It looks great, but I have two words for you: budgeting polygons.
Seriously, it's something to consider. Like, how many segments do your curved sections have and how many do they really need to look good?
Exactly. It pretty much all comes from cylinders. If you use smooth shading properly you can make it look seemless with very few segments and can do stuff in a mere fraction of the polys you'd consume otherwise.
Hi, anystar. Good work so far on what seems to be an extremely ambitious project. I'm interested in seeing how it turns out!
Question for you: Are you modeling the warp plasma intermix conduit based on "Mr. Scott's Guide to the Enterprise"? I say that because it seems you've duplicated some of the same inaccuracies that are in that book.
The single most noticeable one (for me, anyway) is the geometry of the "vanes" surrounding the conduit tubes. On the vertical intermix, the segments directly above and below the tall one are different from any of the others. Instead of starting narrow near the collars and wider in the center, the vanes start wide near the collars and are narrower in the center. On the horizontal intermix, only half of the vanes actually reach the collars -- every other one around each segment end a short distance before the collars.
There are some other differences, but I'll leave it to you to decide if you want to consider them or not. You may not even want to consider the ones I just mentioned.
I've been studying the TMP/TWOK-era engineering set in extreme detail because I'm modeling it as a virtual set for a fan-based production. Here's a link to my thread on this forum:
You can see the segment on the floor below the main level and how the vanes differ. The one above it is exactly the same. On the physical set they used for filming, the lower one was actually a bit shorter because of the forced-perspective tricks they used. However, running the measurements, I calculated that the two segments are the same length.
If you're interested, PM me and I'll send you the data that I gathered prior to modeling this set. Again, great work!
i considered using the concept version of the warp core. but decided in the end to just go 100% with the on-set version. i'll post some wireframes and comparison to my own reference later today. in the current version ive even dropped those metal plates sandwiching the fins just to save a few measly polies! :P they'll probably show back up eventually though.
well, while poking around and trying to discover what you were talking about i did notice one thing that had completely gone over my head. if you look at the "over the railing" shot in the references below, you'll see that its all one long section extending down to the bottom. i had completely missed that before.
one of the concessions i made between set/available space was to make the primary section of core there at main engineering much longer than the 9ish feet i estimated it to be, in order to create sections of equal length to form out the entire main core. as seen here.
so i fixed that and here's a comparison
i also reduced the number of sections leading back to the nacelle junctions to 9. i never got around to counting them before. the set and matte are foreshortened so it looks like on screen that they should be much shorter sections. but i'd rather go with 9 pairs of collars than 2x as many short sections.
im afraid im not sure what you were talking about with the intermix stage, so here's some wireframes of all 3 main components:
and of course this is what ive mainly been working from. i have some notes and sketches from the set that have helped a lot too, but im afraid i cant pass those out.
edit: also, with the reduction of sections (i also got around to finally changing the nacelle shunts from double collars to single) poly count on just he core and plasma conduits has dropped from 510k to 376k.
Silly me! When I was looking at your renders, I was concentrating on the image you posted on January 11th. If I had enlarged the thumbnails you posted after that, I would have seen that you had indeed included the "reverse" vanes.
We don't see what's above the upper one, but from the "looking down" view, it's apparent that there are no other segments of this type below the lower one. Because of that, I've made the presumption that the segments immediately above and below the one on the main level are the only two "reverse" segments. To rationalize the presence of these segments, I'm suggesting that these two segments are "venturi" segments to concentrate the flow prior to its being diverted into the horizontal tube.
I'm at work now, but when I get home I'll put together some of the info I have and post some quick renders of the model I have. Even my model has some inaccuracies, but they're pretty minor. Maybe one day I'll fix even those.
I know what you mean about poly counts! My model includes only what is seen on screen, and it's already at the half-million mark. In addition, the tubes have an animated, translucent texture so that the lights inside the tube can project the plasma stream onto the surrounding set. Plus, all the metal surfaces are textured as brushed aluminum. Also, there's all the transparent components. And I'm not even finished with the model -- I still need the lobby area with its console and the dilithium room off to the right. With oversampling on, each frame takes about an hour and a half to render on my system (quad-core Xeon w/12GB RAM). My goal is to generate about 30 seconds of footage for use in our production. It's been rendering since Christmas, and I've just now passed the 12 second mark (360 frames). Oh, to have a render farm...
