Modified TOS Enterprise

Aft end of the sec hull is pretty much done, not including the doors and some finishing details. And yes, I know there's a seam around the middle; I haven't rejoined the two halves together yet.

Current face count: 10,622

Thats coming along very nicely Vektor. Looks sweet Those smoothing problems always drive me nuts, ie in your previous pics the long poly/triangles... .

How have you been getting rid of those? I've never been able to deal with them in a satisfactory manner...

Owy I wish I did 3D . . .

Originally posted by JasonA
Those smoothing problems always drive me nuts, ie in your previous pics the long poly/triangles... How have you been getting rid of those? I've never been able to deal with them in a satisfactory manner...

Lemme tell ya, I spent most of a year when I was working on the Constitution figuring out exactly how the smooth shader works and how it interacts with different types of geometry. Those pesky smoothing errors just about drove me insane before I finally hit upon a few useful techniques for minimizing or eliminating them.

You have to understand that the smooth shader does what it does by analyzing the angles between a given face and every other face adjacent to it, then creating a blending effect to make them all appear like a smoothly curved surface. But what happens when there is no angle between two of those faces? Obviously, no shading will occur between those two faces, but the shading will still apply to all the other adjacent faces that are at an angle to them. In effect, you've created a "flat spot."

That's where most smoothing errors come from, from faces that wind up getting split or cut in such a way that edges and vertices are created on a flat plane instead of on the curved surface they are supposed to approximate. A long skiny triangle at one edge of such a face--or worse yet, a long skinny triangle in the middle of it--makes the smoothing error worse because the (nearly) parallel edges tend to reinforce each other and accentuate the flat spot.

So what's the solution? Avoid creating edges that are close together and nearly parallel (as in long, skinny triangles), especially when one or more of those edges are between co-planar faces. Take a look at this wireframe of my secondary hull for an example of how I minimized the long, skinny, co-planar faces:



See those extra edges I cut around the cove and the hood over the fantail? Those allowed me to "compress" all the extra triangles that got created by the boolean operations so their edges aren't so close to being parallel. It wouldn't be a problem if so many of them weren't co-planar, but that's what you get when an existing face gets carved up by segmenting one of its edges. The best you can do is try to shorten those triangles and open the angles between their edges as much as possible.

Now, having said all that, there are times when you want to take advantage of closely spaced, parallel edges and edges between co-planar faces. If you want to create a slightly soft edge like the ones I've been working to achieve on my secondary hull, parallel edges like the kind you get from a chamfer operation are ideal. To further control the shading, cut another edge into the faces adjacent to the chamfer. The closer together they all are, the sharper the edge will appear. You can get some pretty good results with reasonable face counts that way.

Hope all of that made some sense.

[Darrell Lawrence]

Originally posted by Dennis
I use boolean subtraction a lot. Spend a lot of time welding points and cleaning up extra polygons, too.

That's what I love about Lightwave... able to zoom right in on a spot and weld/merge points/unify polys and be able to clean up a mesh right there in one shot

Hey Vektor-

I just refined my method of doing windows based on some stuff I learned woing my flying sub. I figure since you have been so gracious as to share some of your methods I thought maybe I can help you and anyone else in return

(or maybe you already know this)

To easily do windows on the curvy parts of the hull. create a spline that approximates the size of your window. Rotate and move it so it sits just above where you want the window. and do a shape merge onto the hull.

That gives you the basic window on the hull.

To add a beveled edge to the window edge:
Go to edit mesh and make sure you have all the window facets selected. Hit lock selection. If you want a beveled edge, go to the extrude control and enter zero and hit enter.then hit bevel and shrink your selected facets slightly. Then go to the top view and move the selection slightly inwards.

now to make the glass:
Add another edit mesh modifier temporarily. While the facets are still selected hit detach. And give the window a pure white color, 100% illumination, and an opacity of something like 10%.

Now to create the thickness of the hull:
Go back to your hull mesh, remove the edit mesh modifier you just made. You should have your original selection back. hit extrude with a zero value again. Go to the top view and move your selection backwards into the hull.

