new features; sets; ramblings

[enetheru]

Hi all,

I've not been completely idle, slowly hacking away at charlie..

I've been wanting to add more variety of features so I grabbed some art from http://opengameart.org/
specifically some of the flora from this guy
http://opengameart.org/users/yughues

you can check out the artwork on youtube
http://www.youtube.com/user/Nobiax

anyway... I was thinking that since he releases the works under public domain, etc it would be a good inclusion into the springfeature archive.. or its own archive etc.

what's the consensus on metal and energy levels for plants, etc?
I was thinking that because they are defined in lua nowadays it could be calculated based on the dimentions of the colission, or occlusion volumes?

I would like to make whatever features I can as compatible as possible with also tools like feature placer and toolbox, so does that mean variables for categories?
should there be discussion on how to organise this?

I'm not really that fussed either way, I just want to be helpful, i can release in a zip the completed features, or working files, or you can download the source material from open game art etc..

[gajop]

I'd really appreciate some work on spring features categories. Can categories only be one string, can also have some hierarchy?
Currently we have this mess (first number represents the amount of features with a certain tag):

Code: Select all

     75 Vegitation
     24 vegitation
    206 Vegetation
      1 urban
     22 Trees
     20 SnowyPipes
      4 SnowyIndy
     32 Rocks
     28 rocks
      1 Resource
      2 millitary
      1 military
      2 Industrial
     51 industrial
      1 corpse
     24 Buildings
      8 arm_corpses

EDIT:
Having all "good" features in one mod which needs to be download for any feature to be used isn't very good.

I don't think that what we have right now with spring features is a permanent solution, but it'll do for now and the current features can always be imported in a better system.

So my suggestion is to feel free to add it, as long as the size doesn't become too large (then we'll need to think of another solution).

[enetheru]

I don't think hierarchy is a good idea, multiple tags are better. but so long as they are managed..

that list is just weird.. snowypipes?, lol

I agree that one single archive isn't so robust, glad that the lobbies have auto dependency fetching.

some sort of scheme could be developed, rules for creation of feature set archives, naming schemes, lua support files, categorisation, licence inclusion etc etc..

at the moment I prefix all my files with the authors name, so in this case its all "nobiax_featurename_suffix.ext" its useful. all the features and model pieces are named the same way.

they all are in obj format, with dds textures too. and all my working files are tga..

i also have a max texture rez of 256x256, so they stay small.. its just features after all.. and plants aren't huge.

I'm still new to creating features though and there is sure to be overlap with units and buidings I don't know about yet.

[knorke]

for engine, category tag does not exist for features.
(it is for units, which unit can shot at what, so you can have dedicated anti air or depthcharges that only target submarines etc)
idk what it does for feature place?

what's the consensus on metal and energy levels for plants, etc?

there is no consensus. imo small vegetation is best when it is simply unreclaimable.

I think preparing model and texture for spring (converting to usefull format, resize texture etc) already helps a lot, the featuredef might get altered anyway. Maybe set a collision volume but even there seems to be big differences what people consider good.

Beside rocks & plants more features like this would be cool:
http://opengameart.org/content/broken-rusty-volvo-car

[smoth]

I will use whatever name I please BWA HA HA.

[FLOZi]

Give games total power over features and their placement, ban maps from including features at all.

...


...



...



...

http://upload.wikimedia.org/wikipedia/c ... smiley.png

Or you know, just give game files precedence over map files like it should have been from the start.

[smoth]

Or you know, just give game files precedence over map files like it should have been from the start.

HUGH BALLS

[enetheru]

for engine, category tag does not exist for features.

nor would I expect it. i was more thinking of a way to organise the hundreds if not thousands of features, so that tools like featureplacer and toolbox can have a usable interface.

[smoth]

Really? so the category tag cannot be read as part of a feature def?

[FLOZi]

Really? so the category tag cannot be read as part of a feature def?

Why would it?

[smoth]

why not?

can featuredefs have a custom table?

[FLOZi]

why not?

It has no purpose in the engine, so why would the engine try to load it?

can featuredefs have a custom table?

Absolutely.

[smoth]

I don't think hierarchy is a good idea, multiple tags are better. but so long as they are managed..

