Update 35: We’re Still Listening

Announcements & media coverage pertaining to the Wasteland series. Only moderators & inXile can make new threads on this forum.

Moderator: Ranger Team Alpha

Post Reply
User avatar
Darkzone
Adventurer
Posts: 508
Joined: December 21st, 2012, 7:37 pm

Re: Update 35: We’re Still Listening

Post by Darkzone » September 27th, 2013, 1:30 pm

@Priest4hire
It's pointless, he has never programmed anything nor did he think more and deeper about it, because he cannot do it. I did let it stand so that the develpers of the "A.I." (As i call it the automated skripted agents), could have samething to laught about. He thinks he is smarter then people who did study this kind of problems and developed algorthms to solve this like Dijkstra.
If you are interested in something like this read a book for beginners like "Introduction to Autonomous Mobile Robots" chepter 6. Or a little bit more complicated but still a classic: "Artificial Intelligence: A Modern Approach". Also a classic: "Computational Intelligence - A Logical Approach". This one i have not read but it is from the same authors: "Artificial Intelligence: Foundations of Computational Agents".
This one is an applet to learn something about search Algorithms.
In the Wasteland every breath i take. I am.

Thank you Brian Fargo for the games i play.
Thank you Fields of the Nephilim for the music i hear.

meganothing
Scholar
Posts: 238
Joined: April 28th, 2012, 4:40 am

Re: Update 35: We’re Still Listening

Post by meganothing » September 27th, 2013, 6:10 pm

ffordesoon wrote: I'm not sure what your point is, frankly.
The point is that you are too pessimistic but even if all your (negative) reasons vanish in a cloud of happy thoughts there are not many (positive) reasons for the devs to be more active on the forums.

User avatar
dorkboy
Master
Posts: 1772
Joined: November 26th, 2012, 10:37 am

Re: Update 35: We’re Still Listening

Post by dorkboy » September 27th, 2013, 8:13 pm

some late feedback with regards to the dialogue UI/NPC text:

1)
in both the 2 demos the text formatting within the dialogue window/bubble is fairly blocky,
so NPCs with a lot on their mind will speak in large, rectangular/squarish chunks of text.

i think increasing the overall "text space" as well as altering the formatting to include more paragraphs & natural pauses (not just between "chunks", but within each chunk of text, as well) could help both readability and better illustrate NPC speech patterns/personality. (without rewriting the actual content).

b)
in the prison demo the coloured keywords are fairly distracting/jarring.
they draw very much attention to themselves - which is good for text that is meant to be skimmed.
not so good for text that is meant to be a narrative experience.
it seems to lean a bit towards dialogue as "info-mining NPCs by clicking unlockable buttons".

&)
insofar as there are descriptive text blurbs (shown in the matrix printout) relating to a specific character, perhaps those should be included in the dialogue window as a preface to talking to that particular NPC?
marmelade & jam

User avatar
ffordesoon
Forum Moderator
Posts: 1724
Joined: March 13th, 2012, 4:39 pm
Contact:

Re: Update 35: We’re Still Listening

Post by ffordesoon » September 27th, 2013, 8:34 pm

@meganothing:

That was a list of reasons not to communicate, not a list of things that would definitely happen if they chose to communicate with us more. Lists of reasons not to do things are pessimistic by their very nature.
I can explain it to you, but I can't understand it for you.
As a poster, I speak only for myself. PM me if you have any problems with the way I moderate.

My Tumblr, if you're into that.

User avatar
Drool
Forum Moderator
Posts: 9757
Joined: March 17th, 2012, 9:58 pm
Location: Under Tenebrosia, doing shots with Sceadu.

