not without egoism, i would say, it let's mine look odd perhaps

i know and i remember that i've said a long time ago that i'm perhaps not the best choice, therefore i loved TomM's decision for the LUA modelling system from the very start. mainly because of the closeness to FE2 and i feel personally that it offers much more as what is possible with "industrial standards".

i remember TomM feared to find any modeler at all who get comfortable with it and he was more then happy to find at least one 😉

like i said before, i know it has limitations and it's true it limits the range of modelers contributing to pioneer.

but on the other hand, i would say let them contribute to whatever they like, there are masses of games under development, some of the comrades of FFED3D doing actually commercial models, they have evolved a lot (compared to me, i got stuck with LUA, but i LOVE it).

perhaps i'm only angry because i will vanish from the scene without pioneers LUA modeling system

besides of that, if that's a fact (...turn away an interested artist every few weeks...), why don't you pipeline them to me? i could help, i would like to help.

i mean if a beginner like me can handle it, what will close out a more experienced modeler? i'm pretty sure they are able at least what i'm able to.

i really would like a close coop with some more experienced modelers, i could learn a lot in this way.

of course i understand to, that some do professional work, freelancing or on the staff and have no interest to waist their time with LUA.

i remember coolhand told me once that he isn't able to do what i do with the LUA system, i don't believe that 😉 , but for above reason it's obvious.

I have nothing to support and much to refute that. If you know something that that is both a) useful and b) possible with LMR that is not with a traditional pipeline, please do share!

At this point there's a lot more people involved than tomm, and we're all trying to build Pioneer into a real game. Having something deliberately designed to make life difficult for people just isn't going to work in the long run.

Sure we could say that, but they come here wanting to contribute to Pioneer. This game is the fulfilment of a dream for so many people that for years have wanted to relive their memories of Frontier. If someone has the skill and the desire to add a little bit of themselves to our game, why should they be denied?

I have directed some to your tutorials in the past. Its hard though - surely you must see that if you're used to working in a highly visual environment, having to start thinking in code could seem daunting? It doesn't matter if it actually is or not, it looks complicated, and that's enough to turn a newcomer away.

You underestimate your own skill. You are the person most experienced with the use of Pioneer's model system by a very large margin. The difficulty is that the skills required to drive LMR effectively aren't quite the same as those that are needed to drive a modelling package.

I'm personally a big fan of moving away from having Lua be the up-front system.

Ideally you don't want to have anything outside of Max/Blender/Maya to get between an artist and the game. Having worked with quite the gamut of artists I've seen some who not only got used to our build pipelines, but then extended them with their own scripts/suggestions/tools... I've also had an artist who couldn't use our build system even after we simplified it to the point where all he had to do was double click on a ".bat" file on his desktop 👿

The simple fact is that most artists will be familiar with exporting meshes; I.e. solid meshes, with texture mapping (diffuse, normal, specular, etc) and it would be good to make it easier for them to get up and running so we can get more peoples ideas into the game.

This doesn't mean that we lose the use of sub-models, there are plenty of games that manage ok with those, but even if it did it would be a minor sacrifice to get more people aboard in my opinion. 😯 I know, I know, controversial.

What we could do with for models even if we stuck with the existing LMR/Lua system is a tool that makes creating, verifying and getting them into the game a LOT easier. Possibly even generates the Lua for you, lets you select models and sub-models from a list - positioning and all. What you eventually end up with though is going to look a lot like a full 3D editor.

The way forward, and I bet it's what the people on FFED3D found as well, is that meshes are better from just about every possible viewpoint. They can be better animated, textured, are easier to create and so open up that creation to more people and you can still have sub-models, plus there are massive performance benefits. Even better than that though, those sub-models aren't there because of a script which decides if they're attached or not, but are instead added as sub-objects in the game - that means you could blow the antenna off if we wanted! Visually as well as in the game logic.