At this point you still have a set of facets still selected, so make it glow with an appropriate texture settings.

Vector: I'm lagging behind the curve a bit compared to everyone else, I think, re: the cutouts one the left/right side of the secondary hull. Everything you've done so far looks just great, and I'm as thankful as anyone else here that you're sharing your techniques with the rest of us. I still have questions re: the sec hull left/right cutouts, however, and I'd truly appreciate your help.

I 've re-lathed my sec hull and created a new "plug" to cut out the inserts on the left/right/bottom of the hull. I do a boolean subtraction, an edit mesh modifier, select the long-ish insert poly that runs on the hull's x-axis, 0-height extrude, neg-bevel and repeat. After target welding the newly created verts in front and back, I'm left with an arrowhead/chevron-shaped aft end of the insert with three rows of polys. Things are looking good at this point.

I'm guessing I'm now I need to make some choices re: smoothing groups & blending those new polys in with the rest of the hull (prior to using mesh smooth modifier). I select the new polys toward the end of the insert and click on the same smoothing group number as the polys which surround them on the hull. Not bad, but not great either. I apply the mesh smooth modifier and it goes a long way towards smoothing out the polys at the end of the insert, but I'm left with two really ugly smoothing errors at the very ends of the cutout itself (on the top/bottome of the insert). What am I missing here?

Do I need to cut some extra edges somewhere? Create some extra faces or polys at/near the end of the cutout? Something else? I've played with smoothing groups till I'm almost blue in the face (not my best color) and I'm no closer to achieving your end result than I was when I started reading about your technique. No matter what I do, I keep getting the same smoothing errors. I'd really appreciate your help because I'm getting bogged down here.

Thanks!

Thomas7g: That was a great tip Re: the windows. I'll give that a try later on when I get to that part.

Forgot to ask something... the "boxes" at the front of the insert, are you doing a boolean union on those before or after you've smoothed out teh polys at the end of the insert? I noticed, in the images that accompany your description of your work, that there are cuts running from top to bottom on the inside of the insert and the box that stands out in front. Have you cut/sliced those before/after doing the boolean union? Is that all part of getting the smoothing right for this part of the mesh?

Thanks again!

Originally posted by thomas7g
I just refined my method of doing windows based on some stuff I learned woing my flying sub. I figure since you have been so gracious as to share some of your methods I thought maybe I can help you and anyone else in return

Actually, what you described is almost the exact method I used to create the windows on my USS Constitution model, with a couple of insignificant differences.

Originally posted by Haggus
I 've re-lathed my sec hull and created a new "plug" to cut out the inserts on the left/right/bottom of the hull. I do a boolean subtraction, an edit mesh modifier, select the long-ish insert poly that runs on the hull's x-axis, 0-height extrude, neg-bevel and repeat. After target welding the newly created verts in front and back, I'm left with an arrowhead/chevron-shaped aft end of the insert with three rows of polys. Things are looking good at this point.

I'm guessing I'm now I need to make some choices re: smoothing groups & blending those new polys in with the rest of the hull (prior to using mesh smooth modifier). I select the new polys toward the end of the insert and click on the same smoothing group number as the polys which surround them on the hull. Not bad, but not great either. I apply the mesh smooth modifier and it goes a long way towards smoothing out the polys at the end of the insert, but I'm left with two really ugly smoothing errors at the very ends of the cutout itself (on the top/bottome of the insert). What am I missing here?

Your method of creating the inset is a little different from what I used but close enough. That part at the back end where the flat face blends into the curved hull is a pain in the ass to say the least. Among other things, it's not a true half-oval as suggested in the Sinclair drawings and I spent a lot of fruitless effort trying to make it into one.

My solution to the smoothing problem was to insert several extra vertices and adjust their positions by eye using a combination of Local and View translation. I also spread each row of vertices out a little to dull the "edge." I didn't use mesh smoothing at all, partly because of the same problem you mentioned with the smoothing errors at the back ends of the cutout, but mostly because Mesh Smooth is just horribly inefficient for this type of modeling. If you go back and look at the difference between my early Meshsmoothed attempts and the most recent version, you'll see that I accomplished very comparable results with about 1/6 the number of faces.

