Yet Another TOS Battlestar WIP

[Senmut]

One question...aren't there alot of critical parts that are exposed? Not a criticism, as we see the same thing in the original show. But it looks like alot of stuff is way to exposed to enemy fire.

[Gunther38Capric]

Bows to you ...

Amazing job, i can only dream of giving such detail to a ship.
My system would crash in no time lol.

The pixels polygons must be murder on a proccesor .
But again amazing detail keep it up.

[Gunther38Capric]

The systems are exposed yes .
But we see the Battlestar as alone depicted in the series or as flag ships.

Look at carriers in the us navy .
they are supported by a whole fleat of destroyers , refuel ships and so on.
Alone a carrier is vulnarable to certain attacks, but with its support vessels a difficult target to hit.

Sadly in the original series we dind get to see that (Afterall it was about the sole surviving battlestar)

Back then budget might have bin at play aswell.

[Senmut]

Good answer. That was my take as well. Carriers are designed to operate in battle groups, not as lone wolves on the prowl.

Good answer. That was my take as well. Carriers are designed to
operate in battle groups, not as lone wolves on the prowl.

My thoughts as well, when i was running my website and writing my, I story came up with some sort of organization for ships and support ships

Battlestar Section

Unit - No. of Capital ships/Support ships - Sub units - Unit leader rank

Element - 1/3 - Cdr
Squadron - 4/12 - 4 Elements - Com
Battle fleet - 8/24 - 2 Squadrons - Vc Ad
Fleet - 16/48 - 2 Battle fleets - Ad

[gmd3d]

My view on the Battlestars is this.....

Yes its a Carrier and its best defence is the Viper craft.

But its also a Battleship... able to go toe to toe with the Cylon Base ship.

once the Viper squadrons have launched the entrance to the bay are closed, (never seen in the show alas)

But I have seen images from a BSG site showing the prototype hangar bay entrance with the deck raised to block the bay.

Then the Battlestar is a battle ship.

I think viewing them only as a carrier type using modern navy arrangement is under playing the ability's of them.

I think they are well able to go it alone when required and it Galactica´s success keeping the slow moving ragtags is an example of that..

also

the lose of the other Battlestars is due to surprise and bad command under the president etc.


So if required can operate as a lone unit very easily like a Imperial star destroyer. but is more effective in small battle groups with the battlestar
as it focus.

anyway they are my thoughts on it

ps I would like to know to if this project is still on

[maudib]

Update 9-18-12. Started on the thruster section.

[gmd3d]

most impressive

[Matador]

Awesome detail... What's the secret for creating those cool vents? The look so cool.

[martok2112]

Very very cool looking detail there!

[Punisher454]

I always look forward to seeing your updates. Really fantastic work.

[maudib]

Update 9-19-12. Hi all! Nice to here from everyone again. Been busy making a guitar last year, and now that it's done...

Another 10 fun-filled hours of work and another greeble/nurnie is done. I can see why not many people have attempted to model this beast! Every nurnie/greeble is a 3D model in its own right! And there are 1000's of them on the big-G.



After working on this Battlestar over many hours, I can see patterns where the model maker re-used the same plastic kit parts over and over. This makes 3D modelling enormously easier rather than trying to make meshes from scratch each time. I made 3 grill meshes just once at the beginning, and modify them in various ways and reuse them over and over, just like the physical modellers did on the studio scale model.

For example here is a pic of the top engine greeble/nurnie



And this is the bottom engine version:



The top version can be made by modifying the bottom version by cutting away the top and sticking a bunch of polys in the hole plus a few other shapes changed out here and there. And engine section looks fantastically detailed to the casual eye, but in reality, just a few number of parts reused and modified over and over.



This piece is 206,000 polys, not bad by itself, but clone it 10times to cover both sides and it becomes 2,060,000 polys! The previous mesh was 194,000 polys, but cloned 26 times for both sides!

Just 4 big nurnies make up 40 percent of the engine section after cloning them many times. The poly count is unbelievable. This section is subject to close animation renders so that's where I put the polys. The top and bottom of the ship is rarely filmed close up, so I have to rethink my modelling strategy in those areas or this could easily end being a 40 million poly project.