I have decided that im not going to finish greebling the big upper complex of the dock for now, so that i can get back into the E by the beginning of next week, if not over the weekend.
but for now, today i worked up the boxes around the light panels, on the prop i believe they masked the transfomers for the flourescents inside the panels themselves. there isnt any real clear refernce to the front side of these things, so for now the front faces are rather simple. when i did the scale blueprints and then modeled my original version, i never really addressed these, i didnt even block them in on the 3d.
How about a photo of the studio model taken during its restoration by Ed Miarecki? It's a pretty clear reference. So are the rest of the photos of the restoration, which can be found here.
yes they are great, Adam over at the Trek bbs owns the drydock now, and was even able to get the 7th section that was removed when it was converted for generations. when i was doing my original mesh, i happened upon him over there and he was kind enough to share all those images with me, and to even took some measurements for me. his collection of studio models is very impressive, here's a link to his website http://www.ncc1701.us/18.html
Sorry to duplicate. From your post I had assumed (wrongly, it seems) that you had not seen these photos.
Yep, I swapped posts with him regarding the drydock on TrekBBS myself, and even referenced one of the other models in his collection, McKinley Station. His collection is, as Darth Vader might say, "Impressive", especially the drydock.
The beauty shot of the Enterprise in drydock has to be one of my favorite images from the film. Shortly after TMP originally came out, I did a 4-foot-by-3-foot oil painting of the Enterprise in drydock for an art show, which won first place and got several offers to buy. I still have it, though. But I digress...
Anyway, good work on the drydock! That's something I'm not willing to tackle in 3D at the moment, so a tip of my hat to you for braving both the drydock and the refit Enterprise at the same time!
i threw it all together, and even tossed in my version-6 saucer and nacelle, just to test out a few camera angles, and see what the poly count would be. even with all the interior stuff its only 1.75m polies. not bad for 2 separate models plus the interiors.
how about another round of "recognize this?" :cool:
i was going through old backups last week digging for an old cad file, and found a lot of old drawings from my college days. this one is something ive always wanted to model, its in my top-10 favorite designs of all time. i've just never gotten around to it.
i may give this one a try the next time i get frustrated with the E.
yea, the wings and any profiles were a dead giveaway, i spent an hour or so this morning comparing it to newer references, and im surprised how close i was originally. i never had any model pics or anything when it was originally drawn, i did it all from stills.
and i mean actually pausing the VHS tapes!!! haha
i've been going through a ton of old drawings like that recently, most of them were done before i even had an autocad liscense, created in a cheap little cad app called Deltacad. i just uploaded the orthos of my old NX-01 to the downloads section. I may start adding more plans if they prove usefull.
All i had time to get to today was finishing the spine and neck, and a little more on the torpedo area.
a few smoothing issues to work with, but its starting to look nice.
added the airlock in too, i can fudge its location with steps, but for now i'll wait on the portholes untill im sure about finished floor placements.
I was looking through IRML's Sovereign thread the other day, and it made me want to work on mine a little, i ended up just scrapping what i had on the saucer and made this today
im modeling it as the First Contact / Insurrection version, i assumed all those stretched out shapes were going to give me fits, but it was all done in about 7 hours. I'm sure the poly count will go up exponentially once i start to recess the windows, but the entire upper sauces came in at under 15k (i think my Refit was about 250k for the same area...)
ive been back to work on the refit this week, not much new to post on it yet. however i was hoping some of you could help me with the E-E. i have re-modeled the saucer a couple times now, i had the proportions all wrong, thanks to using nemesis refs, but sized with FC numbers.
without going into which part im having trouble getting to fit nicely, lets just say im having trouble with the superstructure.
i want to model the FC/Insurrection version, but im having trouble finding good references to the model. basically all i have are the magazine images and about 8 photos from the auction. im amazed i cant find anything more! does anyone have more???
i could have sworn i used to have a couple tiny scans of Sternbach's original model construction drawings, but now i cant finnd those either
not much new on the inside bits to show at the moment, but i did manage this past week to finish the hull neck and torpedo area. and even some more wires this time too. The poly counts in the wires include all of the interior, the actuall hull shown is only about 12.5k polies.
2 weeks ago my son had one of his friend over, and they were geeking out looking through my meshes. When they got to my Trek stuff after a few minutes he asked "So, where's the NEW Enterprise?".
