57 réponses à ce sujet

[Bannor Bloodfist]

Nissa_Red wrote...

@Randomdays

Yes, I would. Actually, even if I'd appreciate it, placeables do not matter that much to me, as I rarely use placeables trees, walls or buildings, and alot of them remain on a flat ground (so their flat PWK is valid and sufficient), but tiles, I use alot. I happened to see whole tilesets with no meshes reflecting walls or tree trunks in their WOKmeshes.

While I can understand why tileset creators would do it to save time and efforts (and spend it elsewhere), I wonder if there is more to it. Perhaps areas with such WOKmeshes would be way too laggy in the context of a PW, for example, because in a LAN/SP context, it doesn't seem to affect us (my players and me) that much.

The lag generated by any wok can be reduced by reducing poly counts.  Most tileset folks generate the terrain/buildings first, then clone that for the wok.  This leads to an enormous number of wasted polys as well issues with unconnected verts or overlapping faces etc, all of which cause sanity check errors.

Woks don't have to match terrain exactly, and can even be used to reach areas that you might other think unreachable.  Someone at Bioware proved this with one of the tiles in Rural (I think) where you have buildings butting up against water, and one of the buildings actually overhangs a small boat parked underneath.  I am still in hospital, so I can't look this up for greater clarity.

Anyway, my main point is that you can reduce the comp[lexity of a wok considerably to help with lag/speed issues.  However, I also agree that a PROPER wok helps with immersion.  I should be able to duck behind some trees in the forest to then be able to set an ambush...flat pwok prevents this, same thing is true in a tile wok.  It is NOT necessary or even all that useful to utilize a flat pwok.

An object that blocks line of sight typically blocks movement as well, a few exceptions can be found in low walls, stone out croppings, cut down trees etc...

The downside to having a more accurate wok (this doubles with pwoks) is all the calculations that the engine makes for every step the pc/npc takes.  Having a more accurate pathnode helps a lot here too.  Tile-set wok files generally follow a "standard" that is not written down by Bioware though, and this is regulated by tilesize, height divisioins, etc.... typically since a tile is 10x10 meters, we divide that into eight sub-pieces (for lack of better wording) each being 125cm by 125cm.  Each of THOSE verts on the outside edge of the wok, should be created if not alreay there.  Aurora uses specific vert locations to tie various geometry together, if one of your outside edge verts is located at 130 and the tile next to it has the corresponding vert at 125, you WILL have a visible gap and may end up with npc's/pc's walking in air or under ground etc. 

For textures?  I like using 1024x1024 where possible for gound cover materials.  I typically UVW map them with the default box type, with size fields set at 1001x1001x3.  Doing it this way you are using a texture larger than the object, shrinking it down to 1001 which is ALSO larger than the object.  This helps hide edges the best way I know (smoothing drives me nuts, and most times does not help with edges like that)

Shoot, I am bouncing all over the place it seems.  I guess these meds are doing more to my mind than I realized. 

If I have raised more questions than answers please give me a direct question and maybe I will be able to keep my mind focused well enough to give you the answer you seek, if I know it.

@nissa_red :  Please re-post your mini-tutorial above in the Custom Content Tutorials Thread.  I am nearly positive that Shadow_m won't mind at all.


EDIT:  Back to poly count and outside edge vert location.  The out side edges of ANY tile should have corresponding verts in the same location as all other tiles in the set, but that is where the mirroring stops. You can reduce the vert count in any vertical object and typically most tree trunks or plants as well.  The WOK does not get seen, but it does affect movement and/or line of sight.  Reducing the polys necessary to get a general feel of the objects shape is all that is really needed, ie you do not need a 16sided polygone shape (and all its resulting polys) to represent the turnk of a tree.  You might want that for the visible section of the tree itself, the one holding the texture, but the wok section has no need for that level of detail. (Except on the outside edges, there, they must match any other tile that might connect to that edge)

Modifié par Bannor Bloodfist, 16 mars 2013 - 05:55 .

[Nissa_Red]

@OTR

Sorry for the delayed reply.

Yes, you got that right, and... Omigod, are you saying there is actually a solution to these ugly, fugly, terribad seams ? That would be so ... wow, after all this pain and suffering I had to go through!

Thank you for your demo module! I will have a look at it, probably not this evening, but tomorrow for sure. I'm sure it will prove to be enlightening and helpful.