I've played with smoothing groups till I'm almost blue in the face (not my best color) and I'm no closer to achieving your end result than I was when I started reading about your technique. No matter what I do, I keep getting the same smoothing errors. I'd really appreciate your help because I'm getting bogged down here.

Well, I only used one smoothing group for the entire secondary hull as it currently exists. Moreover, I did not use a threshold angle for the smoothing. In other words, every angle and surface of my secondary hull is smoothed. To achieve a sharp-edged look here and there, I simply chamfered the edges and then zero-height extruded/inward beveled the groups of faces on either side to create what I call a "break edge," one that constrains the smoothing effect. In a couple of cases, I didn't even do the chamfer first. It depends on how small the detail is and how sharp the edge is supposed to appear.

Maybe later tonight I'll post a wireframe of those boxes and their cutouts to illustrate some of what I just said.

Originally posted by Haggus
Forgot to ask something... the "boxes" at the front of the insert, are you doing a boolean union on those before or after you've smoothed out teh polys at the end of the insert? I noticed, in the images that accompany your description of your work, that there are cuts running from top to bottom on the inside of the insert and the box that stands out in front. Have you cut/sliced those before/after doing the boolean union? Is that all part of getting the smoothing right for this part of the mesh?

I did the boolean union to finalize the basic shape before I ever did anything at the back end of the insert or anywhere else.

As for the vertical edges on the flat face of the inset, I mesh edited those in after the boolean operations were finished in order to simplify the mesh. The first boolean, the subtraction, is pretty clean, but the second one is virtually guaranteed to produce actual mesh errors and all kinds of needlessly complex geometry. The first thing I did was go back clean it all up, and the vertical edges were part of that. It was also necessary to prepare certain parts of the mesh for edge chamfering at a later stage. Without dividing that flat face up along the major lines of the secondary hull, the chamfering tends to get messed up.

After I finish cutting in the windows on one side of the hull and the major modeling is complete, I will probably make an optimization pass to eliminate as many of the unnecessary polygons as possible. I figure I can get rid of at least a few hundred of them. After that, I'll slice off the other half of the hull, mirror what's left and voilla! I'll have myself a complete secondary hull.

I did the same thing for the inset boxes themselves, by the way. I didn't attempt to model all three; I just modeled one and then mirrored/arrayed it into the other two positions when it was finished.

Just reading over that there method for solving smoth errors. I don't get it.

I mean, I get it, but I just... do't get it, if you see what I mean. I can see what you did, and it does work, but I want to understand why it works... see what I mean?

Originally posted by Proximo
Just reading over that there method for solving smoth errors. I don't get it. I mean, I get it, but I just... do't get it, if you see what I mean. I can see what you did, and it does work, but I want to understand why it works... see what I mean?

Heheh, yeah I see what you mean. It took me about six months of trial and error and experimentation before I felt that I truly understood what was going on with the whole smooth shading process.

Let's try a simple illustration: Suppose you want to make a very simple spaceship in the shape of a sphere using a bare minimum of polygons. You start with a sphere object made of only 12 segments. Yeah, I know, that's not enough to make the sphere look realistic, but bear with me. Although the sphere's "horizon" is obviously segmented, the surface will appear perfectly smooth thanks to the wonders of Garoud smooth shading.

Now, pay attention, this is the important part. The sphere we've created is comprised of zero-dimensional vertices connected by one-dimensional edges surrounding two-dimensional faces, all of which form a three-dimensional sphere. The key thing to understand is that only the vertices actually lie on the surface of the sphere; the edges and faces are merely approximations of the true spherical surface. Without the smooth shading, our 12 segment sphere looks more like a cheap disco ball because it's surface is actually faceted not smooth. The same is true for all supposedly "smooth" surfaces represented by a computer model (yes, I know about NURBS and such, but even they are converted to mesh objects at render time).

