SWTC : Death Star II size

[Mr. Oragahn]

Time to give this one its own thread. It's quite baffling that we never did. I ran a search and didn't find anything. I guess we always talked about the second Death Star's dimensions without ever starting a thread solely dedicated to that topic.

Well, now it's done.

I also know we have a thread where we extrapolated upon the number of planetary shields needed to completely protect a planet, but for the love of *** I can never remember where we put that one. >:|

Err... this is going to be some strong attack against the nonsensical 900 km wide figure, and obviously this will be achieved by attacking Curtis Saxton's own webpage dedicated to the Death Star II. But this time, I'll largely deal with the figures he gathered from other sources, and not pay attention to the calculations, since we know anyone has his own set of calculations which are assuredly good enough.

We can start by looking at his summary and see which sources he lists. I won't follow the order from his list though.

First, the concept art destroyer shadow:

star destroyer shadow [DS2]

It may be possible to deduce some limits on the sizes of the Death Stars by observing the geometric relationships between the battle stations and standard star destroyers. This comparison could be made from an image showing a star destroyer and a Death Star in physical proximity, or circumstantially from something known about how these objects interact.

If a star destroyer can dock with a Death Star then it must occur at either a mooring tower several hundred metres high, or a similarly deep hangar. The equatorial trench might be able to conceal a cavity of the required size (a mile long and at least a third of a mile high), although no such cavity is obvious in presently available images. The available images fall into two categories of inadequacy: fully global views at astronomical scales that show the equatorial structures as pattern of barely distinguishable lights; and close views showing limited local regions (showing only a few shuttle-sized hangars). Larger hangars could exist [unseen in the movies] at other longitudes. However if it could be proven that any of the waistband features are star destroyer berths then we could infer that the equatorial trench spans at least 400m (fitting the tightest possible alignment of the destroyer) and probably over a mile (with some margin for safety).

Amongst the concept artworks of ROTJ there are two pieces showing a star destroyer cruising above the surface of the Death Star II. There is one concept sketch in the ROTJ Sketchbook, and a painting in The Art of STAR WARS.] The ship casts a shadow onto the battle station's surface near a trench. The trench is somewhat wider than the shadow by a factor between 1.0 (painting) and 1.7(sketch). Therefore the trench is at least a mile across, the length of the star destroyer.

It is interesting to note that on the second more detailed sketch (which wasn't included in the ROTJ Annoted Sketchbook for some reason), the Star Destroyer's length would barely fit into that trench, if moved into said trench with the same direction as seen on the picture. Amusingly, a one mile wide equatorial trench is precisely the figure RSA got that led him to a size of 155 km.
We can also notice that the DSII is considerably farther from Endor than it is in the film. Here, it is locked in a distant orbit, and makes Endor itself look like a small satellite.
Also, notice the fact that the trench is first drawn curved, then the perspective lines show it's drawn flat until it fades away into the horizon. I guess the Death Star was first designed as a Cube with rounded edges. :)

Hell, perhaps the Death Star II had some flattened areas? That would conveniently explain the Executor crash scene.
Just an idea there, you know. :)

The first sketch leaves no doubt about the planned size of the Death Star, much smaller than on the second sketch, as we can see that the trench is well curved. Clearly, they were in the process of looking for something epic enough, and tried various sizes.

If it is the equatorial waistband trench then we can compare this measurement with the whole-disk views and thereby derive a lower estimate for the global diameter.

Why lower?
There is nothing here that allows us to consider it's a conservative representation at all.

If, rather, it is one of the smaller meridional or latitudinal trenchs (which are invisibly small in whole-disk views) then we have an even stronger lower limit to the station's size: the station would have to be several times larger than implied in an equatorial comparison.

It would be require evidence that those smaller trenches existed as they did on the first Death Star, and that they housed hangars. They didn't on the first battle station.
We don't see hangars on the sketch, but we get to appreciate the curvature of the whole station so it's not important. We see hangars in the trench of the painting, and it's obviously what would turn into the equatorial trench for the movie.

Of course this evidence is not definitive because the artists of ILM could have resized the Death Star II any number of times before settling on the final film version.

Let's point out that he admits the artists would have gone back and forth through the final size.
Stick that somewhere.