PS : Thank you, I really mean it ^.^

Modifié par Nissa_Red, 16 mars 2013 - 06:30 .

[Nissa_Red]

@Bannor

About duplicating geometry and then "streamlining" it for WOKmeshes (and also the subdivision in 8 parts), that's what I am doing, and will continue to do, thanks to the previous teachings of the finer gentlemen among us, like Danmar, OMB and you. Your time did not go to waste (I hope)!

If I only ever had WOKmesh issues, my NWN-life would be made of players adoring their DM, ahem, I mean not half as frustrating as it sometimes can be, because if I am proficient with them (at least I feel at ease, hmm, I actually kind of like that part of the "job"), I cannot say the same about texturing. I always erase my work half-done, because I can never feel satisfed with it.

About adding tree/trunk meshes to existing WOKmeshes not causing particular issues if one is cautious about it (not antagonizing path nodes), that's good to know. Even a simple human sized box at their base would be a significant upgrade to what we currently have by default :



Since I sometimes tend to go overboard with my "ideas", and I could have missed something during my tests or game sessions, I wanted to seek confirmation before resuming this quite time-consuming (if easy) task. I will still keep an attentive eye to my scripts.

About the "tutorial", I am not sure my message really qualifies as tutorial. I mean, it was really meant as a simple reply and "proof of concept" to Tiberius and ShaDoOoW. Also, I probably ought to enhance the formatting before, to make it actually human-readable. The forums here really don't help.

[OldTimeRadio]

@Nissa_Red - What I did was simulate what I thought the issue was, reproduce it, then fix it.  I don't recall ever having come across this issue any other way, though.  Always from me yanking on verts too close to the edges of individual tiles.

[Bannor Bloodfist]

@Nissa_red, ok in your pic? The three red triangles, slicing across the pc's midsection? Those could all be raised or even any ONE of them could be raised and have the new sides turned into type 2, or type 7 faces. Type 2 would allow an archer to stand on opposite side and shoot across if they were raised say half a meter...not much more will work for that, while still preventing them from walking over that section of ground. Type 7 blocks ALL movement, but it ignores the blocking and shooting over etc. This is seen in many interiors where the walls were set to type 7 which blocks you from walking through but allows an archer or wizard to fire right through several of those walls and hit a target.

Anyway, there is a roll-out menu option in gmax and 3ds that will allow you to "break" along the edges of those objects, then you select them all again as a group and you can then "extrude" in positive or negative relative position. Now you have TWO objects for you wok. Select them both, go to vertex mode and select all verts, then "weld" using the default weld strength and you will have a fully attached wok, with no missing verts or gaps, with the blocking faces you raised.

Did that make sense? I don't have my computer here so I can't install 3ds and nwn etc to get clearer instructions/commands for you.

[Nissa_Red]

@Bannor
I don't want my rangers to fire through trees, so I went with option 2 :


(click to enlarge)

I extruded the base of the tree to about 2 meters, then scaled the top to about 95% so that vertices won't overlap, flipped some normals, removed the faces on the base of the tree and applied the correct material to the remaining faces.

Did I correctly replicate what you had in mind, yes ?

EDIT : I never really was certain whether smoothing groups also worked for WOKmeshes. Taking advantage of the thread to ask ^.^ They seem to, but do they really (and is it a good idea to start with) ?

Modifié par Nissa_Red, 16 mars 2013 - 09:26 .

[Zwerkules]

Nissa_Red wrote...
I extruded the base of the tree to about 2 meters, then scaled the top to about 95% so that vertices won't overlap, flipped some normals, removed the faces on the base of the tree and applied the correct material to the remaining faces.

Did I correctly replicate what you had in mind, yes ?

You found an important part of the process Bannor forgot to mention which is to make sure the vertices don't overlap and even if they don't overlap, it is a good idea to scale the top to be smaller than the base because there can be problems with woks if there are vertices that are at exactly the same x and y coordinates but different heights. PCs/NPCs can get stuck at the edge of that part of the walkmesh or even on the top of it.

I also think that it is a wrong assumption that making seamless ground textures is impossible. I don't have time to go into detail now, but maybe I'll be able to write more about it tomorrow.

If you also do what OTR described, you will be able to get seamless ground textures with correct lighting.

