Messages

1

AI War / Re: What do you want the Champion to be?

« on: Yesterday at 11:25:15 AM »

Cross posting is so much fun (for me at least)

I'd be happy with:

• more nebula missions and mission types
• improve balance of existing nebulas
• way(s) to gain unlocks outside of nebulae, even if they were secondary in effectiveness to nebulae. Prefereable, this other way would be more "integrated" into the main goal of taking out the AI
• way(s) to gain noticable experience outside of nebulae, even if they were secondary in effectiveness to nebulae. Prefereable, this other way would be more "integrated" into the main goal of taking out the AI (someone mentioned that starships and now also guardians (and higher tiers as well) are really tough enough to put up a good fight against champs, so it seems fair that they would give some exp too. Also, can the exp gain for guard posts and command stations go up a bit?)
• some way to avoid the "tyranny of the RNG" when it comes to unlocks, especially hull and module unlocks that don't work well together (a somewhat common occurrence that can really hurt the effectiveness of the champion by no fault of your own)
• some sort of "respec" system
• more active abilities
• better balance of existing active abilities

Yes, not trivial by any means, but doesn't require "ditching" huge amounts of stuff already in the game, and wouldn't require a whole "expansions worth" of work. Also, it still keeps the current system, mostly, but just expands on it.
Something like unlock trees and a "fallen spire, but with champions" type thing would be cool too, but that seems like it would take too much more work than is reasonably worth to consider at this time.

2

AI War / Re: The future of Champions

« on: Yesterday at 11:24:37 AM »

I'd be happy with:

• more nebula missions and mission types
• improve balance of existing nebulas
• way(s) to gain unlocks outside of nebulae, even if they were secondary in effectiveness to nebulae. Prefereable, this other way would be more "integrated" into the main goal of taking out the AI
• way(s) to gain noticable experience outside of nebulae, even if they were secondary in effectiveness to nebulae. Prefereable, this other way would be more "integrated" into the main goal of taking out the AI (someone mentioned that starships and now also guardians (and higher tiers as well) are really tough enough to put up a good fight against champs, so it seems fair that they would give some exp too. Also, can the exp gain for guard posts and command stations go up a bit?)
• some way to avoid the "tyranny of the RNG" when it comes to unlocks, especially hull and module unlocks that don't work well together (a somewhat common occurrence that can really hurt the effectiveness of the champion by no fault of your own)
• some sort of "respec" system
• more active abilities
• better balance of existing active abilities

Yes, not trivial by any means, but doesn't require "ditching" huge amounts of stuff already in the game, and wouldn't require a whole "expansions worth" of work. Also, it still keeps the current system, mostly, but just expands on it.
Something like unlock trees and a "fallen spire, but with champions" type thing would be cool too, but that seems like it would take too much more work than is reasonably worth to consider at this time.

[very]

3

AI War / Re: AI War Beta 6.032 "Tumbleweeds In The Stars" Released!

« on: Yesterday at 09:29:43 AM »

So, I was playing some AI War when suddenly my favorite TV show came on:

When ARSs Attack!



Yup.  I think something went a little off in the recent AI logic updates.

Haha, that's awesome.

I'm guessing that you were in the process of building the command station but hadn't finished it?  Otherwise the ARS would have been captured already.

Anyway, I just need to make that branch not apply to stuff with 0 attack

So fair WARNING: this release has some significant \"foundational upheavel\", mostly in the AI code, and may have some very strange and wonderful bugs (more strange than wonderful).

So, is this one of those "strange and wonderful" bugs you were talking about?

4

AI War / Re: AI War Beta 6.032 "Tumbleweeds In The Stars" Released!

« on: May 23, 2013, 02:27:50 AM »

Awwww now I won't be able to sleep until I can see it 
Dam you AI you ruined too mane of my nights already !

Ah you are falling prey to one of the most insidious plots of the AI. Changing up the stuff the do and have to maintain your interest, to the point where you forgo sleep and eventually your performance starts to suffer due to sleep deprivation.

