C2C : Game Speed

[EldrinFal]

Pushed the changes to the SVN.

I used Snail as the reference speed keeping (most) settings as close as possible to what is already there. For the other speeds, I adjusted the values to make them proportional.

Turns and increments were NOT changed, as to protect current save games. However, they DO need adjustment, and can hopefully be corrected at some point.

 

[BlueGenie]

Okay... I'm going to demand/request all the people complaining eternity is too short a game by speed and the average tech takes 6 turns explain what their person skill level is set on, asuming custom games.

Playing on Immortal. Gone down from Deity since Koshling's AI modifications have made the AI better. Not the top researcher in my current game (Huge Eternity game with std amount of Civs, Minor Civs on, Tech Diffusion on).

Of the Techs I can choose now (after finishing Optics) I have:
2 techs for 1 turn.
3 techs for 2 turns.
2 techs for 3 turns.
5 techs for 4 turns.
2 techs for 5 turns.
3 techs for 6 turns.
1 tech for 7 turns.
18 techs for 69 turns is 3.8 turns average.

What you have to remember too is Tech Diffusion. Choosing to research more advanced techs gives me some Tech Diffusion on the back end techs that I'm leaving a little while (the 1 through 3 turn techs).

The lone 7 turns tech is Astronomy. One AI has had that a while, two others have recently got it so I'm going to be #4 to it.
Tech Diffusion helps me from roughly 8.5 turns to 7 turns. Roughly because I'm not sure what the Era and Difficulty itself does to my research points.
Oh, 3181 tech per turn. Rising.

Hope that helps some.

Cheers

 

[Questdog]

Report on my Eternity game with research set to 2000:

It is now January, 1943 and I just completed researching Priesthood. THIS is what an Eternity game should be! Currently teching about once every 12 turns. Still not enough time to build EVERYTHING, but most things can now be built, at least in my capital.

Playing at Monarch level on a Gigantic map with 40 civs (though we are now down to around 30).

 

[45°38'N-13°47'E]

If I can add my 2 cents... (note, I'm not playing with the latest SVN)
I've just made a test game on a large map, 6 or 8 civs starting, no rev and snail speed.
I've left the game autoplay for 4500 turns (almost 24 hours) and I've discovered the last tech. Since snail speed has 9900 turns, what am I supposed to do for the remaining 5400 turns?
I wonder, has ANYBODY every finished a game with C2C at snail or eternity speed?
From what I've seen in the xml speed file, Eternity, Snail, Marathon and Epic games are supposed to end in 3000AD (180.000 months); Normal speed is off-scale as it ends in 3060AD (if I'm not mistaken). Anyway the point is that with Snail speed, for example, I finish researching every tech well before half the turns I'm supposed to play. I guess NOBODY can really desire to play 5500 more turns without researching anything...

@Questdog, researching Priesthood in Jan 1943 isn't an eternity game: it's a game completely off scale. It's obvious there should be more turns before 1AD and less turns after; or better, turns should be completely reworked to at least try to match real world research. Researching Priesthood in 1943 is totally unacceptable: either we remove the year and leave only the turn indicator or I think we have to find a solution to this unrealistic year/research match.

One more thing: at snail speed, 95% of research with EVERY other rival civ being vassal to my civ (I have no more rivals so I have already won although I need to wait 5500 turn to reach victory...), I'm researching more than 400.000 per turn: nonetheless, I needed around 50 turns to research Euclidean 5-Space Geometry. Are you sure this is what you want when playing at snail speed (not to mention eternity...)?

 

[Questdog]

@Questdog, researching Priesthood in Jan 1943 isn't an eternity game: it's a game completely off scale. It's obvious there should be more turns before 1AD and less turns after; or better, turns should be completely reworked to at least try to match real world research. Researching Priesthood in 1943 is totally unacceptable: either we remove the year and leave only the turn indicator or I think we have to find a solution to this unrealistic year/research match.

I upped the research times, but did nothing to try to get the date to match techs. That would not be a difficult matter to fix once I settle on what the proper speed should be. Actually, that game has turned out to be a little to fast for me. I am now teching about once every 10 turns, which is too quick for Eternity, in my opinion.

 

[Necratoid]

I managed to finish an novel game on eternity... granted I bought a fusion generator outright in a new city (my contient had no aluminum or sulfer... yeah!) and eventually ran out of things to build and did this to counter FB issue is v17. I would have won anyway eventually. I completely forgot cutural victory was on as I was going for space race. That was May 1996.

I do wonder now that it came up... am I the only one with tech diffusion disabled? If so that would explain a few things.

 

[Questdog]

I managed to finish an novel game on eternity... granted I bought a fusion generator outright in a new city (my contient had no aluminum or sulfer... yeah!) and eventually ran out of things to build and did this to counter FB issue is v17. I would have won anyway eventually. I completely forgot cutural victory was on as I was going for space race. That was May 1996.

