matthew180 Wrote:Greetings, wwwmwww's post a few months ago has inspired me to revisit my lightcycle and tank models again as well. http://www.tron-sector.com/forums/default.aspx?do=top&id=329467

I'd like to get back to mine myself. Between work and home I've been flooded lately. Soon I hope...

matthew180 Wrote:I'm a bit of a perfectionist and would like to reproduce the lightcycles exactly as MAGI made them. I know this may not be totally possible, but maybe someone here can get me a little closer. * The info of 57 primitives is good to know, but does that include the primitives used in the boolean operations to cut the other primitives?

Yes, I believe so. I was able to do it with 54 POV-Ray primitives here:

http://www.tron-sector.com/forums/default.aspx?do=top&id=303566
and check some of my later threads I've tweeked this model already and I believe I have a more accurate one posted here elsewhere already.

matthew180 Wrote:* Does anyone know the list of CSG primitives that MAGI/Synthavision had available?

I'm not 100% sure but I've used:

sphere
cylinder
box
plane
cone
prism
torus

matthew180 Wrote:* On the bonus disc of the Anniversary release there is a clip that shows some of the actual lightcycle construction and boolean operations happening. Is there possibly a complete version of this video from beginning to end?

What you see is all that I'm aware of. Dave might be a good person to ask. For a while I've thought about trying to make my own version of that video but that's low on a long list of other things I don't have time to work on at the moment.

matthew180 Wrote:I really like wwwmwww's models (and the others I have seen here as well), but I would like to make my own just for fun (I'm sure you understand ;-) ), but I also want to be accurate to what MAGI created. Thanks, Matthew

Thanks. I too have had alot of fun making these models so I understand. Have fun and share the images you create.

Carl

*Races in with fake goatee only to discover Todd is already there* NOOOOOoooooooooo! I can't steal Todd's greeting now! Oh well! Next time, Todd, I'll get you and your little greeting too! *Throws a corrupted primitive at Matthew180*
~Kamui.EXE

TheReelTodd Wrote:Greetings, matthew180 and welcome to TRON-Sector! You've got light cycle questions? Carl's got answers! That's wwwmwww, who's already responded I see.

Thank you for the welcome. I've been to the site several times in the last few years, but it seems that it has really changed a lot (for the better) since the last time I was here.

TheReelTodd Wrote: I'm hoping we get some more light cycle images and in-depth discussion about them. I love the light cycle imagery threads! I may not understand all the math, but I dig all the cool imagery and never get tired of light cycles! Once you start rendering your light cycles (or any other TRON imagery), please submit them via the member upload or just post them in the thread. They'll get archived in your own TRON-Sector fan art section in the image gallery.

I'll do my best to add all I can and know about the lightcycles. I was 12 when the movie hit the theaters and it made an ever-lasting impression on me, particularly the lightcycles, tanks, and recognizers (in that order.) I think the CG and animation still holds up today! I love the perfect computer generated look of the whole movie. Modern CG can keep their real-life looking organic charaters and effects, personally I like TRON. It was kind of like the first STAR WARS (episode IV, 1977), dirty and purposful, not all the junk looking CG and bad writing they use today.

TheReelTodd Wrote: A couple of threads you may want to hit are: What's with your name? And... Don't forget to PING in every time you have yourself Another dose of TRON I think you'll enjoy your time here at The Sector

I'm not sure what "PING in" is, but I'll probably figure it out soon enough. :-) Thanks again!

Matthew

wwwmwww Wrote: Yes... you do need a torus. May I ask what software you are using. Aside from POV-Ray I'm not aware of many ray tracers that still use CSGs.

I use a program called Realsoft 3D http://realsoft.fi. It was originally called Real|3D and started life on the Amiga which is what I had when I bought it back in 1993. They ported to Windows/Unix when the Amiga died, so I've stuck with it. It's not really well known, and the development team is small, but it has some very unique features and is very capable (plus I'm used to it... ;-) )

In Realsoft CSG's are called "analytics" and have been supported from the first versions. Also very well supported are boolean operations, so it's perfect for recreating a lightcycle. The only CSG they don't have is the torus, so I'll have to make one by sweeping a small circle around a larger circle to create a mesh.

wwwmwww Wrote:I script everything by hand. I like coding and I like having exact control over everything which you don't often get with point and click front ends. And again most of the modeling front ends that are out there are mesh based... I think. If you want to get into POV-Ray I'd certainly encourage you. Its a great programing language and I'm starting to get very tempted to share even more of my code. If there were others out there that could learn from it or use it that would be enough to push me over the edge.

