66 réponses à ce sujet

[Pstemarie]

What they are trying to say is that if a placeable has a pwk that is 3-dimensional you can drop the placeable below z 0 in the toolset and still have it block.

[NWN_baba yaga]

that is new and cool to know, thanks:)

[FunkySwerve]

_six wrote...

FunkySwerve wrote...

then go on about Bioware's style beter then the REAL style??

Actually, six called it the REAL style. I scarequoted it to show my skepticism about the label, which you obviously share. Again, pays to read the thread.

I guess the fact I've done work in other engines that actually have 3 dimensional movement collision plenty before somewhat skews my view towards 'practise'. I tend to apply rules that make the most sense from a realistic physical perspective rather than what happens to be convenient. I'll happily give you a list of things I've done that I
think are wrong in that sense - I'm not saying it's universally bad to do such things. Just unrealistic and liable to cause people who don't have a thorough understanding of how your models work to run into
issues.

I guess it does. Building for other engines is not a particularly good idea in my book. If this were a FPS, where objects needed to have three dimensional substance to block bullets, I could understand it.

From the perspective of someone making custom content, I'd rather my models worked as they would appear to from first glance. See pstemarie's post for a perfect example of where that's completely necessary as
opposed to an alternative.

That's an interesting point. Of course, since the community norm (according to Bannor, though I agree) is flat pwks, I think how most people expect models to work is NOT how you are describing. I expect any place I use to be walkable when I sink or raise it past the margin of error (is it 0.1? I forget). In fact, I rely on it, and anything different would be somewhat of a nuisance, for the reasons I've already illustrated with the bridge. Except, of course, when large objects require filler places, which simple three dimensionality can avoid, and which is the nuisance on the other end of the spectrum.

As an aside, please don't invent things that I've said and then accuse others of not reading the thread properly. My post is clearly visible above if you'd like to double check it. It's understandable to confuse my use of 'realistic' with 'real' inferring proper, but my post didn't even use that word.

Actually, I didn't invent anything - I just accidentally misattributed Bannor's label to you - apologies, it was late when I posted. I wasn't trying to put words in your mouth. Actually, the original use has been repeated in quotes in this thread, to describe a position similar to yours. Your position seems  less extreme than his, calling for a compromise, but still three-dimensionality. For my part, I tend to agree with that, with the caveat that 3d pwks should only be on objects that need them for normal use - ie, walls you could otherwise walk through sections of. Things that are relatively contained, and therefore likely to be within the margin of error  of the tileset section they're sitting on, just don't need them, and shouldn't have them (that is, they should have 2d ones instead). That category encompasses the vast majority of the places I lifted from bioware tiles. Even on the larger stuff I ripped, I I generally just went with flat ones, adding more than two dimensions in only a half dozen cases or so. In practice, I haven't run into much trouble using them that way - just one large boulder thus far which didn't block as it should (though that was admittedly quite a nuisance).

Funky

Modifié par FunkySwerve, 23 novembre 2011 - 07:37 .

[Bannor Bloodfist]

The question was whether or not a fully shaped pwk (matching object shape) was worth the effort.

Some folks think that a flat pwk is all that is needed... and that does work for blocking MOVEMENT of the pc/npc, but has issues on uneven terrain and does not block arrows or spells from overflying the "blocking" pwk.

The issue is whether or not a fully realized pwk is needed, useful, or even practical.

The answer is: Depends on the type/usage of the placeable. Some, like walls, should have at least a simple raised pwk that does block ammo, where an object designed to be placed on top of a table, or in an otherwise non-accessible area does NOT need a raised pwk, since the engine won't allow you to get to it anyway.

So, it boils down to the intended use of the individual placeable.

I think I have covered pretty much any type of object you might require in a game, those that are not used/modified/moved by a player, and those that should interfere with the players movement and those that should block incoming/outgoing spells or ammo.

Each placeable has a specific purpose. Blanketly creating a flat pwk is a waste of time if you have to go back and add an additional HIDDEN placeable to block something when all that was really required was for the original object to have a properly created pwk in the first place.

The engine does NOT flatten pwks as has been mentioned in this thread. The creator of the object had to do that. Creating a pwk that perfectly matches a complicated object is also a huge waste of resources/processing power, but something that comes close to the overall shape of the object really provides the best of both worlds. It allows you to block movement and ammunition, while not overloading the engine if the pwk is created with simple shapes.

Having to go back and create a 2nd hidden placeable to block something that should have been blocked in the first place doesn't make sense either as it increases item counts and processing time.