The trench width is 1.4 ± 0.4 times the one-mile length of the shadow. The global diameter is 153 ± 7 times the width of the Death Star II's equatorial trench. Assuming that the sketch is equatorial, and that it accurately reflects the final scaling of the movie, the diameter of Death Star II would be

D = 345 ± 100 km

This estimate is probably good to within an order of magnitude, since althogh the sketched scene was cut from the movie, it is unlikely that the artists could have revised the scale up or down by more than a factor of a few times.

This estimate is solely derived from his former waistband trench width figure.
RSA had found a ratio of 128.099.

With the trench being 1.4 times wide than the SD is long (one mile), that's around 286.9 km, not 345 km.
With 1.8 and still going with RSA's ratio, that's 368.92512 km.
It becomes more or less 80 km, not 100.

Besides, the sketch is very... sketchy. Between the trench that obviously never meets the horizon, and the other one which begins curved but ends flat... well...

Next source.

Cross-sections [DS1]

STAR WARS Incredible Cross-Sections shows the interior structure of the original Death Star, in both schematic form and a coloured perspective drawing. The book states a diameter of 160km, which agrees with results calculated here, but one of the most interesting aspects is the treatment of the main reactor system. Compared to the Death Star II schematics used by the rebels in Return of the Jedi, this Death Star's core is much larger in proportion to the whole sphere. In the first Death Star the main reactor chamber is 36% of the total diameter, and the actual power core is about ten percent of the diameter. In the second Death Star, the reactor chamber and core are 13% and 3% of the diameter respectively.

There's no firm size we can get from canon about any Death Star's core, first or second.
I even proved that if we did like some warsies do and took the hologram at face value (necessary for the 900 km value), the 2nd Death Star would be about ~17 km wide, if you use the X-wing width to measure the core's larger ring and then the bulb itself.

The second station was unarguably larger and more powerful, and therefore must have had a reactor of at least the same size as its predecessor. As a lower limit, if we assume that the power systems were identical, then the diameter of the Death Star II must be at least 2.8 to 3.1 times the original's 160km diameter. Accounting for the need for a bigger reactor, the size ratio of five or six times (due to independent physical evidence and the CINEFEX interview above) is comfortably allowable.

Why didn't he mention other EU sources regarding the first Death Star schematics?
Why go for the ICS one, when he does know that other schematics exist:

http://www.theforce.net/swtc/Pix/given/rb/ds21.jpg
http://www.theforce.net/swtc/Pix/magazi ... ds1tj2.gif

Schematics showing the size of the first Death Star's main reactor in relation to the surface. General Dodonna's (canonical) tactical animation [ANH] showing a growing chain reacton propagating towards the bulb of the main reactor. Secondary but more detailed reference schematic [Star Wars Technical Journal vol. 2].

Here's a zoomed in view of the DSI core:

________________________________________

This one makes it clear that it's very similar in shape to the core seen inside the DSII. Of course it makes the station absurdly small by the same token. It's also, in my opinion, a much better exposé of what the DSI reactor assembly would have looked like. It has many things right, from the spherical core and the shaft above it, as seen during the briefing, to the large room which is similar to the DSII's. The real problem, though, is that the spherical core cannot be as big as the one seen in the briefing, otherwise, just like for the DSII, the DSI would be absurdly smallish, perhaps worth an Executor's length, more or less.

Nevertheless, can't he even see the difference in core size?
I guess he goes with the ICS drawing because these DK books have a tendency to be detailed and accurate enough. Doesn't mean they're always right though.
The ICS artist who drew the Death Star, Hans Jenssen, was faithful to the core's dimensions as seen during the briefing, and even bothered encasing a gigantic hemispherical bulb inside the superstructure, on top of the spherical core, in order to establish a continuity with the DSII core design, again by respecting the proportions as seen during the briefing. Yet, he refused to place the dish on the equatorial trench, despite the fact it was seen as such during the same briefing.
That's cherry picking on Jenssen's part, and probably one of the major problems of the old ICS' cutaway of the DSI.
Being certain of the core size for the first battle station would be necessary before attempting to use the first Death Star's core as a yardstick for the second one.
Saxton cannot, and doesn't even really try, since he ignores several sources. Really, his methodology is broken.

