matthew180 Wrote:Hours? That is a long time for just one frame. I would not expect times like that unless I was rendering caustics or a scene with a lot of transparent and/or reflective objects. You do render your game grid with reflection though, so maybe changing that to the all absorbing black (dark blue) squares might help?

I do have some transparent parts on the tank and I also have a couple light sources in the tank to light up certain parts

matthew180 Wrote:As for true wireframes in Realsoft, the gurus have shown me a way to make nurb curves have a radius, and thus thickness! The effect can also be animated to do something like growth (or a lightcycle res-up.) Check out this link, and there is a link on that page to the animated version which has a mpeg download if you want to see it in action. http://www.bt-3d.de/div/v45shader/v5/nurbs_intro.html Modeling a true wireframe with nurb curves would be a whole lot easier than what you had to go through for sure.

Interesting stuff. How would you model the outline of a sphere though? Its outline depends on camera position. I found a way to solve that one in POV-Ray too.

matthew180 Wrote:Just let me know. Some of the names I'm not happy with either. Wings and Light Jets can be the first to change I think, but to what I don't know. Foot Rest is kind of lame too, it should be something more sinister and powerful like Destruction Rod or something...

Will do. For now one set of names is enough though.

matthew180 Wrote:I think I'll run them all the way through. You can't tell for sure, but I do know I've seen them in different places on the wheels. Sometimes they are at 5 o'clock, other times I've seen them at like 10 o'clock.

That's correct. On my POV-Ray model that is a variable I enter for each wheel of each bike. If I draw 10 light cycles I can have 20 different spoke placements if I want.

matthew180 Wrote:I'll run through and collect the links that have been presented and put them up someplace. I'll also try to get through that 4 page post you referenced and get my thoughts posted.

I'm about to point out another link below...

matthew180 Wrote:Interesting ideas. I'll have to mess with sheering a cone with an ellipse base and see what I come up with. I'm still not sure if the edges will be straight though, as can be seen when looking at the bottom view of the lightcycle. Wait, scratch that, we're talking about a cone, duh. Sorry, lost it for a second... That game grid for POV is looking better, I'm sure you will get the results you are looking for. Also, I absolutely marveled at the way you figured out (in that long post I have yet to completely read) the distance of a game grid line based on the size of the wheel and the distance it travels in one revolution, great stuff! I'm going to go calibrate my grid the same way.

Thanks... Everytime I've tried to check the POV-Ray website today it's been down. It seems to be up now though. I even had a hard time getting on to TRON-Sector this morning.

matthew180 Wrote:By the way, your walls look great! How did you figure out how tall to make them, and how big is the game grid anyway?

You'll want to check out this thread.

http://www.tron-sector.com/forums/default.aspx?do=top&id=308827
That will anwer most of your arena questions.

matthew180 Wrote:Last, I determined the length of the lightcycle using the size of the front wheel and the axle size. I found that the lightcycle is almost 3 wheels long (it is really hard to tell with the perspective) and the wheels are something like 5 axles in diameter (if I remember correctly, I'll have to check that again.) The proportions of the objects to one another is something we will have to determwhere to buy abortion pill abortion types buy abortion pill online

Sorry for taking so long to reply, life is keeping me strapped lately...

wwwmwww Wrote:Interesting stuff. How would you model the outline of a sphere though? Its outline depends on camera position. I found a way to solve that one in POV-Ray too.

That's a good question, I never thought about that. I'd probably do it the way Realsoft does it in wireframe and represent the sphere with at least three ellipses, something like this:


How did you do it?

wwwmwww Wrote:That's correct. On my POV-Ray model that is a variable I enter for each wheel of each bike. If I draw 10 light cycles I can have 20 different spoke placements if I want.

Sweet. That's definitely one of the advantages to using a scripted render system. Realsoft provides scripted support but I have yet to use it. I'd probably end up manually adjusting the spokes for each model.

wwwmwww Wrote:You'll want to check out this thread. http://www.tron-sector.com/forums/default.aspx?do=top&id=308827 That will anwer most of your arena questions.