Re: Update 35: We’re Still Listening

Post by Drool » September 27th, 2013, 10:58 pm

Priest4hire wrote:Since you didn't provide a grid, I measured two sides and came up with roughly 8cm x 7cm. Assuming one AP per cm it allows us to estimate the movement cost. Path B is simple enough. It's just (8+7=15). Path A is a tad more complex since the character will need to mix diagonal and normal movement as he'll hit the top before reaching the target. It's (7*1.5+1=11.5). Even rounded up, that's still less than 15. So how exactly does a diagonal movement cost of 1.5 result in diagonal moves taking more AP?
Presumably a diagonal movement wouldn't be calculated via stair-stepping, but rather via Pythagoras:

a^2+b^2=c^2
8^2+7^2=113
sqrt 113 = 10.63014581273465

Which could be rounded a few ways. If it's a floor() function, then we'd get a distance of 10 with a movement cost of 1.5 which would give us 15 exactly. If it's rounded up to 11, we get 16.5 which is more costly. If it's kept at 10.6 we get 15.9. So, in 2/3 of the most likely rounding methods, we get a 1-2 AP premium for diagonal movement.

Or they could just calculate the cost via path B and animate via path A.
Alwa nasci korliri das.

User avatar
Priest4hire
Explorer
Posts: 480
Joined: May 9th, 2012, 12:38 pm

Re: Update 35: We’re Still Listening

Post by Priest4hire » September 27th, 2013, 11:33 pm

They've already stated that diagonal movement is possible, and that it costs 1.5 per square. Why put that in only to toss it out again in favour of a different scheme the moment you actually move diagonally? Wait, you seem to be saying the 1.5 cost is added to the distance traveled, so that movement in certain directions costs half again as much as movement in other directions. That seems ass backwards to me. I assumed the 1.5 represented the greater distance moving diagonally from square to square; rounded up from 1.41. Thus it's 1.5 per square, not 1.5 times the total distance.

User avatar
sparkee666
Adventurer
Posts: 663
Joined: June 15th, 2012, 10:32 am

Re: Update 35: We’re Still Listening

Post by sparkee666 » September 28th, 2013, 12:28 am

I always chalked movement up to board games, the movement along the edges would be 2 action points because a turn was required to travel the rest of the distance. The hypotenuse is always longer than the edges of a tile, but because there is no turn required traveling the length of the hypotenuse it wouldn't take the full two action points it would have to take 1.5 action points because of the Pythagoras theorem.

User avatar
Hiver
Adventurer
Posts: 743
Joined: March 13th, 2012, 4:17 pm
Contact:

Re: Update 35: We’re Still Listening

Post by Hiver » September 28th, 2013, 3:11 am

Priest4hire wrote:Since you didn't provide a grid, I measured two sides and came up with roughly 8cm x 7cm. Assuming one AP per cm it allows us to estimate the movement cost. Path B is simple enough. It's just (8+7=15). Path A is a tad more complex since the character will need to mix diagonal and normal movement as he'll hit the top before reaching the target. It's (7*1.5+1=11.5). Even rounded up, that's still less than 15. So how exactly does a diagonal movement cost of 1.5 result in diagonal moves taking more AP?
Yes, of course.
Your calculation is - naturally - correct.

What i dont understand at all is - why in the hell would we calculate movement as if we were moving on a square grid - if we are in fact NOT. Since there is no f/ing grid at all.

Then how and why in the seven hells would we move "mixing diagonal and normal movement"? What do you mean "hit the top"? What top?

What the hell does that even mean?

There is no "diagonal" at all. Youre only moving in a straight line - which you may only call "diagonal" relative to other directions.