We started talking about the new E, and the parts i really like in the design and then we got to the part where i really despise those nacells. I really liked the kid's response to it all; "Ok, sooooooo, why dont you just do better?!"
i got a kick out of this, and thought about all the versions i've seen of others doing a similar take, so i started digging and screencapping and studying, and even started modeling on the saucer.
but then you guessed it, my Cg-ADD kicked in again and i got sidetracked! HAH
as i worked i had been thinking about all the different nacelle designs that have been on the E's, and started playing around with all the one's i'd already modeled trying to make a unified design that would integrate aspects of each into a single nacelle.
then i started going even farther, and well this is the result, i started a whole new ship trying to create an all new heavy-cruiser type incorporating my favorite aspects of all the E's from the Original all the way up to the '09 ship.
there isnt much detail yet, its just experimental still, but im starting to like it!
i got a good laugh out of this too, in the 2nd image, the 1701-e style cutout viewing gallery on the underside of the saucer, as a happy surprise, that is were 10-forward will be! :cool:
edit: almost forgot, its sized proportionately to the D and reboot ships, e.g. the saucer rim is 4 decks tall. this might change as i play with it though, its a little "chunky" when seen head-on, but on the other hand i like the happy accident where i ended up with the 10-forward bit.
Posts
Exactly. It pretty much all comes from cylinders. If you use smooth shading properly you can make it look seemless with very few segments and can do stuff in a mere fraction of the polys you'd consume otherwise.
Question for you: Are you modeling the warp plasma intermix conduit based on "Mr. Scott's Guide to the Enterprise"? I say that because it seems you've duplicated some of the same inaccuracies that are in that book.
The single most noticeable one (for me, anyway) is the geometry of the "vanes" surrounding the conduit tubes. On the vertical intermix, the segments directly above and below the tall one are different from any of the others. Instead of starting narrow near the collars and wider in the center, the vanes start wide near the collars and are narrower in the center. On the horizontal intermix, only half of the vanes actually reach the collars -- every other one around each segment end a short distance before the collars.
There are some other differences, but I'll leave it to you to decide if you want to consider them or not. You may not even want to consider the ones I just mentioned.
I've been studying the TMP/TWOK-era engineering set in extreme detail because I'm modeling it as a virtual set for a fan-based production. Here's a link to my thread on this forum:
http://www.scifi-meshes.com/forums/showthread.php?78440-Star-Trek-Movie-Era-Main-Engineering&prefixid=3d
You can see the segment on the floor below the main level and how the vanes differ. The one above it is exactly the same. On the physical set they used for filming, the lower one was actually a bit shorter because of the forced-perspective tricks they used. However, running the measurements, I calculated that the two segments are the same length.
If you're interested, PM me and I'll send you the data that I gathered prior to modeling this set. Again, great work!
i considered using the concept version of the warp core. but decided in the end to just go 100% with the on-set version. i'll post some wireframes and comparison to my own reference later today. in the current version ive even dropped those metal plates sandwiching the fins just to save a few measly polies! :P they'll probably show back up eventually though.
one of the concessions i made between set/available space was to make the primary section of core there at main engineering much longer than the 9ish feet i estimated it to be, in order to create sections of equal length to form out the entire main core. as seen here.
so i fixed that and here's a comparison
i also reduced the number of sections leading back to the nacelle junctions to 9. i never got around to counting them before. the set and matte are foreshortened so it looks like on screen that they should be much shorter sections. but i'd rather go with 9 pairs of collars than 2x as many short sections.
im afraid im not sure what you were talking about with the intermix stage, so here's some wireframes of all 3 main components:
and of course this is what ive mainly been working from. i have some notes and sketches from the set that have helped a lot too, but im afraid i cant pass those out.
edit: also, with the reduction of sections (i also got around to finally changing the nacelle shunts from double collars to single) poly count on just he core and plasma conduits has dropped from 510k to 376k.
We don't see what's above the upper one, but from the "looking down" view, it's apparent that there are no other segments of this type below the lower one. Because of that, I've made the presumption that the segments immediately above and below the one on the main level are the only two "reverse" segments. To rationalize the presence of these segments, I'm suggesting that these two segments are "venturi" segments to concentrate the flow prior to its being diverted into the horizontal tube.
I'm at work now, but when I get home I'll put together some of the info I have and post some quick renders of the model I have. Even my model has some inaccuracies, but they're pretty minor. Maybe one day I'll fix even those.
