilization.
Click to expand...

Always try to go for complete wipe-outs if possible. Especially with REV on.
 
Koshling said:
Always try to go for complete wipe-outs if possible. Especially with REV on.
Click to expand...

Ha, it isn't always that easy. My game is at about the same point his is (I just got Christianity), and I currently, on Immortal, have 4 rivals bordering me with 20 cities a piece (I have 21), and another couple farther away that have 25. It could be a really competitive game, thanks to all of your AI improvements. The only thing that is lacking is that the tactical AI almost never sends an invasion at me when I'm at war, and when it does it lingers on damaging terrain (I thought I'd fixed that).
 
ls612 said:
Ha, it isn't always that easy. My game is at about the same point his is (I just got Christianity), and I currently, on Immortal, have 4 rivals bordering me with 20 cities a piece (I have 21), and another couple farther away that have 25. It could be a really competitive game, thanks to all of your AI improvements. The only thing that is lacking is that the tactical AI almost never sends an invasion at me when I'm at war, and when it does it lingers on damaging terrain (I thought I'd fixed that).
Click to expand...

Yeh, the economic aspects of the AI are in a pretty decent state now, but the military aspects still suck. Always more to do...
 
ls612 said:
Ha, it isn't always that easy. My game is at about the same point his is (I just got Christianity), and I currently, on Immortal, have 4 rivals bordering me with 20 cities a piece (I have 21), and another couple farther away that have 25. It could be a really competitive game, thanks to all of your AI improvements. The only thing that is lacking is that the tactical AI almost never sends an invasion at me when I'm at war, and when it does it lingers on damaging terrain (I thought I'd fixed that).
Click to expand...

I ran across a section of code in CvUnit that will force units that are beyond 50% damage to stay put on a damaging terrain if they are also surrounded by damaging terrain. This could be one reason for that lingering - I have put some notes there so after my next update, scan through the changes in CvUnit to find this section.
 
Thunderbrd said:
I ran across a section of code in CvUnit that will force units that are beyond 50% damage to stay put on a damaging terrain if they are also surrounded by damaging terrain. This could be one reason for that lingering - I have put some notes there so after my next update, scan through the changes in CvUnit to find this section.
Click to expand...

All AI for that should be in CvUnitAI, not CvUnit.
 
The fact that its NOT AI is what's probably giving you trouble. It's set up like a game rule that only applies to AI units.

Under 'bool CvUnit::canMoveInto(const CvPlot* pPlot, bool bAttack, bool bDeclareWar, bool bIgnoreLoad, bool bIgnoreTileLimit, bool bIgnoreLocation, bool bIgnoreAttack) const'

you have:

Code: 
	if (!isHuman())
	{
		if (plot()->getTerrainTurnDamage() <= 0)
		{
			if (getDamage() > 50)
			{
				if (pPlot->getTerrainTurnDamage() > 0)
				{
					return false;
				}
			}
		}
	}
Could that be causing you just a bit of trouble? Tell you what... I'll just comment that whole bit out because I know its not right.

Then again, on second glance here, looking at it a bit more closely, it's supposed to make it only checks against this IF the plot its on is not damaging itself. Huh... maybe has little to do with it then.
 
Thunderbrd said:
The fact that its NOT AI is what's probably giving you trouble. It's set up like a game rule that only applies to AI units.

Under 'bool CvUnit::canMoveInto(const CvPlot* pPlot, bool bAttack, bool bDeclareWar, bool bIgnoreLoad, bool bIgnoreTileLimit, bool bIgnoreLocation, bool bIgnoreAttack) const'

you have:

Code: 
	if (!isHuman())
	{
		if (plot()->getTerrainTurnDamage() <= 0)
		{
			if (getDamage() > 50)
			{
				if (pPlot->getTerrainTurnDamage() > 0)
				{
					return false;
				}
			}
		}
	}
Could that be causing you just a bit of trouble? Tell you what... I'll just comment that whole bit out because I know its not right.

