PC Developer Discussions: Alpha 17

Developer Discussions: Alpha 17

  • Newly Updated

    Votes: 1 100.0%
  • Check out the newest reveals by Madmole

    Votes: 0 0.0%
  • Over 100 new perk books with set collecting and bonuses

    Votes: 0 0.0%

  • Total voters
    1
Status
Not open for further replies.
So no chance of a truly high-def texture pack? That's gonna annoy at least two people.
If we can find where the altas mapping is, I know a few people who are willing to make a larger atlas.

Here is a small shot of the atlas:

deleteblockatlas.png


...an here is what I /think/ is the mapping for brick in said atlas. The numbers would be coordinates, but I don't know how to apply them to the atlas.

Code:
    <uv  id="7" x="0.004150391" y="0.004150391" w="0.04162598" h="0.04162598" blockw="3" blockh="3" color="0.281089,0.2328958,0.2147789" globaluv="True" material="stone" texture="brick.tga" />
...but I'm sure it involves math. :)

 
pillars.....

I'm not against that technique in and of itself. I just wish the gap between pillars didn't visually appear as though the zombies should be able to slip between them-- and yet they can't. I think they would be balanced better if there were a couple of smaller zombie types that could squeeze between them so you'd have to be on the lookout for that...putrid girl and the screamer chick for example-- or dogs.

 
Last edited by a moderator:
pillars.....
I'm not against that technique in and of itself. I just wish the gap between pillars didn't visually appear as though the zombies should be able to slip between them-- and yet they can't. I think they would be balanced better if there were a couple of smaller zombie types that could squeeze between them so you'd have to be on the lookout for that...putrid girl and the screamer chick for example-- or dogs.
Yep, the moment dogs can slip through that gap it'll be time for a design change. I'd use arrow slits but they're nowhere as useful in-game as they are in real life, if you put them on the "right" way around you haven't got enough ability to pan your weapon. Putting them on backwards is better, it funnels the zombies towards the gap although looks bloody weird for someone who has grown up around real castles.

 
Are you talking about outsmarting Zombies? Seriously I don´t expect them to be super smart at all.

But if I build a tunnel into a mountain with a big entrance and defences etc. ... I would realy like to see them take the entrance.

 
pillars.....
I'm not against that technique in and of itself. I just wish the gap between pillars didn't visually appear as though the zombies should be able to slip between them-- and yet they can't. I think they would be balanced better if there were a couple of smaller zombie types that could squeeze between them so you'd have to be on the lookout for that...putrid girl and the screamer chick for example-- or dogs.
Agree completely. And I hope they can fit through them in the future.

The pillar base and toothy are the 2 that IMO need to move on. And I hope that A17 will help that happen.

 
Ok, so im a complete Noob in IT/Programming Stuff, but is it that hard to expand the Atlas? Not skeptic, just trying to understand.

IMO Both, the Building and exploring Aspect, would highly benefit from more Variety and more detailed Textures, so maybe TFP could invest some Ressources in this ?

 
Reason I ask this, is I swear there were some exploit(ish) base build styles that used the Door targeting priority to essentially mess with Zombie AI and Pathing.
I'm guessing no, or at the very best, minimal depth and overall health of depth checking at best could be implemented. Just wondering if we're reintroducing behavior that was "resolved" at one point.
Not sure I get the scenario you describe, but doors are now the same as other blocks, they just are considered to have less health than they actually do. Depth is not exactly what it is doing. It is path of least resistance (lowest health) in all directions to you.

 
Can I presume then, that doors now are "paintable"? :)
He meant the same as other blocks in terms of targeting. There used to be an AI Task to break doors specifically as a priority. He probably got rid of that task and now they just follow the path of least resistance which also happens to be doors compared to other blocks. But they cannot be painted as of now.

 
Ok, so im a complete Noob in IT/Programming Stuff, but is it that hard to expand the Atlas? Not skeptic, just trying to understand.
IMO Both, the Building and exploring Aspect, would highly benefit from more Variety and more detailed Textures, so maybe TFP could invest some Ressources in this ?
Depending on how it's implemented you can have larger atlases but that suffers from the exponentially increasing resources associated with large textures, an atlas is simply a single texture that you use little chunks of. If you double the size of a texture you're actually using four times the system resources to hold it, you're doubling both width and height.

In theory you can have a single atlas per texture and have 8k textures but having a single texture per atlas negates the advantage of using an atlas.

- - - Updated - - -

Thanks! Little more to it to convert to pixel coordinates in a paint program, but this let me crack the nut.
Gimp has a feature to do this, a lot of people use it to make atlases.

 
Not sure I get the scenario you describe, but doors are now the same as other blocks, they just are considered to have less health than they actually do. Depth is not exactly what it is doing. It is path of least resistance (lowest health) in all directions to you.
Hi, I have a question about the pathfinding: What is the cost of moving over an unobscured tile compared to breaking a block/door?