To his benefit, he does list the early blueprints which support a width for the DSI that's slightly above 67 km:

blueprints [DS1]

At some time during the first-generation classic era of STAR WARS publishing, long before the advent of the Roleplaying Game, a set of blueprints were published, including diagrams depicting the Death Star and star destroyer. The plans in The Technical Book of Science Fiction Films indicate a diameter for Death Star I of only 67.375km. This is less than half of the size measurable from the trench run and General Dodonna's [intratextual] schematics, and only slightly more than half of the value quoted by West End Games and its derivatives.

The reliability of the scaling in these blueprints can be judged from the contemporary blueprints of a star destroyer, a better known and measured vessel. The 1978 destroyer blueprints underestimate the ship's length: only 486.5m rather than the one mile (approximately 1609m) that is apparent in the movies (e.g. measured in comparisons with destroyers with the Millennium Falcon and Lord Vader's Lambda-class shuttle.) and stated in most second-generation publications.

The error may have arisen through the misuse of a preliminary scale diagram drawn during the initial design of vessels in A New Hope. (For example, there exists a preliminary scale chart that had an underscaled blockade runner and equivalently underscaled star destroyer.) In its (extratextual) design evolution, the star destroyer grew from a small-scale gun-bristled prototype up to the final vast vessel with invisibly-small weapons [e.g. documented in From STAR WARS to Indiana Jones].

Despite the wrong size, the star destroyer blueprints retain some credibility. The proportions roughly match those of the canonical vessel. Some important features differ (e.g. the forward hangar), but other features are well done, including dorsal details that were invisible in ANH. This implies that the draftsman did work from some priviledged information. Perhaps the blueprints show a hybrid of the model seen on film, and superseded concept design sketches. In summary, the star destroyer blueprints are conceptually good in many respects but show an over-dependence on early designs which were superseded in the final film version. (In those days it would have been technologically difficult to collect adequate reference photos, even within Lucasfilm itself.)

It is harder to guess the exact reason why the Death Star blueprints underestimate the battle station's diameter. It may have something to do with early production art of A New Hope which is not available to public examination at this time. Perhaps the 67.375km figure was invented arbitrarily without considering any evidence, like the later statistics from WEG roleplaying games.

The Death Star blueprints are easier to dismiss as evidence, because they condemn themselves with more serious inconsistencies than the star destroyer blueprints. The "typical docking bay" in the equatorial trench is drawn many times too large: it takes up nearly the entire span of the trench, whereas the relatively tiny lights seen in the film are mere indentations visible only at close range and fitting in several rows within the trench width. The polar trench has been exagerrated until it is half the width of the equatorial trench. In the film, no polar trenchs is wide enough to be visible from an astronomical distance.

If we upscale the Death Star I blueprints by the same magnification as the star destroyer blueprints need, the global diameter becomes 222km. This is slightly beyond the size range suggested by the target scope and trench measurements. Perhaps there is a reason why both blueprints are underscaled by a similar amount? After all the intervening years, it may now be impossible to reconstruct what happened.

He arbitrarily assumes that both the Death Star and the Star Destroyer were scaled up (or down) by a very similar ratio.
There's no proof of that.

Next one, the cheat:

Inside the Worlds [DS2]

Inside the Worlds of the Star Wars Trilogy [hereafter ITW] states the diameter of the Death Star II as 900km or 550miles. This is consistent with the measurable movie evidence and with the CINEFEX report of Richard Edlund.

Surely, the fact that he was a consultant in the writing of "Inside the Worlds of Star Wars Trilogy", the very first EU book to ever propose the 900 km figure, has nothing to do with this convenient consistency?

I also like how he favours the CINEFEX report over the ROTJ Annoted Sketchbook:

...However there is firm evidence that the artists of the film eventually dropped this early unambitious figure, upgrading the battle station to a greater set of dimensions matching the results of the physical study above.
...