5

AI War / Re: Champions

« on: May 23, 2013, 02:25:57 AM »

Yet another repost on yet another relevant thread.

I think to "fix" champions, I think adding more nebula missions, possibly new types (like something that requires going out of the nebula at some point in the mission), balancing rewards, (including the module unlocks may not be compatible with hull unlocks that seems to crop up sometimes), and possibly adding someway to unlock stuff outside of nebulae missions, though possibly in a less "time effecient" manner. Nebula missions should remain the primary way to unlock stuff, but some sort of fallback just in case something goes wrong making nebulae too hard with the unlocks you got or something, or you want to reward helping out normal players more directly would be a great idea.



*fix is in quotes, as I'm not sure if there is something broken, just merely something lacking

Something I mentioned up there but didn't really expand on is the unlock system. I can understand randomizing rewards some, as it fits with the procedurally generated nature of the game. But right now, for something so fundamental to whether the unit will be awesome or just "decent", there is too much randomness. Combine this with a lack of ways to get unlocks (nebulas only right now, which I talked about above) and no way to "respec", you can often times get punished for bad upgrade decisions before you really know what you need or be screwed over by the RNG in truly unfun ways and magnitudes. Yes, this sort of thing exists for the rest of the game, but not to the magnitude where it can almost ruin the whole point of a mechanic, and not in such unfun ways.

I would love for some sort of tweaks, small-ish additions to the unlock, experience, and upgrade systems to reduce the magnitude of these effects. The "randomness" and "punish for unlocking before getting relevant info" aspects should most certainly stay, just like the rest of the game has. But they should be toned down some to keep champions fun and "feeling fair", even in "pretty bad cases" of RNG rolls or minor to moderate mistakes.

6

After Action Reports / Re: Advanced Paranoia

« on: May 22, 2013, 03:20:47 PM »

Yes, I don't think cutlasses are OP because the aren't all that terribly durable nor are they fast (especially by melee unit standards). But yes, if you can actually get them to their destination, they can inflict massive pain.

7

AI War / Re: Which role do you play as in AI War: Ancient Shadows?

« on: May 22, 2013, 11:50:29 AM »

Reposting here as it is also relvant:

I think to "fix" champions, I think adding more nebula missions, possibly new types (like something that requires going out of the nebula at some point in the mission), balancing rewards, (including the module unlocks may not be compatible with hull unlocks that seems to crop up sometimes), and possibly adding someway to unlock stuff outside of nebulae missions, though possibly in a less "time effecient" manner. Nebula missions should remain the primary way to unlock stuff, but some sort of fallback just in case something goes wrong making nebulae too hard with the unlocks you got or something, or you want to reward helping out normal players more directly would be a great idea.



*fix is in quotes, as I'm not sure if there is something broken, just merely something lacking

8

AI War / Re: Champion Options

« on: May 22, 2013, 10:46:29 AM »

I think to "fix" champions, I think adding more nebula missions, possibly new types (like something that requires going out of the nebula at some point in the mission), balancing rewards, (including the module unlocks may not be compatible with hull unlocks that seems to crop up sometimes), and possibly adding someway to unlock stuff outside of nebulae missions, though possibly in a less "time effecient" manner. Nebula missions should remain the primary way to unlock stuff, but some sort of fallback just in case something goes wrong making nebulae too hard with the unlocks you got or something, or you want to reward helping out normal players more directly would be a great idea.



*fix is in quotes, as I'm not sure if there is something broken, just merely something lacking

9

AI War / Re: Dump of galaxy wormholes?

« on: May 21, 2013, 10:15:51 AM »

The best thing I can think of is to ask the game to not compress it. That way, both the gzip and the base64 steps are skipped, and you can just open it up directly in notepad/wordpad/vim/emacs/gedit/nano/textedit/whatever.

And AFIAK, the game can load up totally uncompressed games as well.

10

AI War / Re: Too many generals, not enought privates

« on: May 20, 2013, 08:52:38 PM »

Ah, oops, I assumed the boost was cumulative

Still, does that look like a reasonable number of flagships?

