3rd Century BCE Wars, a 270 BC Mod - Release Thread

classic Classic list List threaded Threaded
42 messages Options
123
Reply | Threaded
Open this post in threaded view
|

Re: 3rd Century BCE Wars, a 270 BC Mod - Release Thread

Eschelon
This post was updated on .
redrum wrote
@Eschelon - Alright. The AI actually wasn't handling the territory effects no blitz property but I just committed a fix and updated the pre-release. I made some significant improvements to the AI combat move especially when dealing with neutral territories so your next game should be even better!

Looking at the unit prices I think Hoplites, Hypaspists, and Spartan Hoplites are all over priced. I personally wouldn't really build them either. The cheaper 1 hp units are much better for fodder and the 2 hp units only cost a little more than those expensive 1hp so it better to focus on them when looking to buy fire power. I would recommend decreasing the cost of Swordsmen, Hoplites, Hypaspists, and Spartan Hoplites by 1 PU to better balance them.

In terms of map balance, I feel like the nations in the middle of the map need a little buff (Romans, Greeks, Macedonians, Seleucids). They end up having to fight on many more fronts then the other nations which seems to cause them to eventually collapse due to attrition.
@ Redrum,

That explains your previous inquiry r.e. the blitzing thing.  I didn't realise there was an issue, thanks for fixing it!

The 'mid unit' prices are why I bumped the base values of all units by 1.  Swordsmen are still getting built, as are archers, but I think in the case of the Macedonians & Greeks, they have sooo many units to pick from, that the Hoplite-style units are getting overlooked.  Plus their economies are a bit tighter than say the Romans...

I'm hesitant to lower the unit costs of the Hoplites, as (according to the attack+defense with no adders for support or move), they should be inline with the other units.  I'll continue experimenting with this.

Plus, this is the era of the Phalanx, so I would prefer that the AI builds those first when possible (so I don't mind them being slightly undervalued).  I may try bumping their values a bit more to see what happens.  The other two hit units are currently getting the appropriate amount of love by their respective empires (Immmortals, Warelephants, Legates, Barbarians, and Carthaginian Phalanxes).

R.E. map balance, that's a continued work in progress.  I've seen each of the empires hit a zenith/postion for the win win in one game or another (or they would have if I as the human player weren't in the mix), and I can usually eek out a win from any of the 10 starting positions, but you are right the middle of the map tends to be less forgiving of bad strategic choices.  

Mind you, we are talking over a LARGE number of turns for said potential for AI wins from any position (40+ turns usually).

I'll play around with some of the territory values (some more) in the middle region of the map, and see what happens.  I've also been debating adding some defensive value to cities, since the AI isn't taking advantage of walls.  Walls are also a great way to picket frontier territories on my map, if you are trying to slow down an enemy advance, but the AI's have no clue about this usage for them currently.  I'd hate to need to remove walls from this map entirely (don't like humans having unfair advantages over AI's), but one thing at a time.  They also allow empires to build up city defenses more quickly (i.e. exceed the build limit), which can be important when trying to hang on to your gains.


Back on topic, Rome in particular has the opportunity to dominate, but as an AI it gets distracted by shiny things easily, leaving it's other flanks open for exploitation.  This is why I have a couple of territories with massive neutral stacks, to keep Rome (and some other empires) from taking those routes before they are ready to deal with the consequences of said routes.  Those neutral unit mixes are of course subject to further review r.e. balance issues of course.

Thanks for the continued hard work on your AI!  I'll grab the latest jar as soon as I finish up this game I'm in the middle of.
Reply | Threaded
Open this post in threaded view
|

Re: 3rd Century BCE Wars, a 270 BC Mod - Release Thread

Frostion
In reply to this post by Eschelon
Ok I have tried to take a screenshot of your map with your colors. On the website it is 100% like on my screen colorwise. I tested is with a hex color picker Tool, after taking a screenshot of this website post and comparing it with the screenshot of the map.


The colors I found for you are the same brightness/dimness. I got them from here
http://www.rapidtables.com/web/color/RGB_Color.htm
… note that all colors in the same row should be the same brightness. Not that I think is a MUST rule of nation colors, but I think it is a good way to ensure color “uniformity”.
Reply | Threaded
Open this post in threaded view
|

Re: 3rd Century BCE Wars, a 270 BC Mod - Release Thread

Eschelon
@ Frostion

I can achieve the colors you have simply by deactivating map blends (and I'm very acquainted with those colors from playing 270BC back in the day pre 1.3.2.2).  The question is, how the heck did I get the pastels to show up then?!?