I do wonder now that it came up... am I the only one with tech diffusion disabled? If so that would explain a few things.

I play with no Tech Diffusion. The game already includes this by making a tech cheaper the more civs that research it. I also play with no tech swapping.

I basically try to turn off all options that make the game harder on the AI. That's why I don't play with Revolutions on. It's always the AIs that are kicking my butt that get hit with revolts and just makes it easier for me to catch up.

 

[45°38'N-13°47'E]

I managed to finish an novel game on eternity...[cut] That was May 1996.

Sorry, my question wasn't clear: I was wondering if anyone has ever won a mastery victory on snail or eternity. Of course it's pretty easy to win other victories; if you've won in 1996 it means that you were not at 20% of the game yet. My point was that there are too many useless turns at the end of the game because with probably every speed longer than Marathon you can obtain every tech before you reach 50% of the turns in the game. So there's no point in a mastery victory unless you want to play at least 5000 turns doing absolutely nothing...

 

[AIAndy]

Pushed the changes to the SVN.

I used Snail as the reference speed keeping (most) settings as close as possible to what is already there. For the other speeds, I adjusted the values to make them proportional.

Turns and increments were NOT changed, as to protect current save games. However, they DO need adjustment, and can hopefully be corrected at some point.

There is very little usage of game year so it should be fine to change the increments.

 

[Necratoid]

I have to play with Revs on... becuase if I don't I end up with a spiritual leader with 14 cities when I'm making my fourth.... by medival era I'm looking at a 24-30 odd city abomination (on a large or huge map) that I have to destroy 3-8 Civs to get to. Eventually I end up with that Civ doing absolutely nothing but making 3-5 units a turn from every city and fighting a stupid war of attrition that involves all my terrian improvements being obliterated and any unit that attacks the stack of Doom (one of 4) of knights and elephants (usually riflemen for me) and the RNG lets me kill none of them.

Granted this was v17... but I got a 120+ neanderthalls and 20+ other units in a city in the early ancient era in v18... so I'm expecting that to be the same. The main issue I find AIs have with REV is that they keep switching civics constantly (v17 again)... this means they keep getting large amounts of disorder as the rank and file can't even remember what kind of society they live in. The Civ Leader reacts by changing the civics again.

This means that they never bleed off the instability and control it through spamming troops into their cities for control. The Spiritual leader has 0 turn civic switches so they never accumulate the disorder and what little does get made is bled off by their leadership trait.

Anyway, to be clear the only reason I could get done so fast was v17 fusion plants added 10k to the city and that can be sent directly into research. Without it it would have been the late 20 teens or so before I was to the tech level I got to. So that distorts the game math a lot.

So as so many people think the tech is to fast... at which point is it that you think that starts?

 

[Questdog]

So as so many people think the tech is to fast... at which point is it that you think that starts?

For me it is too fast right from the get go on Eternity, but really gets too fast at the Ancient Era when I am teching every 6 turns. I mean, in Vanilla Civ on STANDARD speed if you can tech every 6 turns you are cranking it. I would like to play at a speed where it is not possible to tech in under 10 turns with 15 to 20 being the usual to expect if you are going all out for science.

 

[JosEPh_II]

For me it is too fast right from the get go on Eternity, but really gets too fast at the Ancient Era when I am teching every 6 turns. I mean, in Vanilla Civ on STANDARD speed if you can tech every 6 turns you are cranking it. I would like to play at a speed where it is not possible to tech in under 10 turns with 15 to 20 being the usual to expect if you are going all out for science.

Lets compare apples to apples please. (red)Vanilla CIV IV on Standard has 84 techs to research (and you don't have to research them all either to win). How many techs are in C2C? At last count almost 4X as many, with preq's included so that you have to research most of them, but not all yet. How many turns per Era do you get? Should Era turn increments even be adjusted for different game Speeds? Is there too many turns at the slower levels giving those players a 'false" sense of time?

(blue)Maybe Eternity is Not the speed you should be playing? Maybe you should go back to Epic to refresh your expectations? Change your play style perhaps?

JosEPh

 

[Questdog]

Lets compare apples to apples please. (red)Vanilla CIV IV on Standard has 84 techs to research (and you don't have to research them all either to win). How many techs are in C2C? At last count almost 4X as many, with preq's included so that you have to research most of them, but not all yet. How many turns per Era do you get? Should Era turn increments even be adjusted for different game Speeds? Is there too many turns at the slower levels giving those players a 'false" sense of time?
JosEPh

By that reasoning, it would be okay to assume that tech rate in C2C at Marathon speed could be consistent with Vanilla Civ at Standard. I'm not talking Marathon speed; I'm talking ETERNITY, for goodness' sake. How can anyone argue that 6 turns per tech is a reasonable approximation of ETERNITY?