Moveable objects also affect things, and definitely prove that the "baked down into the wok" idea is flat out wrong. If you can block access through a doorway with a box or nailed up board whatever, and can destroy that object to then be able to get through the door, then the "baked" bit is flat proven wrong. I know this can be done as we did it in WyvernCrown. The stones blocking the hidden walkway get destroyed and the placeable walkway moves up into place where the real walkable surface already existed. The area is not reloaded, it is just animation and destruction of objects.

[Pstemarie]

3-D pwks are a natural evolution resulting from the way tileset construction has evolved over the years. When NWN first came out all the tilesets were flat, therefore a 2-D pwk for placables was enough. However, with the addition of grades (e.g. sloping hills) into the geometry of tilesets, the flat pwk no longer was enough.

I would say I have to disagree with Funky about expecting a placeable to become nonwalkable if I drop its Z-axis below the tile plane. No matter how far below the tile plane I drop a placeable, I'd expect something like a wall to still serve as a blocker. If I wanted something walkable, I'd just copy the mdl, kill the pwk, and add it in a hak.

Modifié par Pstemarie, 23 novembre 2011 - 07:49 .

[Bannor Bloodfist]

FunkySwerve wrote...
<snipped for brevity>
That's an interesting point. Of course, since the community norm (according to Bannor, though I agree) is flat pwks, I think how most people expect models to work is NOT how you are describing. I expect any place I use to be walkable when I sink or raise it past the margin of error (is it 0.1? I forget). In fact, I rely on it, and anything different would be somewhat of a nuisance, for the reasons I've already illustrated with the bridge. Except, of course, when large objects require filler places, which simple three dimensionality can avoid, and which is the nuisance on the other end of the spectrum.
Funky

Bannor Bloodfist wrote...

Yes, exactly. However, a flat pwk is a waste as well.

Best
bet is to not clone the geometry of the object, but to create a shape,
as simple as possible, that reflects the general shape of the object. 2
faces per side type. Cone/angled etc, with as few faces as possible.
In fact, a pyramid reflects the general affect of a cone, while only
requiring 4 faces. It is perfect? No, but it approximates the shape of
the cone it replaces.

Really complex shapes are basically useless
as the object affects walkable location finding from the base or widest
parts of the object. A simple box for a very complex object is better
than a flat plane though for targeting purposes.

What is all
boils down to, is how realistic you want your objects to affect
targeting, and path-finding. The engine can, and does, calculate for
each face, but are all of those faces really necessary?

This is
where low-poly generation is important, and 99% of the time, it is
assumed by any serious cc creator for nwn. The lower the poly count,
the easier the engine handles it. However, this doesn't mean that low
poly is always the answer either. Some things, creatures for example,
(pcs/npcs/creatures) should be higher poly. Even then though, high poly
does NOT mean it is actually better looking. It is a blend of poly
count and proper texture application that determines how good any
object/creature is in game.

My complaints about how the
Placeables are generally created is that no thought has been given to
what the object actually is, most of the placeables available are using
the simplest, 2 faced, flat plane, to block walking. No thought given
to targeting etc. No thought has been given to the objects actual
geometry etc.

Now where does that differ from your statements?  It points out exactly the charge that YOU are not reading the entire thread before posting and making accusations AND assumptions on a subjest that you just stated above you actually agree with?

Get your act together please.

I have stated that a realistic PWK for a blocking placeable is better than a flat one.  The caveat being the destined use for said object.  

My complaint is the lazybutt way of creating flat pwks simply because it requires no skill or time so it makes it easier. You have stated that you have created thousands of placeables that only have a flat pwk, and those are the ones that need to be adjusted, just as you mentioned in that post.  So, where am I wrong?

[s e n]

its possible to get 3dimensional pwks that block sight as if they werewoks with faces set to obcuring?

[s e n]

maybe giving the placeable a wok instead of a pwk does the trick?

[Bannor Bloodfist]

s e n wrote...

its possible to get 3dimensional pwks that block sight as if they werewoks with faces set to obcuring?

No.  The engine reads the visibilty node from the underlying tile's wok object.  That does NOT mean that you can't hide a creature from behind a wall, it just means that depending on the view of the camera, the creature will be seen anyway.  IE top down will show whatever is behind a placeable wall.

s e n wrote...

maybe giving the placeable a wok instead of a pwk does the trick?

No, Placeables do not have woks, only pwks.  You can't create a wok for a placeable object.  This is an either or type of choice, either the object is a tile, thus a wok, or a placeable, thus a pwk.