Once again, only the vertices lie on the true surface of the sphere. Smooth shading works by looking at the "normal" of each vertex, which is really just an average of the relative angles of all the faces it's connected to, then blending the illumination values of those vertices across the faces between them. This has the effect of making the flat surfaces between the vertices appear curved, thus eliminating the facets, but it's really just a convenient illusion and it does have its limits.

Now, having created the basic sphere, suppose you want to cut a small, square window into it. A simple shapemerge will do the trick nicely, but you'll find when you go to render it that the newly created window is surrounded by ugly shading errors. Once the faces inside the window are deleted, it will look a bit like someone tried to cut the window out with a pair of dull tin snips. Why does this happen? Because the vertices defining the window do not lie on the true surface of the sphere.

In fact, given how low-poly our sphere is, the window was probably created in the middle of a much larger face, which is actually flat, not curved. That's why the vertices aren't on the surface of the sphere but rather slightly below it. If the window wound up overlapping an edge then the shading errors could actually be even worse, for reasons I'll get to shortly.

Remember, smooth shading works by blending the illumination values of vertices that lie on the true surface of a curved object, like a sphere. It works best when the curvature of the surface is fairly regular and the vertices are fairly evenly spaced. It's also important that the angles between adjacent surfaces be fairly consistent. The window we created in the surface of our spherical spaceship just trashed every one of those principles.

More than likely, we now have a few long, skinny triangles connecting the sides of the window to the nearest vertex of the original sphere. The problem with long, skinny triangles is that they force the smooth shader to blend between one set of vertices that are close together and another set that are far apart, relatively speaking. This is where the smoothing error that looks like a warp or ripple comes from.

I should also mention that these long, skinny triangles only present a problem when they are coplanar to an adjacent face, such as what happens when an existing face gets carved up for some reason (like shapemerging a window). To be fully accurate, they still don't present a problem unless their vertices happen to disrupt the regular pattern of the surface. Unfortunately, such worst-case results are virtually inevitable in most boolean and shapemerge operations. I'll make a brief attempt to explain why this is true by eliminating one dimension from the spherical spaceship example.

Suppose you draw a circle on a piece of paper, but instead of making it perfectly round, you draw it as a series of short, straight segments. This is analagous to how the computer represents curved objects. Next, draw a little dot at the corner between each line segment. These are analagous to vertices. Now draw a little arrow leading off from each vertex, pointing straight out from the center of the circle. These are analagous to the vertex normals.

Okay, to illustrate where smoothing errors come from, draw a new dot in the middle of one of the line segments and a little arrow perpendicular to the segment. If you drew the dot precisely in the middle of a segment, you won't have any smoothing problems, but if you're significantly closer to one end or the other then you're going to have trouble. Why? Because the normal of the new vertex remains perpendicular to the original segment no matter where it's located along the length of the segment. Because the new vertex does not lie on the true edge of a circle, it's normal is not consistent with those on either side of it unless it's exactly halfway between them. The closer it is to one end or the other, the more inconsistent it is.

When the smooth shader does it's work, it blends smoothly from one vertex to the next, producing a 50% difference in the illumination values exactly halfway between. But consider what happens in our example when it encounters a new vertex with an inconsistent normal right next to a consistent one. In effect, it's forced to move the 50% point off-center. The shading between the two closer vertices gets compressed and the shdading between the two that are farther apart gets stretched. The result is a very noticeable smoothing error.

Now, if you can mentally extrapolate that example back into three-dimensions, you should be able to see why long, skinny triangles cause trouble when they are not a part of the regular pattern of a "curved" surface.

So what's the solution to all of this? Simply put, you need to do everything possible to locate all your vertices on the true curved surface of the object, or as close to it as you can manage. You also need to minimize long, skinny triangles that don't fit the overall pattern, usually by breaking and turning edges to make the triangles shorter and less skinny. You want the mesh density to blend smoothly, not abruptly.

I should also mention that the shading compression effect you get from closely spaced vertices and long, skinny triangles can be exploited to great effect. That's basically what I've done on purpose with my secondary hull to create all those slightly smoothed edges and corners. Every one of those edges actually consists of three or four parallel edges spaced closely together, with no angle on the outermost edges to constrain the smooth shading. It works wonderfully well when you understand how and why it actually works.