Gernot, I find these arguments against moving towards meshes to be weird. Whenever I've given an artist bigger textures, a higher polygon budget, easier animations and simplifed getting their work into a game... they're generally really happy rather than threaten to quit ❓

ok, that's a argument

still i would miss it 😥

it's not that i'm completly against solid meshes, i only FEEL it will limit things a bit.

because, certain things like real dynamic changes of a mesh arn't easy possible, as example the wiggling tail of the "Turner Class".

yes, it can be done i know, you can animate a vertex group of course to do such, but 1st you need a fitting game engine, 2nd a fitting file format, 3rd i'm not sure if this is really better as a animated bezier in a scripted geometry.

besides all of that, making models with 4LOD's with meshes needs to make 4 meshes, while the scripted geometry has it's divisions set by the LOD and you don't have to care for such.

nostalgic? certainly yes 😀

i guess i can get comfortable with whatever modeling system, i mean i've hacked them as example for FFED3D to make some stuff appear in the way i like or to work around some limits given by XNA (dx6 or 7 i guess) dreamzz used to create his wrapper.

and i said that before, i really feel i'm not good enough to compete with guys like coolhand or solcommand. perhaps i shouldn't idolise them that much but....

i know its egoism of me, because i know as long as we have the LUA models i havn't much competitors. 😉

btw, underestimating isn't a bad thing, no?

at least better rather the opposite (yes, it can be a handicap, no one has to tell me that, that's why i'm only a construction worker, only helper often and perhaps one reason why i'm still unemployed, or why i never made a cent with my artworks at all, i'm not good in selling myself. sometimes this simply lets me think i'm not good enough, i can't help, it's me. i remeber i presented some of my paintings to a artist in the netherlands and he enforced me to do more and that he can see i could get some out of it, but i'm still not convinced. i guess i'm cowardice one could say. farther every "i don't like it", which is common and i have to expect such, tears me down ten times more rather a compliment could upright me).

argh, it's shitty to have a inferiority complex, you get nowhere and often useless agressive...

ok, i found out why i couldn't get it running properly now.

yes, i remember i've read such on the issue tracker (i guess), you removed the .mtl parameter.

very rarely used and never fully supported anyway.

i know that i'm by myself often recommended not to use it, but actually i don't know if it's good or not 😉

i guess good, since it never was fully supported, i only stumbled over that.

on the other hand...

it's not that bad, i mean perhaps i would have liked it if it would have been fully supported.

of course this means that the same texture will get used for all LOD's, but anyway you have to restrict them from LOD1 and make a different mesh for LOD1 at least, else you get a remarkable performance hit. farther i stopped since a long time to use different sized textures, this since we have mip-mapping and i have been told that it's as good as to use different sized textures (right?).

still a voice in the back of my head is telling me that it perhaps might be still clever to use a very basic small texture for LOD2 instead of the full sized sheet.

e.g. like what you've got in many games, a little "box" on which you plaster a "lookalike" texture (like a shot from each side of the model).

though i'm a little torn... was it good or not 😕

anyway it's fact now and i'm happy that i found the reason at all. 😀

[/hr]

not without to squeeze a little tear out of the corner of my eye, i think it's not a complete bad idea to give up the scripted geometry.

but if you do so, do it quick and painless

that would mean some models won't work anymore,

actually since my PC isn't running yet i'm not able to make quick conversions of the scripted models, else it would be no problem to snapshot them with OGLE.

Not sure exactly what you're referring to, but nothing has been deliberately removed from LMR for a long time (if ever). If you could point me to the issue or post where you read that this was changed I can investigate it and see if a bug has been introduced somewhere.

Potsmoke, would you please elaborate a bit on this? A Google search didn't turn up anything useful.

As you know, I'm trying to detail Philby's pad station. Your help with the lod and collision mesh has been invaluable. I had to put it aside for a while, and now that I've started to work on it again my mind is a bit boggled. I look at it and think, "What the hell was I doing?" I've got a collision mesh for the tower that's a little bit simpler than the one you did for me, but trying to do a simple mesh for the pad has been slow going. I'm not a programmer, so I guess that's my problem. I've never really gotten into a CAD program either. I can draw and paint fairly well but doing things on a computer is a major learning curve for me whichever way I'm going to do it. I'm inferring that OGLE is a way to convert a scripted model to CAD, am I right? That could be very useful. I'm planning to download Blender once I get my newer computer, probably early next week. I don't know whether I'll be using lua or CAD in the future, but I do want to continue contributing to Pioneer. Both forms of modeling can coexist within the game, I think.

OGLE is a tool to capture geometry from an OpenGL program by intercepting it on its way to the graphics card and saving the data out. Its very much a reverse-engineering technique, so you'll need to do some cleaning and fixing of the data you get this way.

http://ogle.eyebeamresearch.org/ appears to be down, so I'm not sure where to get it from. There's a little bit more information here though: http://eyebeam.org/projects/ogle .

