Author Topic: Is there a possibility of a new AI War expansion anytime soon? And Why Not?  (Read 18081 times)

Offline Kahuna

  • Core Member
  • *****
  • Posts: 2,222
  • Kahuna Matata!
Why are there hacking sub-units at all?  Why not just fly your Hacker up to, say, a Fabricator, and target-click the Fabricator to start hacking it?
Yeah, the fact that it's possible to learn is no excuse for it being harder to learn than it needs to be.

If there's 10 different hacking options then there is unless the game gets dumbed down.
You have to learn what each hack does regardless of how the hacking system works.

How to hack in 3 steps:
1. Build a hacker
2. Move the hacker on the planet you want to hack
3. Select the hack(ing sub unit) you want and place it where you want
-->You have successfully started a hack

That's how hacking works at the moment. If that's too complicated then I don't know what to say. What would be an easier alternative?

Even if it worked as Elestan described it would still have 3 steps:
1. Build a hacker
2. Move the hacker on the planet you want to hack
3. Click an object you want to hack

Which would actually be more tedious than the current system because you would have zoom in and out and find the object on the planet. Now you can just drop the subunit and the hacking starts automatically.
« Last Edit: April 03, 2016, 12:54:17 pm by Kahuna »
set /A diff=10
if %diff%==max (
   set /A me=:)
) else (
   set /A me=SadPanda
)
echo Check out my AI War strategy guide and find your inner Super Cat!
echo 2592 hours of AI War and counting!
echo Kahuna matata!

Offline Draco18s

  • Resident Velociraptor
  • Core Member Mark V
  • *****
  • Posts: 4,251
Not to mention running multiple simultaneous hacks.

Offline keith.lamothe

  • Arcen Games Staff
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 19,505
unless the game gets dumbed down.
Just wanted to say: I will be doing no dumbing-down :)


On the UI stuff, there's certainly a lot of room for improvement. There's no need to defend the way it is from the perspective of "this is the best way it could be". It's not, by a long shot, and one of my big motivations for the rewrite is so I can do a new UI that's not encumbered by the way things were done six years ago. I also hope to make some of the UI (like the top resource bar and the ship tooltips) at least partially based on xml scripts so you can mod it (both for personal taste and to experiment with possible improvements for later implementation), but that's not really going to address things like the usability of hackers.

If necessary I could defend the current UI from the perspective of "this was the best available trade-off", but I don't think that's needed either.

What is helpful, and I'm glad to see it's happening here, is to identify problems and throw possible solutions around and refine them. Those solutions will probably not be better than the current approach on the first try, and simply switching to another not-great solution isn't going to help. But with refinement there's good potential.
Have ideas or bug reports for one of our games? Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline Cyborg

  • Master Member Mark III
  • *****
  • Posts: 1,957
Summary, keep discussing! I agree.
Kahuna strategy guide:
http://www.arcengames.com/forums/index.php/topic,13369.0.html

Suggestions, bugs? Don't be lazy, give back:
http://www.arcengames.com/mantisbt/

Planetcracker. Believe it.

The stigma of hunger. http://wayw.re/Vi12BK

Offline Elestan

  • Full Member Mark II
  • ***
  • Posts: 158
Even if it worked as Elestan described it would still have 3 steps
The difference is that "Select my unit, then target-click the enemy unit for it to go after" is how every other unit in the game works, so it's what new players are going to instinctively try.  If we can make it work that way, it's one less time they need to stop playing to look something up.

I could be wrong, of course.  I'd encourage Arcen to do actual usability testing on the game when evaluating their design.  It shouldn't be too hard to find a pool of 16-24 year olds who haven't played AIW before who could be lured in to your office with a promise of free pizza and LAN gaming.  Drop them in the game, watch them as they try to learn it, and see what they stumble over.

Not to mention running multiple simultaneous hacks.
Just bring in multiple Hackers.  It's a nerf, but a minor one.

