DaveP's Luke RotJ V2 Research/Development Thread 2021

Don't want to see this ad? Sign up for anRPF Premium Membershiptoday. Support the community. Stop the ads.


Don't want to see this ad? Sign up for anRPF Premium Membershiptoday. Support the community. Stop the ads.

Zenkai

Well-Known Member
RPF PREMIUM MEMBER
Now the tricky (and most important) part... I scale the rough model to the correct scale and then move to a perspective/camera view. I then rotate the camera around the model to replicate the angle of the camera that took the original reference photo and adjust the field of view accordingly to match the perspective. There are also tools for correcting lens distortion.
Could you make a video of this part of the process?
 

DaveP

Master Member
RPF PREMIUM MEMBER
Could you make a video of this part of the process?
I'll certainly give it a go. I'm going to be lining the model up with a totally different view to focus specifically on the pommel next week, so I'll see if I can get a video of the lining up process. It can be a long process though, with tiny little adjustments, so could potentially be too long to record. I'll see what I can do though :D
 

Don't want to see this ad? Sign up for anRPF Premium Membershiptoday. Support the community. Stop the ads.

Triin

Active Member
RPF PREMIUM MEMBER
Amazing modeling work!! . have you thought about bringing this model into substance painter to paint on it in real time? I find it easier to get the super fine details ( texturing wise) on my models.. I thought if this when you mentioned the stencils..
 

Pedro

Sr Member
RPF PREMIUM MEMBER
Thanks Pedro! Not derailing at all. :) I went over it briefly at the start of the thread, but didn't go in to much detail. You're right. It's not easy! The trickiest part is getting your camera set up to match the scene. I'm a freelance games artist by trade, so I have licences for 3DS Max and ZBrush out of necessity. I'm sure many (if not all) of the tools available to me in Max can be found in Blender though (which is free).

I start by applying a side profile of the object to a material and then applying that material to a plane in the left orthogonal viewport.

I then divide the plane up along horizontal and vertical reference points. Trying to cut through the centres of any ellipses. The edges on this plane can then be snapped to in the orthogonal view to build a rough profile of the object.

View attachment 1488822

Now the tricky (and most important) part... I scale the rough model to the correct scale and then move to a perspective/camera view. I then rotate the camera around the model to replicate the angle of the camera that took the original reference photo and adjust the field of view accordingly to match the perspective. There are also tools for correcting lens distortion.

Once the angle of the camera is locked in, you can then switch between orthogonal views (or have different views open on different screens) and adjust the rough model to match the reference perfectly. It's best to make any adjustments in the orthogonal viewports whilst watching the model update in real time in the camera view.

View attachment 1488823

For the textures, I build a simple model that only has the areas I want to texture. I then UV unwrap these and flatten the sections out in to UV islands (A bit like you would if you were making a cardboard model). I then make a template from this that can be textured in Photoshop using as many references as possible. It's important to remove any cylindrical or perspective distortion from the references before piecing them all together and then overlaying them on to the texture sheet in the correct place.

View attachment 1488827

View attachment 1488826

The ZBrush model that you see in the previews is based on the models that will be sent to the machinist, but it has been subdivided in to an incredibly dense mesh which can then be sculpted with surface details like a piece of virtual clay. Details from the low poly model (like the stencil detail) can then be projected on to it. If you look closely, even the paint has physical depth on this model. ZBrush is incredibly powerful visual tool, but these models are really just for presentation and would be no use to the machinist.

View attachment 1488824
View attachment 1488825

Anyway, I hope that gives an idea of the process I go through when I'm building these. To get the models as accurate as humanly possible, it's a good idea to repeat the process with multiple references and views, moving back and forth between them, correcting as you go until you arrive at a model that fits with all viewing angles.

Hope that's been of interest anyway. I'm going to be going through all the above again over the next couple of weeks as I put the models through one final round of accuracy checks.

