Slitherine

the text descriptions for the other factions are interesting - but
I don't see a description of my own faction, such
as used to exist in previous versions.

The descriptions for military, economy and research are relative to your strength, so you can immediately see how strong they are in comparison and don't have to look up two entries.

void wrote:The descriptions for military, economy and research are relative to your strength, so you can immediately see how strong they are in comparison and don't have to look up two entries.

Not too fan about this change, it seems the mechanism is exactly the same than previously but the text makes it less obvious to visualize the situation, so I am not sure this is an improvement. Perhaps the text could be added as an hoovering text over the previous dot if user want to understand better the meaning of the dots?

I just (nearly) finished a game with the religious faction (hard difficulty, medium pangea) on 1.1.8 and I wanted to share my comments :

Basically I haven't seen any difference in the army composition of the AI, so I am not sure the modification on refit actually changed anything. AI still spammed loads of crappy infantry and basic aircraft (around strength 11). Imperium was about to dominate the map and I intervene to stop him. He had been warmongering since the very beginning, so he didn't had to much basic colonial trooper left, however he only had unit with around 10 strength while I was pumping out unit of strength around 20. Solar dynasty didn't made any war during the game and their army was made at 100% of basic colonial infantry and strength 6-8 aircraft.

Both of them had no mecha era tank or plane, no artillery, no anti aircraft units and no fully upgraded units. They were both at early transcendence era but still using colonial chassis / armor and early mecha weapon.

On the diplo side it was one of those game where from the start I got good relation with all faction, so I was absolutely not worried the entire game and was able to pick the time of my choice to strike, which of course made thing even easier.

The Ai did this nonsense city spamming wherever a spot was available, making the map totally chaotic with no clear border and units running all around the place with no unity, once again making thing easy for a human player to pick them up. On this topic AI will never hesitate to send his plane to a guaranteed suicide right in the middle of my lands with no support, so picking them up is really easy, and later on when the tanks and infantry arrives you already slaughtered half their army with little problem. Some kind of scripts making the AI regrouping their troop before attacking and diversify their troops type would be welcome to spice things up.

I created a thread, then remembered that this thread was already created so I deleted it. Here is what I said:

I like the changes to the diplomacy window, except for one thing. I really miss the the ability to quickly see the progress among factions. I'm okay with getting rid of the "dots" but I think replacing it with "all text" is a step backwards.

Also, the only way to see how my military, tech, or economy compare, I have to go to those individual windows and read the text of the advisors. Now, I don't mind reading, but for something like this, I like to be able to glance over real quick and grab a lot of information in a few seconds instead of playing "Where's Waldo, Text Edition" to find key words like "outdated techs" to get an idea of how things are going.

Other thought: You know how there are two tabs in the economy window? There should be two in the diplomacy window; the second tab would be charts/graphs. At the end of every game, there is a progress graph that shows the factions over time. Why do we have to wait till the end of the game to see that chart? There could be a military, tech, economy, territory, and an overall chart in this new tab.
-Military Chart: accumulative sum of all unit power (3-turn rolling average) vs. time.
-Tech Chart: total number of research points accumulated over the course of the game vs. time.
-Economy Chart: monthly income (6-turn rolling average) vs. time.
-Territory Chart: number of territories controlled vs. time.

Before someone complains and says that some of these (such as the research chart) will be an exponential curve and that I wont like it... I'll just say that the SMAC progress chart also had an exponential curve and it never bothered me. In fact, I loved it. There is also the possibility of bar charts (not over time) that just show the current state of each faction against each other. Such charts can convey a ton of information in 2 seconds which is a lot better than playing "Where's Waldo."

jdmillard wrote:.

...

Also, the only way to see how my military, tech, or economy compare, I have to go to those individual windows and read the text of the advisors. Now, I don't mind reading, but for something like this, I like to be able to glance over real quick and grab a lot of information in a few seconds instead of playing "Where's Waldo, Text Edition" to find key words like "outdated techs" to get an idea of how things are going.

...

