[NerdZilla]

thats because it's duke nuken... i meen if this was halo? that dude can't clime a hill, even with his super sute. you have to remeber, this is duke we're talking about.

[Mikko]

 DeeperThought, on Nov 9 2009, 06:09 PM, said:

That looks really nice. How do you imagine that mappers will make use of this without having a terrain generator?

Not to mention the amount of resources a realistic looking terrain consumes.

[Geoffrey]

 DeeperThought, on Nov 9 2009, 09:09 AM, said:

Another thing that occurs to me is that in Duke, player speed is not adjusted for slopes. I mean, the player continues to move at the same horizontal speed regardless of the degree of slope, so the result on that terrain would be be pretty strange.

That always kinda bothered me, it's the one thing that makes realistic terrain in Duke feel like an unnessecary gimmick: you clip over it and it never becomes an actual in-game obstacle. That, and the fact that Duke's distance above the ground is a straight vertical line, instead of somewhat using the floor's curve to push Duke back. So on a really steep slope you're walking around with your face in the grass.

[Danukem]

 Geoffrey, on Nov 9 2009, 10:11 AM, said:

That always kinda bothered me, it's the one thing that makes realistic terrain in Duke feel like an unnessecary gimmick: you clip over it and it never becomes an actual in-game obstacle. That, and the fact that Duke's distance above the ground is a straight vertical line, instead of somewhat using the floor's curve to push Duke back. So on a really steep slope you're walking around with your face in the grass.

At some point I was going to fix this in Duke Plus by changing the player's velocities based on the degree of floor slope. But then I realized that it was more complicated because it depends on which way he is facing relative to the slope (going DOWN a hill isn't slow) and I wasn't sure how to calculate that so I never bothered with it.

EDIT: Typing that out made me think of a cheap way to do it that might work ok.

[Stabs]

meh chews to many walls & sectors, dosnt really bring any new tech in, its just stuff anyone could do if they gave it the time, but most mappers dont, because its way to costly, there is no way it could be used to make massive terrains either, which would be the point of something like this.

it looks good yes, but thats only one aspect of mapping

[Daedolon]

Not sure how well does this count, but I started working on a cover version of Grabbag, I only got as far as recording the beginning, then spent the rest of the day trying to come up with a decent mix, it's a bit quiet atm. It's on one of my MySpaces.

[Antal]

I'm helping with a 'HD' soundpack for Duke. About 50 new sounds here in improved quality, note that the pigcop sounds aren't very close to the original and only here as placeholders

Includes a CON file for the sound to work.

http://www.romes.nl/...d_soundpack.rar

[TerminX]

 DanM, on Nov 9 2009, 11:59 AM, said:

meh chews to many walls & sectors, dosnt really bring any new tech in, its just stuff anyone could do if they gave it the time, but most mappers dont, because its way to costly, there is no way it could be used to make massive terrains either, which would be the point of something like this.

it looks good yes, but thats only one aspect of mapping

Indeed. It's really important to not render thousands upon thousands of walls simultaneously, for performance considerations.

[Stabs]

that and the fact you make this landscape put half a building or something in it and are back on here crying about how there is not enough walls in mapster

[The Commander]

It could work well for a small DM map or something similar.

[NerdZilla]

DeeperThought, on Nov 9 2009, 10:21 AM, said:

At some point I was going to fix this in Duke Plus by changing the player's velocities based on the degree of floor slope. But then I realized that it was more complicated because it depends on which way he is facing relative to the slope (going DOWN a hill isn't slow) and I wasn't sure how to calculate that so I never bothered with it.

EDIT: Typing that out made me think of a cheap way to do it that might work ok.

yeah it was alot of work. some parts took whole day to finesh. i had to brake down a few numbers by hand with a cululator. to move the wall to get the right angle.

im still serprized at how well duke runs these hills. i get a GRATE FPS in my game off my computer =D. the part im looking forword to is getting a team to try to take hold of the dich between each base. i really thought out how the hills play out. so i look forword to leading a team into VICTORY!!

EDIT : i just realised i miss read that... sorry, for the strage reply. i pulled an all nighter last night at my work. it would be interesting to see duke 3d behave to those standerds. but it is kind of awesome how cerrent settings play out in that map.

[Gambini]

Fully animated and almost ready to be used

[Stabs]

looks good who is coding for it? and whats happening with the laser gun?