Thanks for looking and all the best,

Dave
Thank you Dave for this detailed information, exactly what I was asking for! The good news is that it sounds like at least part of this is "brute force", which makes me feel better, having tried Sketchup photo matching without great results. I was decent in 3DS Max about 20 years ago, back when I didn't feel obligated to pay for it. ;) I spent most of the time between doing no 3d modeling, and man have things changed! I appreciate the tips, thanks! Still a bit in awe of the matched damage geometry, amazing work sir.
 

DaveP

Master Member
RPF PREMIUM MEMBER
**UPDATE**

Apologies for the lack of updates guys...

Rest assured that I've been insanely busy working on a variety of things relating to this project.

My time has been split between:

  • Further research and development, (resulting in a couple of small but important design changes).
  • Updating models
  • Creating artwork (to show design changes)
  • Testing new vinyl cutter and stencils
  • Working on a complete Starkiller build (which is nearly finished). **
  • Testing the solution for the ANH Blade/motor

** It should go without saying that no measurements have been taken from this, and my models were pretty much finished by the time I bought it from a friend. The reason for buying it was so that I could get a completed V2 build done to show as an example of my commission work before the run goes ahead, and to test my stencils. The thought of even raising a tape measure to someone else's work goes against everything I believe in and all of my measurements have been taken from references of the original prop (with the clamp as the scale reference). That should go without saying, but as we know from recent events, it's not unheard of for people to take the easy road unfortunately.

I've just been so busy spinning all of these plates, that I haven't had a chance to stop and sit down to talk about everything.

Well, I thought it was high time I filled you in!

Starting with the new insights in to the prop, and what this means for the design.

The nipple has now been extended down in to the emitter, and the bearing sits below the emitter set screws.

ROTJ CUTAWAY 1.jpg

ROTJ CUTAWAY 2.jpg


This design change came about because the old design didn't satisfy some of the facts that I'm now 100% certain of regarding the emitter:

The ROTJ Nipple is fixed to the emitter (NOT the rod). This is quite a significant change from my previous thinking. I noticed when working on the stencils that there is a distinctive lens shape on the face of the nipple. It was necessary to reference two different images in order to get the full picture of the paint detail on the emitter plate. That's when I noticed the nipple was in the same exact orientation compared to the emitter!

NIP ORIENTATION.jpg


This got me thinking about the fact that Brandon couldn't say for sure if the nipple was even a separate part, which implies that he hasn't been able to separate the two, or even move them.

If you look at the above references, the internal rod HAS changed position.

On reflection, from a design point of view, it makes sense that the nipple didn't spin with the blade.

I have arrived at the conclusion that the rod was NOT a part of the blade. (There would have been no way to connect it to the motor) *it had to be connected to mechanism prior to housing the motor inside the hilt.

As the emitter screws are presumably holding the nipple in place, and not the bearing, it figures that the nipple screws hold the whole emitter on to the rod and the hilt together when the blade is not attached. These would need to have been loosened in order to allow the rod (and blade) to spin independently of the emitter.

It could be that the nipple screws also served to protect to mechanism when pressure fitting the blade to the exposed rod which must have extended out through the emitter for the blade to then be attached.

So, the blade was attached to the rod after the hilt was fully assembled, which explains why the entire hilt was swapped in, rather than just tricking in the blade.



I'm yet to see any definitive evidence that the blades had collars.

I'm fairly certain that the nipple we see on the prop today is what we see in ANH. It matches perfectly in diameter, height, and the unusual positioning of the set screw. There are also signs of the narrowing step in some of the references, which I believe was there to transition from the diameter of the bearing to the thinner diameter of the blade.

NIPPLE REFERENCE.jpg


So... if what we're seeing is the nipple, and therefore part of the emitter/hilt, it would appear that, on this particular prop at least, the blades had no collars.

Absence of evidence is not evidence of absence of course, but when you consider that these blades were frequently replaced due to breakage, perhaps it just wasn't worth fitting them with collars, which would have needed to be perfectly flush with the blade and hidden below the reflective tape?

"I had to teach them if that was the sword, to stop before the touch because the blades were breaking. We broke so many blades! They just kept snapping" - Peter Diamond (Stunt Coordinator)

So, based on the above, I believe that the blade itself consisted of just a wooden dowel that had been bored out at one end, in order to pressure fit the tang, which which already mounted within the hilt and protruding from the emitter/nipple.


