Version 1.4 discussion thread

I eliminated the prestige penalty entirely.

Also started on the computer AI, took away a lot of its bonuses at higher difficulty levels and am replacing it with just better strategy code.

Made unit training time be effected by population:

If I can, I'll try to put this stuff so that it's accessible via XML.  But that sort of thing is tedious and time consuming. So I'll wait to see how this plays out first.

Changed the city hub names of Kingdom and Fallen to make them a bit more distinctive.

Changed the city hub populations so that they store more population.

Changed prestige so that it now increases population as a percentage rather than a whole number. I.e. N% per turn. Will see how that feels.

Sovereign now provides 1 research per turn. So no reason why research should be put off.

Lowered the cost of spells to level up.

Lowered turn time to build many improvements to 1 turn. If you've got the money you shouldn't have to artificially wait around.

I posted a bunch of bugs here: https://forums.elementalgame.com/410743

Also, there are issues with how you implemented the new limits on repeatable buildings, which I comment on here: https://forums.elementalgame.com/410744 

I like where it's going, but be careful so the player can't reach 0.00 training multiplier through buildings, ie instant training for units.

Instead of reducing some improvements to 1 turn, how about something similar like what you did with training for building improvements? Ie larger cities build improvements faster.

Thanks!

Fixed a crash issue caused by multithreading.

Any chance you could throw in a fix for not seeing "goodies" on the cloth map? Just mentioned it in another thread, but seeing as you are rolling solo now, this may be the best spot for such a request.

Thanks for the update by the way. Good to see you still grinding away on this!

Brad,

Are you only focusing on bug fixes or are you looking for suggestions?

If so, are you only going to do mod-type fixes and no interface changes?

Thanks!

Just gameplay. Only things that I can do. I can't change UI.

Often times, there's a handful of "simple" things that are really destructive to gameplay.

Let me give you one huge example I'm working on: The pathfinding in the game aborts on the AI all the time without informing the AI. So the AI tells a unit to "go here" to found a city but the path finding code in WOM is very fancy but strict about how complex you can ask from it (to avoid long delays in path finding). PRoblem is, the AI tells a unit to go somewhere and it simply doesn't and never knows why.  

I only noticed because the AI is so crappy even though it's working as designed. The problem is the units aern't going. No units moving no attacking, no coordinating, no city expansion -- except, of course, when it does work (which is often).

I only discovered it because I have so much code around the sovereign now to figure out why it does stupid stuff that I raelized that it was not stupid but that it couldn't stop itself because the pathfinding wouldn't let it get out of danger.

Love you Frogboy!

To me, this is a huge improvement for the game.  Like I said in another post, there are so many bottlenecks already to building things, we don't need time to be one also.  Thank you.

frogboy , it sounds like you are getting wom to head in the right direction at last, but even humans can be effected by the path bug, I have had several times in many of the games where I tell a stack(or single unit) to go to a particular location IN my territory from elseware in my territory, and it does NOT go there, and have to tell it to go part way ,then later tell it to finish the journey(and it was(and is)mildly irritating.

harpo

Cool, following.

Booked marked this thread.  I'll be popping in over the next few days to see whats happening.  I like that this is happening - a lot of the issues that still persist with WoM are simply gameplay stuff, so I still have hope that WoM will get to where it needs to be.

Thanks for letting us know!

Great thread! Really appreciate you 'blogging' your way through 1.4s development, this makes the wait for 1.4 and FE easier. I really like to follow this kind of thing.

Anyways, very interesting changes so far. I think they sound good.

How does this play with the Brilliant trait? Is Brilliant just an extra point on top of that?

Great idea because research is really slow right now. This should help jump start things a little.

Bugs are a big one. I'm not running into anything as game crippling as earlier versions, but the more polish the better.

I agree that the current system for buildings is a bit off. I think that better buildings should be saved for higher levels. That would be a nice incentive for leveling up a city instead of spamming lots of little ones. We already have that with some buildings, so I say maintain that system and maybe remove repeatable buildings or increase the amount per city level, so instead of 1,2,3, per level, do 1, 2, 4 or something.