Now I'm skared to shut the blends back off, because I like them...

Y'all are using the same zip file I use (I upload the zip file directly from the triplea 1.8.0.5 map folder after I'm done dragging in the modified .xml into that zip file).

I mentioned the border thing just in case that might have some effect.  I didn't think it had much to to with this situation, other than with how the blends change the territory borders when borders are set to low, which I like.

I can always set the default map colors to the pastels, but then the borders won't look as pretty.  You'll note that the borders on my map tiles are wider than 1 pixel (about 3 pixels-ish), which looks cool when the blends come into play.


BTW, in that same menu, I found the 'unit count font color' thing under the edit map font and color'.  I tried some other colors, but white is really best with these unit icons.  I am probably going to add a circular shadow to the unit icons, so that the white can contrast better.
Reply | Threaded
Open this post in threaded view
|

Re: 3rd Century BCE Wars, a 270 BC Mod - Release Thread

Eschelon
Sure enough, the millisecond I disabled map blends, the pastels went away... and now I can't get them to show up again!

Ahhh, the great mysteries of life!
Reply | Threaded
Open this post in threaded view
|

Re: 3rd Century BCE Wars, a 270 BC Mod - Release Thread

Eschelon
@ Frostion

OK, so it appears to be related to the map alpha settings.  I'm not sure why the old alpha value was 'lodged' in TripleA's memory, but upon doing some playing around with the alpha settings, the 'pastels' have reappeared, after I bumped the alpha to .35 or so.   I had previously been playing around with the alpha settings a while back.

I've arrived at a .40f setting for now for this setting under map.properties:
map.mapBlendAlpha=0.40f

This might be a little more washed out than my last screenshot, but it seems to have toned down the Greek neon green a bit.  I think that my 'old' alpha setting must have been around .50f, comparing the screenshot to what I have now.  I like the .40f look, though.

So, if you are still game, Frostion, here's my current map.properties file (I haven't added any color changes to this as of yet).

map.properties

Or you could just bump the alpha in your own (edited) version of the map.properties to .40f, and restart TripleA.  Let me know if this gives you the pastel look I've been describing.  You may need to toggle the map blends off and on to get this to appear...

Much appreciated for the help!
Eschelon
Reply | Threaded
Open this post in threaded view
|

Re: 3rd Century BCE Wars, a 270 BC Mod - Release Thread

Frostion
Can't download. Try to zip it an then I can download. I can't download any rar files posted here on the website either
Reply | Threaded
Open this post in threaded view
|

Re: 3rd Century BCE Wars, a 270 BC Mod - Release Thread

Eschelon
Try this:

map.zip
Reply | Threaded
Open this post in threaded view
|

Re: 3rd Century BCE Wars, a 270 BC Mod - Release Thread

Irinam
I did have the same colors as Frostion.
Will test your zip, when back at home...

Greetings,
Irinam
Reply | Threaded
Open this post in threaded view
|

Re: 3rd Century BCE Wars, a 270 BC Mod - Release Thread

Frostion
In reply to this post by Eschelon
This is how it looks on my PC with the mapBlendAlpha=0.40f
Reply | Threaded
Open this post in threaded view
|

Re: 3rd Century BCE Wars, a 270 BC Mod - Release Thread

Eschelon
Frostion wrote
This is how it looks on my PC with the mapBlendAlpha=0.40f
@ Frostion & Irinam

So yeah, we are now on the same page.  Sorry to make y'all jump through hoops, I honestly have no idea what had happened (gotta luv glitches)!  No idea why the alpha value wasn't showing properly for me, and I'm glad that we were able to track the problem down.

Do you like this color presentation (with the higher alpha) better? Or do you prefer the darker colors (lower alpha)?

If so, any suggested color tweaks at the higher alpha?  (I have the darker color tweaks copied for testing).

I may try out using the resulting colors at the higher alpha as the baseline (using the eyedropper tool in photoshop to get the new RGB values), although I like the map blend presentation better (the borders are purdier).

Thanks in advance guys!
Eschelon
Reply | Threaded
Open this post in threaded view
|

Re: 3rd Century BCE Wars, a 270 BC Mod - Release Thread

Irinam
OK,

i didn't have the map blends turn on...
With blends on now and your newest presentation i'm fine...
but to tell the true... for me colors are not important,
the game mechanics are what makes a good game.

If grafics are really important for a player,he would probably not play triplea at all...
most lousy free to play games on playstore have better graphics,
but the are worse games...