ANH CUTAWAY 1.jpg

ANH CUTAWAY 2.jpg

*Vintage Motor setup

ANH CUTAWAY 3.jpg

ANH CUTAWAY 5.jpg

ANH CUTAWAY 4.jpg

*Modern Motor Setup

Based on my observation, there is a slight taper towards the end of the blade. This is not obvious in many references but there are a couple where the blade is angled towards the camera, and where you would expect to see some perspective on an un tapered blade, it appears that the tip is slightly narrower than the base:

Tapered.JPG
Tapered2.JPG


The black line on the blade appears to me to get thinner towards the top, which could indicate that the strips of reflective material are a uniform width, applied to a tapered blade.

ANH BLADE.jpg


There does appear to be some variation between blades, which is worth noting.

I'm open to suggestions/debate with regards to all of this. I know a lot of people have spent a lot longer than I have thinking about these blades, and there is every chance I've missed something. This is my thinking right now though.

I'll be posting another update soon to demonstrate the motor/battery/blade setup. I'm just waiting on the reflective material to arrive.

Stencils!....

I'm just about finished working on the Stencils! I'm really pleased with the vinyl cutter so far and the vinyl I've chosen is really easy to weed out and apply/remove. The only problem is that it isn't inkjet printable. I really like the vinyl though, so rather than go for a different, printable vinyl that might not be as suitable, I've decided to print separate labels for marking up the sheets and identifying the various sections.

I've gone over the process of making the Stencils in an earlier post, but here are some images of the Starkiller build that I've been working on this week:

PXL_20210830_143328387~2[9150].png
PXL_20210831_113406403~3[4753].jpg

IMG_20210903_174130_310.jpg
IMG_20210903_174130_261.jpg


I've been working on this at a leisurely pace alongside everything else I've been working on. It's very nearly finished though. I'm just waiting on the jig to bend the emitter, which is currently "in production".

I'm also in the process of rendering some artwork to supply along with the stencils to serve as a guide:

NATE RENDER.jpg


This particular view was requested by Natesamlord to use as reference for a tattoo! :)

The stencils will be supplied with 4X different views taken at 90 degree intervals.


Anyway, that's all for today's update.... The next update will include the new accuracy checks (which are now happening this coming week), as well as a preview/demonstration of the ANH motor solution and blade (hopefully with reflective tape if it's arrived). I think it'll be a couple of weeks before the jig arrives for the emitters unfortunately, so my Starkiller build is on hold now for the time being. I'm expecting to have it in hand and tested before the run goes ahead. (y)

Thanks for looking everyone! I'll be going off-grid again now whilst I tie up these last few loose ends and get the final models CAD drawn up.

All the best, and MTFBWY!

Dave
 
Last edited:

Don't want to see this ad? Sign up for anRPF Premium Membershiptoday. Support the community. Stop the ads.

Filandrius

Sr Member
RPF PREMIUM MEMBER
Those updates of yours read like a fine novel. Seriously, I’m floored by that amazing Starkiller. It’s easily one of the best V2 I’ve ever seen. I already couldn’t wait to see a finished one in person, now the wait is going to be unbearable. ;)
 
Last edited:

thd9791

Master Member
RPF PREMIUM MEMBER
I love what you've done - and have some relevant information

The collars help keep the blades from splitting when the inner bushing is installed. They're fiberglass or some other composite. The dowel thing was hashed out a while back, too heavy for the motor and impossible to taper without machinery. Now, they're fragile so they did break a lot I bet..


Anyway the reflective tape is not uniform. I tested this in my blade threads and a rectangular strip covers more at the base in an angled pattern. Given the strip visible, I found they trimmed the edges of the reflective material making a trapezoid. I've done it repeatedly and it gives this same exposed black stripe when done over a black painted blade. Thing is, I suck, so all my black stripes have bad edges and twist because applying it is a bitch. At least on some it may have been spray adhesive and fabric

I'm also pretty sure there were multiples so make sure the pictures you're using are of the V2 and not the others. ;)
 

DaveP

Master Member
RPF PREMIUM MEMBER
I love what you've done - and have some relevant information

The collars help keep the blades from splitting when the inner bushing is installed. They're fiberglass or some other composite. The dowel thing was hashed out a while back, too heavy for the motor and impossible to taper without machinery. Now, they're fragile so they did break a lot I bet..


