Remaing models to beta:

Post by **Rebel** » Sat Oct 09, 2004 6:34 pm

Ok... give me a moment.

Try this one

Post by **Remdul** » Sat Oct 09, 2004 6:51 pm

This one worked.

Got the idea, I'm on it.

Post by **Remdul** » Sat Oct 09, 2004 8:08 pm

Aaaaaaargh! I was just gonna export the .tpm file when 3dsmax crashed and dragged the scene file into it's grave! Damn app, keeps corrupting files on me. Still have an auto-backup from 5 minutes, ago so lemme remodel some things quick.

In the mean while, enjoy watching the render:

http://home.planet.nl/~buijs512/trescom/thing.jpg

Gee, I always rename the existing file when saving in my apps, and after the new file has been saved I delete the backup. Hope they did the same thing with more software...

* Here's the thing:

http://home.planet.nl/~buijs512/trescom/cavern_end.zip

Includes the plate at the end, which I gave a bit of hollow shape for parallax.
If you like I can create a few elements that go in front of it to simulate rays in the water (just a single transparant texture that scrolls horizontally) or perhaps fish (same technique). I'm not sure how long the animation lasts, and whether that would be worth the trouble.

But we can add some underwater bubbles inside the cavern, using the same stuff we used for the water drops, that shouldn't take much trouble.

Post by **Rebel** » Sat Oct 09, 2004 8:35 pm

The only thing inside the zip is a bmp file, Remdul.

Yea, the render looks pretty good though. If you could come up with 1 scroll texture to simulate water waves, go for it.

Post by **Remdul** » Sat Oct 09, 2004 8:48 pm

Oh. I forgetting stuff again...I'll take a break after this tonight.

Here's the file:

http://home.planet.nl/~buijs512/trescom/bubbles.zip

Cavern model included.

Includes a transparant bubble texture, put it on a 1x2 board and move several it up on cue during the end animation.

Also includes the ray textures. Just copy the backplate model twice, put this transparant texture on them, and let one scroll to the left, the other to the right (ocean object scrol stuff) and you got nicely animated light rays.

Post by **Rebel** » Sat Oct 09, 2004 8:51 pm

Believe it's all there now, Remdul. You have a good night.

Post by **Remdul** » Sat Oct 09, 2004 8:58 pm

Thanks.

Btw, I got the Spitter entirely rigged, and I'm almost done tweaking the Parasaur. TPDC is doing the job well, really putting it to the test now. The 'new dino' wizard has been cleaned up and tested thoroughly, and I actually started using it myself

. Being able to edit the dino after rigging makes the process a whole lot easier and faster.

Anywoo, I'll probably upload those two dinos tomorrow.

Post by **Rebel** » Sat Oct 09, 2004 10:16 pm

Sure. I could always import them into test levels, I just won't do a tc_isle import until I'm ready for them. I still want that raptor.

TPDC is a little confusing to use, but the few times that I've messed with it the learning curve isn't too bad, it's moreso that editing dinos isn't my thing. I'll leave the creation and editing up to experts such as yourself.

Post by **Remdul** » Sun Oct 10, 2004 10:19 am

Yes, I'm gonna include some more documentation in the 0.2 release, but everything has become so simple that everyone should be able to use it.

Got the Dilophosaur aka Spitter here:

http://home.planet.nl/~buijs512/trescom/spitter.zip

I haven't renamed it yet (still named RaptorB, we can change that right?) and in some poses its neck looks a bit odd, but it looks much better than the original dinos.

Post by **Rebel** » Sun Oct 10, 2004 11:52 am

Downloaded. (Check it out later on today, got some crap to do this morning)

I'll do the renaming before import, not a problem. (one of the niceties about tpm format) More docs would be helpful for the users of the software. As Martha Stewart use to say 'before her jail life', 'it's a good thing'.

*Think you'll be satisfied with the swim exit.

Post by **Rebel** » Sun Oct 10, 2004 11:31 pm

Hey, Martijn. Sorry it took so long for me to give some feedback.

Overall, I'm impressed. He's a helluva lot scarier and cooler looking than our now oldtime tres enemies. I noticed that neck genuflex you mentioned, but I wouldn't worry about it, it's no big deal at all. The way the original dinos contorted, their anomalies were much more apparent. He eats well, stalks well... I like it, damn good job.