Reply | Threaded
Open this post in threaded view
|

Re: 3rd Century BCE Wars, a 270 BC Mod - Release Thread

Frostion
In reply to this post by Eschelon
Well, the new ”pastel” colors are more pleasant for the eyes, and they are more harmonic with each other.

But, why not use the pastel colors as hex code in the map properties file, and then play map without map blends?

I know the borders look/are a bit thinner without blends, but you could paint the baseTile picture up to be thicker or something. Then you won’t have the problem of players not having blends on.

It would be nice if there were a setting to automatic turn blends on. I do not know if this is an option.
Reply | Threaded
Open this post in threaded view
|

Re: 3rd Century BCE Wars, a 270 BC Mod - Release Thread

Eschelon
Frostion wrote
Well, the new ”pastel” colors are more pleasant for the eyes, and they are more harmonic with each other.

But, why not use the pastel colors as hex code in the map properties file, and then play map without map blends?

I know the borders look/are a bit thinner without blends, but you could paint the baseTile picture up to be thicker or something. Then you won’t have the problem of players not having blends on.

It would be nice if there were a setting to automatic turn blends on. I do not know if this is an option.
So I've already thickened the borders to taste, and with blends on the black borders are blended with the 'empire color' to create the darker colored borders.  With blends off, this tends to go away (solid black lines, or if map details is also disabled then you only get 1 pt black borders).

While it'd be great if I could have multiple map properties files, right now I think you are stuck with one properties file per map folder.  So what I'll probably do is release two versions of this map in the future.  One with blends in mind, and the other with the pastels as the default colors for those who don't want to bother with blends.

I finally finished the aforementioned game, in Round 41.  Scythia was stopped cold on the Parthian Front.  Egypt did indeed get wiped out, and Parthia and Numidia were simply too strong to advance on Parthia without losing Antioch in the process.  But Scythia did manage to capture Athens and the two Macedonian capitals, with the Germanic Tribes taking 3 of the four African Alliance capitals for the win.  They captured the Roman and Seleucid capitals a while back, and those were pretty much secured, which allowed them to hit the magic 14 VC number.

Oh, and Greece actually built, over the course of 41 turns 1 (count them, 1) Spartan Hoplite!  Yeah, I have some more cost tweaks in my future.  I'll try bumping the 2 hit unit costs slightly in the next release (once I do some more playtesting on my end) and see what happens.  I may also take regular hoplites out of the Greek and Macedonian mixes to see if that makes a difference.

I'm also looking at bumping the PUs slightly on a couple of territories, in the middle region of the map, to see how that plays out.  I don't want to increase the build rates in said territories, so these will probably be odd to even bumps (i.e. 5 to 6, 7 to 8, 9 to 10, etc.).

On another note, my 'reasoning' for having 10 'owned' sea zone territories (to coax the AI to value one capital over the other) doesn't seem to apply to the HardAI at all, plus I think this is slowing down the purchase phase for destroyed empires (game spends a minute or so contemplating builds for empires that hold no land territories and have no units on the map), so I will try deleting those in my playtesting and see what happens.  If the 'build phase hang for destroyed empires thing' persists, I will report this behavior, but as of right now I think it's a map specific thing, not necessarily an AI thing.

I want to do some playtesting of my own before the next release, so sit tight.  In the meantime, as usual, suggestions are always welcome!
Reply | Threaded
Open this post in threaded view
|

Re: 3rd Century BCE Wars, a 270 BC Mod - Release Thread

Irinam
Eschelon wrote
On another note, my 'reasoning' for having 10 'owned' sea zone territories (to coax the AI to value one capital over the other) doesn't seem to apply to the HardAI at all, plus I think this is slowing down the purchase phase for destroyed empires (game spends a minute or so contemplating builds for empires that hold no land territories and have no units on the map), so I will try deleting those in my playtesting and see what happens.  If the 'build phase hang for destroyed empires thing' persists, I will report this behavior, but as of right now I think it's a map specific thing, not necessarily an AI thing.
I always change destroyed AI empires to "DoesnothingAI"...
it's a lot faster on my PC than "HardAI", if no units are left...
Reply | Threaded
Open this post in threaded view
|

Re: 3rd Century BCE Wars, a 270 BC Mod - Release Thread

Eschelon
Redrum has introduced some code on the 'no buildable territories' issue, which I'll take for a spin when I get some time.

I just got real busy with a side project, so the next update won't be for a bit.  But a quick preview for what I have planned.