Eg; at what point is walking tiles on open terrain (detour) have a higher cost than breaking a simple level 1 door.

 
He meant the same as other blocks in terms of targeting. There used to be an AI Task to break doors specifically as a priority. He probably got rid of that task and now they just follow the path of least resistance which also happens to be doors compared to other blocks. But they cannot be painted as of now.
Thanks Roland :)

 
Pillars

pillars.....
I'm not against that technique in and of itself. I just wish the gap between pillars didn't visually appear as though the zombies should be able to slip between them-- and yet they can't. I think they would be balanced better if there were a couple of smaller zombie types that could squeeze between them so you'd have to be on the lookout for that...putrid girl and the screamer chick for example-- or dogs.
Roland: For the pillar scenario is it possible to designate the pillars as a polygon vs a voxel? Because as you wrote a voxel is a complete 3X3 block with only the texture being visible to represent the pillar, but the remainder is still impenetrable. Or for pillars designate them as 1/4 1/3 1/2 voxel volume to free the air above and below. Referring back to one of the original builds the shutters and boards looked like polygon objects when they broke up.

 
Hi, I have a question about the pathfinding: What is the cost of moving over an unobscured tile compared to breaking a block/door?Eg; at what point is walking tiles on open terrain (detour) have a higher cost than breaking a simple level 1 door.
Depends on the length of the detour and what the values are for the type of door. For example, let's say a wooden door has a value of five, a detour of four blocks or less would be taken. If the door has a value of twenty-five then a detour of twenty-four or less would be taken. Things get a little more complicated when you factor in diagonals which are worth roughly about 1.41 if memory serves.

- - - Updated - - -

Roland: For the pillar scenario is it possible to designate the pillars as a polygon vs a voxel? Because as you wrote a voxel is a complete 3X3 block with only the texture being visible to represent the pillar, but the remainder is still impenetrable. Or for pillars designate them as 1/4 1/3 1/2 voxel volume to free the air above and below. Referring back to one of the original builds the shutters and boards looked like polygon objects when they broke up.
The collider for a pillar follows the same shape as the pillar, although it's placed on a square it's actually a cylinder for all intents and purposes.

 
Depends on the length of the detour and what the values are for the type of door. For example, let's say a wooden door has a value of five, a detour of four blocks or less would be taken. If the door has a value of twenty-five then a detour of twenty-four or less would be taken. Things get a little more complicated when you factor in diagonals which are worth roughly about 1.41 if memory serves.
Hm, I actually thought the game does not use diagonals for the pathing (there are those edge cases then, having two blocks at the adjacent blocking, while the diagonal is free..)

 
The Gronk:

The collider for a pillar follows the same shape as the pillar, although it's placed on a square it's actually a cylinder for all intents and purposes.

Okay that part I understand, But what confuses me is the volumetric calculation of a solid visible object in a voxel. I see it in my head as a lot of tightly packed cubes like a Rubik's. and the pillar is the shaded area the collider is the damage designator, but the non damage are is invisble yet unpassable because the block volume mathematically is still greater than 0. sort of like a long island iced tea as far as layer representation.

 
Okay that part I understand, But what confuses me is the volumetric calculation of a solid visible object in a voxel. I see it in my head as a lot of tightly packed cubes like a Rubik's. and the pillar is the shaded area the collider is the damage designator, but the non damage are is invisble yet unpassable because the block volume mathematically is still greater than 0. sort of like a long island iced tea as far as layer representation.
Get your head away from the idea of sub-voxels, they would be far too resource intensive to implement. The collider is a cylinder in a very literal sense and has very little interaction with the underlying voxel engine. Once the pillar is placed the only real interaction between the pillar and the voxel engine is that once the pillar is destroyed the block it is placed inside can then have something else placed inside it.

 
The point of my original pathing post was not to complain about the digging/knock on the new AI; more to help me determine if I needed to spend the resources to upgrade my floors to the same material as my walls. As of now my walls remain stout and I paint less resource intensive floor blocks to match the walls, or make it "prettier" with the other paint options. Seems I will be fine continuing my normal play. I am always pleasantly surprised when an AI entity deviates from what I expect it to do. Definitely more immersive that way regardless of how "smart" I perceive the threat to be. To me unexpected=realistic. You can only plan for so much IRL.

As always thank you Fataal for taking time to answer questions.

 
Hm, I actually thought the game does not use diagonals for the pathing (there are those edge cases then, having two blocks at the adjacent blocking, while the diagonal is free..)
It does not use diagonals in the path grid, but I make diagonals from the resulting path and then I do some random smoothing.

 
Status
Not open for further replies.
Back
Top