Here, let me push a little... ;-) I'm actually a programmer too, and have this desire to have complete control like you do. However, I try to let go sometimes and use normal tools like everyone else. In the case of 3D, I figured a good modeling and animation program would be nice to have instead of trying to program or script everything. The nice thing about Realsoft is that it does provide very accurate control if you want, as well as a scripting language (currently JavaScript) for direct programming control.

I compiled POV-Ray on my FreeBSD box a few days ago after your first reply. I started breaking your model down to see the individual parts as you have them. The scripting seems very straight forward, so I'll probably have to start working with it more! ;-) So far playing with POV and your model has been pretty fun!

A couple of notes about your current model:

Rear fender (primitive 14): Currently you have it as a cylinder and use POV's "shear" macro to get it to the correct angle. However, I believe it is a cone which can be seen below. There are a few other shots I have where this can be seen clearly as well. But certainly, as you pointed out in a previous post, there seems to be several revisions of the lightcycles and we may never know which one made it as the final version.

Rear side cone (primitive 17): A very interesting way to create those, but again they look more like actual "cones" to me, maybe rotated and sunk into the side a little. I'm going to play with those at little and see what I can come up with.

I'm also trying to think of a way to avoid the torus. I wish I knew for sure what CSG's they had to work with. If they had the ability to create a cylinder with rounded edges, that may work. But I was looking closely at the front-on view in the image below, and it looks like the rounded canopy edges start to curve in a little before merging with the top-most part of the canopy. That could only be done with a torus...

wwwmwww Wrote:There may very well have been more of that footage at some time. If it still exists... and who may have it... I have no idea. Dave is DaveTRON. He's the most prolific poster in these parts and isn't too hard to find. He probably knows personally half the people you emailed a year ago.

Sweet. Well, let's hope he reads this and makes a few calls... [please!] :-)

wwwmwww Wrote:The<

Well, I finally got my lightcycle model done. Here is a link to some images:

http://digitalstratum.com/programming/tron
I didn't want to post them in the forum because they are 1024x768.

Whew, making that model took longer than I thought! I think I nailed most of the geometry though. The one thing that is so hard to tell is the rounded canopy shoulders. They look so different in almost every image of a lightcycle. It does not help that all the images from the movie and such are all rendered with perspective, so it is very hard to determine size relationships and placement. I also think there are several versions of the lightcycle with subtle differences. I did end up using 61 primitives though, but probably due to the way my software works. I'm going to see if I can manage to get the primitive count down (was it 54 or 57?), but it will be tough.

There are some interesting facets of the lightcycle that I'd like to share, but I need to prepare the images first, so that will be another post in the near future. I'm also going to start working on the assembly animation.

Matthew

By the way, you'll want to read Cueball's thread here:

http://www.tron-sector.com/forums/default.aspx?do=top&id=331276&pg=1
It looks like you have the same basic shape for the front tire that I started with. There was enough evidence presend in the above thread to convince me I was wrong.

Carl

matthew180 Wrote:First we'll start with another glass rendered image since you liked it so much. :-)

Nice!!! Are you sure the back wheel is touching the game grid? It looks like its hovering above the game grid by just a small amount.

matthew180 Wrote:Links to my new stuff which includes part reference, notes, and complete assembly details: http://digitalstratum.com/programming/tron_refhttp://digitalstratum.com/programming/tron_noteshttp://digitalstratum.com/programming/tron_assembly

WOW!!! There is enough stuff there to keep me "debating" for days. Give me a while to get some pictures together and I'll try to address each point you've made in your notes.

matthew180 Wrote:Cueball's stuff does look nice, especially the lighting and highlights, they are very close to the original!! I'll have to work more with my render settings and materials to try and achieve the same results since most modern software is designed for making photo realistic images.

I'm in 100% agreement with you there. I have the same problems/issues with POV-Ray.

matthew180 Wrote:As for the front tire geometry, I know it is correct (one sphere cut with two slightly smaller spheres) since the front tire is one of the parts shown being created in the "Beyond TRON" video that is part of the anniversary DVD.

Yes, that is correct. What I think you have wrong and what I had wrong initially is the shape of the front hub per your tron_ref page. Go back and do a critical reading of Cueball's thread and I think there is enough evidence there to convince you.

matthew180 Wrote:I know the front tire looks very different in many of the clips and images around, almost barrel-like sometimes, but I believe that is just the perspective and camera angle. I'm sticking with what I saw on the DVD as the final the proof of the geometry, unless someone can come up with more solid evidence.

I'm in 100% agreement with the shape of the part you've called front wheel on your tron_ref page. We are talking about different parts.

matthew180 Wrote:Anyway, let me know what you think.