While we're on the topic of Hacking, I might as well post some other thoughts I'd written down on it.  These go deeper than UI, and might be a bridge too far for what's planned, but they might be thought-provoking:

1)  I'm not fond of the way the hacking response works, for a couple of reasons:

  - It lets the AI conjure ships out of raw vacuum, coming in from the system edge (as though it had counter-posts), and pouring in from the CS at far higher than the normal reinforcement limit, even if the warp gate has been taken out.  If the AI has the ability to do this, then why doesn't it do it in response to other provocations?  In my opinion, hacking responses should not allow the AI to break its normal ship deployment rules.

  - The whole point of hacking using the security weaknesses represented by HaP is that you're hiding it from the AI.  Yet the AI immediately detects your hacking anyway and starts sending ships at you.  An interesting alternative would be to have the AI's response go through phases, perhaps based on the players' excess HaP and the number of simultaneous hacks they are trying to execute

  Phase 1:  The AI is unaware of your hack, and makes no response.  Status message:  "ARS Hack in xxxxx vs AI 2 proceeding".
  Phase 2:  The AI knows what kind of hack is being executed, but does not know what system.  Systems with the appropriate kind of asset may get hit with Tachyon bursts.  Active Hackers may take ship damage from "countermeasures", and risk destruction if not repaired.  Status message:  "AI 2 executing Network security sweep.  Hacker in xxxxx damaged by countermeasures."
  Phase 3:  The AI knows what system is being hacked.  All reinforcements are diverted to the target system (up to its reinforcement limit), and waves start getting launched through the nearest gate.  AI taunt:  "Found you."
  Phase 4:  For hacks against important installations, if the AI doesn't feel it can make an adequate response in phase 3, it launches an exo-wave led by a "Counter-Hacker Golem"; a force-field immune ship with planetary tachyon scanners and a preferred target of Hackers.

2)  It would be nice if one could have multiple Hackers work together to make the hack go faster.  This would also burn HaP faster and increase the likelihood of detection/severity of the response.

3)  It would be nice if one could research better Hackers.  Maybe the level 1 hackers don't even have cloaking.  Higher level ones might be able to do paralysis or reclamation attacks.

Offline Kahuna

  • Core Member
  • *****
  • Posts: 2,222
  • Kahuna Matata!
another (and perhaps the largest) source of complexity is the number and semi-arbitrary nature of hull types and attack multipliers
Agreed about the hull system. Some of the hull types could go away. Also ammo type vs hull type could replace the attack multipliers.

There's no need to defend the way it is from the perspective of "this is the best way it could be". It's not, by a long shot, and one of my big motivations for the rewrite is so I can do a new UI that's not encumbered by the way things were done six years ago.
Not defending per se, but being unable to understand how something consisting of 3 steps in overly complicated. But hey I could be a human like computer/cyborg without realizing it.

The difference is that "Select my unit, then target-click the enemy unit for it to go after" is how every other unit in the game works, so it's what new players are going to instinctively try.
Most of the objects in the game can't be hacked so how would the newbie know which objects can be targeted with the hacker?

instinctively
When I select a unit I instinctively check out the new window/information/tool tip that pops up. The one where you select the hacking sub units. How could anyone not see it. And if the player sees it why wouldn't the player read what the tool tips say. Too much text in the tool tips?

If we can make it work that way, it's one less time they need to stop playing to look something up.
What does the newbie have to look up when starting a hack? (I assume you mean something other than tool tips)

1)  I'm not fond of the way the hacking response works, for a couple of reasons:
...
[snip]
...
3)  It would be nice if one could research better Hackers.  Maybe the level 1 hackers don't even have cloaking.  Higher level ones might be able to do paralysis or reclamation attacks.
These changes sound good to me. Although simply making the AI redirect all the reinforcements into the planet being hacked wouldn't be much of a challenge.
« Last Edit: April 03, 2016, 05:25:47 pm by Kahuna »
set /A diff=10
if %diff%==max (
   set /A me=:)
) else (
   set /A me=SadPanda
)
echo Check out my AI War strategy guide and find your inner Super Cat!
echo 2592 hours of AI War and counting!
echo Kahuna matata!

