[Galemp]

Post the 3DS and whoever does it will post.

[mikhael]

3ds? I'm afraid to do that. It'll get decomposed into triangles. Right now its a combination of quads and triangles.

[mikhael]

I'm working on the LODs and debris as I write this. I'll try to have the mesh uploaded as an LWO in the next couple of days.

[Lightspeed]

I might texture it

[Drew]

ud better do a kickass job m8

[Stryke 9]

Doesn't FS only take triangular polys? I'd imagine so, at least, since that's the base for everything else. In which case, it'd be a good idea to find out what the real polycount is.

[spaceman spiff]

oh he will...(i hope he does) lightspeed's the best. just take a look at all his shinemaps.

[spaceman spiff]

whoops... forgot to quote

[karajorma]

Originally posted by Stryke 9
Doesn't FS only take triangular polys? I'd imagine so, at least, since that's the base for everything else. In which case, it'd be a good idea to find out what the real polycount is.

I'll agree with you there but the HT&L build automatically triangulates models so you don't need to triangulate models yourself (In fact it leads to bigger file sizes and longer load times so it's a good idea not to).
 The only reason you need to triangulate a model is just to check the poly count.

[Ulala]

Let's see some textures!

[Lightspeed]

Someone needs the model to have a try texturing it. Best would be to have a model with placeholder textures (solid green, red, blue, etc) so you can mess around without actually having to edit the model

[StratComm]

woah, woah, wait a minute here... FS does and always has taken polygons with greater than 3 sides.  In the HT&L builds they get triangulated, yes, but in regular freespace they do not and you just have to follow the geometry rules.  Now with a model this complex, you might want to triangulate it and see what happens to the polycount.  But it still sounds like it's within limits.

[Stryke 9]

But it's built for HT&L, which does triangles. And he's been saying that his model has a lotta quads, and that he doesn't want it to get triangulated- i.e. because that'll increase the polycount substantially. So, as I was saying, the polycount with quads might not be very useful in determining how useable it is. Just sayin, y'know, so that we don't end up with a lotta crashing later and a buncha people stuck doing it all over again from scratch.

[karajorma]

Originally posted by StratComm
woah, woah, wait a minute here... FS does and always has taken polygons with greater than 3 sides.  In the HT&L builds they get triangulated, yes, but in regular freespace they do not and you just have to follow the geometry rules.  Now with a model this complex, you might want to triangulate it and see what happens to the polycount.  But it still sounds like it's within limits.

As Stryke said the model is for HT&L. Anyone trying to run this on non HT&L will get crashes, shards of death and all other kinds of problems cause I doubt that Mikhael has stuck to the <~800 polys rule that you need to stick to in non-HT&L builds.

It is worth doing a triangulation just to test the poly count but the Mjolnir is closer to a capship than a fighter (even in Kings Gambit there were only a handful of them on screen at a time) and that means that it can use the capship limits of over 20,000 triangles. I think it's a fair bet that it's under that

[mikhael]

Right now, the model is done as quads, tris and 25 higher order polys (hexs and octs). I will devolve the hexagons and octagons into tris and quads manually. I prefer to leave models with as many quads as possible, so the engine and/or model packing process can decide how best to triangulate the polys.

My main concern about converting the model to 3ds has nothing to do with the engine though. I just find the 3ds format is... untrustworthy when it comes to model conversion. Given the choice, I'd rather send it to Obj format and import from that. 3ds is just crap.

[Drew]

Originally posted by spaceman spiff
lightspeed's the best. just take a look at all his shinemaps.

Dude! Lightspeeds got nothing on GE!!!!!!!!!!!

[Galemp]

Originally posted by Drew

Dude! Lightspeeds got nothing on GE!!!!!!!!!!!

..?

I'm flattered...

[Stryke 9]

Mik: Ah, makes sense. And I agree- MAX ain't so hot when it comes to communicating with non-MAX software...

[mikhael]

The README.TXT from the zip file:

Code: [Select]


First, a dedication:
To the Freespace community at large, and the FS:Source
Code Project team in particular. Without you lot, this wouldn't
have been made and I'd probably still be doing some really
pathetic stuff in Truespace3.


This archive contains an updated version of the RBC Mjolnir
model from Volition, Inc.'s FreeSpace2. Though the concept
of this mesh is not original, the entirety of the mesh included
in this archive is modelled from scratch.

Permission is granted by the creator to use, modify and
distribute this mesh under the terms of the GNU General
Public License. For questions regarding the license, refer to
the LICENSE.TXT file within the archive.

The files contained in this archive are:
             README.TXT -- You're reading it
            LICENSE.TXT -- The GNU General Public License
  MjolnirSCP-1layer.lwo -- The RBC Mjolnir in LWO format on a single layer, exploded
 MjolnirSCP-10layer.lwo -- The RBC Mjolnir in LWO format on 10 layers, assembled
         MjolnirSCP.3ds -- The RBC Mjolnir in 3ds format, exploded

If you choose to texture this model or convert it for use in
Freespace, or any other project, please send a copy of your
work back to me so I can include your work in this archive.

Questions about the mesh can be directed to [email][email protected][/email]

Thanks, and enjoy.

Mikhael@HLP




Get the zip file.

Currently lacks debris. I'll worry about that later. Now people can get to work trying to map her.

[Sticks]

Yes!

This is what I want to see! And Bob is right, with HT&L we can push tons of polygons even on older cards. I've almost been tempted to start putting together some models myself because the old models are far too blocky and really don't put the new code to the test at all.

The other reason is that once high poly ships start floating around, people will stop complaining that specular is up too high. The reasons ships look to bright right now is that there aren't enough vertices to get lighting normals from, and thus you have these massive polies getting lit.