[martok2112]

Yeah, your battlestar is going to be ....well....countless polys. (Oh, sure, the computer can count, but until then, all I can do is boggle at just how many polys your complete ship is going to be.

The entirety of my current rendition of a battlestar is just under 3 million faces. Now, I don't know what that is in polys, but I can tell you it will be the merest fraction (if that) of what you're putting together.. I eagerly anticipate your finished product.

Bi La Kaifa!

[Titon]

Lightwave will choke if you push 40 million poly's in a mesh. Not to mention animating it will be almost impossible. That Rico tank you just did should not be even be close to that high of a number. My entire engine section as it sit's now is 623,000 but is nearly 85% complete.

You have to be realistic about this beast in some aspects. Cut corners where you can get away with it. The poly's you put in the opening of the Rico tank for instance were not there on the original model. You do not need them there and i would take them out and follow the original. In cg you would have to push in really really close for anyone to notice there is a difference between the original model and the cg version.

What i would suggest is make concessions on the base mesh and if you want to get really close on a certain area build it seperately for close up shot's. Otherwise even opengl will give you problems.

Other than a few nit picks good work!

[maudib]

@Titon...yes, good advice on all points! Will look at the needs for detail vs the time wasted on poly reduction efforts. However too much obsessive/compulsive behavior with poly counts and the model will never get done in my lifetime. Model it, be done with it and move on!

Here's an effort I put in today and wasted 6 hours just to save 100K polys.

Top is the original mesh at 205K. The bottom is the reduced poly mesh at 100864 polys. Looks 95 percent detailed from the original, but cost me a lot of time. In this case, since I will be cloning it 10 times to cover the engine sides, it was well worth it to save a million polys.

[Titon]

Can't tell the difference. Nice job on the reduction.

I know it takes a lot of time correcting poly's and such but it will be worth it. Again the tank looks good but in all honesty it doesn't have to look that good. When you render you can compensate with anti-aliasing even if you have to push the number upwards it's still quicker than having to many poly's that it won't even load. On my Roco tanks the more visable detail is there but again if i want to get really close to it i will rebuild it and use it as a seperate mesh.

A good habit to get into is reduce poly's as you build. Bandglue is a good plugin to get into.

I see your pipes look really really smooth. When i build them i start out with a 6 or 8 sided poly and extrude it, delete the ends and sub patch them. When they are subpatched you can keep smooth shifting them to make your bends. Reduce the subpatch levels before you freeze the tube. That way you do not have to make a tube with 40 plus to get them to look absolutely smooth. I can tell those tube's are pretty high poly's?

[maudib]

For complex bending pipes or pipes that have to fit a precise curve against a hull or surface, I create a complex curve using point-to-curve function. Then I create and place a disk at the start and end points using the smallest number of faces (24-48 is usually enough) I can get away with a minimum of texture smoothing. I then use the rail function and again use the min number of segments to achieve a smooth curve with min smoothing used.

I don't like high degrees of smoothing as distortion occurs when rendering. I don't have to spend hours trying to fit a pipe to my location, the pipe fits perfectly in an instant, and I move on to the next mesh.

I am however having the devil of a time trying to fit these finished pieces to the engine framework! It doesn't fit. Either I have made a mistake in measuring the overall shape of the engine section, or my parts are off, or both. Ugh.

[maudib]

Update 9-20-12. Another huge greeble mesh added and cloned 8 times. Section is starting to fill out nicely. A couple of others greebles in the top front corner. Looks like a concrete road barrier.

[Matador]

Awesome work man!

I'm currently working on the side gun turrets... I was hopping you may have some reference pics of this area or a link.

[maudib]

No gun turrets were ever described or placed on the TOS studio scale model. The TV show never made any attempt to explain where the guns were located. Sorry.

[martok2112]

Subakh ul kuhar, Muad'Dib? (How are things, Muad'Dib? )
Yeah, we never really got a sense of just where the guns were placed (just like there's no official word on ship length, and other aspects...at least for the classic Galactica), but I had a few ideas which I implemented on my battlestar.

