Wasteland Wasteland 2 Pre-Release Discussion Thread [GAME RELEASED, GO TO NEW THREAD]

[Darkzone]

@Temaperacl

I was the first one to suggest at least the cost of 1.5 for the transitions. (Why i did this you will not understand.)

Incorrect - Jim The Dinosaur suggested it before you made your first post.

You should have read further the post:
You can easy verify this if you look into the official Wasteland 2 forum, at the Update 34.
This releates to the InXile Wasteland 2 forum. And i stated it at 31.8.2013, which comes before the 04.09.2013, at least in the normal Calendar.
What do you think imweasel was arguing about? If you will answer this then i will show you where are there the mistakes.

 

[Gord]

Give it up already.

 

[imweasel]

weasel said that movement costs should be closely proportional to the distance travelled, regardless of direction. This was the entire point of squarecirclegate, which was a pic I found prudent because the main argument in favour of hexes were that they made the apspent/travel ratio more consistent. I really, really don't get how anything he said was controversial in the least o_O

Edit: Actually I might have misrepresented imweasels point a bit, since the discussion is so long. But the circle represents all the points which are equally far from point A, so clearly any movement function that aims to have a consistent AP/movement ratio should strive to be as close to the circle as possible, isn't that what you tried to show...?

Exactly.

It never came up because it's solved by having 3 AP for diagonal, 2 AP for normal, movement.

Yes, this came up a few weeks ago (I think you mentioned it first, not sure).

According to Brother None in the main forum, they are looking at tweaking this system a bit too.

 

[Darkzone]

@herostratus

imweasel said that movement costs should be closely proportional to the distance travelled, regardless of direction.

And even this is a mistake, any obstacle any terrain modifier changes this. And in the really world it does not function at all. It is only one and only one case from all the inifinite possibilities that correspondes to this case a plane without any features.
Drawing a circle around some squares does not solve any problem of navigation. The reason why i always speak about the A* is that it is the most common algorithm to solve the pathplanning problem, but there are hunderts more of algorihtms to solve this problem, and non of them is draw a circle around some squares. The reachability graph has to be developed by a greed algorithm which takes the terrain features and other modifiers into the account, and mostly it is not the distance between two points, or the distance that one can travel in a certain time.
How do i explain it to someone who does not understand the problems of navigation? Think this way you are standing on a mountain, you can travel up / down or stay at the same altitude (go around to the sides). Up you will travel slower, down you will travel faster, or travel to the sides but you will never travel the same distance in any direction, besides the symetrical mirrowed by the top/down axis. The only thing imweasel has done is to say that a circle can include a symetrical figure made out of squares, but the same applies to the hexes and does not mean anything. So to say he did say really nothing concerning the navigation. Navigation contains obstacle avoidance and pathplanning. His examples and thinking generates only mistakes in pathplanning and reachability graph generation.
Further his first example as an answer to felipepepe was in all matter false. Why this is so should be obvious.

 

[Gord]

Ban Darkzone already (and imweasel, too, if necessary)

 

[Brotherman Bill]

nd even this is a mistake, any obstacle any terrain modifier changes this. Drawing a circle around some squares does not solve any problem of navigation.[...]Navigation contains obstacle avoidance and pathplanning

He isn't trying to solve any problem of navigation. He's not trying to write an algorithm.

The discussion was about which methods - squares, hexagons, squares w/1.5x diagonal costs, etc, would keep movement/ap ratio equal no matter which direction you move. It is true that no one specified the premise "given equal movement costs for each square" but it was implied, and pretty obviously at that...

 

[Jim the Dinosaur]

I brought this up earlier - having "variable spending" of AP on walking is going to generate fractions, which is a terrible way to go. Hopefully they're not actually going to proceed with that.

I really doubt they'll end up with fractions, especially because movement and other actions are calculated the same way (non-fractional attacks combined with fractional movement would give really weird results, not the least of which being that you'd probably almost never automatically finish the turn due to being out of AP). So I think they'll be rounding, which has a lot of the same issues (having to always hover your mouse around to see how much AP movement costs because you'll never be sure how much it costs), in addition to creating some pretty arbitrary values.

 

[tuluse]

They could easily hide the fractions, and only display whole numbers. If they don't do fractions, and movement costs are not 2/3, it's gonna be fucked up.

 

[Jim the Dinosaur]

They could easily hide the fractions, and only display whole numbers. If they don't do fractions, and movement costs are not 2/3, it's gonna be fucked up.

How do you hide fractions? Either you have a movement cost 2.3 AP and then have an attack that costs 3 AP fail when you have 2.7 left or you don't. If you have it fail, then you have to display it as 2.3 or else the player won't understand why it failed, if you have it succeed then you're rounding.

 

[tuluse]

1) You don't round the shown number, you just display the integer part. So 2-2.9 all show the player 2
2) You don't let actions "fail" you just don't let the player use an action that takes more AP that is available. Just like Fallout. You can't fire your 4 ap gun if you only have 3 ap, so why would it let you "fail" if an action costs 2.3 and you only have 2 or whatever.
3) You do round AP and just let characters have up to an extra .5 AP per turn

 

[Gord]

Ninja'd.

What tuluse said.

 

[Jim the Dinosaur]

2) You don't let actions "fail" you just don't let the player use an action that takes more AP that is available. Just like Fallout. You can't fire your 4 ap gun if you only have 3 ap, so why would it let you "fail" if an action costs 2.3 and you only have 2 or whatever.