+1

Hi boulugre,

Do you per chance have a save for these aircraft situations? I'm worried that the pillaging results in units suiciding too much.

Also for the refiting, saves for situations where you think there should have been refits performed would go a long way. It could be that the refit budget is simply too low (he starts allocating for refit once he has units equal to or more than than the recommended minimal unit count, which is at least one unit per city, and more for larger cities).

boulugre wrote:

jdmillard wrote:.

...

Also, the only way to see how my military, tech, or economy compare, I have to go to those individual windows and read the text of the advisors. Now, I don't mind reading, but for something like this, I like to be able to glance over real quick and grab a lot of information in a few seconds instead of playing "Where's Waldo, Text Edition" to find key words like "outdated techs" to get an idea of how things are going.

...

+1

Is this just the text instead of the dots being a problem?

The advisors tell you how you compare against the average of the other factions, the diplomacy panel shows how you compare against the individual factions.

void wrote:Is this just the text instead of the dots being a problem?

That is part of it. It's not that the text itself is that bad, it's that I'm so used to being able to quickly grab a lot of information with a single glance. In my current game I'm struggling to know who is doing the best and who is lagging behind.

void wrote:The advisors tell you how you compare against the average of the other factions, the diplomacy panel shows how you compare against the individual factions.

The concept is great. But the if the diplomacy panel is there to show how we compare to other factions individually, then the text ought to be worded as such. For example: instead of "outdated techs" it should say "their tech is years behind us" (keyword = "us").

If the other advisors are there to show how we compare to the average of other factions, add a simple bar chart in each advisor window that depicts the relative position of the factions with respect to that advisor's topic. You wouldn't have to put numerical values on the bar chart y-axis (though the heights of the bars would be derived from real numbers), it would show where everyone stands relatively (and thus give an intuitive average as well). You don't have to clutter it with labels, just color each bar the appropriate faction color. It wouldn't have to be huge, put it in the bottom-left corner below the advisor's comment. In my example below I only did 4, but it's the general idea.



I like to be able to grab a lot of information quickly. If some people don't like it, make it disable-able (it's a new word).

Excellent idea with the bar chart. Although if they ever get around to adding
more factions, more space might be needed.

And let's add data for total populations for each faction. I'm trying to win
a pacifist "military" victory by controlling 75% of the population by building more
cities and growing them; but I have no feedback on my growth relative to other
factions. I would appreciate some info on opponents' number of cities and total
population.

void wrote:Hi boulugre,

Do you per chance have a save for these aircraft situations? I'm worried that the pillaging results in units suiciding too much.

Also for the refiting, saves for situations where you think there should have been refits performed would go a long way. It could be that the refit budget is simply too low (he starts allocating for refit once he has units equal to or more than than the recommended minimal unit count, which is at least one unit per city, and more for larger cities).

Erf no sorry, I didn't saved before starting the war, and the latest autosave I have is just a few turn before I stopped the playing once 90% of AI army was turned to dust. Ill pay attention to that in future game to make saves at relevant moments.

I am not sure if the plane behavior is due to to them wanting to pillage cause I had some aircraft well in my territories but haven't lost any improvement.

Concerning the army composition, I think the trouble is not only on refit but also on AI unit productions choice. In this game it was quite obvious Solar dynasty kept pumping out colonial infantry while having researched much better chassis/armor/weapon.

Also I don't understand why an AI who suffer loses during war will be more keen to produce better unit and more or less follow the tech tree improvement than a faction who sits in peace and just pump outs colonial infantry and plane??

IMO you should try to :

- Make the AI start producing their new chassis/armor/weapon as soon as they get them researched and obsolete older design in their workshop (at least obsolete outdated chassis) and see how AI would behave.

- Lower the costs of refit for all factions (human and AI) to allow an easier upgrade of troops. This will give a better advantage to teching factions and avoid large quantity of outdated troops running around.