I noticed that the creature's face textures don't hold well at a distance, but I'm not sure that there's much we can do about that. In a smaller level (like in the testlevel I ported the dilo into), I could go mipmap false to clear up the texture distortion, but I'm not sure that we'd be able to do that in tc_isle. Unknown to me how much memory is used by storing entire textures in memory, but I imagine it's enough to concern ourselves with.

Beyond that, the community is definitely going to enjoy your creations.

Grab this file I renamed your dino for you.

Contains tpm of dino in 2 unique names (Spitter/Dilophosaur -- I checked out both renamed dinos ingame, no problems), choose whichever you prefer. You should probably create a new test level and import these guys into it, that way when you import the raptor you won't have to worry about name clashings. (Omit the mipmap value inside the values text before import if you want, or perhaps import it the way it is to view the difference)

Grab this file as a Mesh Render of the Dilophosaur model. Basically, this was the dreamworks' way of doing their dinos. They'd import their master model first, which in this case is the Dilophosaur, then they'd import the render that would use the master mesh (in this case, the render here is DilophosaurB) Well, you probably know all this, but I wasn't certain if you ever tried to do an import like this before. It actually has advantages, as each mesh render can be assigned different values than the master mesh.

Post by **Remdul** » Mon Oct 11, 2004 12:00 pm

Thanks Rebel.

Yeah, I noticed the texture problem on the head as well. Most likely has to do with the mipmaps themselves. I'm not sure how GeomAdd generates these, but I remember Andres mentioning problems with that (blank mipmaps etc.). I think Tres grabs the lowest mipmap for some reason in this case, or the mipmaps themselves may be messed up. Vertex normals are used as well to determine the mipmap level (though Trespasser uses per-face normals instead of per-vertex). TPDC re-normalizes those when loading models, so that shouldn't be the problem (I'll check on this to be sure). But it only seems to happen with the head texture, so its most likely an issue with the engine itself.

Setting Mipmap = false shouldn't hurt the performance to much I think, I suggest you use it, unless it causes other textures in the level to go limbo.

I'll add dino renaming to TPDC as well, just a matter of looping thru all bones and other elements and replacing their names.

Post by **Rebel** » Mon Oct 11, 2004 12:13 pm

I was playin' with the little critter again this morning while sippin' my coffee -- nice work, man.

Going to try n' import the venture's bridgeworks and swim end into tc_isle in a bit. Hopefully, that'll go smoothly (though, more often than not, it doesn't

)

.... other than those, we're down to two statics.

Yea, I recall Andres mentioning the mipmaps, though don't remember the specifics.

Post by **Remdul** » Mon Oct 11, 2004 4:09 pm

Been playing with the para but I've had some problems with the mouth. For the new skin it opens to far, causes some warped polygons.

http://home.planet.nl/~buijs512/trescom ... saurus.jpg

Now on bones there are values that can influence how much vertices are rotated, "RotationRatio" and "Ratio". Not sure what these actually do, they don't seem to make any difference.

Anyway, I've uploaded the para:

http://home.planet.nl/~buijs512/trescom/parasaurv1.zip

I thought maybe our 'script doctor' may want to have a look at it. The jaw joint is number 09.

I've added the ActBit values purely for testing btw.

If we can't solve it, I'll just assign the lowerjaw vertices to the head, but then it won't be able to open it's mouth (which doesn't seem to be very often anyway).

Other than that, these values should be used on all dino models:

float Diffuse = 1.000000
float Specular = 0.000000
float SpecularAngle = 32.000000

This disables specular, which didn't look really good, but more important it removes popping black vertices. Note that these have to be set before import, Tres may crash otherwise, according to TresEd.
And it also seems that after changing these values (when you already have the models in the level), GeomAdd adds, the textures as new, doesn't replace them anymore. Not sure this is a bug or something...

Post by **Rebel** » Mon Oct 11, 2004 8:49 pm

I'll give it a look, but if you can't locate problem, it doesn't bode so well that I will.

I'm not really up to date on these riggings, but it seems that there's a vert assignment some ways past the actual mouth opening available. Wouldn't that by itself cause the distortion? It appears that it's attempting to scretch open a section of mouth that's closed off by attached faces. It'd have to distort, yes? Perhaps recheck the assignments?

It'd be a shame if you had to unassign mouth movement, as when it opens its mouth partly to eat, it does look quite cool and naturally adds to the realism of the dino.