[Chip]

Quote

i get a GRATE FPS in my game off my computer =D.

Think Polymer lighting and high res textures.

[Gambini]

DanM, on Nov 10 2009, 03:39 AM, said:

looks good who is coding for it? and whats happening with the laser gun?

Nobody is coding for it yet, but i put all my tokens in DT´s expertise

And about the lasergun. Well, i´m learning little by little what is all about this of importing and exporting models. The lasergun will be probably redone and merged with duke´s hand for Dukeplus, later of course. Because i already have at least three totally different projects in progress, a Half life 2 ep2 mod, a single user map for vanilla duke, and all this idea about the 8bits looking models (without forgetting some models i want to make/modify for DP), all them deserve to be finished.

[NerdZilla]

Chip, on Nov 10 2009, 08:01 AM, said:

Think Polymer lighting and high res textures.

Your right, I have a working polymer right now I'm sure it'll cut the FPS of one my labtops. I'll just run that one in software.
It was all readdy becomeing a day time map in the middle of that landscape design to takes up walls.
But now I'm planning on doing a night version of the map with polymer. I'm making sure its a totol use secters and walls for best GFX. so of course, a slower GFX will lose FPS. So, for those slower GFX PC gamers: Hey my friend, I'm vary sorry about my dope map designs. You mite have to turn off polymor or cut to a lower res to play then. As my designs will all ways push vary limits of duke3d... Software mode?
But anyways, well it was comming along. I got the idea to put a camra that flys around the map that both teams can look at (so no hidding spots).
What im asking is; Can I make it so theres a spotlight on that cam? That would be awesome to see in game.

[Tea Monster]

Generally it looks good, but the 8-bit terrain textures look very out of place with the 3D models. Grab some upgraded grass and dirt textures to go with the models.

[Sebastian]

Still just having fun with some old work on new software. The soldier sprites are just player dummies scattered around to get a sense of what could be.

[Gambini]

Looks great!

Are those enemies from some alien game or they´re brand new sprites? first time i see them, although i would dare to say they´re from Alien Trilogy .

[Sebastian]

Thanks! These fellas are from Aliens vs Predator 2 (by Monolith) on PC. I custom skinned both of them to make the conversion to Duke a bit smoother. The dome on their head is actually blue now, incase I needed more than one type in the future (this was for the idea for the old TC).

[NerdZilla]

those spites look awesome. it kind of reminds me of the spites for AvP on Jag64.

[Sobek]

Yeah that Alien sprite is bloody scary, I ike it.

The last week or two have been pretty shitty for me, so I haven't had time to do one damn thing I wanted to. I did however find the occasional bit of time here and there to do some modelling work. What I'm working on is implementing a shuttle bay into my Voyager mesh, mainly just for cutscene use (there's a seperate map of the shuttlebay itself for when you're actually walking around in there). I started by taking the (insanely) high quality mesh made by 'Sarod' (you can find that on Trekmeshes) and basically creating my own replica shuttlebay from scratch, using his mesh as a visual aid / base. Most of what I've done so far is VERY rudimentary, and there are so far no textures in place, so bear that in mind.

I think another week and I should be about done, but it's tricky getting everything to sit just right in my Voyager mesh. I opted to make the shuttlebay itself a seperate model that just slots into the back of my Voyager mesh, rather than try (and mostly fail) to physically integrate it. It's just more flexible this way. In the end it will have proper railings, 'warning' wall details, entrance / exit doors, and a hatchway on the floor that, at least in the actual map, will be lowerable.





*edit* Just as a test, I put in a shuttle and some simple details. It kinda shows how well it'll gel with the overall mesh in the end. Irection it'll look great in action;

[Daedolon]

Wow that's impressive, but how do you keep everything on screen without the models disappearing, or has the Model Occlusion been fixed?

[NerdZilla]

just some pics of what you mite expect from my work. the buildings are made from 2d mode friendly spites. 2 floors with a roof. buildings (bases) will get more dynamic as i go along.

[Sobek]

Daedolon, on Nov 16 2009, 10:39 PM, said:

Wow that's impressive, but how do you keep everything on screen without the models disappearing, or has the Model Occlusion been fixed?

It hasn't been 'fixed' yet, no (I got the impression that it wasn't exactly fixable really, and the only 'solution' was honestly not ideal so I don't know what to do about it all). Since this will only be used in cutscenes, I do have a few ideas with regards to player position & sector design to keep what I need visible even while moving... But it sure is annoying. Sometimes you just CAN'T keep the model constrained to it's home sector, and even when you do, simply standing in a seperate sector negates that anyway (if the sector the model is in is far larger than the one you're standing in).