Thanks for the update.  Any idea when you'll release a 1.4 beta?  As you can see by the response to your post there are still quite a few of us more than willing to take it for a spin. 

I want to get the first beta out right away really. Not for the game play reasons as much as for stability.  v1.3 was a lot more multithreaded so that people with more cores could take advantage of the game better.  However, there's a lot of thread conflicts that come up depending on the # of cores you have (i.e. crashes) that I've been able to reproduce and fix.

The other big thing is the pathfinding stuff.  This will "magically" make the AI much smarter without having really updated the AI much so far because now when I tell a unit to go to X,Y it should go there rather than saying "Sure" and just not.  This *really* affected pioneers. In case you ever played a game where the AI wasn't expanding, there you go.

It's really unfortunate that our industry has a distinct lack of network programmers and thread developers.  I was talking to some of my friends at Firaxis recently about this recently.  Finding good network coders is difficult now and the ones that are out there have migrated to the MMO companies.

Here's the world I'm testing with for today. I used one of the random terrain maps (rather my own testing maps) to get better real world conditions to see if path finding works the way I would expect.

I'm at the bottom right.  Everything is on normal and I've toned down the AI XML advantages massively from v1.3

What I'm going to simply do is see how quickly I can conquer the AI and how well the AI does with the new training system and economic changes.

Going to just send this group to conquer the AI and see if I can do it. If I can, then I have my work cut out for me this weekend.

Frogboy, I suggest you fix unit HP (they need something of the order of a 10fold increase) - and maybe look into making the tech level progression in Armor and Weapons less... sharp.

This might require a change in the way heroes HP/attack works.

Unit balance has been an issue for a while now, and the unit HP change in 1.3 wasn't very well received, because it made no sense and killed any potential fun in combat.

First of all I'm glad WoM is still being worked on. The AI improvements and fixes also look sweet, and I imagine they will be applicable to FE as well.

If you really want WoM to improve though there are some other glaring issues that need fixed. The games balance has been shoved to the side in favor of adding features and fixing bugs for way too long, and it's a real drag on the fun factor.

Everything he just said is very true. The in game balance right now is very bad. A significant hp bonus to troops, a weapon dmg nerf, and a weapon dmg scale nerf are all good ideas. Unfortunately to fix the balance you would probably need something like a comprehensive spreadsheet of all equipment and monsters, which would be a whole whack of work.

Another balance issue is the limit of 1 per city level on unlimited improvements. This limit kills city specialization and only encourages city spam. Maybe something like a limit of 4 of any type of unlimited improvement per city level instead. This would be much better in my opinion. Or just ditch the limit entirely.

The decrease in improvement build time is interesting. Not sure why you would put in so many artificial limits on them only to decrease a natural limit like build time. Now instead of having cities spending all their time slowly building improvements most the time the will be doing nothing until they reach the next level and then bam a lot of buildings will just pop up. Also won't that seriously Nerf other buildings like construction yards, and the spell magical hammers? Maybe you could tell us how you see the current building system and what you hope to achieve with it?

I have always felt that the AI in elemental is actually pretty good when it comes to building armies, expanding(when not frozen), and leveling cities. It just has bad pathfinding, tactics, equipment use, and makes some dumb moves from time to time. I suspect from what you have said that that pathfinding will be greatly improved in 1.4 and the AI will be making significantly less dumb decisions.

Thus it's weak spots will be poor unit design(which is tied up with poor equipment balance), poor champion equipment, and a bad tactical AI. I am aware the tactical Ai is really a different issue but when it comes to beating the AI a tactical AI so bad that your units are at least twice as effective as the AI's is a huge issue. Combine that with bad equipment use(they don't maximize armor, which is 90% of combat) and I routinely kill AI armies that outnumber me 4 to 1. Routinely as in one after the other for entire games.