That's what I meant by failing. Also, I'm either a complete idiot or your example doesn't make any sense: 4 is more than 3, 2.3 is more than 2, so of course both should fail. Unless you're rounding :aiee:.

3) You do round AP and just let characters have up to an extra .5 AP per turn

I don't know about you, but I'd find rounding less arbitrary than this.

 

[tuluse]

I thought by failing you meant: the player would attempt an action, the action wouldn't happen and the player would lose AP. Civ2 had a thing like this. If you moved on a road it would only take 1/3 of an AP, so if you tried to move off the road on the same turn as 1 road movement, you had a 1/3 chance of failing. You wouldn't move and would stay in the same square and your 2/3 of an AP would be spend. After 2 road movements, it was a 2/3 chance of failing.

I'm really not seeing a problem here. 2 AP listed when you have 2.5, action takes 2.5 or less, you can do it.

Also, the final answer isn't really arbitrary at all. What it means is that you have [AP listed] + .5. So 10 AP is actually 10.5 AP. It's not the best form to hide that from the player, but it's not arbitrary.

 

[Jim the Dinosaur]

I'm really not seeing a problem here. 2 AP listed when you have 2.5, action takes 2.5 or less, you can do it.

Because if the action cost 2.9, and you have 2.5 left, they'll both be shown as 2, and yet you get a message that you can't do it. How is that not confusing?

Also, the final answer isn't really arbitrary at all. What it means is that you have [AP listed] + .5. So 10 AP is actually 10.5 AP. It's not the best form to hide that from the player, but it's not arbitrary.

Then I misunderstood you, but that doesn't really fix anything. You want to do two actions costing 5.9 AP (shown as 5) and yet you can't because you only have 10.5 AP.

 

[tuluse]

The player AP would be shown as an integer, not the action AP. That would be listed as the amount it costs.

 

[Lhynn]

Why not multiply the AP numberx10? that should allow enough flexibility. Not a fan of big numbers, but if they give me better feedback for my actions i could live with it.

 

[Jim the Dinosaur]

The problem isn't with the fractions themselves. If you go from 10 to 100 AP per turn, and attacking costs 30 AP, then you're still searching around for that one route costing 70, and not 71 AP. Also, you get the annoying thing that if you go for the route costing 69 and then attack for 30, you don't get the auto end turn at 0 AP because you have 1 AP left (though I suppose you could work around this partially by calculating some minimum AP requirements per attacker).

Maybe they can find some solution to these things (or I'm just missing a very obvious one), but it really sounds like they built their AP system around straight movements, which allowed them to have variable movement costs without too many complications, and then had to switch to a system which doesn't cooperate at all with these variable AP costs at the last minute due to the criticism.

 

[Lhynn]

The problem isn't with the fractions themselves. If you go from 10 to 100 AP per turn, and attacking costs 30 AP, then you're still searching around for that one route costing 70, and not 71 AP. Also, you get the annoying thing that if you go for the route costing 69 and then attack for 30, you don't get the auto end turn at 0 AP because you have 1 AP left.

Maybe they can find some solution to these things (or I'm just missing a very obvious one), but it really sounds like they built their AP system around straight movements, which allowed them to have variable movement costs without too many complications, and then had to switch to a system which doesn't cooperate at all with these variable AP costs at the last minute due to the criticism.

Why not go the ToEE route then?

 

[Jim the Dinosaur]

They're in polishing phase right now, I'm even surprised they could go the step of adding diagonal movement at this point. Going to free movement would be impossible (rebuilding everything basically). Would be funny though if they went from hex to square to free though.

 

[tuluse]

ToEE is 3E with move action and standard action, why do that instead of good old action points?

Also, they only need to be as specific as .5 AP, so doubling action points should be fine and then do 2/3 move cost.

 

[Darkzone]

Jim the Dinosaur

They're in polishing phase right now, I'm even surprised they could go the step of adding diagonal movement at this point.

If you know how those things work, then this is not a surprise, and you know that this is not a problem. I have been bashing at them to do this in the InXile forum from day 1 after the Prison Demo.

 

[Jim the Dinosaur]

Darkzone I do know exactly how these (ALL OF THESE) things work.

Like they've pretty much admitted themselves ("the calculations get pretty complicated"), their movement system was based on the idea that there would be straight movement and that this simplicity would allow them to play around with movement costs via the speed stat. It's weird to switch that around last minute without also changing the way the speed stat works. But maybe I'm wrong and I don't know how these things work.

 

[Gord]

I would probably just have used the grid for placement and nothing else, but in the end I have faith that those guys know more about designing the game and it's systems than the average forum guy/codexer. In other words, I'm relatively sure that they will find a working solution to those concerns.

 

[Darkzone]

"the calculations get pretty complicated"

This can be very misleading, because it depends on many factors.

their movement system was based on the idea that there would be straight movement and that this simplicity would allow them to play around with movement costs via the speed stat. It's weird to switch that around last minute without also changing the way the speed stat works.

Not necessary. They can still apply it, the same way, without any change. In worst case they have to change the algorithm for the reachability graph with the terrain modifiers which have to be extra evaluated for the new edges of the graph. The movement range is already given, by the APs and the speed modifier.
The movent range goes then in to the evaluation of the reachability graph.

 

[tuluse]

I would probably just have used the grid for placement and nothing else, but in the end I have faith that those guys know more about designing the game and it's systems than the average forum guy/codexer. In other words, I'm relatively sure that they will find a working solution to those concerns.

If you're not going to use a grid for movement, why would you use it for placement?