User avatar
Darkzone
Adventurer
Posts: 508
Joined: December 21st, 2012, 7:37 pm

Re: Update 35: We’re Still Listening

Post by Darkzone » September 28th, 2013, 3:48 am

Priest4hire wrote: I assumed the 1.5 represented the greater distance moving diagonally from square to square; rounded up from 1.41. Thus it's 1.5 per square, not 1.5 times the total distance.
Basically yes, but more a first approximation, towards the square root of 2, which by the way would not do very much better.
The problem of the a square grid is that you cannot move as realistic as we would wish, but it offers many benefits for the automated agents. Like the use of the search algorithms, position value evaluation, reachability of certain fields, etc. In other words it make it more easy to make an automated agent, with lesser degree of flaws ( a lesser chance of stupid behaviors). Yes the cover system that we currently have may even require a square grid.
A gridless system is very problematic towards the mentioned above functions, that is why we even make a gridless map to a square grid map, besides the collision avoiding. For a colision avoidence are the voronoi diagram much better, but we prefer the square grid. And in a game movement on the voronoi diagram would be very confusing, and this would look very artificial.
Priest4hire wrote:Why put that in only to toss it out again in favour of a different scheme the movement you actually move diagonally? Wait, you seem to be saying the 1.5 cost is added to the distance traveled, so that movement in certain directions costs half again as much as movement in other directions. That seems ass backwards to me.
They cannot toss it out in that way. A grid is represented by a graph, and the transition cost would be in the diagonal direction 1.5x the transitions costs for the vertical and horizontal direction. But that is not so easy, if you have terrain factors that influence the movement the 1.5x have to calculated towards the entry cost of this square, for the adjacent square of the both squares.
On the graph each node has a value, the use of a coversystem as an example gives some nodes a higher value, then others. An automated agent for example evaluates this value and action possibilites, to perhaps max a certain desired actions, like: Do as much damage as possible, but stay in cover. The grid gives the automated agents a very different possible behaviors. Yes the enemies may act very different, not only melee and ranged enemies like in the most known gridless systems. The explanation why this is much more difficult on a gridless system, i will spare myself.
Now to a deeper explanation of the movement of the automated agents, before a movement is made a reachability graph would be made to have the movement borders, each field (node) would then have assigned then a cost value, to reach it. The automated agent will then evaluate the action possibilities and take what he was scripted for.
Then most possible a A*-algorithm will be taken, to get a path from the current field (node) to y (the desired node), if naturally the nodes are different. The path can be then executed.
Now why not a simple pythagoras to calculate the movement because:
First the calculation overhead would be extreme, you got to calculate from each node the distance to an destination node. How great is the cost for counting the distance or to safe the values: O((n^2)/2) with n equals the number of squares in that map. Now think if this map has only 100x100 nodes, this means: FUCK YOU!!!.
Second you don't know if between two existed nodes is a straight connection at all.
Third you cannot have any terrain factors.
I will stop here because i could go on an on and on.
In one of the answers to Hiver i told a story about an accident, what happend as someone tried to optimise the path of a robot, by very simple manners. Second and Third were the reason for what and why it has happend. In rpgcodex i argue with an other "genius" why it does not function what he assumes, there i suggested solutions for a more natural movement with functions at least of the second degree, as an approximation to the path and to smooth it out. I did also suggest this solution in this forum a long time ago.
Yes no straight lines, to cut transisted fields (nodes) of the path off.

To anyone who has managed to read this entirely through.
Check out the search.jar applet from aispace, to gasp at least some understanding about this things, it is very interessting and i love it to simply play around with this things.
For any interese and questions about robotics and navigation, just ask me, in the worst case i have no time, to answer it.

And by the way you can be about one thing certain: the developers of the "A.I." at InXile do know exactly what they do, and they surely know the pythagorean theorem. They are not stupid, this are people who know mathematics and how to programm complex things, most probable they even got computer science or physics or mathematics degrees, and are not newcomers.

Edit:
I have just seen that under the Pathfinding is everything necessary listed, to understand this. I should have looked before i wrote the whole text, because i could have spared me much of explanation work.
In the Wasteland every breath i take. I am.

Thank you Brian Fargo for the games i play.
Thank you Fields of the Nephilim for the music i hear.

User avatar
dorkboy
Master
Posts: 1772
Joined: November 26th, 2012, 10:37 am

Re: Update 35: We’re Still Listening

Post by dorkboy » September 28th, 2013, 9:10 am

fascinating stuff, Darkzone.
i've always wanted to know more about pathfinding, but i've also always been too lazy to find out. ;)
marmelade & jam

Militant
Scholar
Posts: 120
Joined: September 28th, 2013, 10:42 am

Re: Update 35: We’re Still Listening

Post by Militant » September 28th, 2013, 11:02 am

I havnt been following W2 too closely, havnt checked on it in about a year, but I have to say I'm IMPRESSED by the progression and the early quality! I was very excited to learn of W2 as few games have matched the gameplay since.