Offline Pumpkin

  • Hero Member Mark III
  • *****
  • Posts: 1,201
  • Neinzul Gardener Enclave
Is Keith Arcen willing to revamp the Starship status by putting in all the big fleet ships with a cap of 5 and immunity to insta-kill? There would be a possibility to make a "starship" perk that would pack several perks, like immunity to reclamation, translocation, ion, etc, and ability to be targeted by OMD and arachnid GPosts (there is also two kind of swallowing, one targeting only fleetships, one targeting only starships). I think it would be both much clearer and more streamlined.

Similar question to buildings: what about a perk-pack with 0 speed, immunity to translocation, tractor, swallowing, reclamation, etc?

Other question: is Arcen willing to sort out the AI "personality" in plots, starting ship's blueprints and wave intensity/delay settings?

I'm willing to put together a detailed list of these features (like a list of what should be a starship and what are the exact mechanical differences, or how to sort the lobby's MF/Plot/Personality options) but I wanted to know if this AIW2 project would encompass that kind of revamp.
Please excuse my english: I'm not a native speaker. Don't hesitate to correct me.

Offline Draco18s

  • Resident Velociraptor
  • Core Member Mark V
  • *****
  • Posts: 4,251
Is Keith Arcen willing to revamp the Starship status by putting in all the big fleet ships with a cap of 5 and immunity to insta-kill? There would be a possibility to make a "starship" perk that would pack several perks, like immunity to reclamation, translocation, ion, etc, and ability to be targeted by OMD and arachnid GPosts (there is also two kind of swallowing, one targeting only fleetships, one targeting only starships). I think it would be both much clearer and more streamlined.

IIRC there's already an isStarship flag that confers all of those bonuses if the ship didn't already have them.
It's kind of why those low-cap fleet ships aren't starships: because they don't have the full set of starship immunities and they're not supposed to.

Offline Pumpkin

  • Hero Member Mark III
  • *****
  • Posts: 1,201
  • Neinzul Gardener Enclave
Is Keith Arcen willing to revamp the Starship status by putting in all the big fleet ships with a cap of 5 and immunity to insta-kill? There would be a possibility to make a "starship" perk that would pack several perks, like immunity to reclamation, translocation, ion, etc, and ability to be targeted by OMD and arachnid GPosts (there is also two kind of swallowing, one targeting only fleetships, one targeting only starships). I think it would be both much clearer and more streamlined.

IIRC there's already an isStarship flag that confers all of those bonuses if the ship didn't already have them.
It's kind of why those low-cap fleet ships aren't starships: because they don't have the full set of starship immunities and they're not supposed to.
I think they should. And also I think the flag must be displayed. Moreover, IIRC, the arachnid GPost targets only starships and big, low-cap (5) fleetships without it being clearly written somewhere.

I really want to know the position of Arcen/Keith on the 5-cap fleetships.
Please excuse my english: I'm not a native speaker. Don't hesitate to correct me.

Offline keith.lamothe

  • Arcen Games Staff
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 19,505
I don't think it's necessary that all low-cap fleet ships be moved to the starship category. But I'm not opposed to reevaluating them individually for clarity.

In general with the rewrite I'm not trying to redesign the game except where I see it's necessary from a foundational perspective (to make higher performance, multi-core, modding, etc less impossible) to avoid getting right back into the "well, it's already this way and it'd be really hard to change..." situation.

Stuff where it'd be straightforward to change after the lava's cooled (like the starships thing), and especially stuff that like that where the community may or may not even want it, I'm generally  not messing with during the main rewrite.
Have ideas or bug reports for one of our games? Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline Elestan

  • Full Member Mark II
  • ***
  • Posts: 158
The difference is that "Select my unit, then target-click the enemy unit for it to go after" is how every other unit in the game works, so it's what new players are going to instinctively try.
Most of the objects in the game can't be hacked so how would the newbie know which objects can be targeted with the hacker?

The tooltip of the hackable unit could indicate it, preferably with a prominent icon that looks like a Hacker.