[sarcasm]Clearly, right off the bat, we can painfully see that the material presented in the "ROTJ Annoted Sketchbook" is in disagreement with the movie evidence of a much smaller DSII:[/sarcasm]

________________________________________

________________________________________

________________________________________

________________________________________

________________________________________

No, really, that CINEFEX report is the absolute Holy Grail. It really trumps everything:

The July 1983 issue of CINEFEX contains an interview with Richard Edlund of Industrial Light & Magic, regarding the effects of Return of the Jedi. In this interview, on pp.7-8 he says:

" The Deathstar, I think, will be a lot more interesting than the one in the first Star Wars — mainly because it is under construction ... Plus, it will be MUCH bigger. In Star Wars, it was really difficult to establish the scale. It was supposed to be miles in diameter, but with a full sphere it was hard to tell. The NEW one is SUPPOSED TO BE MORE like FIVE HUNDRED MILES in diameter, but since we're not dealing with a sphere all the time, we'll be able to establish landmarks and get a better sense of scale. "

This is in excellent agreement with the physically-derived estimate of the Death Star diameter, which is equivalent to somewhere between 530 miles and 590 miles. CINEFEX flatly contradicts the 100-mile diameter estimate in many second-generation sources [e.g. The Death Star Technical Companion]. As a final report by the special effects artists who made the movie, the CINEFEX article supersedes the earlier, preliminary scaling sketch reproduced in the ROTJ Sketchbook.

First of all, Saxton focuses a lot on the idea that that CINEFEX extract so proves that the DSII was more than 500 miles wide, yet misses the fact that Richard Edlund said the first one was merely miles wide.
Not tens, dozens or even a hundred miles wide, you know. Just "miles wide".
I know they'd say miles can mean anything, but from the mouth of someone who a few words later, goes to the length of being overtechnical about the new size, why pretend that he wasn't being technical just a few seconds before?
This is where it gets fantastic, actually, because during the entire report, Edlund is deadly technical about a variety of things, from the intricate devices used for shooting, special effects and post production montages, to the various budgets (he even bothers putting the budget into perspective because of the inflation).
I know that because I have read the report, and it seems that as usual, the same group of warsies may have once again decided to omit some facts.

Secondly, Saxton presents the CINEFEX report as if it mirrored the immuable final status of the movie, as if we could reliably take what he said without the slighest particle of salt, because the movie was done and packaged, ready for the audience.

But was it?

2 February 1982 - RICHARD EDLUND

This is the date of the first report. If you know a few things about the production of the movie, you may already sense that something's fishy about Saxton's position on the quote's value.

Now, with the third act, George has to really drive the nail home, and I think Revenge of the Jedi is going to do that. Right now, I'm still totally immersed in Poltergeist, but I have read the Jedi script and in a few weeks I'll be going over to Englang to supervise some of the effects shooting.

Yes, so basically Edlung had just read the script at this point. He had yet to join the group at Elstree Studios.

BUT EVERYTHING IS DEFINITIVE!!!!1!

The script is very good, and many things are resolved -- but Goerge is very secretive and there are certain pages that you know are not the right pages, even though they read very well. So there will be surprises, no doubt, for all of us.

Wait. Does that mean what Edlund knew of the movie was not 100% certain?

George has a lot of chips riding on this one. He's over in England now working with Richard Marquand, and will probably stay there until the picture is finished shooting.

Movie industry: live action shooting -> special effects -> polish.
They were stuck around phase 1 back then.

We've done five films here at ILM since the last Star Wars episode, so we've got some more experience under our belts and some better technology. And we'll be focused together on a single project -- although it looks like we'll be split into three separate units. Dennis Murren, Ken Ralston and I well each be heading one, and we'll be working more or less autonomously.

And they were. There basically was a tripartite system, with Edlund covering the Sarlacc and the space battle.

Joe Johnston and George Henson have been drawing storyboards for about six months -- hundreds and hundreds of storyboards -- drawing and redrawing them as George refines his sequences and changes things around.

So there was lots of pre-production material, and Lucas still was looking over their shoulders.

More stuff about the reliability of Edlund's knowledge of the movie:

As with the other pictures, Revenge of the Jedi starts out in space again. This time, the rebels have a great big cruiser of their own and there'll be a need for some grand matte shots there. Then the action opens up as Luke and Leia go back to Tatooine to rescue Han Solo from deep freeze.

Never made it to the movie.
A bit earlier on he said that Lucas really was tight on the overall budget.

So we're going to keep our locations secret as much as possible.
[...]
We're also doing some location work in north California, up near the Oregon border.

Again, confirmation that they had lots of live action to shoot.

The two-legged scout walkers that were seen only briefly in a couple of shots in Empire will also be featured -- though they're camouflaged, so they'll look a little different.