- Raise the refit budget of the AI (and if necessary give some 'cheat' gold to AI in high difficulties level)

- Make the AI use all the weapons type available, especially artillery and Anti aircraft. They should also diversify use of unit devices, I think AI tends to mass product the latest one available

-Teach the AI to balance its troops between attack and defense. AI will always commit 90% of its troop to the front line, just keeping one or 2 units in their back cities. While in some situation a strong attack is a sound strategy, AI often end up wasting its troops in futile assault (especially against a cunning human who will vaporize them with nukes and artillery) and then counter attacking & taking over its entire lands is often a walk in the park. If the AI kept one Infantry, one tank, one artillery and one anti aircraft unit in each of their cities that would make them waaay more harder to conquer.

Perhaps AI could take an account a ratio of 1/3 of its troop kept in defense, 2/3rd in attack? Also Perhaps an AI at war with an opponent who has a large superiority in troops should know to stay 100% in defense until the strength ratio gets more in its favor ?

- Concerning Pillaging, perhaps a solution would be to lower the priority of AI human faction for this strategy (or maybe even suppress that behavior, keeping pillaging for alien units). After all a human faction wants to take over its opponent economy as untouched as possible. This would eliminate suicide behavior of AI units spreading themselves around to pillage and being easily destroyed by human player. If you want to keep some kind of damage to lands improvement by combat and war , why not introduce a X% chance of a tile improvement being destroyed when a unit is being destroyed on that tile?

Very sorry for the wall of text, but I think that it is the combat system that makes this game really good so making the AI a worthy opponent is absolutely key to keep the game interesting and the re playability value high. A very high difficulty AI should have a clear advantage in terms of number and quality of units, forcing human player to out brain them on the battlefield in order to win.

void wrote:...It could be that the refit budget is simply too low (he starts allocating for refit once he has units equal to or more than than the recommended minimal unit count, which is at least one unit per city, and more for larger cities).

(I could be wrong because I haven't seen the code) I think what you've created is a passive upgrade system where the AI looks each turn at the obsolete units to refit, but doesn't have enough money that turn, so it spends it on something else (perhaps another cheap unit, which is another problem in and of itself). An active upgrade system would be where the AI "targets" a unit and considers how much it will cost to refit, then proceeds to save for a few turns in order to upgrade it. I know that such an implementation might require a huge change in the way the AI spends its money, but I don't know... just throwing out my limited analysis.

Also, I don't think the AI should should set a goal to go and pillage territories. They should only do it if the odds are against them in attacking the city (so as to draw out the defenders). Pillaging shouldn't be the target, rather the "plan B". It's okay if the aliens randomly pillage though.

boulugre wrote:...Lower the costs of refit for all factions (human and AI) to allow an easier upgrade of troops. This will give a better advantage to teching factions and avoid large quantity of outdated troops running around.

I agree with everything you said, boulugre. I even agree with the above quote about lowering refit costs. The only thing is that the change has to be carefully made in order to avoid creating a situation where it's more advantageous to purchase a cheap colonial trooper and refit him the next turn into something amazing and have it cost much less than the original purchasing cost or the equivalent production cost of that awesome unit alone. Don't get me wrong: I totally agree with you, it's just that we don't want to destabilize the production aspect of the game. Idea: What if there was a non-linear relationship between the difference in production for a refitting and it's cost? For example, if there was an exponential relationship, then the formula could be carefully picked such that incremental upgrades are pretty cheap, but upgrading a colonial trooper to a decked-out precentor trooper would cost a fortune. Picking the right delicate values in the formula would make or break this method. Just an idea, it makes realistic sense too.

Then nothing would prevent you from doing several incremental upgrades for cheap...

BlueTemplar wrote:Then nothing would prevent you from doing several incremental upgrades for cheap...

Right, and taking many turns to do it. Time is a resource too.

Agreed, gold rushing unit and refitting should be more expensive than slow producing trough cities in order to avoid unbalanced strategy as you mentioned Jdmilard.

But I don't think the refitting formula needs to be very complicate either. Right now refitting is extremely expensive, its costs more to refit an unit than to rush buy it from a city. That doesn't make sense because it do not take in account the already existing value of the unit you want to refit.