Since you have announced a beta coming next month, A suggestion... Spruce up your website, it's pretty damn sad with minimal information on the game and leaves alot to be desired (i want as many people working on the game like everyone else! but Betas generate Buzz, and some prospective gamers wanting to checkout the site to learn about the game will likely immidately be like WTF is this?)

Also, I tried searching, but couldn't find anything as to the size of the world, only topics on how many hours of gameplay.. So approximately how big is the world? Fallout2 sized, Fallout3, Skyrim?

User avatar
Darkzone
Adventurer
Posts: 508
Joined: December 21st, 2012, 7:37 pm

Re: Update 35: We’re Still Listening

Post by Darkzone » September 28th, 2013, 11:20 am

@Milittant
About the size of Fallout 2. 40 to 50 maps smaller and larger.

@dorkboy
I am glad that my writing has found at least, one reader, and that you liked it.
In the Wasteland every breath i take. I am.

Thank you Brian Fargo for the games i play.
Thank you Fields of the Nephilim for the music i hear.

Militant
Scholar
Posts: 120
Joined: September 28th, 2013, 10:42 am

Re: Update 35: We’re Still Listening

Post by Militant » September 28th, 2013, 1:59 pm

That's not bad, I was hoping for more, but that's a respectable size considering they don't have a huge budget. And if the game does well as I hope it does, hopefully there will be more content down the road :D

User avatar
Darkzone
Adventurer
Posts: 508
Joined: December 21st, 2012, 7:37 pm

Re: Update 35: We’re Still Listening

Post by Darkzone » September 28th, 2013, 2:05 pm

Hope for the best, expect the worst. I believe that it will have additional sales between 200k and 300k, and that is enough for an extension.
In the Wasteland every breath i take. I am.

Thank you Brian Fargo for the games i play.
Thank you Fields of the Nephilim for the music i hear.

Militant
Scholar
Posts: 120
Joined: September 28th, 2013, 10:42 am

Re: Update 35: We’re Still Listening

Post by Militant » September 28th, 2013, 3:59 pm

Have they stated how much money was donated? Or what they hope to get in sales?

Edit: Found a destructoid vid on youtube that said they got over 3mil last year, 6th most money on a kickstart, pretty good
Last edited by Militant on September 28th, 2013, 5:44 pm, edited 1 time in total.

meganothing
Scholar
Posts: 238
Joined: April 28th, 2012, 4:40 am

Re: Update 35: We’re Still Listening

Post by meganothing » September 28th, 2013, 5:40 pm

ffordesoon wrote:@meganothing:
Lists of reasons not to do things are pessimistic by their very nature.
Counter-example: Reason not to kill someone on your birthday party: "You very likely will go to jail." This reason not to do something is not pessimistic, it is realistic.

"Secondly, anything a dev says is automatically treated as a promise", "Fifth, devs don't settle arguments by stating that they're coming down on one side or the other. They reignite them." or "This never stops."

In their finality and inevitability these statements were very pessimistic. But ok, it was more entertaining this way, so I probably just didn't notice the artistic flavour.

User avatar
Gizmo
Grandmaster
Posts: 3715
Joined: March 6th, 2012, 6:25 am

Re: Update 35: We’re Still Listening

Post by Gizmo » September 28th, 2013, 7:38 pm

Hiver wrote:for the devs:
Just make a visual representation that shows me the movement range i can achieve with a selected character and AP cost - WHICH SHOULD BE LESS for going in any straight line - whether its "diagonal" or not.
The point of combat in games like these is to win by using the rules, not whether or not it follows reality in the least. If it does ~cool; if it does not, then it's simply the rules of the game... you play according to the rules ~the challenge of winning the fight is unaffected by this.