I think it unlikely that Pioneer will have both a scripted and a traditional model system at the same time. It might have been a possibility if LMR wasn't such a damn mess, but as I keep saying, it needs something approximating a rewrite. There's no interest in doing that, so it will likely be removed once the new system is done.

We may still keep some kind of scripted model system around to support things like procedurally-generated geometry (eg buildings), but if it does happen it will likely have quite a limited/focused scope. It most certainly will not have the ability to regenerate geometry each frame (ie the dynamic function).

We of course won't be replacing the current system until we have converted or replaced enough models. It would be rather foolish to release a space game with no ships in it 😆 Its also worth noting that its likely going to take months before this is ready for use, so the existing system will be a part of our lives for a good while yet.

@robn

i guess it's no bug, it's only that i had a sub-model (unreleased you ca say) which still used the .mtl lib, i thought really you (or else of the dev.) must have removed the support for it completely. that's also because (i don't quite remember which issue, but i think it was in conjunction with the new buildings) i have read that you will remove it, or at least that it isn't wise to use it because you can't restrict textures from the collision mesh then if there is no special collision mesh.

anyway i also recommended always not to use it, because of the above described problem and that if you can't specify a material in the .mtl lib it din't makes much sense only for textures (yes you could once specify a material, two years ago, but i recognized a problem with that, the material appeared inconsistant, means in different colors from different angles, tomm or j.j. must have removed it then).

@marcel

i think as a precaution i uploaded at least the handy GLXtractor to SSC, which is a frontend to OGLE (ogle is hard to setup if you don't know exactly what's behind. i see it as little risk for copyright infringement* if you can snapshot a mesh from any Open GL program),

anyway if you once have downloaded and installed OGLE, you can use GLXtractor to let it do the job for you.

*but we don't like to rip for sure

i will search for the link(s)

further OGLE is also available on the page from ETH zürich (eigenössische technische hochschule, zürich)

it's a cool tool imo, i extracted once a whole scene from pioneer, you can later watch or render the scene in blender from any angle or position.

on the other hand, i cant tell you what OGLE will change on your OS, it's at least a violation of Open GL.

and marcel, in the case for the groundstation "beam" it to me, i will throw a eye on it and this time comment what and why i did this or that.

❓ "was'n das?" (what is that good for? not ment as criticism)

[attachment=1169:Bildschirmfoto 2012-04-20 um 17.44.23.png]

well, i have some idea, but shouldn't there be a 100% too?

"well, i have some idea, but shouldn't there be a 100% too?"

Read manual. Explains it well.

Now that's a criticism. 👿

Wouldn't be a low thrust power level if it was full thrust, would it? (-:

oh yes, manuals.

i guess the last manual i have read was SimEarth's, 300 pages and the author stated that he didn't know why he wrote it, it won't get read anyway. 🙄

that's not true.

i must have read a lot of them, but usually i play and if i encounter a problem then i start to read.

that was a bit different for FE2, but it was fun to read (and even simearth's manual was fun to read and a source of knowledge).

without RRT2's manual i wouldn't have been able to make the missions of course (else i played the original many years before, so i knew the game).

i had books filled with "manuals", from my daddy, such as "ich sag dir alles", or "wie funktioniert das", back from the 50's.

there was manuals in it how to plaster a street, or how a lathe works, how a orchestra is scored, what size a soccerfield is, ... and, and, and.

and sometimes on the throne i read manuals of all kind, this since childhood 😆

do we have a lecture which i can take to the throne?

and if you ever think i have confusion in mind, now you know why

still i don't know what's the purpose of it.

ok, perhaps i can't imagine why i should use 50% power, i mean i never drove my motorcycle at 50%, vice versa i would say, at 150% 😆

[/hr]

to sad not all of you can read german i found a "manual" for "Staats- Stadt- Haus- und Landwirthschaft" from the 18th/19th century.

when i start to read it i knew where "old shatterhand" has it's knowledge from.

anyway here's the link

http://kruenitz1.uni-trier.de/

it's a must read imo.

and for those who can't read german, well they can have at least a good laugh from the pictures in it

http://kruenitz1.uni-trier.de/postkarte ... tz_pk.html

i like this one especially

this ones of special interest for us