[s e n]

but what about dis?
http://nwvault.ign.c....Detail&id=6639
thats something always misleaded me

[Lord Sullivan]

s e n wrote...

but what about dis?
http://nwvault.ign.c....Detail&id=6639
thats something always misleaded me

Such attempts at fooling the engine into thinking your placeable is a tile can lead to issues. It's better
and safe to make an object part of a tile and create the proper walkmesh for it's intended use.

@All who are learning

To be clear on the "3D PWKs" and the drop below "0" Z Axis. The placeble will continue to block as long
as the max height of the PWK remains on top of the Z Axis. If the whole of the 3D PWK ends up below the
the Z Axis, it will not block movement anymore just like a 2D Flat PWK would not. I just thought I'd throw this
in for it to be clear.

[OldTimeRadio]

s e n wrote...

thats something always misleaded me

If this uses the method I'm thinking of, Danmar wrote a decent breakdown of the limitations about 7 years ago in a thread titled "WOK vs PWK", which you can find in the Omnibus.  My experiments indicated he was wrong about not being able to make a hole but I haven't played around with it all that much.

[Bannor Bloodfist]

That placeable and useable bed is using a 3d pwk, which is exactly what I am stating that a 3d representation on a pwk is desirable. Yes, you can use a pwk to block movement and targeting. A wall can prevent a creature from shooting arrows at the pc, until the creature walks around the side of the wall anyway.

A useable bed is adding a useable/walkable surface that is on TOP of a walkable surface on the flat tile itself.

What you can NOT do is add walkable on top of non-walkable. IE, if the flat tile beneath that bed was made non walkable, the bed would not be usable.

So; flat tile, completely walkable, you can add a placeable that blocks the ability to walk, OR to change the height (going upwards only) that the creature walks on. IE you CAN make walkable steps that will raise the pc up when they walk across it.

So, 3d PWKs make sense depending on the object in question. This has been my argument throughout this entire thread. Flat pwks placed at the bottom of that bed placeable (instead of a raised pwk) would allow the pc/npc to laydown on the floor under the bed, where if you had a correct 3d pwk, it allows you to walk up to, onto the top of the placeable and thus to lay down on top of it.

[NWN_baba yaga]

you can create a "hole" OTR. If you want an example im happy to give you an example

[Rolo Kipp]

<raises the...>

Bannor Bloodfist wrote...
What you can NOT do is add walkable on top of non-walkable. IE, if the flat tile beneath that bed was made non walkable, the bed would not be usable.

So; flat tile, completely walkable, you can add a placeable that blocks the ability to walk, OR to change the height (going upwards only) that the creature walks on. IE you CAN make walkable steps that will raise the pc up when they walk across it.

So, possibly, you could create a bridge tile that had *geometry* for a walk above it, but walkmesh at ground level going *through* under the bridge and put in place (with triggers and scripting) an invisible block with mesh that can be raised to bridge level for walking across the bridge and lowered to ground level for walking under the bridge?

I like.

<...ante>

[OldTimeRadio]

NWN_baba yaga wrote...

you can create a "hole" OTR. If you want an example im happy to give you an example

Thank you for the offer! I've got a few examples on this end, though. 

[Bannor Bloodfist]

Interesting, I was under the distinct impression that you could not assign a wok to a placeable.

So, you are stating that your placeable's wok overwrites the the original tile's wok?

How are you attaching this wok to the placeable? What are the limits? Does it require that the placeable fit entirely inside a single tile on the map?

[Lord Sullivan]

OldTimeRadio wrote...

NWN_baba yaga wrote...

you can create a "hole" OTR. If you want an example im happy to give you an example

Thank you for the offer! I've got a few examples on this end, though. 

This is not a good idea as it is as easy to create a tile that does just that without the quirks an improper
placeable presents and from what I've read over the years the test I did a while back, it's actualy pointless
to create improper placeable+Wok as you have to place/align them manually in the position of a tile.

defeats the purpose.

[Rolo Kipp]

<filling a bucket with water...>

Lord Sullivan wrote...
This is not a good idea as it is as easy to create a tile that does just that without the quirks an improper
placeable presents and from what I've read over the years the test I did a while back, it's actualy pointless
to create improper placeable+Wok as you have to place/align them manually in the position of a tile.

defeats the purpose.

I am by no means trying to fan the flames... But, if this can be made to work reliably without a major performance hit, there is one thing that would make this worthwhile... being able to create, destroy and move walkmesh by manipulating "exceptions" placeables.  

This is something that can not be done in the game with tiles.  If we could do this with placeables, this experiment will be worth it.

My opinion.

<...just in case>

[Pstemarie]

Bannor Bloodfist wrote...

Interesting, I was under the distinct impression that you could not assign a wok to a placeable.

So, you are stating that your placeable's wok overwrites the the original tile's wok?