Edit: I always remove all the smoothing groups from walk meshes. It makes finding problems with them much easier and helps when texturing hilly terrain. If you look at the walk mesh without smoothing groups you'll notice faces that are so steep that they'l get smeared when the ground is textured as a plane far easier than if you look at the ground mesh which has smoothing groups.

Modifié par Zwerkules, 16 mars 2013 - 09:41 .

[Shadooow]

Nissa_Red wrote...

So without any further delay, here is how I proceed to add missing "use nodes" on placeables, without ever touching GMax/NWmax, even if it's of course strongly recommended to do so in a second time, be it only to check if the model is otherwise "valid" and that your "use nodes" are properly positioned.

Painful process, but managed it to work. Unfortunately, I wasnt able to export anything from GMax (either it does nothing when I press export all or it throws sanity check even on unmodified file) so I had to add use node manually, open model in gmax find good coords, change in pwk manually. Unfortunately I wasnt even able to open one (yet) placeable, throwed some kind of "get" error.

Now, another issue that bothers me: highlighting. Is it possible to control it somehow?

Link to image

Thi highlighting on the image is very annoying, when moving around with character player often target the placeable via Z axis even if turned by back so the city placeable is not visible at all - you know, the same what might happen with (scaled) dragons...

[Wall3T]

what an interesting discussion

im glad to see everyone giving there different ideas on this. id like to add an answer of my own to # 3: as a builder i agree with this! but at a different perspective.
+++++++++++++++++++++++++++++++++++

Rather than saying to look for tilesets that appear historical or ones that would "make sense", i use what is given to me as the advantage.

one of my more favorite styles of building is to consider how actual medieval european cultures lived, and create structures, or items, that would relate to what im doing.

ive played quite abit of strategy games that did well with trying to immerse you in these experiences (medieval total war, age of empires 2, stronghold) and ive based my building experience from these and a few others

as an example, i created a quarry for use in a city that was made mostly of stone(which would be almost every bioware city set ever . i thought of every aspect of its process from: the stone masons, to the cutters, to the warehouse they get transported to

overall, i find its much easier to immerse yourself (the builder) into this world setting and in return would allow players to experience a different style from the usual norm
++++++++++++++++++++++++++++++++++++

its definitley something that most pw's seem to want to miss out on for the exception of "space". i find that many owners do not wish to go the extra mile, thinking it would slow there server space.

its something that we all know about with the toolsets limitations (i mean after all it is an almost (12?) year old system, correct me on that but i know its old!) so most of them tend to ignore the nicer tilesets thinking that the extra texture style or nicer looking buildings will slow there servers down (which in some cases, is true)

im not trying to say that most of the nicer tilesets are a bad idea, in fact its a great idea!  

but based on my online experiences, i find that though yes some tilesets do have a slower fps than others. its easy by limiting the area size and type of that area during that time online. this way you arent forcing the player to be immersed in constant flora.

either way, its something that will not change. a good question to ask is, why change something if what you have just simply works?

i think what most builders think, is choosing the best option from the source. id like to see more work in this area myself (perhaps maybe even some historical building tilesets, workshops, etc) but its merely wishful thinking

Modifié par oOKyeOo, 16 mars 2013 - 10:35 .

[Pstemarie]

Removed...noticed Shadoow was already using NWMax - gotta learn to read more carefully 

Modifié par Pstemarie, 17 mars 2013 - 01:02 .

[Shadooow]

Pstemarie wrote...
Sorry Shadoow. You need to download the GMax version of NWMax to import and export .mdl files correctly. Once you got that installed (it has a self-installer) you should be all set.

I did that, cos with clean GMax there was no way to open nwn models at all., so I found nwmax and was able to import nwn models, but yet not export them. I would send you PM and ask for help privately, but you put me on block list so I cant.

Sorry for OT.

Modifié par ShaDoOoW, 17 mars 2013 - 01:01 .

[Pstemarie]

A "get" error? That's new to me. Drop MDA a note, he'll know a lot quicker than me what's up and get you sorted out.

[Rolo Kipp]

<getting a bit...>

Can't remember where I saw it but did some one ask about a Revolution based serious historical game?