I know what you mean about poly counts! My model includes only what is seen on screen, and it's already at the half-million mark. In addition, the tubes have an animated, translucent texture so that the lights inside the tube can project the plasma stream onto the surrounding set. Plus, all the metal surfaces are textured as brushed aluminum. Also, there's all the transparent components. And I'm not even finished with the model -- I still need the lobby area with its console and the dilithium room off to the right. With oversampling on, each frame takes about an hour and a half to render on my system (quad-core Xeon w/12GB RAM). My goal is to generate about 30 seconds of footage for use in our production. It's been rendering since Christmas, and I've just now passed the 12 second mark (360 frames).
but for now, today i worked up the boxes around the light panels, on the prop i believe they masked the transfomers for the flourescents inside the panels themselves. there isnt any real clear refernce to the front side of these things, so for now the front faces are rather simple. when i did the scale blueprints and then modeled my original version, i never really addressed these, i didnt even block them in on the 3d.
Yep, I swapped posts with him regarding the drydock on TrekBBS myself, and even referenced one of the other models in his collection, McKinley Station. His collection is, as Darth Vader might say, "Impressive", especially the drydock.
The beauty shot of the Enterprise in drydock has to be one of my favorite images from the film. Shortly after TMP originally came out, I did a 4-foot-by-3-foot oil painting of the Enterprise in drydock for an art show, which won first place and got several offers to buy. I still have it, though. But I digress...
Anyway, good work on the drydock! That's something I'm not willing to tackle in 3D at the moment, so a tip of my hat to you for braving both the drydock and the refit Enterprise at the same time!
so for now its back to the big E, but before any more interiors i'll be finishing the outer surface of the Engineering hull and neck!
Various Work: U.S.S. Constellation - Matt Jefferies Concept Shuttle
i was going through old backups last week digging for an old cad file, and found a lot of old drawings from my college days. this one is something ive always wanted to model, its in my top-10 favorite designs of all time. i've just never gotten around to it.
i may give this one a try the next time i get frustrated with the E.
Various Work: U.S.S. Constellation - Matt Jefferies Concept Shuttle
yea, the wings and any profiles were a dead giveaway, i spent an hour or so this morning comparing it to newer references, and im surprised how close i was originally. i never had any model pics or anything when it was originally drawn, i did it all from stills.
and i mean actually pausing the VHS tapes!!! haha
here's a little more complete layout of it
Also, nice avatar. The Transformers rule.
Various Work: U.S.S. Constellation - Matt Jefferies Concept Shuttle
All i had time to get to today was finishing the spine and neck, and a little more on the torpedo area.
a few smoothing issues to work with, but its starting to look nice.
added the airlock in too, i can fudge its location with steps, but for now i'll wait on the portholes untill im sure about finished floor placements.
im modeling it as the First Contact / Insurrection version, i assumed all those stretched out shapes were going to give me fits, but it was all done in about 7 hours. I'm sure the poly count will go up exponentially once i start to recess the windows, but the entire upper sauces came in at under 15k (i think my Refit was about 250k for the same area...)
without going into which part im having trouble getting to fit nicely, lets just say im having trouble with the superstructure.
i want to model the FC/Insurrection version, but im having trouble finding good references to the model. basically all i have are the magazine images and about 8 photos from the auction. im amazed i cant find anything more! does anyone have more???
i could have sworn i used to have a couple tiny scans of Sternbach's original model construction drawings, but now i cant finnd those either
The mesh is very clean.
Al
We started talking about the new E, and the parts i really like in the design and then we got to the part where i really despise those nacells. I really liked the kid's response to it all; "Ok, sooooooo, why dont you just do better?!"
i got a kick out of this, and thought about all the versions i've seen of others doing a similar take, so i started digging and screencapping and studying, and even started modeling on the saucer.
but then you guessed it, my Cg-ADD kicked in again and i got sidetracked! HAH
as i worked i had been thinking about all the different nacelle designs that have been on the E's, and started playing around with all the one's i'd already modeled trying to make a unified design that would integrate aspects of each into a single nacelle.
then i started going even farther, and well this is the result, i started a whole new ship trying to create an all new heavy-cruiser type incorporating my favorite aspects of all the E's from the Original all the way up to the '09 ship.
there isnt much detail yet, its just experimental still, but im starting to like it!
i got a good laugh out of this too, in the 2nd image, the 1701-e style cutout viewing gallery on the underside of the saucer, as a happy surprise, that is were 10-forward will be! :cool:
edit: almost forgot, its sized proportionately to the D and reboot ships, e.g. the saucer rim is 4 decks tall. this might change as i play with it though, its a little "chunky" when seen head-on, but on the other hand i like the happy accident where i ended up with the 10-forward bit.