Well, I hadn't really intended to write a whole book here, but I had some time on my hands and it seemed like something worthwhile to talk about. Hopefully it will prove useful.

Originally posted by Vektor


Your method of creating the inset is a little different from what I used but close enough. That part at the back end where the flat face blends into the curved hull is a pain in the ass to say the least. Among other things, it's not a true half-oval as suggested in the Sinclair drawings and I spent a lot of fruitless effort trying to make it into one.

I think I see what you're saying re: how you did it. I'll have to play with it a bit more and see what happens when I have time. Yes, that rear area of the inset is QUITE the pain in the ass! If you take a look at footage from the series you may see that this area isn't perfectly smoothed at all. I think it appears that way (and, in my opinion, looks better that way), however, due to studio lighting, the way the model was painted, camera angles, etc. I have a few screen grabs that definitely show that area at the end of the insert is a partial ellipse/oval. Its somewhat smoothed into the rest of the hull, but one can tell its an oval shape there. I can post the screen grab to illustrate if you'd like to see it. (I'm just mentioning an observation here, not trying to start an "Its an oval/Its not an oval" flame war or anything )

My solution to the smoothing problem was to insert several extra vertices and adjust their positions by eye using a combination of Local and View translation. I also spread each row of vertices out a little to dull the "edge." I didn't use mesh smoothing at all, partly because of the same problem you mentioned with the smoothing errors at the back ends of the cutout, but mostly because Mesh Smooth is just horribly inefficient for this type of modeling. If you go back and look at the difference between my early Meshsmoothed attempts and the most recent version, you'll see that I accomplished very comparable results with about 1/6 the number of faces.

I learned after some trial and error (and after reading some of your extremely helpful posts here, by the way ) that moving the verts at the end of the insert (spreading them out, as you did) was a big help to dull the edge of the curve back there. This all has to be done until it just looks right and smoother before using Meshsmooth or any other method. After all of the steps I wrote about previously, and after moving the verts around to dull the edges of the ellipse at the rear of the insert, I used Meshsmooth with one smoothing iteration. That, I found, was sufficient to give me an adequate amount of smoothness at the rear of the three inserts. After that, it was a simple matter of boolean-unions for the front blocks. I did a few (very few) welds and finally had a mesh with no errors and around 7K polys. Dunno how many I'll have after I add the shuttle bay hood, opening, and cutout (in addition to the windows). Dunno about doing all of the chamfering around the front of the secondary hull, either, since I'm not sure I;ll be pulling that close to it with the camera at any time.

Well, I only used one smoothing group for the entire secondary hull as it currently exists. Moreover, I did not use a threshold angle for the smoothing. In other words, every angle and surface of my secondary hull is smoothed. To achieve a sharp-edged look here and there, I simply chamfered the edges and then zero-height extruded/inward beveled the groups of faces on either side to create what I call a "break edge," one that constrains the smoothing effect. In a couple of cases, I didn't even do the chamfer first. It depends on how small the detail is and how sharp the edge is supposed to appear.

That is quite an impressive feat, to be certain! I'm glad I saved my secondary hull once before I did any of the other work to it so I can try your method out, too. You've really given me a lot of food for thought with your posts, for which I'm very thankful!

Well this is no conversation for a 2D man. But I hope all gets fixed and we can see the updates! LOL! :thumbsup: :devil:

Progress!







Believe it or not, this is yet another scratch rebuild of the entire secondary hull, using yeat another technique for softening all those sharp edges just a tad. The last version wasn't bad, but the edge chamfering process tended to warp and distort a little bit, especially on the curved edges, so I went back to the drawing board and figured out a way to make it come out clean.

The current face count for everything you see here, including the modeled windows, is 14,051. Given the level of detail and how close you can get without any visible segmentation, I'd say that isn't too awful bad.

Unfortunately, that's probably as far as I'm going to get for the next few days. Just thought I should post at least one update before you all forgot about me completely.

Crazy.