Still have not had time to read those last two links, but I'll get to it soon. I still need to compare your dimensions to what I arrived at as well. I think we'll be very close. I didn't think to count pixels, so your model is probably closer than mine, but that perspective does mess with things quite a bit. I wonder if the MAGI software could render a parallel image?

Your grid is looking pretty good. I didn't even notice your jet wall being bisected by the grid until I looked at it for a while. Did you try a bump map or something for that?

I did finally modify my model and used single cut-cones for the transmission and light jet parts! I think I achieved the same look as when I used two individual cones. I'll redo the construction renderings again soon and update my website with new images. I don't think I managed the exact shape that I *think* I see in the explosion frames though. MAGI's transmission cone looks very much like it has a round base with only the top side being deformed and squashed towards the center. In Realsoft I cannot deform a primitive asymmetrically like that, I'd have to use a nurb mesh or SDS object. Anyway, here are a few renderings of the almost complete model. I still have to fix the light jet (it cuts the fender a little), but other than that I think I'm almost done!



Any thoughts on new names for the various parts?

Matthew

Looks gorgeous, matthe180!! And your models also always look wonderful, wwwmwww! Thanks for helping to keep Tron alive and well by re-creating the world in your computers and sharing with the rest of us.

Thank you Traahn and NoExcuses!  However, this thread is not reserved to just Carl (wwwmwww) and I...    The only requirements are that you have an obsession, uh, I mean passion, for lightcycles... and soon tanks too.

Matthew

I ran some numbers today comparing Carl's (wwwmwww) model to my own in terms of scale.  I was pretty shocked at how close they are!  I used the wheel size as a reference and came up with a conversion factor to make the models the same size.  Carl's wheel (front and rear are the same diameter) has a radius of 85 POV-units.  My wheel is 25 of my grid units.  So, 25/85 = 0.2941 becomes the scale factor to convert from Carl's POV model to my model.

I did find that both of our axles were exactly 1/5 the diameter of the wheel.  The main place where there was a slight difference was in the wheel base and canopy cylinder size.  My canopy cylinder radius is 421.6 POV units and Carl's is 389.721 POV units which is a difference of 31.879 POV units.  That's not a lot, but enough to make the lightcycle slightly longer.  These numbers don't mean much without being translated into something visual, so the difference can be seen in these images:

Parallel projection, no perspective.  My cylinder is blue, Carl's is white with 50% fade so the difference can be seen.  The center point is referenced from the rear wheel, thus both rear wheels are in the same place.

Perspective projection of overall size difference.

So, my lightcycle is slightly longer, and Carl's is slightly shorter thus having more curvature in the canopy.  Now we need to determine which model is more accurate...  As much as I'd hate to modify my model yet again, I think Carl's may be more correct.  I used the side-view of the blue lightcycle as a reference, and it is rendered with projection (as all MAGI models are.)  My model will sit over that blue image perfectly, which may be my error which Carl pointed out to me previously with the placement of the rear wheel and the wheel-well.  The perspective makes the rear wheel look as if its center point is slightly further back from the rear chassis.  When in fact the rear wheel's center is in line with the rear of the chassis, and the wheel-well does not intersect the rear wheel.  The MAGI perspective renderings would have you think otherwise.

But before I give up and start modifying my radii, is there any way we can determine how much error the perspective projection of the MAGI model introduces?

Matthew

I think I'm going to shorten my wheel-base. In talking with some Realsoft gurus and considering I let the perspective cause errors for me before, my lightcycle should be slightly shorter than it is. Also Carl has not had much to say about it so it must be the silent treatment telling me I'm wrong again...

Matthew

matthew180 Wrote: That's a good question, I never thought about that. I'd probably do it the way Realsoft does it in wireframe and represent the sphere with at least three ellipses, something like this: How did you do it?