I really do hope that some kind of 'priority' tag or system is implemented one day that lets you assign specific models to be rendered at ALL times. It seems like it shouldn't be TOO hard to accomplish

NerdZilla, on Nov 17 2009, 05:19 AM, said:

just some pics of what you mite expect from my work. the buildings are made from 2d mode friendly spites. 2 floors with a roof. buildings (bases) will get more dynamic as i go along.

Hey I really like that. I don't know, something about the design of the area is just right. It almost reminds me of something from Tribes 2... I'm loving the snow

[Stabs]

have you tried what i suggested?

rotate the model 180 and flip it horizontally (i think this is how you would do it) in a model editor, then put it into build and make it rotated / flipped to face the complete opposite direction of the original sprite and place on the original sprite, it should still look the same in-game, and then when one disapeers another is there to take its place, i think this may work

other alternative is to go one further and use 4 different models covering all 4 sides

[Gambini]

I dont know whether it´s the same method or not. But for me worked pretty well to make the sprite wall relative. If the dummytile is big enough for taking the same size than the model, it would not disappear.

[Sobek]

DanM, on Nov 17 2009, 12:39 PM, said:

have you tried what i suggested?

rotate the model 180 and flip it horizontally (i think this is how you would do it) in a model editor, then put it into build and make it rotated / flipped to face the complete opposite direction of the original sprite and place on the original sprite, it should still look the same in-game, and then when one disapeers another is there to take its place, i think this may work

other alternative is to go one further and use 4 different models covering all 4 sides

I recall trying that, however I never could get it to work quite right. And when dealing with models that have a decent poly count, it's probably not ideal from a performance perspective to double or even quadruple what's effectively being rendered. I still think some kind of capability to force the rendering of any given model or sprite on demand would be a PERFECT solution.

Gambini, on Nov 17 2009, 01:21 PM, said:

I dont know whether it´s the same method or not. But for me worked pretty well to make the sprite wall relative. If the dummytile is big enough for taking the same size than the model, it would not disappear.

That can help, but there are many situations (especially for myself while using large-scale models) where this still doesn't help. The problem is that the model has that queer priority about rendering itself where if you're in a separate sector to it, so it'll vanish on those extreme angles, regardless of whether the sprite & model are of an equal size and fit within their sector etc... Look at the image I just whipped up as an example;



The red line indicates the size of the sprite that the model is 'overlaying' on (the dummytile basically). The player is situated beside the sector the model is in, in his own sector. The two highlighted green areas indicate the two most extreme viewing angles I would be able to look at before the engine just stops rendering the model. If I were to turn any further to the left or right, it simply vanishes... Even though the base sprite & model sizes match, and it's not protruding out of its 'parent' sector. The engine just culls it when you turn your view too far (but NOT if you stay within the same sector. If you're in the same sector you can turn on any damn angle you please and you're good to go).

Obviously making the model smaller to fit within the confines of those two most extreme viewing angles often 'fixes' the 'issue', sometimes that just isn't an acceptable solution. Sometimes, I NEED the model to be a certain size in order to generate the right scale and atmosphere in a part of a level.

I'm not trying to give you a 'dummy' explanation of this, but I'm just trying to make it clear what it's doing...

*edit*

As you mention though, making the sprite/model wall aligned does sometimes do the trick... But not always. I don't get that. Also, sometimes it just randomly doesn't happen too. Just now I recreated the same map that I used in the above picture, and for whatever reason the model refused to vanish on those extreme angles, whereas it was doing it earlier today when I tried to do the exact same thing. It's so bizarre.

[Stabs]

the only other possible solution i could see is if you code the model to work like the blue deck chair sprite 680 - 684 and gave it a whole bunch of different tiles for different angles ???

MAYBE, dunno, probably not

[Chip]

Perhaps you could use the "showview" command (I think its that - the one that draws another viewpoint on screen) point the angle of view at the center of the model and have the window to be drawn on screen set at 1x1 and in the corner so you see nothing out of the normal but since the showview command is forcing what it sees to be drawn then the model will be always drawn regardless of where the true player's view is looking.

However I have never used this command so I don't know if a window of 1x1 on screen is possible.