Don't make a massive feature file dir. It is important to sort them by author and/or pack. It is important for maintenance

that list is just weird.. snowypipes?, lol.

Features were originally intended to belong to specific maps. It was up to the mapper as to whether or not the feature name would be resolved and how.

some sort of scheme could be developed, rules for creation of feature set archives, naming schemes, lua support files, categorisation, licence inclusion etc etc..

at the moment I prefix all my files with the authors name, so in this case its all "nobiax_featurename_suffix.ext" its useful. all the features and model pieces are named the same way.

what is suffix for?

Feature defs could include a custom params table which could contain the folllowing:

author name
feature set

these could be concatenated to the internal name of the features and thus prevent collision.

they all are in obj format, with dds textures too. and all my working files are tga..

i also have a max texture rez of 256x256, so they stay small.. its just features after all.. and plants aren't huge.

I'm still new to creating features though and there is sure to be overlap with units and buidings I don't know about yet.

Seems unrelated. Obj can be imported to upspring. Convert the TGA files to DDS. The sizes are no big deal really, it is irrelephant. Good luck with that stuff.

[enetheru]

I don't think hierarchy is a good idea, multiple tags are better. but so long as they are managed..

Don't make a massive feature file dir. It is important to sort them by author and/or pack. It is important for maintenance

yeah totally, I think the archive should be based on theme, with the layout inside like the current spring features one.

so you could have a winter pack, a summer pack, an alien pack, tropical pack, etc. and inside would be relevant objects from various authors.

what I was referring to was categorisation for when you want some interface to display all the features for map editing, I expect something like genre tags for music, eg. plants, trees,, grass, rocks, building, wreck, temperate, tropical, arid, alien etc. and being able to filter the list based on the categories, with boolean search results.

some sort of scheme could be developed, rules for creation of feature set archives, naming schemes, lua support files, categorisation, licence inclusion etc etc..

at the moment I prefix all my files with the authors name, so in this case its all "nobiax_featurename_suffix.ext" its useful. all the features and model pieces are named the same way.

what is suffix for?

diffuse, specular, normal, tex1, tex2, etc...

Feature defs could include a custom params table which could contain the folllowing:

author name
feature set

these could be concatenated to the internal name of the features and thus prevent collision.

featureset + author + featurename?

[smoth]

yeah totally, I think the archive should be based on theme, with the layout inside like the current spring features one.

so you could have a winter pack, a summer pack, an alien pack, tropical pack, etc. and inside would be relevant objects from various authors.

Entirely against such a short-sighted initiative.

what I was referring to was categorization for when you want some interface to display all the features for map editing, I expect something like genre tags for music, eg. plants, trees,, grass, rocks, building, wreck, temperate, tropical, arid, alien etc. and being able to filter the list based on the categories, with boolean search results.

CUSTOM PARAMS.
customparams = { category="vegitation" }

" boolean search results." you are kidding right?

what is suffix for?

diffuse, specular, normal, tex1, tex2, etc...

except that model textures when it comes to s3os are bound to the model. That doesn't include shaders and whatever kind of crazyness the shader introduces. or whatever the hell the ASSIMP stuff will introduce.

featureset + author + featurename?

I feel as though I lost you.

[enetheru]

you must have completely lost me.

--

so you want a single archive? with directories within sorted by pack and author.. not separate archives?

I agree with directories within an archive separating authors and sets
But I want it in multiple archives. so if there are going to be multiple archives how to make the split, hence the previous description.

--

keep in mind i was only thingking category to be used for map editors, not game data.

category being a list of tags, because some features contain multiple things.
say for instance rocks and grass together.. so you categorise them as rocks and as grass, then you can search for rocks minus grass,

or also add a climate tag, so vegitation and temperate climate,
or all vegitation minus trees.

having multiple tags for the category allows you for greater filtering of the features, only showing what you want rather than most things.

is that better described?

--

At least you are replying...cheers :)

[smoth]

Multiple archives make it harder on players. Either we do several archives:

IE: SmothHarbor.sd7, smothBannanaTREES.sd7, Smothgundamfeatures.sd7

or one mega archive. Why a mega archive, because it can be maintained. It is hard enough to get people to stick to a single standard. Whoever maintains the archive can keep the standard in working order. Otherwise it becomes chaos.