You don't see anyone complaining that mounted knights don't really charge their opponents in 'L' shaped leaps from point to point; it's a function of the game design, and they serve a purpose. In innumerable RPGs the wizard [aka the artillery unit] is vulnerable and near helpless in close melee combat ~it's a function of the game design, not plausibility... A wizard is presumably brilliant, and there is no credible reason they could not train with a sword & shield... but allowing that in the game diminishes the fighter classes whose specialty is combat, not spells. Game rules always trump the fiction and even all semblance of believability if it comes to it.

'Realms of Arkania' also uses a square grid and APs, and it plays just fine like that; so will WL2. They won't release something they are unhappy with.

________________________

I agree with what I've read posted about dev interaction on the forums...but I know at least one forum that must be near magical for the environment... Devs post all the time, answering questions, and even giving suggestions to modders. I've even seen them sit down with a beer and call for feature suggestions, and code them on the spot, and they appeared in the next patch. One guy asked for a UI change and they Dev asked him why, and it was because he only had one serviceable hand... And so they changed it; added mouse-keys to optionally replace the WASD if desired.

Twice I asked for features in their level editor ~and twice those features showed up; Great place, great forum; great crowd on that forum. They made the Legend of Grimrock game.

User avatar
Blacky
Explorer
Posts: 272
Joined: January 25th, 2013, 4:04 am
Location: Provence, France
Contact:

Re: Update 35: We’re Still Listening

Post by Blacky » September 29th, 2013, 7:59 pm

Gizmo wrote:In innumerable RPGs the wizard [aka the artillery unit] is vulnerable and near helpless in close melee combat ~it's a function of the game design
Mostly because of incredible bad game design, because:
not plausibility... A wizard is presumably brilliant, and there is no credible reason they could not train with a sword & shield
Or find a close combat application of magic.

Well, it depends. In a very unreal game, like Tetris, Pac Man, etc. that's not an issue. Having blue square do something and red circle do something else, nobody has an issue with it. Switch pieces by wizards and knights, same thing.

In a simulation game (not the genre, but the definition), when the game is introduced by something like “In this age of war, YOU are the Talent born, you will conquer lands, make peace, build alliances, you'll be the end of an age and the beginning of the next [cue Hollywood music]” or whatever, it's mostly badly game design.

There is a profound need for believability. It may be a land of magic, or a stellar system in a distant future, or whatever. But internal coherence is paramount.

Brian Fargo said it quite recently:
Brian Fargo wrote:If things make sense in a real-world sense, I don’t think people give too much flak for it not being game-y enough, I think keeping the world sense together is the most important thing we can be doing
If you don't do that because you're lazy or you spend $10b on a graphic engine but $2K on rule mechanic (that's mostly it, this should cover 95% of the bad games out there).
Last edited by Blacky on September 29th, 2013, 8:10 pm, edited 1 time in total.
Chez Blacky, jdr et autres
If things make sense in a real-world sense, I don’t think people give too much flak for it not being game-y enough, I think keeping the world sense together is the most important thing we can be doing.—Brian Fargo

User avatar
Blacky
Explorer
Posts: 272
Joined: January 25th, 2013, 4:04 am
Location: Provence, France
Contact:

About stances

Post by Blacky » September 29th, 2013, 8:06 pm

About stances:

One thing I would like to see first is crawling. I mean, stances as a mobile soft cover, a way to steady your shot, etc. that's all good but it's far from critical. If the dev team after reviewing it thinks it's too much work, I can live with it.

However, crawling seems much more important. It's critical to do recon, and plan&execute ambushes. And it's by far the easier stance to implement, since there's much less AI to do with it.
Chez Blacky, jdr et autres
If things make sense in a real-world sense, I don’t think people give too much flak for it not being game-y enough, I think keeping the world sense together is the most important thing we can be doing.—Brian Fargo

User avatar
Drool
Forum Moderator
Posts: 9757
Joined: March 17th, 2012, 9:58 pm
Location: Under Tenebrosia, doing shots with Sceadu.

Re: Update 35: We’re Still Listening

Post by Drool » September 29th, 2013, 10:01 pm

I think it's pretty clear that any stances are a long shot.
Alwa nasci korliri das.

Post Reply

Who is online

Users browsing this forum: No registered users and 3 guests