How are you attaching this wok to the placeable? What are the limits? Does it require that the placeable fit entirely inside a single tile on the map?

I believe that you change the model type to TILE and the extension of the PWK to WOK. However, if I recall correctly, the placeable's WOK would have to be the same size as a tile's WOK.

Furthermore, if you set the model classification to TILE while editing it in GMax or 3ds, when you export with NWMax the exporter will create a wok file for you. Albeit, an empty one - but you could always go in and add the walkmesh geometry later. However, if you've gone to that extent, why not just make it a tile in the first place. In the case of the quicksand vid, you could easily "destroy" the quicksand just by animating it to move up and down. Some time back I experimented with water tiles where the water was animated so that it could be raised or lowered.

Modifié par Pstemarie, 24 novembre 2011 - 12:12 .

[OldTimeRadio]

@Bannor -

Bannor Bloodfist wrote...

Interesting, I was under the distinct impression that you could not assign a wok to a placeable.

So, you are stating that your placeable's wok overwrites the the original tile's wok?

I don't have enough empirical evidence to state much beyond what the video shows.  It's a positive sign that the new WOK's information shows up with the renderaabb 1 console command but I strongly doubt it would universally replace all tile WOKs it is dropped onto.  I'm looking at using it for a no-tile implementation of the airship dock I spoke about some months ago, so mainly I'm exploring implementation on bare, ground-level surfaces.  Again, needs more testing.   

How are you attaching this wok to the placeable?

Modelbase set as Tile classification with visible geometry (if any) and WOK both parented to the model base directly.  Easy peasey, just like what you'd expect.  There may be other arrangements which work better.

What are the limits?

Good question.  I don't entirely know.  If you cruise through the Omnibus, you'll find a surprising number of threads on the topic of placeables with WOK, and which seem to indicate this or that, but I don't think many of the practitioners 'got' why it seems to work in the first place and so didn't really use it correctly- like placing it correctly at the center of the tile it is to be used in.  BTW, this appears to be the case because Aurora ditches (at least to some extent) the previous WOK, meaning any replacement can't be partial and must be a full 1000cm x 1000cm WOK.  You can see this visually using renderaabb and placing it improperly.  Again, in the context of a featureless, ground-level tile.

Does it require that the placeable fit entirely inside a single tile on the map?

Yes.  Pseudo tile groups appear to work, whether sliced or using something a bit gentler, like Velmar's TileLink.

@Lord Sullivan - I look forward to your empirical findings on this or any other topic. 

@Rolo Kipp - This only works in the toolset- and as a way to quickly add a walkable tile without having to modfiy the tileset, itself.  My use for this right now is (hopefully) as an easy way to include an airship docking tile without having to require modification of all tilesets in order to use it.  This has been a big problem such as with the addition of Some_UX's airships.  In general, tilesets are a little too 'locked in stone' for my tastes and I'd like a way for placeable creators to more easily affect them.  There are other uses and maybe some other tricks available, but that's it right now.

[NWN_baba yaga]

lord sullivan: the way i did specific things are ok for me. nobody thought you can bumb the camera down like i did with my spider caves tileset so just take my word on what i say

a "hole" is possible for a pwk, just not a closed circle ya know eh!!!

and btw. im sick of this nitpicking realy, we are the lat ones of doing stuff so we work together or we do not at all

Modifié par NWN_baba yaga, 24 novembre 2011 - 12:18 .

[Frith5]

+1 to Baba's sentiment. Could this be useful for simulating good pit traps then?

[Bannor Bloodfist]

OK< OTR, I understand now what you are creating. It is a hybrid of a tile that is unassigned to the tileset and used as a placeable. This would mean that yes, this "quasi-placeable" is limited in usage. If you don't have a completely flat tile to paint this placeable onto, you will end up with holes in the wok, this means that your character/npc can go UNDER a given tileset or popup into the air if the edges of the placed "quasi-placeable" do not exactly line up with the underlying tile's edges. Ver dangerous, but likely usable for the situation of an airship dock, provided that the given tile looks right and meets the min specs of being flat etc.

A tile as a placeable.. not real sure how useful that would be though. I am truly not understanding why it would be necessary... your so called "dock" will work on a flat tile anyway, provided you create the pwk correctly. You can add raised terrain via the pwk to allow stepping up onto the dock and walking across it etc, without having to have an entire tile sized placeable.

[Frith5]

So, Banner, would this technique be suitable for use in confined areas that would help with placement and with creatures/PCs going through a gap? For example, a pit trap placed in a hallway. The corridor walls would still stop movement, or would they not?

JFK