Anyway the reflective tape is not uniform. I tested this in my blade threads and a rectangular strip covers more at the base in an angled pattern. Given the strip visible, I found they trimmed the edges of the reflective material making a trapezoid. I've done it repeatedly and it gives this same exposed black stripe when done over a black painted blade. Thing is, I suck, so all my black stripes have bad edges and twist because applying it is a bitch. At least on some it may have been spray adhesive and fabric

I'm also pretty sure there were multiples so make sure the pictures you're using are of the V2 and not the others. ;)
Thanks for the comment Tom, I've responded to your DM too. I'll post an update soon in more detail based on what we discussed. (y)
Looking great dave!
Thanks Ariston! (y)
Just out of curiosity, is the weathering scheme based on how the prop looks now, or how it looked during production?
Hey there. That's a good question! The weathering on day one of filming was quite different to how it looks now. It ended up looking pretty much how it is today by the end of production though. I'm going for how it is now/towards the end of production.

It would be really cool to try one that matches the sandstorm scene at some point, but I think most people would prefer the more recognisable look. Also, there aren't really any high res references of the early production V2 for making the stencils, so there would be a lot of guess work/artistic licence.

Hope that's ok?

Cheers,

Dave
 

Don't want to see this ad? Sign up for anRPF Premium Membershiptoday. Support the community. Stop the ads.

Filandrius

Sr Member
RPF PREMIUM MEMBER
My 2 cents: I 110% agree with your interpretation. As it looks today, which is equivalent to the end of production, is by far my favorite version, and the one I want. :)
 

Eruonen

Member
Don’t get me wrong, I’m a stickler for accuracy too, so having an accurate version of something over guess work is always preferable!

But there’s that side of me (and I suspect some of you can identify with this) that always wants what no one has done before, so I just thought I’d ask. I’ve recently been poring over all the old pages of research just to get myself up to speed on OT sabers and the Luke ROTJ sabers specifically, and I was intrigued by how much more paint was on the booster even in the Throne Room and Endor scenes, so I just thought I’d ask.

All that to say, I love the attention to detail, and your willingness to hear people out. I’m incredibly excited for this run, man!
 

DaveP

Master Member
RPF PREMIUM MEMBER
Don’t get me wrong, I’m a stickler for accuracy too, so having an accurate version of something over guess work is always preferable!

But there’s that side of me (and I suspect some of you can identify with this) that always wants what no one has done before, so I just thought I’d ask. I’ve recently been poring over all the old pages of research just to get myself up to speed on OT sabers and the Luke ROTJ sabers specifically, and I was intrigued by how much more paint was on the booster even in the Throne Room and Endor scenes, so I just thought I’d ask.

All that to say, I love the attention to detail, and your willingness to hear people out. I’m incredibly excited for this run, man!
Totally understand. I love the thought of doing a sandstorm version. It would be an interesting challenge for sure. I can't promise anything, but If I get some free time before the run goes ahead, I could do a variation on the stencils. It would be an artist's impression though. In theory, I could expand the paint areas in photoshop until it approaches the right look. I'd be curious to see how that turns out.

Leave it with me! (y)
 

DaveP

Master Member
RPF PREMIUM MEMBER
I love what you've done - and have some relevant information

The collars help keep the blades from splitting when the inner bushing is installed. They're fiberglass or some other composite. The dowel thing was hashed out a while back, too heavy for the motor and impossible to taper without machinery. Now, they're fragile so they did break a lot I bet..


