Most of the X-Wings people model are based on the 1/24" Hero models used for the film, and further, usually Red 5 (Luke's X-Wing). This model has begun its life the same way, albeit a lot of the little details on/inside the wings are things I never knew about until I spent a lot of time with (what's left of) Red 5 in the Lucasfilm Archives. I took gigs of photos and reference measurements last time I was up there, so if I wanted to, I could probably finish off a "definitive" Red 5 1/24th model, but that's not the purpose of this gig.
This gig requires a 1:1 X-wing render, so I'm simply starting with this level of 1/24th-scale detail, but then I'm going to have to up it significantly, and use creative license, etc., to create something that doesn't actually exist. Even the "full-scale" X-Wing setpiece from Star Wars was only 90% full-scale, and its level of detail was less than the 1/24th, and generally unimpressive. So this will be a good chance to see whether 30+ years of study of the aesthetic will guide me properly in preserving the characteristics of the 1/24th models, while increasing detail to make it suitable for a 1:1-style rendering/context.
Rhino is my favorite modeling application. I use it mainly for high-detail architectural models, and I love how flexible and easy to use it is. (That is, for a guy who uses AutoCAD 9 hours a day). Like Mike said, it plays really well with other applications as well.
Keep up the good work, sir! I can't wait to see how this project develops.
Just curious with the detail level being high being the ratio 1:1 how much ram would it take up in you application? Would you need to run a 64 bit app with more memory when doing a job like this?
"Need to"? I don't really know. I run 64-bit kind of whenever possible, but ultimately I think the memory needs would really only show up at rendertime. I think there's a low possibility of actually maxxing out the modeling app. Also, too, "level of detail" in this context might actually mean more about what kind of detail, versus how much.
For what it's worth, I have never successfully done what I'm trying to do here. I've wanted to do it for a long time - thought about attempting it on the TOS version of the Enterprise. But when you study real, say, jet body construction, or rocket body construction, a lot of their scale comes from the little dimples in the panels around all the rivets, or countersunk rivets. Body panels on jets and planes and boats are always sort of dimpled and wrinkled, almost like fabric that was frozen into position. And of course, the panels themselves are never totally straight or perfectly aligned. Also, we tend to WAY exaggerate the depth and width of panel lines to make them "read," but it also kills realism - real ships don't need to do that to make lines read, and I think part of it is because we "infer" the panel divisions from the dimpling of rivets, and other cues that we get from a surface at the contact points.
With something like the X-Wing, it is tricky, because we still expect the panel lines and patterns to be a certain way, so there has to be something to justify that, without it being unrealistic, and if you're going to add something, it can't steal from that primary read/expectation at all. I really don't know yet what my workflow is going to be, but I've started doing tests to find the "x-factor" for my workflow: what are the best ways to make this ship's scale believable, through manufacturing cues? I'm sure as hell not putting rivets all over it. Rivets read too prominently from too far away for that to work, I feel.
We'll see. In any case, I can absolutely tell you that the model will need a pass in Modo or something similar, to create the proper ripples/dimples in the surface. This one won't be a bunch of displacement map tricks.
yay, I'm going to follow that one closely. I'm very curious about what it's going to be used for. Looks like a good start, but considering your previous xwing(s), I know it's only a very early beginning
The original model is 1/24. If you try and blow that level of detail up to "actual" size, it falls apart. This model is being built as 1:1, like R2 was, only there is no actual 1:1 source for it, unlike R2. So I have to try and increase the detail, without losing the feel and look of the 1/24 version.
It would be interesting to look to some of the kit bashed parts (all of the junk in the R2 strip, for example) and their real world counterparts as a means to impart some of that extra detail that you're looking for. Obviously it's not always applicable or appropriate - the main plate in the R2 strip for example is from the deck of a ship, not exactly what you'd expect to find as part of the exterior of a spacecraft. Some of the other pieces (ie. Saturn rocket) might be worth a look.
Yes, I'd considered that. I have all the original kit parts modeled, and know where they come from. But I've found more often than not that once you begin detailing them like their 1:1 counterparts, a gun or a hubcap actually looks like a gun or a hubcap, and not just slightly abstract enough to pass for something else...
Posts
Most of the X-Wings people model are based on the 1/24" Hero models used for the film, and further, usually Red 5 (Luke's X-Wing). This model has begun its life the same way, albeit a lot of the little details on/inside the wings are things I never knew about until I spent a lot of time with (what's left of) Red 5 in the Lucasfilm Archives. I took gigs of photos and reference measurements last time I was up there, so if I wanted to, I could probably finish off a "definitive" Red 5 1/24th model, but that's not the purpose of this gig.
This gig requires a 1:1 X-wing render, so I'm simply starting with this level of 1/24th-scale detail, but then I'm going to have to up it significantly, and use creative license, etc., to create something that doesn't actually exist. Even the "full-scale" X-Wing setpiece from Star Wars was only 90% full-scale, and its level of detail was less than the 1/24th, and generally unimpressive. So this will be a good chance to see whether 30+ years of study of the aesthetic will guide me properly in preserving the characteristics of the 1/24th models, while increasing detail to make it suitable for a 1:1-style rendering/context.
_Mike
Lucas isnt planning more new scenes for the blu-ray releases is he
_Mike
heh, yeah ... I know what you mean. Mine is on my "overwork todo list" too
Chris
PS: your working in Maya, right?
_Mike
ahh, I see. Rhino is a NURBS modeller as far as I know. Does it export good to other apps? I'm thinking about to test Rhino ...
_Mike
Rhino is my favorite modeling application. I use it mainly for high-detail architectural models, and I love how flexible and easy to use it is. (That is, for a guy who uses AutoCAD 9 hours a day). Like Mike said, it plays really well with other applications as well.
Keep up the good work, sir! I can't wait to see how this project develops.
(maybe you can post some of the LF Archive photos of the prop?)
For what it's worth, I have never successfully done what I'm trying to do here. I've wanted to do it for a long time - thought about attempting it on the TOS version of the Enterprise. But when you study real, say, jet body construction, or rocket body construction, a lot of their scale comes from the little dimples in the panels around all the rivets, or countersunk rivets. Body panels on jets and planes and boats are always sort of dimpled and wrinkled, almost like fabric that was frozen into position. And of course, the panels themselves are never totally straight or perfectly aligned. Also, we tend to WAY exaggerate the depth and width of panel lines to make them "read," but it also kills realism - real ships don't need to do that to make lines read, and I think part of it is because we "infer" the panel divisions from the dimpling of rivets, and other cues that we get from a surface at the contact points.
With something like the X-Wing, it is tricky, because we still expect the panel lines and patterns to be a certain way, so there has to be something to justify that, without it being unrealistic, and if you're going to add something, it can't steal from that primary read/expectation at all. I really don't know yet what my workflow is going to be, but I've started doing tests to find the "x-factor" for my workflow: what are the best ways to make this ship's scale believable, through manufacturing cues? I'm sure as hell not putting rivets all over it. Rivets read too prominently from too far away for that to work, I feel.
We'll see. In any case, I can absolutely tell you that the model will need a pass in Modo or something similar, to create the proper ripples/dimples in the surface. This one won't be a bunch of displacement map tricks.
_Mike
Good luck on such a good (and dare I say) Epic task. Realy sounds like a massive undertaking..but you never disappoint.
_Mike
_Mike
_Mike