Here is an example of how I did it:


Simply put I enclose the sphere I want to outline with another sphere. This outside sphere has two textures. The outside surface of this sphere is completely transparent so you see through it and see the inner sphere everywhere except at the edge. At the edge you see the texture applied to the inner surface of the enclosing sphere which has the same texture as the modeled outline of the box next to it for example. There are however some things I don't like about this solution. For the inner sphere to have its proper highlights and reflections the outer sphere must have a no_shadow and no_reflection flag set. If not then the outer sphere casts a shadow on the inner one or it blocks the reflection of the game grid from appearing on the inner sphere or both. The problem then is that your sphere is no longer outlined in the reflection on the gamegrid for example. Still I think its the best solution I've found in POV-Ray at the moment and I like the fact that the physical model doesn't have to be tied to the camera location.

Carl

P.S. I should add that in the example above the box itself is outlined with cylinders along its edges and with spheres at the corners. The intersection of the box and the sphere is just that. It's an intersection of a slighly larger box and a slighly larger sphere. All these shapes just have one texture applied to their outside surface and they do cast shadows and have reflections on the game grid for example.on line abortion pill misoprostol dose abortion medical abortion pill online

matthew180 Wrote:Any thoughts on new names for the various parts?

Actually yes, but I want to hold off on this until after we've fixed the geometry. To introduce more name right now I think could create confusion.

Carl

wwwmwww Wrote:Well I do like the curve on mine more but I don't think I can say you are wrong. And I'm most certainly not wanting to give anyone the silent treatment. After taking last Friday and Monday off work I'm just now getting caught back up. I put in better then a 12 hour yesterday just to get caught up and my reward today is a pounding headache and a sore throat. I'll try to get some more said today but right now what I want to do is go home and get back in bed. I can't atleast not yet, there is a piece of equipment here still running at the moment that I'm babysitting as the person that usually operates this tool just went home sick. Believe me I'd much rather be rendering light cycles but you do what you have to do. Carl

You can say I'm wrong, it's the only way we will get it right.

The whole "silent treatment" thing was just me joking around. I saw the post that you were out of town, and by all means go home and rest! Otherwise you'll get delusional and try to tell me a lightcycle is square or something like that...

I did work out a shape for both the single object lightjet and transmission, check out the new construction and assembly pages on my site (which I see you have done since you have posted while I was writing this.)

wwwmwww Wrote:To be honest I think your "basic canopy" is my "completed canopy". In short I don't think you need to cut a hole for the windshield. Just make the windshild from a cylinder and cones with a slightly bigger radious and place it on top of the canopy. The difference in radious can be very small, so small in fact that you'll never see it, it just needs to be big enough to avoid coincident surfaces.

That is true, but I like to keep things grid snapped, which I have done with all but a few parts. Also, if for some reason I want to render the windshield as transparent, I'll need the hole inside. Last, I would be down by 1 primitive and that would make me below the 54 part count...

After I change my wheel-base and canopy cylinder I think I'll be happy, unless you see something in my construction or assmebly that needs correcting. I'm going to make some dimmensions for construction so anyone could make an accurate model in just about any program.

After I change my wheel-base our models will be almost identical with only very small differences in things like the exact shape of the lightjets, transmission, wings, rim size (the part of the front wheel between the black inner surface and the outer edge), etc.. But these pieces should fit pretty well into place in anyone's model if they get the major objects to the correct size and placement.

Then we can finalize the part names, and finally on to the fun stuff like animation and reviewing those "rules of engagment"!! Woo Hoo!

Matthew

Well, I concluded that the front of the canopy is slightly set back from the front edge of the front wheel (that's a lot of "fronts"...)  I didn't see any comments about it either way, so I hope I figured this one out correctly.  It was a total pain to modify the canopy since it affects almost every other aspect of the lightcycle!

I'm hoping I've completed my last bit of changes which included reducing my wheel base, modifying the canopy, shoulders, sides, etc. with a smaller radius, and tweaking the transmission yet again.  I think this version is pretty accurate and is almost identical to Carl's model in terms of scale and most geometry.

I started working on my jetwall material as well, reproducing the banding of the color fade of the jetwall.  I think I'm close enough to be happy with it, see below:

Screenshot of original lightcycle from DVD

My reproduction of that screenshot

Same model, different angle

Unless there is any other geometry issues, I think that about does it.  If anyone is still following this, I'd like to nail down some names for the parts as well as theoretical functions for those parts.

Matthew