The upper and lower guns we sometimes saw, I placed them in between the launch tube groups along the landing arms of the battlestar. Also, I think I placed a couple of turrets in those little notches on either side of the battlestar's head section. And also, decided to place some of them on the support pylons, and in pairs at the upper and lower/fore and aft of the landing arms. Placed a few on the forward and aft of the crossbar underneath the battlestar, the bar that connects the lower center (which I think has been pretty popularly designated as the pilot barracks). Even placed a couple on the bridge module itself, to prevent suicide attacks from Cylon Raiders that might try to go for the bridge.

And then of course, there are the big guns that I decided to run with.

[maudib]

Update 9-23-12

Another big greeble done and cloned 10 times, plus a few others. This one was time consuming since I had to make most of the detail from scratch. The good news is that I have more stuff in my library of parts to reuse again. The bad news is that the greeble is backwards! It's easy enough to flip in Lightwave. I'll do it later.

[maudib]

update 9-26-12

Made another large greeble and cloned 2 versions of it to cover the sides and inside the groove of the engine section. Getting everything to fit was a nightmare, and I just can't reconcile the pieces I modeled with reference images of the original. It's close enough for me. This is the last major update for awhile again.

[mikala]

Wow just does not cover it!

[gmd3d]

I would like to throw in a wow there also

[Titon]

Good job on the update.

Getting the nurnies to fit is a nightmare. That's why it is taking so long for me to finish my Galactica. You dang near have to have the kit bash pieces on hand to measure in order to get everything to fit. Not to mention a lot of reference from guy's that have actually built this beast physically. Otherwise it just never looks right in cg.

Keep it up!

[martok2112]

Yes, my jaw is irreparably dropped at the stunning detail you are putting on this maudib. I don't think I can even pronounce "WOW" without it sounding like "auuaaaaoooowwwwooouaaauueeehhhhhhh!"

It will be REALLY impressive to see if this monster (and I say that with all due affection for this model) can be animated.

[David Kerin]

Every update you have makes me realize how much I love the design of the battlestar. Beautiful work!!! I remember on September 18th, 1978, start of 5th grade class and I drew a big side view of the Galactica on the chalk board. I could have never imagined the creative tools that computers would bring us.

And I must say it puts my chalk board drawing to shame!

[maudib]

Thanks everyone for the comments. Renders at this point with full antialiasing at 1920x1080 HD resolution (17 passes in LW Layout, and radiosity turned on) takes 3 min and 3 sec. I spent the last 3 days poly trimming as Titon suggested and now practice reasonably high poly modelling on the big G. I used to just shrink down the little nurnies from the big versions to make details on a big mesh, but now I poly reduce it first for a small version. This has saved me over 4 million polys so far on the engine section alone! The total number is 9.8 million so far.

I anticipate at this rate in staying below my 20 million goal before completion. Like I said, only 1/2 the mesh needs to be loaded at a time for one-sided render passes, so only 10 million polys will need to be loaded and rendered in most cases.

Fitting the parts had me scratching my head for days! Did I make a mistake in measuring the basic hull? My main reference was noticing that 4 basic greebles make up most of the engine detail. So I started with the 1st one, and using that as a "ruler" for determining the basic shape and proportion of the engine section. The 1st greeble then had to fit on the next location without resizing, so the basic engine hull was resized until it fit the part.

For example, here is a section of the engine hull which I didn't notice until I tried to fit the 1st big greeble above it. I did not model in this angle initially, so the thickness of the side was too narrow to fit the next greeble without re-sizing it (which is a big no-no!)


And here is the corrected mesh with all the part now fitting, much to my relief!


I then made the 2nd greeble, and put it next to the 1st and cloned the sequence 4 times, and that was how I arrived at the overall length of the engine section. I find that my method was so accurate, that no resizing of the midsection and front section was needed, and looking at the reference pics, the battlestar looks good to my "eye". I wouldn't dare measure it though! I'm just going to pretend that everything is good and NEVER do this again!

[Thunderstruck]

That is looking awesome....

Wished I could model, but am gradually picking it up.

curious though, 17 passes?

why so many on a still ?

never used LW seems like a lot.


Great job so far.

Regards,

Randal R.