<...absent-minded in his old age>

[Master Jax]

Nissa, so sorry about hijacking your thread to ask some things interesting me, but you see, I can't send private messages to other community members right now. Some browser issue, I'm sure. But I would like to thank both Bannor and Randomdays for taking the time to share a bit of their expertise with me. I would also hope they could contact me themselves via PM, given my current predicament, so I can ask them some other related questions without messing up your thread any more, or opening a whole new one. Thanks in advance, and yet again, sorry for the inconvenience. -___-

[Nissa_Red]

@Zwerkules

Thank you, Zwerkules.

I like to understand what I'm doing, I mean I like when something I'm doing makes sense to my little head. When I see someone else doing it, I sometimes try to do the opposite, see why that wouldn't work and set it back to the original if nothing conclusive comes out of it, but in the case of smoothing groups for WOKmeshes, which I first saw it on some DOA tiles I think, I couldn't find anything obvious or worthwhile to ponder. That is always rather frustrating.

Anyway, I think I'll follow your advice from now on.

Please be well.

[Nissa_Red]

@oOKyeOo

Every perspective is welcome here, as far as I am concerned ^.^

While it's true that I sometimes have strong opinions about some topics, I equally value any other take on it, provided it comes as motivated.

In the case of historical veri-similitude (or however one may call it) in NWN, I am a strong proponent of it. I am fond of medieval settings, and history with a big H more generally in any setting : Oriental (Japan, Vietnam, China), Latino or South-american, North-american, the Middle East, etc. I like to travel in my head while playing! That doesn't mean I can't appreciate games which settings like Steampunk (Arcanum), modern/futuristic times (D20), but they ought to make sense, be consistent if you will.

So reading you is encouraging. Yes, more often than not, we ought to make the best with what we have, but that is an interesting challenge, a mission that we implicitely accept when wrapping our sleeves up, donning the metamophorical builder garb +1 and click on that toolset icon. As you said, there are several ways to go about that though, and paying attention to stuff like you mentioned (how do people get to eat in the town we're entering during our adventures, where does the water they need to drink and wash themselves come from, what kind of resources they use or lack to subsist and prosper, do they have places to host visitors, like merchants or pilgrims, do they have places where they can worship their divinities) are at the very core of the discussion I intended to bring up here. Thank you for participating in it, and sharing your point of view.

About using more recent custom content, I also feel strongly in favor of it. While I never got to manage a PW with 60+ more players, and Mielikki preserve me from that (no offense intended), I have had my share of online experiences as player and builder on some social servers, and as I said I also have the chance to share gaming time with some very good friends in LAN (or however it is called). I've never had to regret using "newer" resources like tilesets or creatures (since they're probably which generate the most "lag"), provided I was *very* selective about it. Comparatively, the BW resources are not inherently bad (some of them are actually technically remarkably good even by modern standards), but they sorely lack that little "extra" (I call it a soul) which makes me *want* to use them. Slapping down stuff with little to no attention to people that enjoy the historical, fantastical or artistical candy that is so ever-present in our collective PnP memories (at least I think so), with a strong bias towards purely technical considerations, and be done with it, is not part of my nature. At the end of the day, you can only do so much with near endless artistical freedom that role-playing allows, but limited initial resources.

So to answer your question, why change ? Because it's worth it of course, even if comes at a price sometimes (slightly more taxing on my computer, or spending inordinate amounts of time filtering/sorting/catalogueing/customizing). Bugs are unforgivable, but the rest depends solely on our determination, as many cc artists and builders here keep demonstrating almost daily ^.^

So, hmm, yeah, there is a lot I'd like to tell you more about this topic, but that would probably extend over to game design itself, art in computer games, which I enjoy discussing don't get me wrong (!), but I don't want to bore you to death just yet (or you'll regret ever replying to this thread, and hold me grudge >.<).

[Nissa_Red]

@Rolo Kipp

Let it be known that one should never say Impossible in front of Master Rolo.

I had spent a huge amount of time on the Vault I thought, even on other NWN or game sites scouring and searching for that hakpack, without ever meeting the mystical fairy of success during that quest...

Impossible is sometimes... just not impossible enough for some of us.

PS : Thank you for digging it up for me ^.^

[Nissa_Red]

@Master Jax

Please, by all means. This is not *my* thread. Even if it was so, I have such a bad reputation already in derailing the threads of other community members here that I really would have a heart of purest obsidian to object.