I want to play a game where when something upgrades it is a big deal and reason to close the banks and send everyone to the pubs to celebrate. When I get a new unit I want to be able to use it for a good long time before it obsoletes. I want to be able to complete construction of buildings before their upgrades come on line. In short, I want to experience all the goodness and savor it; not rush through the eras sampling bits and pieces along the way. And when the building of units takes longer than the research for the next tech, it is impossible to accomplish this.

 

[Thunderbrd]

The point you seem to be making then is that we need a slower speed setting still beyond eternity. I actually agree. I've often thought we should just take our current game speeds and shift the names over one category such that eternity now gets the snail speed settings, and we create a new eternity setting while establishing a 'Lightning' speed setting just faster than blitz for the new blitz.

This might make it feel a bit more like the game players are expecting when they select a particular speed option.

 

[AIAndy]

The point you seem to be making then is that we need a slower speed setting still beyond eternity. I actually agree. I've often thought we should just take our current game speeds and shift the names over one category such that eternity now gets the snail speed settings, and we create a new eternity setting while establishing a 'Lightning' speed setting just faster than blitz for the new blitz.

This might make it feel a bit more like the game players are expecting when they select a particular speed option.

That sounds good. Both renaming and adding a new speed at the end are easy to do and do not break existing save games.

 

[EldrinFal]

There is very little usage of game year so it should be fine to change the increments.

Ah. I had not done so on your words of warning before. But if you think we're all good in changing it, I'll do so.

The point you seem to be making then is that we need a slower speed setting still beyond eternity. I actually agree. I've often thought we should just take our current game speeds and shift the names over one category such that eternity now gets the snail speed settings, and we create a new eternity setting while establishing a 'Lightning' speed setting just faster than blitz for the new blitz.

This might make it feel a bit more like the game players are expecting when they select a particular speed option.

I brought this up just recently (see previous page in thread) and AIAndy said removing speeds could cause issues. I think I proposed the exact same thing you suggested. Great minds think alike I suppose

On the topic in general, I just barely started a Snail game and there is a definite difference in the ratio of research to building than there was in my pre-modified GameSpeed. So we're on the right track. The values still need tuning though. And it might not just be the GameSpeed but the Era file also because there are mutlipliers there as well. i.e. if the ratio is good in Prehistoric and Ancient, but then it gets way too fast in Medieval and Industrial+ then those eras can be slowed down.

This is all great feedback by the way. For such a game-wide issue, it's impossible for just 1 person to check when games take so long.

 

[Questdog]

That sounds good. Both renaming and adding a new speed at the end are easy to do and do not break existing save games.

It won't break save games, but it will change the speed they are currently playing.

 

[AIAndy]

It won't break save games, but it will change the speed they are currently playing.

The speed won't change because for the existing game speeds it is just a rename.

 

[Thunderbrd]

When I did the first edit on gamespeed for the mod, I noticed that the afforess version hadn't maintained equality in research and production etc... ratios. I thought that odd and that it may cause trouble down the road.

But I didn't change it cuz I was just trying to make it so that:

a) the tech progress tended to match fairly close to Theocracy achieved on or around 0 AD.
and
b) we no longer had month increments being introduced before AD (causes a nasty effect as bad or worse than a crash) in any gamespeed

Now, since then, I believe we've had a few more adjustments made to the gamespeed file and I've been trusting they've been made with competence as I had faith in Praetyre's abilities there. I'm not sure if anyone else has touched the file since.

But aligning the % of production, research, etc... to the same ratio difference as the game speed increments is something we should streamline still yes.

So the idea to create a shift doesn't cause any trouble because we AREN'T removing a gamespeed... we're adding a new one - just, in name, doing so at the fast end and calling it lightning (but we're not making a faster speed, just changing the gamespeed increment NAMES to adjust them all one category so yeah, individual ongoing games would just suddenly have a different gamespeed name, not an actual new gamespeed definition (unless we do some tinkering there too.))

Then we're adding a new gamespeed at the end of it all and making it, what, twice as long? I mean, is this enough though? Should we have two increments difference? One being 1 and 1/2 as long and one being twice as long as the current longest gamespeed (eternity)? In the latter, we also add one at the end (in name) and call it Neverending.

Now, if we can come up with a consensus, in all fairness, who would be the one to make the final decision that the idea is a go and who would make the adjustments in the gamespeed file?

 

[AIAndy]

b) we no longer had month increments being introduced before AD (causes a nasty effect as bad or worse than a crash) in any gamespeed

I fixed that bug so you can have month increments before AD now.