Will do. I disagree with some of your notes but it will take me some time to find the frames or render the images I need to show you why I disagree. Granted I may be the one that is wrong but I'll try to get my view of things on the table for you to think about.

Great work by the way... I love these king of discussions.
Carl

matthew180 Wrote:As for the front hub, yes, I agree I could have it wrong! It is also one of those parts that is very difficult to determine. I'll re-read that post. What did you and/or queball determine it may be?

In effect its the inner surface of the two small spheres used to cut out the sides of the front tire. Look at the pov-ray code I have posted here:

http://www.tron-sector.com/forums/default.aspx?do=top&id=331276&pg=3
As that should spell out exactly how I'm currently making the front hub.

matthew180 Wrote:Yes, the lightcycles do hover slightly above the grid, I'll try to find some movie clips that support this.

I don't think they do but I'm curious to see what you have that may prove me wrong. What you may be looking at is the jet wall that extends under the rear tire. It has width and is nearly as wide as the rear tire itself. This white part of the jet wall sticks out to the side just where the rear tire contacts the game grid. I don't actually think there is a gap between the rear tire and the game grid itself.

Look at these 2 pics:



You can see a shadow on the top of the rear tire created by the over hang of the jet wall. Its this width that you also see where the rear tire touches the game grid.

matthew180 Wrote:If you can find images that support different geometry, I'd love to see them! I'm all for finding the "real" geometry of the lightcycle, and not at all worried that I may be wrong, so let's figure it out. If you can give me examples that will reduce my geometry by about 5 primitives, I'll be really happy! :-)

I'll sure try... In fact some of your notes I may be able to use to reduce the number of primitives in my model as well.

Carl

Now on to the third note...

Some images look as if the wings are higher up and may actually touch the shoulder torus. This shows that they are in fact set lower and intersect the side of the canopy and not the shoulder.

I agree the part you are calling a wing shouldn't contact the exposed part of the torus however I'm not sure how your image shows that to be the case. For it to be clear from that perspective the side of the canopy and the torus would have to be different colors as I used here for example.

http://www.tron-sector.com/forums/default.aspx?do=top&id=329467
Note the small blue strip between the "wing" and the yellow torus.

Also note that the wings are made from two ellipses, and not a single ellipse crossing both sides. This is proven by the fact that the cross-section of the wing is a circle, as can be seen in the rear view (see #8.) Also, the "Beyond TRON" lightcycle clip on the Anniversary DVD shows the wing being created using an ellipse for one side.

Here I disagree. I think that one ellipsoid you see on the Anniversary DVD created BOTH "wings".


Can you prove the cross-section is a perfect circle? And ellipse seems to fit quite nicely. Keep in mind the plane of this cut isn't verticle and there is perspective present in this image. Even if this were a perfect circle or even an ellipse for that mater it would get distored a little by that perspective. That said I can't prove that it isn't a circle either.


But one ellipsoid renders faster then two and in my opinion I think the shape of one looks more accurate. Seeing how rendering times were a big issue back then it makes more sense to me that they'd have used just one if they could have.

Carlorder abortion pill http://unclejohnsprojects.com/template/default.aspx?morning-after-pill-price where to buy abortion pillabortion pills online abortion pill online purchase cytotec abortion

Nice detective work! Well, if the front tire does not have a hub, then I'm down one primitive! :-) It just does not seem like it's just the cutting sphere surface though. There seems to be a hard edge between the wheel color and black. I suppose that could be shadow, but at times it appears to be symmetrical all the way around the wheel... Wouldn't it be nice to have a wireframe render of MAGI's geometry?!?! :-)

I'll drop my hub and see what the renders look like. Now, what about the rear wheel?

The ellipse used to make the wing on the DVD sure looked like it only went half way through, but your argument about keeping primitive count down is a good one. I suppose that lightcycle could have been work in progress as well.

I wonder is anyone on this board knows the guys who actually built the lightcycle model at MAGI?? Anyone?

I do see that line shaped shadow you are talking about, and I cannot tell what it is either. I'm firing up my DVD now to single-frame through the lightcycle scenes. :-) Also in your animated images, look at the area just under the wings where engine cone is... The white outlines look totally wrong.

I can't find the image where I saw the lightcycle slightly above the game grid, but I'll keep looking. Interesting notes about the jet wall though, it does seem to be almost at wide as the tire and cast a shadow on it.

The last time I watched the lightcycle sequences I also noticed that the wings are actually transparent like the windshield! The actors look through them and the torus shoulder passes just above the actors head on both sides.

Matthew