I do hope you've found the answers you were looking for ^.^

[Nissa_Red]

@ShaDoOoW

About the nwMax export thingie, yes, that is what I tried to warn people about :

that also it would allow us to import faulty models but would not let us export us faulty models (even if we're not really interested in fixing those, like unwelded vertices or missing anim keys, see the trebuchet model I "fixed" above), which could frustrate/confuse novice users, but I discarded it. Honestly, it's not what matters to me in the end.

nwMax not allowing to export faulty models is both a curse and blessing. A curse, because sometimes we just don't want (or don't have the time) to "fix" whatever could be wrong in the model (or don't know how to go about it).

In your case, I'd say it still was positive, since you followed PsteMarie's advice and made your first steps with that tool, and managed to meet your goal, which is good news, and probably better than what I achieved when I started with it ^.^

Please be assured that your time and efforts working/learning nwMax are NOT wasted, however you might feel about it right now.

About the highlighting issue, it very much looks like an issue in the texturing and alpha punch through settings :

OMNIBUS
"alpha punch through"

I am far from being an expert in texturing however, so I'll leave it to the experts here to confirm or deny, hopefully getting you back on the right track.

If you still need help with the "use nodes", or something else I am hopefully more savvy at, please don't hesitate.

[Wall3T]

@ Nissa_Red

thanks for the inciteful reply !

the answer you took was precisely why i had asked the question. i wished to see what everyone considers on the topic, and i agree that changing is always worth the hardships that come with it

there was perhaps so many avenues for me to take on that subject, but to keep it brief i think (now more than ever) ive noticed more builders are seeking newer content than before, which is to say exciting!

which had brought up another question i wanted to ask everyone. if most builders still choose to consider in what ill say as the generic (ie: standard texture quality, standard ui, etc) over newer content (ie: overrides, new textures, archiecture, etc), would it be better to create *newer* content more associated with the generic?

i know it may still sound like im stuck on the whole "stay with the old" idea, but i wanted to have it considered as an option to still consider

as cc creators, i realize that we/they are among the few groups who can still help shape the direction in which the community develops (so you can see why im quite intrested in this topic

i wanted to share my question with an example of how im looking at the topic by using Borden Ha'elven Iron foundary set.

nwvault.ign.com/View.php

should it be a good idea to create more content like this? or perhaps continue developing and refurbishing  more elaborate and greater quality content?

Modifié par oOKyeOo, 17 mars 2013 - 11:59 .

[Wall3T]

@Rolo Kipp

has anyone ever managed to get that module to work? i was never successful and wanted to know if anyone else had been

Modifié par oOKyeOo, 17 mars 2013 - 11:58 .

[s e n]

About the lighting issues that happen between tiles: as Baba and others said one can be fixed (it depends of smoothing groups) by having the edge verts AND the verts next to that (in the adiacent column/row) of a given terrain tile share both z and x or z and y coordinates (it depends on orientation)

there is another issue that apparently cant be solved because its dependant of the tileset area orientation and lighting settings and doesnt depend of tile specific lights. this effect is a lot more evident using white terrain texure as dummy, and it increments with hieght steps (so on plain flat tiles there is no such effects, but when you start use tileset height transitions of 5 or more meters, it can really become disappointing when you use light textures)

[OldTimeRadio]

s e n wrote...
there is another issue that apparently cant be solved because its dependant of the tileset area orientation and lighting settings and doesnt depend of tile specific lights. this effect is a lot more evident using white terrain texure as dummy, and it increments with hieght steps (so on plain flat tiles there is no such effects, but when you start use tileset height transitions of 5 or more meters, it can really become disappointing when you use light textures)

Does it happen with Bioware default tiles?  Can you provide a screenshot of whatever that looks like and a test area with tiles arranged to show that problem?

I also assume you're not talking about something to do with shadows but the shading (vertex lighting), correct?

[s e n]

OTR: you're correct, I decided not to deal with shadows, so its shading issue. about ss, i havent any, but if you want you can see it clearly overriding grass texture with a white one on my terria wip hak (i should still have a dload link available, send you via pm)

[OldTimeRadio]

That's a really pretty concept for a tileset, SEN, and thank you for sending me material to try to reproduce the issue.  I whipped up what I think is the situation you're describing.  Are we talking about the things I've marked as A, B, both, or something else?



FWIW, I noticed a lot of what I think technically qualify as "sparklies" on that before I replaced the grass texture.  I'm not sure how much that would have to do with things yet, though.