Quote
When I select a unit I instinctively check out the new window/information/tool tip that pops up. The one where you select the hacking sub units. How could anyone not see it. And if the player sees it why wouldn't the player read what the tool tips say. Too much text in the tool tips?

Yes, and too many subunits to read through.  I remember the first time I clicked on a Hacker, I decided to put off learning about Hacking until I'd gotten a better understanding of the basics of the game.

Quote
What does the newbie have to look up when starting a hack? (I assume you mean something other than tool tips)
My immediate reaction to the Hacking process was "Hacking seems big and complicated.  I shouldn't try to do it without having read the Wiki first."  Most of the information I needed actually was in the tooltips, but it took effort to sift through it.  If the key bullet points (HaP cost, time, one-sentence effect description) were presented on the target, the player would be getting just the information they need, when they need it.

Offline Elestan

  • Full Member Mark II
  • ***
  • Posts: 158
So, another simplification request on my list is Warp Gates.  Right now, there are a bunch of different kinds of gates, but I've never found the differences to be very helpful.

In my opinion, it would be better to consolidate them down to one, or perhaps two types of gates:

The first acts just like an artificial wormhole between two systems; ships can fly through it just like a normal wormhole, use it in route planning, etc.  It requires building a gate on both ends, and cannot be re-pointed once built.

The second (perhaps higher tech level) is a fully networked gate.  It would also have a default (but changable) destination where ships go if just sent through the gate, but ships on a route that enter one will automatically emerge from the closest gate to their destination, to avoid the need to micro-manage the gate.

In all cases, the player would need Supply and a Builder on both ends to build a gate.
« Last Edit: April 05, 2016, 03:56:48 pm by Elestan »

Offline TechSY730

  • Core Member Mark V
  • *****
  • Posts: 4,570
unless the game gets dumbed down.
Just wanted to say: I will be doing no dumbing-down :)

Hooray! (party emote that isn't present on this forum instance)

The ridiculous custom-ability (except for the notable lack of mod support, which may be fixed if this pans out) and huge scale (both breadth and depth) are key features make AI War, well, AI war, IMO.

Offline Pumpkin

  • Hero Member Mark III
  • *****
  • Posts: 1,201
  • Neinzul Gardener Enclave
I'm generally not messing with during the main rewrite.
Got it. Thanks for the answer.

My main concern was to:
1) make advanced starship constructors more desirable;
2) improve clarity and coherency;
3) simplify tooltip by merging perks into one "starship" status.

Speaking of merging perks under a status, I think the "building" question got packed with the starship matter, but I feel there is a more subtle concern here. The way AI War currently works: everything is a "unit", and buildings are just units with 0 speed and perks like immunity to swallowing, translocation, tractor, etc (everything that would move it). Which is fine. But maybe there is something to do with the core of the game's engine about the buildings, more than with the starships that can be modded fleet- or starship easily. Would it be interesting to split "unit" between "ship" and "building" more deeply than in the stats? Like the starship status the tooltip would be simplified by merging the various immunities, but maybe there are optimizations or something more to gain with this. I don't know.

Or maybe I'm wrong and it's no different from the starship type and would be easily modded latter.
Please excuse my english: I'm not a native speaker. Don't hesitate to correct me.

Offline Pumpkin

  • Hero Member Mark III
  • *****
  • Posts: 1,201
  • Neinzul Gardener Enclave
This sounds to be the place about AIW2, so here I am (again) for (yet another) question.

Do you, Arcen, plan to add mod support for galaxy generation? For instance, I would see something like a Lua interface or so where each graph is a file with a function that defines how vertices link to each other. User input would be defined by such "mod" and could be on/off checkbox (do you want your clusters to be microcosm?) or integer sliders (total number of planets, width and height of a grid map, etc) or float sliders (probability for a link between A and B, etc).

Does this sounds feasible? Or too complicated for too few added value?
I'm curious.
Thanks in advance.
Please excuse my english: I'm not a native speaker. Don't hesitate to correct me.

 

SMF spam blocked by CleanTalk