Anyway the reflective tape is not uniform. I tested this in my blade threads and a rectangular strip covers more at the base in an angled pattern. Given the strip visible, I found they trimmed the edges of the reflective material making a trapezoid. I've done it repeatedly and it gives this same exposed black stripe when done over a black painted blade. Thing is, I suck, so all my black stripes have bad edges and twist because applying it is a bitch. At least on some it may have been spray adhesive and fabric

I'm also pretty sure there were multiples so make sure the pictures you're using are of the V2 and not the others. ;)

Apologies for the delayed reply to this comment for anyone that's curious about what my response is.

I've already replied to Tom via DM but think it's relevant to the thread, so I'm hoping that Tom doesn't mind me sharing my thoughts here too. (I've pasted some of what I've already said to Tom word for word as, like I say, it's relevant to the project)

I hate to disagree with anyone, and I'd especially hate to appear as though I'm not open to suggestions (or the idea that I might be wrong). I'm absolutely sure that the blades on the motorised stunt didn't have collars though.

I've already mentioned why I think so, but Tom's comment encouraged me to dig a little deeper and try to prove it (or disprove it). I'd be happy with either outcome, so long as we arrive at the truth.

I can understand why it would be preferable to have a collar on there, but as we know, the blades DID break. Who's to say they weren't breaking at the base (for that reason)?

So, as I've mentioned, my reason for believing that there were no collars on the Obi stunt blades is that I'm convinced that what we see at the base of the blade is the nipple (as it is today on the V2).

I've put this little slideshow together to demonstrate:



There's no doubt in my mind that that's the V2 nipple, and not a collar.

Tom shared this image with me, which I have to admit, made me sit up straight when I saw it!

Screen Shot 2021-07-04 at 12.19.10 AM.png


After thinking about it though, the above image actually confirms (to me at least) that there ISN'T a collar. I believe that what we're seeing here is gaffers tape. Could it be that this was applied because the blades were splitting at the base? It's no thicker than the actual blade, and the material doesn't appear to be metallic to me.

As for the blades themselves... I was under the impression that it was well documented that the blades began as wood, before switching to fiberglass, and then later aluminium for the prequels?

Hardwood blades would probably be a little too heavy for the mechanism, but pine or redwood are actually quite a bit lighter than fiberglass (and even carbon fibre):

Pine = 0.5g/cm3
Carbon Fibre = 1.75g/cm3
Fiberglass = 2.5g/cm3

I've worked out a method for boring them and tapering them using just hand tools, but that's a post for another time as I'm waiting on a couple of things to arrive to set things up. It basically involves setting up a dowelling table that consists of a fixed horizontal drill that has been perfectly aligned with some Angled moulding to act as a guide for the dowel. I'm looking forward to setting this up and sharing the results.

Anyway, I didn't want to just disregard what Tom has said without giving good reason. He's put a lot of work in to researching the blades and I'm especially grateful for the advice on wrapping the blades (I'm still waiting on my material to arrive!). I just think the Obi stunt was a special case (as seen in the video with the full switcheroo).

As always, I'm open to disagreement (and especially welcome any evidence I might have missed)

All the best, and MTFBWY!

Dave
 

PoopaPapaPalps

Master Member
RPF PREMIUM MEMBER
If it means anything, I'm also in the camp that doesn't believe the Obi stunt blades were collared at all. At least, the majority of them. The collared blades may have been used for the Vader/Luke stunts at this time, and may have been switched out onto the Obi at some moments, but considering the anecdotal evidence of what was being used on set during the time of the duel on the Death Star, as well as the documented photos we have available: the majority of the stunt blades for the Obi stunt I believe to be wood, dowels or squared off.
 

Don't want to see this ad? Sign up for anRPF Premium Membershiptoday. Support the community. Stop the ads.

Your message may be considered spam for the following reasons:

  1. Your new thread title is very short, and likely is unhelpful.
  2. Your reply is very short and likely does not add anything to the thread.
  3. Your reply is very long and likely does not add anything to the thread.
  4. It is very likely that it does not need any further discussion and thus bumping it serves no purpose.
  5. Your message is mostly quotes or spoilers.
  6. Your reply has occurred very quickly after a previous reply and likely does not add anything to the thread.
  7. This thread is locked.

Don't want to see this ad? Sign up for anRPF Premium Membershiptoday. Support the community. Stop the ads.

Top