Some key territory values will be getting bumped up slightly, but I'm dropping the ratio of # of units built per territory a bit.  Instead of a 50% reduction (i.e. 1 build per 2 PU's, round up), I'm planing on trying out 1 per 2.5 (i.e. 3-5=2 builds, 6-7=3 builds, 8-10=4 builds, 11-12=5 builds, 13-15=6 builds).  This is to address an issue I'm seeing in my current game where the game is going slinger/peltast happy.  I'd like to see a few more 'better quality' units getting built, hence the change.

A number of 'middle' empires have peltasts & slingers (Greece, Macedonia, Seleucia, Egypt), which is a nice balancing feature vs. the other countries that are in better defensive positions.  That being said, I'm seeing WAAAAY too many of these currently in the mid to late game (which is slowing down AI turns considerably as a result), hence why I think this change to the PU/Build ratio will help.

Also, I plan to bump the city/factory build cost to 30 (maybe a bit higher depending on how this goes), and may incorporate some minor defensive support bonus & attack penalties if a city is present.  I'm also contemplating some mechanism that adds a small number of walls to the building cues each turn, since the AI isn't paying attention to these.  It's either that or removing walls entirely since the AI is ignoring them, which I'd rather not do as they have their uses on this map.

I'm also thinking about broadening the 'build location restrictions' for some special units.  As it stands currently, Spartan Hoplites and Hypaspist builds are limited to 1-2 territory(s) that they may be built in - which ends up being a bit behind the front lines in the mid-late game.  So the AI doesn't see any reason to build these (would take too long to march them to the front).  I'm also still continuing to experiment with the build costs, although I've seen at least one of every unit get built at some point in my current game, except for the aforementioned walls.

I hope to get back to this within the next couple of weeks and post up the next build.  In the meantime, comments are always welcome!
Reply | Threaded
Open this post in threaded view
|

Re: 3rd Century BCE Wars, a 270 BC Mod - Release Thread

Mandalon
This might be slightly off-topic here, but I have a problem with the Dynamix land-only AI. It does not build any cities but builds way more units than it can place.
When loading a game and switching to hard AI, it immediately builds more cities for production.

Can this be triggered by the high cost of the cities?
Reply | Threaded
Open this post in threaded view
|

Re: 3rd Century BCE Wars, a 270 BC Mod - Release Thread

redrum
Administrator
@Mandalon - Don't use the Dynamix AI anymore. Please try the Hard AI as it should perform much better.
Reply | Threaded
Open this post in threaded view
|

Re: 3rd Century BCE Wars, a 270 BC Mod - Release Thread

Eschelon
In reply to this post by Mandalon
Dynamix has unfortunately been abandoned.  Post 1.3.2.2 some tag changes essentially 'broke' Dynamix for this map.

I'd love to see someone spend a day or so bringing Dynamix up to date, but no one wants to, so it's an abandoned AI at this point.

If you DO want to play this map with Dynamix, you'll want the 1.3.2.2 version (and you'll need to download TripleA 1.3.2.2).  the link to the 1.3.2.2 version of my map is in the first post of this thread.  I probably should update the link at the bottom of the first post to the most recent version... note to self.

There are some definite differences in the approaches the two AI's take.  I still think that Dynamix can be a competitive AI even today with some tweaks, but since it's a 'Land Only' AI, there simply hasn't been anyone willing to pick up where Wisconsin left off (since it doesn't know how to use navies...).

Hard AI does pretty well on this map, but it does take a while to get through turns in the later game.  I haven't tried the latest version (I'm about a month behind the curve at this point) so I don't know how many speed improvements have been made since my last release.

FYI, I'm in the middle of a coding project not related to TripleA, so it'll still be a while before I get around to the next update.
Reply | Threaded
Open this post in threaded view
|

Re: 3rd Century BCE Wars, a 270 BC Mod - Release Thread

Mandalon
I was asking about the AI because I often get exceptions for the AI while purchasing units (when using the hard AI), see below. I'm using engine 1.8.0.5 and game version 0.9.1. Does anyone else suffer from these problems too?



Reply | Threaded
Open this post in threaded view
|

Re: 3rd Century BCE Wars, a 270 BC Mod - Release Thread

Irinam
Make sure to use the latest development version:

redrum wrote
Latest Development Version
https://sourceforge.net/projects/tripleamaps/files/TripleA/prerelease/
make sure you download the latest stable version first, then download this prerelease jar file, and copy it over the old jar file (in the /bin/ folder)
123