I've never seen a TOS model detailed so beautifully. You'd think we were looking at a WIP for a the refit Ent. People who complain about the looks of this ship really need to take a second look.

Vector;
You are the MAN!
Seeing your latest update has left me reeling with envy and despair. I am now afraid to do
anything on my TOS Enterprise because all I can see are my failings and inadequacies. What I'm trying to say is:yikes: :yikes: :yikes: :yikes: I wish I could do what you are doing. Is there any chance at all of a forthcoming tutorial? Or, at any rate could you name subjects that I should study to produce similar results?

Originally posted by jingram_98
Seeing your latest update has left me reeling with envy and despair. I am now afraid to do
anything on my TOS Enterprise because all I can see are my failings and inadequacies.

Dude, you've got nothing to be envious about. Your own TOS Enterprise model is fantastic!

What I'm trying to say is:yikes: :yikes: :yikes: :yikes: I wish I could do what you are doing. Is there any chance at all of a forthcoming tutorial? Or, at any rate could you name subjects that I should study to produce similar results?

Well, I have given some thought off and on to making tutorials on various techniques I've discovered or refined, the kind of things that are really hard to find, like how smooth shading actually works and why. Problem is, I barely have enough free time to do the modeling work, let alone build tutorials.

For the time being, a simpler approach might be for you to ask specific questions, how did I do this or how did I do that? I'll try to answer all of them as best I can.

Wow . . . U kive ut (In Dave's typo language) "I love it"

Okay, I have made an important decision about this model. I have decided to go back to the original idea of a modified version of the TOS Enterprise instead of a precise recreation of it. The reason why is because other people have done the recreation about as well as it can possibly be done (Nova Class's Version 3 being a prime example) and there's not much point in me churning another one.

Instead, I'm going to build my own interpretation of the Enterprise with some of her details and dimensions subtly altered. As much as I love that old design, there are a few things about it that I feel could stand some improvement and that's what I'm going to do. I'm not talking about drastic changes, just a few tiny but significant ones.

Thus far I'm about 85% finished with the secondary hull, having gone back to incorporate a few of those modifications since the last images I posted. Here's what it looks like currently:



If you look closely, you can see that I have increased the maximum diameter of the secondary hull just slightly, enough to give it a little more of an outward bulge in the forward third or so. The original nearly-straight tapered cylinder looked too much like a 1970s cocktail mixer to me, so I gave it some more curvature. Not much, just an extra meter or two of diameter.

I also added a little bit of curve to the taper of the deflector cowl. Again, not much, just a little. The ring just behind the cowl now has a slight lip on the front edge as well.

I never cared for the way the deflector dish just kind of stuck out there, especially on pilot version of the filming model, so I moved it back toward the cowl by two or three meters and extended the inner cowl forward a little bit. The idea was to unify the dish into the main hull a little bit more.

Here's another image:



Here you can clearly see that I've gotten creative with those grills on the sides of the "boxes" surrounding the deflector cowl. The same basic look is there, but I've extended the grill to the deflector cowl and offset them in a slightly curved line. I rather like the way this looks but I haven't decided yet whether to keep them this way or try something else.

You can also see from this image that I went with a pretty high level of detail so really nice close-ups are possible. The secondary hull has 96 segments around its circumference and twenty or more along its length. Even at that, I've managed to model everything you see here in something less than 16,000 faces.

And now the windows:



Yes indeed, all of those windows have rounded corners. I've hit on a new technique for creating my windows that is a lot less work than the old method and generally produces better results. Smoothing errors are virtually undetectable now.

Eventually, I'll do the same textured box interior behind these windows that I did on the Constitution.

Front view:



Here you can see another slight change I've made. The "boxes" surrounding the deflector cowl now have a slight taper toward the front instead of being perfectly parallel. This is partly to accommodate the increased diameter of the secondary hull and partly because I just liked the look. You'll also note I've added a grill to the bottom box as well as the sides.

Rear view:



Here you can see the alignment of the windows, which is more precise and regular than on the filming model. The basic pattern is the same, but I've lined some of the rows up better to make them look less scattered in the obliquely angled views like this one.