Then again, on second glance here, looking at it a bit more closely, it's supposed to make it only checks against this IF the plot its on is not damaging itself. Huh... maybe has little to do with it then.
Click to expand...

It doesn't explain ls612's case, but well spotted - this code absolutely should NOT be here - I think it's an original sticky-plaster for the terrain damage issue that has long since been superseded.
 
That particular code doesn't actually do anything. The first if is passed only if the terrain damage value is <=0. The second, inside the first, only if it is > 0. It can not both be <= 0 and > 0 therefore it will never do the "return false". The entire block of code is just a waste of time, especially since it must be checked every time a unit moves which happens a lot.
 
Koshling said:
It doesn't explain ls612's case, but well spotted - this code absolutely should NOT be here - I think it's an original sticky-plaster for the terrain damage issue that has long since been superseded.
Click to expand...

I'll get rid of it.

Edit: Also, I just saw God-Emperor's post, I'll see how much performance I net from this.
 
God-Emperor said:
That particular code doesn't actually do anything. The first if is passed only if the terrain damage value is <=0. The second, inside the first, only if it is > 0. It can not both be <= 0 and > 0 therefore it will never do the "return false". The entire block of code is just a waste of time, especially since it must be checked every time a unit moves which happens a lot.
Click to expand...

No... plot() is the plot the unit is currently on while pPlot is the plot that's being evaluated to see if the unit can move onto that plot. So this COULD strand a damaged unit on a non-damaging terrain if otherwise surrounded until the unit heals, which isn't really all that bad a thing.

The basis of the problem is that its basically AI in gamerule form and that doesn't seem to be the correct manner in establishing AI decision making.
 
They mess me up often (forgetting them usually... lol)

I'm just surprised I spotted something YOU hadn't considering you're usually amazing at seeing the Devil in the Details!
 
Ok I am thinking of continuing this lets play with a single modification in the way I record it... Basically in the past I used to record everything including ai turn time then spend a painfully boring hour after recording each epsiode cutting out the ai turn time in movie maker.

Now, to make things easier I am thinking of pausing the recording after my turn so I am not recording the ai turn time then unpausing it when it's my turn again. By doing this I can simply upload the epsiode after recording rather than spending an hour editing which puts my off the game entirally. The downside is that by doing this the turn cut off points will be choppy rather than seamless as they have been in the first 25 epsiodes.

So what do you all think? Do you accept my proposal of continuing the lets play with choppier turn transitions?
 
That sounds like it would be great to me. Anything that lessens the burden of editing and makes the Let's Play more fun for you to record.
 
Ok after a long time here is a new episode... the last one was nearing 100 views so I guess it this series is popular enough to continue now.

Remember that I now don't record AI turn time. Before I used to record them and chop them out but now I don't record them at all. This means choppy turn transitions from now on. Also due to some boring balancing I skipped a few turns of recording every now and then.

Episode 27: http://youtu.be/T5CGPbaalxU
Wiping out all those civilizations and taking over their cities has left me financially insecure. In this episode I work towards stabilizing my empire and working towards building a fleet of ships to explore the oceans and discover new lands.
 
You must log in or register to reply here.

Similar threads

BlueTemplar
[Stream] [YouTube] Hadriex plays Neanderthals
Replies
0
Views
604
BlueTemplar
BlueTemplar
CanusAlbinus
PolyCast 439: More to Believe In [Part 2]
Replies
4
Views
666
Trav'ling Canuck
Trav'ling Canuck
CanusAlbinus
PolyCast 440: PAX and Direct
Replies
4
Views
1K
TheMeInTeam
TheMeInTeam
CanusAlbinus
PolyCast 438: Something to Believe In [Part 1]
Replies
2
Views
552
CanusAlbinus
CanusAlbinus
CanusAlbinus
PolyCast 435: So It Shall Be
Replies
0
Views
623
CanusAlbinus
CanusAlbinus
Back
Top Bottom