An easy and rational way to implement refitting could be the following formula :

Taking a base of one production costing 2 gold (or whatever the credit production is)

(Cost of wanted model - cost of existing model) + 10 %

So you have a tank which costed you 100 hammer to initially build, you want to upgrade it to a model which cost 150 hammer, the refit cost would be
(300 - 200) + 10% = 110 credit.

This would make refitting more expensive than slow producing, so building cheap chassis to refit them could not be really exploited, while making refitting way more affordable. Of course the 10% would need to be tested to find the right value.

1 production makes 2 credits, but to rush-buy you need 8 credits for 1 production. Also when you sell a unit, you get 1 credit for 1 production (that's 12.5% of the credit cost and not 25% like the tooltip says).

So, in your example, you could disband the tank mk1 for 100 credits, but it would cost 1200 credits to rush-buy the tank mk2, making it 10 times less expensive to upgrade (according to this formula) than to disband, then rush-buy.

But I don't think it's normal to be able to refit a unit to a different unit at no cost, even if they have the same production cost, like your formula makes it.

BlueTemplar wrote:...But I don't think it's normal to be able to refit a unit to a different unit at no cost, even if they have the same production cost, like your formula makes it.

Yeah, it's probably more complicated than we think. But I've never really analyzed the patterns of the current system, only made passive observations (like how dang expensive it is). I think the devs came out and gave some details about how the refit stuff works (back during a discussion about producing Wealth). But i don't remember what they said and I'm too lazy to look it up right now.

BlueTemplar wrote:1 production makes 2 credits, but to rush-buy you need 8 credits for 1 production. Also when you sell a unit, you get 1 credit for 1 production (that's 12.5% of the credit cost and not 25% like the tooltip says).

So, in your example, you could disband the tank mk1 for 100 credits, but it would cost 1200 credits to rush-buy the tank mk2, making it 10 times less expensive to upgrade (according to this formula) than to disband, then rush-buy.

mmm yes you are right, refit needs to be calibrated by taking in account other game function to keep it balanced. Whatever balance the dev's decide to implement, ultimately it should make refitting easier and more common for the AI.

IMO the relative costs of each hammer from cheap to most expensive should be

- Slow produce units in cities
- refitting units
- rush buying them

When you slow build something you use you own producing means and personal, so it makes sense for it to be cheaper. When you refit a military unit you basically already have the manpower, the organization, the barracks, logistics etc etc... you are just buying new equipment and train the soldier to use it. Instantly rush buying a unit makes more sense to be expensive as you have to build a new unit from scratch with all what it involves.

Disbanding shouldn't provide you much money, you are basically dumping stuff and firing people, those moves are rarely instantly profitable ( the real gain being to save the upkeep in the long run)

BlueTemplar wrote:
But I don't think it's normal to be able to refit a unit to a different unit at no cost, even if they have the same production cost, like your formula makes it.

You are right again, a minimum cost for refitting should be applied even to refit to same costs or lower costs units

Perhaps formula could be

(Cost of wanted model - 50% or 75% cost of existing model)


Another solution for refitting, which I already mentioned in several post, is to implement a function to refit units via the production of a city. Immobilizing the unit in the city for the time of the refitting (or removing it from the game, the disbanded unit production value directly transferred to the amount of hammers needed to create the new unit) , the cost being in 'city hammers' and be comparatively cheaper than credit refitting it. I really think tha would be a cool feature but I do realize it requires much more work on dev's side so I am not sure they will be willing or can afford to do it.

Oh, I should have been more precise : 1 production also implies 1 mineral!

BlueTemplar wrote:Oh, I should have been more precise : 1 production also implies 1 mineral!

Sorry I don't get your point?

Of course you need mineral to get a production point to build unit, but you also need this mineral to build wealth so its not really to be taken in account in the refitting discussion.

Just something to remember, as Pandora's economic system differs in this from other games. Implicitly, when rush-buying, you buy production, but also the corresponding minerals with the (usually) tax money.