I understand why you are doing it. The spring map features archive was something I was working on. I had intended on going much farther but instead a community member put out the archive before I had finished sorting everything. There are MANY issues with the current one and no it isn't how I wanted it to be done vs what was done. The cat's out the bag and now I am just like hey you guys go do what you want.


big rant below
Featureplacer was going to be a tool for all mappers. The design intent was to have it handle everything for you, build the mapcofigs, export the feature files with it(should you desire to NOT use a feature archive).

I had intended on a UI with tabs for different categories with check boxes to hide or show certain elements, everything from scale(which is fing relevant) to author and theme. After I got it all sorted in the ui, the plan was an independent release.

I had a thread where I was asking for feedback on things and received very little and was maintaining feature placer actively. Being that no one used it, after I shut gundamrts down. I put it on a shelf, on hold while I worked on my new project. I have been improving the code behind it in between this crazy unit and shader stuff I have been working on.

Keep in mind that I was the only one working on the feature placer crap for YEARS. We had worldbuilder which I really won't discuss beyond it massive maintenance issues. Gajop is doing an editor but isn't maintaining an archive and forb decided on his own to do the feature archive before talking to me about why I was holding back on it.

Otherwise yeah there might have been more of a standard but boo fing hoo and now I would rather spend my time doing a project for me rather than community service. I get you have some grand design for how features should work blah blah but you are not going to implement them in such a tool. You want to define standards but don't consider the people who actually work on the tools might have a logical plan. Just like everything around here, everyone wants to lead but the people leading already have pretty good ideas. It isn't put in the form of feedback, it just, hey go do this now! Of course this is partly me venting about the feature archive. Which I already bitched about to forb so I am beating a dead horse but hey, you know I think I deserve at least to be able to vent! And partly bitching about how everyone wants to tell me how shit should be while not working on it themselves.

I do think the discussion is good it would have been great to have you around 2 years ago, and that IS NOT your fault. I am SO TIRED of backseat developers around here who never want to do this or do that. However, when you ask for feedback but happily swoop in afterwards. I did my map options code, blueprint comes out invalidating months of my development and planing, was anything said while I was doing it? NOPE, was any consideration done to the fact that I was improving my map options stuff? NOPE, I had a newer better version coming that I had been revising for months. Instead bam sucks for you smoth, have fun being replaced. I discuss featureplacer to a specific community member, worldbuilder happens then the author of worldbuilder rages at me for creating featureplacer when he didn't meet my needs.

I sit down and archive every spring feature and am working on tools to place them, display them by category and do clever stuff with them, randomizers, spread placement, random rotations etc.. and someone puts out a massive archive insisting I now use HIS svn to maintain the core archive to MY project.

I am just a little pissed about the whole deal with mapping and trying to get stuff better. I fought for years about it and probably am just too embattled to care. It is a bunch of drama but I wanted to get that off my chest(hope you don't mind).

I hope that explains why I am cross

[enetheru]

haha, wow that's awesome, thanks for the insight.. I empathise with your frustrations.

I'm glad that these problems have already been visited.

I don't really want to lead, I'm more of a facilitator of discussion and a helper, since there is no way I can code up an tool or help with current tool efforts, without spending time I simply don't have. I just want to make assets for people to use, and when I see inadequacy I pipe up and say something.

I want someone to lead me, tell me what extra stuff I should include in my feature defs, provide some docs so that i can follow them.. but that hasn't happened.

I'm happy to add them to springfeatures when I've got the set complete, but there are a lot of extra tags in there that i don't know how they relate to anything.

As far as the mega archive goes, I don't like it. I want to be able to work independenty, create a set of features and release an feature set archive to spec that map editors can use in whatever tool supports it.

[smoth]

Nothing saying you cannot release entherupack.sdz/sd7.

I am not the boss of you.

a few pointers though:

at least use entherupack1,entherupack2, etc so people will know what version they are using. Do contact me in lobby, I am always willing to chat if I am there. Just forgive me if there are some times where I say I am actually doing something and cannot chat.

[Forboding Angel]

Why not just send them to me and I'll add them to spring features?

Spring features is on rapid, which makes the most sense as you can easily have multiple versions without big filesize increases. Moreover new versions is only be a moderately sized download.