If they properly sacrificed the opportunity to get additional starships or guardians or fleetships or a golem, then yes, that seems reasonable. (What difficulty are you playing again? And what scaling of the exos?). However, knowing whether it payed for it properly or not is hard to tell without logs.

Now, it could be argued spending most of its points into 59 Mk. I flagships was a rather dumb choice for a unit composition by the AI. Maybe that could be addressed.
Or it could be argued that the exo-cost of the Mk. I flagship is too low for how much it brings. Well, it wouldn't be the only one. Exo-costs are really in a strange place right now TBH.

11

AI War / Re: Dump of galaxy wormholes?

« on: May 20, 2013, 08:48:03 PM »

Before anyone asks: yes, we are aware that there are massively more efficient ways to do this, and in fact our later stuff is much more efficient, but this particular bit really doesn't need the optimization and the chance of breaking something outweighs the benefit

Hey, it's more space efficient than XML is.

12

AI War / Re: Poll: Bonus Ship Type ideas for the Vengeance expansion

« on: May 18, 2013, 12:19:37 PM »

Shifted a vote to the laser pods.

However, it would be nice if their self damage wasn't so high that they oneshot themselves. Attack power could go down and/or cost could go up a bit in return.

13

AI War / Re: Dump of galaxy wormholes?

« on: May 17, 2013, 07:28:53 PM »

There is a thread somewhere with some info on the save game format, but I can't find it at the moment.  The save format is pretty straight forward if I recall.

You need to turn off savegame compression first, so you can see the plain-text format. But yes, it's pretty easy to figure out.

The things of interest here are the planet list (which has the planet IDs and the planet names) and the adjacency lists of planet connections.

In fact, I could pretty easily write up a program that cross references these and spits them out in a nicer format, or even as a graph for whatever analysis you want to do with it.

14

AI War / Re: A two part question:

« on: May 17, 2013, 05:03:43 PM »

Considering the amount of work this is going to take, I think it is better that the values be lowered by 100, even if superweapons also need to be reduced by a smaller amount.

Keep in mind this whole thing was brought about because we are running out of ceiling for numbers. The advantage of reducing values by 100 over 20 or 25 is it gives 4 to 5 times the wiggle room for relative power.

Since the amount of work of going to 100 compared to 20 or 25, in the big picture, is about the same, better to do that value even if golems need their power reduced as well.

I mean reduce the superweapons by only 20x or 25x. The rest of the non-superweapons would still be reduced by 100x.

Reducing both by 100x would just leave things in almost the same relative place they are now, with superweapons only being "pretty good" instead of, you know, super.

15

AI War / Re: A two part question:

« on: May 17, 2013, 04:19:00 PM »

The reason I am advocated chopping "slightly more" than one 0 off of superweapon stats is that a 10x relative shift is a absolutely gigantic shift, much more than I think many people realize. It could easily overcome the bit of cost-effective lacklusterness of them compared to starships, so much so that it could even dominate to the point of being game-breakingly (as in, "I win button" even on difficulty 9) OP even for superweapons. Yes, a 10x shift can easily be that huge, especially if what you shifting wasn't that far behind to begin with.


However, scaling it down by the same amount would not accomplish anything. This is why I suggested dividing the superweapon stats by 20 or 25, so they get only a 5x or 4x relative shift. Yes, that is still huge, but less likely to break the game. If it still isn't enough, then maybe we can try something more. However, I really don't think superweapons are that far behind that they need an order of magnitude (base 10) relative buff.

EDIT: For the mathematically inclined among you, dividing the stats by 20 (5x relative buff) would be a ~1.30 order of magnitude reduction (~.70 orders of magnitude above relatively); dividing the stats by 25 (4x relative buff) would be a ~1.40 order of magnitude reduction (~.60 orders of magnitude above relatively)
As a base, dividing the stats by 10 (10x relative buff) would be a 1 order of magnitude reduction (1 order of magnitude above relatively)