If you're observant, you might also notice that I have smoothed the sides of the fantail into the back end of the hull, which I prefer over the sharp corner on the filming model.

As always, comments and criticisms are welcome. I'm especially interested to hear what eveyone thinks of my modifications.

Oh my god.

I like what I'm seeing! A lot. Your changes are not obvious (to me, anyway), but I can imagine that they will have an effect on the final model.

More, please.

Well, I'll say this right now, I think the appraoch you are taking is very nice! I love what you did to the grills on the side boxes near the deflector. That is so 50s scifi and I love it! You should definitely keep that
Rounded windows, very nice! Everything else is so subtle that if you hadn't mentioned them I couldn't have seen it, though.

One thing I did notice, because I'm changing it on my model, is the proper look for the deflector dish. Engraved outer rings and a raised inner one. That's one of those speculative things, that I'm only now becoming convinced of, but after speaking with a few people at length about it, I'm convinced that that's the way it is.

Are you keeping this the Enterprise or will you be naming her differently as well?

Anyway, all very nice. Thanks for your compliments as well. I'd be proud to fly my E along side yours. :thumbsup:

Vector:
It's so goog to see some more of your work here again. I like the approach you you are taking with the new version. I've stalled on my Enterprise because of those smoothing errors. Would you mind describing the steps you take from the point of lathing the initial cylinder on?

Incredibly perfect Bud...need more updates

Originally posted by Nova Class
Well, I'll say this right now, I think the appraoch you are taking is very nice! I love what you did to the grills on the side boxes near the deflector. That is so 50s scifi and I love it! You should definitely keep that

Thanks, Nova Class.

Yeah, I wasn't sure about it at first but it's been growing on me. I will probably let them stay in their present form.

Rounded windows, very nice! Everything else is so subtle that if you hadn't mentioned them I couldn't have seen it, though.

Well, I do like the basic design and didn't want to change it overtly.

One thing I did notice, because I'm changing it on my model, is the proper look for the deflector dish.

I'd love to take credit for it but the dish profile is modeled directly off the Sinclaire schematics. I have to say I don't like the spindle, though; looks too much like one of the legs off my grandmother's dining room table. That's probably gonna get "modified" as well.

Are you keeping this the Enterprise or will you be naming her differently as well?

Good question. I haven't decided yet. This could wind up to be the USS Endeavor I promised to build so long ago. If I actually release the mesh, it almost certainly will be. I have no plans to release an Enterprise mesh.

Anyway, all very nice. Thanks for your compliments as well. I'd be proud to fly my E along side yours. :thumbsup:

The compliments come easily. You're one of the best and fastest modelers I have ever seen and I would love to be ranked in your class one day.

Originally posted by jingram_98
Vector:
It's so goog to see some more of your work here again. I like the approach you you are taking with the new version. I've stalled on my Enterprise because of those smoothing errors. Would you mind describing the steps you take from the point of lathing the initial cylinder on?

Phew! That's a tall order, I'm afraid. If smoothing errors are your biggest problem, I can probably offer a whole slew of suggestions about how to deal with them. I consider myself somewhat of an expert on the subject at this point.

Tell you what, why don't you post some wireframe screen captures or renderings in your own Enterprise thread with the problem areas highlighted and I'll see what help I can offer. The reason I'm asking for wireframes is because that's the only way to see what's really going on as far as what is probably causing your smoothing errors.

I think I will also post some wireframes of my own this evening to show how some of the techniques I use actually work at the mesh level.

Originally posted by EBOLII
Incredibly perfect Bud...need more updates

Workin' on it.

I thought some wireframe screen caps might help to illustrate some of my modeling techniques:

blah blah...little lonely are you?.....nobody to talk to huh?

so Um....ANYthing else you would like to say??????

no no it's okay us the little people have nothin else to do but read your "novel"...ooop I mean post....

Actually just razin ya....chill

On a serious note...I think it is a seriously perfect mesh thus far..What are your intensions for the Nacelles?..I personally like the 2nd season end caps that have the same deflector type tip and the hole exhausts

BUT that is me.....give it a thought or 2