They did look a little different, but that had nothing to do with the non-existent camouflage. Camo on such noisy things btw?
That one is interesting because it refers to a given vehicle design which we know did get altered.

Meanwhile, the Empire is building yet another Deathstar, and this one is even bigger and grander than the last one. At the end, there'll be a furious space battle and dogfight that takes place in and around the new Deathstar -- which will be really interesting to deal with because there'll be a lot of wild flying scenes through girders and partially completed structures.

There has not been much dogfighting at all inside the Death Star, I'd rather call that a one way pursuit. And the combat didn't take place around it either, but far from it.
Actually, I think it's obvious that is sounded like ANH's space battle redux, with some combat inside added.

We're going to have some great new models. There'll be a rebel cruiser this time and a really nice-looking single-wing fighter. It's just a long wing with a cockpit -- and the cockpit rotates so the thing can fly around. I think that will allow for some really interesting battle maneuvers in terms of compositional geometry.

We got several new cruisers and the B-wings were never shown fighting. :(

The Deathstar, I think, will be a lot more interesting than the one in the first Star Wars -- mainly because it's under construction, so we'll be flying down through this half-finished Deathstar which will be a lot more exciting. Plus it will be much bigger.
In Star Wars, it was really difficult to establish the scale. It was supposed to be miles in diameter, but with a full sphere it was hard to tell. The new one is supposed to be more like five hundred miles in diameter, but since we're not dealing with a sphere all the time, we'll be able to establish landmarks and get a better sense of scale. Lots of visual perspective.

Doesn't look like we got anything of that. Everything we got in ol' ANH should have been enough to get a sense of scale: fighting above the Death Star, flying between defense towers, skimming the armored surface (remember, we're talking about the pre-Special Edition).
Now, look at ROTJ's space battle. In the end, did we get a better sense of scale, just because the damn thing was not finished?
Absolutely not. It's like he's talking of an entirely different section of the movie.

There are much more bits over the next pages which deal with the problems they had to deal with and everything they would use to shoot some special scenes for the space battle, and how they'd merge all that with printers and computers, etc.

They kept working, and by December 16th (compared to February 2nd, date of the first report that included the 500 miles figure), they had 500 shots for the rebel attack on the Death Star.

So I'll stop here. The rest of the issue's material is of the same vein, following the crew through the entire production of the movie.
The point is clear though: whatever Edlund knew of the movie by the time of the 500 miles figure was pre-production. The teams weren't even set up, he had only got an eye on the script, and many things were to change.

Now, let's be a bit fair towards Saxton. It is necessary to know that he gives credits to someone else for that CINEFEX quote. The culprit would be this other individual.

In the Acknowledgements section, he calls:

Anthony Tully for his extensive and intelligent stream of suggestions, for the reference to CINEFEX regarding the size of the Death Star II and for an assortment of other valuable and rare images.

There ends the fairness. Saxton is no rookie when it comes to scientific logic and the methodology of relying on auxiliary sources.
Before accepting a single figure from an extra source, he would need to know more about it, get the context, when it was published, if there were other versions of it, some evolution or some correction.
Now did he do all that? By the look of it, not much.
Can we believe that with an issue like this CINEFEX one, he'd have satisfied himself with just that one single figure, when there were chances that there could have been considerable amounts of pre and in-production information to pick for his website? Even if Tully had simply forgotten to point out the chronological relevance, it's hard to believe Saxton would make the same mistake.
Tully obviously didn't find this single figure by pure dumb luck. He has read a given amount of the magazine, most likely a large amount of it. I'd even say all of it, at leat once. You don't find such a detail in a magazine issue from the 80s just *like that*.
Now, chances that he'd forget to point out any of all the other details that litter the pages of this magazine (dates, state of the production, other details)? Very little.

Then, chances that Saxton, scientist that he is and avid fan who has spent an absolutely mind boggling amount of time analyzing Star Wars material, would not even ask for any further information? 1 in 400 billion.

Heck, we can see in the acknowledgment alone that Curtis and Anthony conversed a lot ("extensive and intelligent stream of suggestions"), the later also providing a variety of "valuable and rare images".

In all honesty, I can't call that anything but cherry picking.
This is rather consistent, mind you.