3DShipWright

Okay, I've stripped down the stern, only keeping the corner ornaments and the basic trim board. The rudder chain will be re-positioned once I get a general idea of how the steering rigging will go. Tiller has been modelled, but I still need to rework the back section on a large scale

@Martes - Belay the images request - I just noticed you included he link. Thx!

@Martes - Ahh, gotcha! So I can incorporate a wheel if I cosmetically like the look of it (which was kind of the whole reason I did one), you're saying I just need to model the tiller on the top of the rudder as well.. Perfect, consider it done! The model you're showing is really good - Would you mind sharing a few more images of the back-end detail for reference? Any actual diagrams or dimentions of the tiller? As a general rule, if I eyeball something it comes out wrong, so i'd like to start as accurate as possible. Regarding the rudder chain, no problem - I'm using curves and the array modifier so its easy to reposition. Something felt wrong about the way I'd done it anyway, I just couldn't put my finger on it. @druxey - Yeah, as both you and @Martes have alluded to - I think I need to make a decision on whether I'm going for real or creative, historical or retro-fitted. Tell ya what - I'll first complete the model realistically and historically, then when I'm done, I'll duplicate the project and go creatively nuts! Does that work for everyone? Lol Thanks guys, much appreciated, as always. -N.

@Martes And the taff rail I did is pretty low, I think the angle of the image may have been deceptive. Here's a character in the scene for reference. She's 5-1 if that helps. Best, -N.

Hey @Martes - No offense taken, I always appreciate constructive feedback, and hopefully you noticed that I already incorporated some of your previous suggestions. To unpack your comments a bit: Metal Rings - I believe those rings are sockets for supports when the ship is in drydock. Google real images and you'll notice that they can be seen on the HMS Victory in Portsmouth, the HMS Surprise (Rose reconstruction, renamed for the Master and Commander movie) in San Diego, the USS Constitution in Boston, and so on. The Anatomy of the Ship illustrations also show them on the Pandora and Essex, so I'm guessing they're not a new invention. Wheel vs. Tiller - The picture you just shared shows a brig with a helm's wheel, yet you told me previously there was no room - J'Accuse! Taff Rail - You're kind of right, kind of wrong here - Brigs don't typically have taff rails that are really distinguishable from the cap rail. If they do have a bridge deck or a poop deck, yes, they will usually use metal rods with rope instead of a wooden rail. However, no where is it written that a perminant wooden fixture couldn't be used instead. You can believe me or not, but I used to live in Galveston Tx and have personally set foot on a couple of operational brigs that did have raised wooden taff railings along the back. Best Regards, -Nate

A few bowspirt updates as well: I removed the secondary boom segment (Not sure what this called... Flying Jib Boom, maybe?) Thanks @Martes for the tough love lol - it sucks to remove a piece I worked hard on, but I gotta admit it does look much more in proportion without it. The spirt was modelled from the HMS Pandora, so it's always going to be a bit big for this Brig, but its definitely within the realm of possibility now. The mesh safety netting has been added as well. This was a first attempt, so not ready for feedback just yet - It was mostly an experiment to see if I could get away with a 2D plane that uses a mesh texture with alpha (transparency) instead of modelling every tiny rope that makes up the netting. I'm happy with the results, I just have to make the sides connect beneath the boom and model the gasket rails along the sides of the main bowspirt - still a much easier approach though. Finally, I've introduced a bump map onto the sail cloth that I'm hoping will simulate the tiny wrinkles that real sails form around the seams. Right now you can see this effect on the jibs, but it's only randomly placed using a noise texture. I will eventually 'paint' the effect onto where these wrinkles actually occur, which should produce stunning results. Sadly, this screenshot also points out all the work that remains to be done on this part of the ship, including: Add the downhaul lines to the jibs Cut out the sheeves and add the underlying support blocks for the bowspirt bees Add the double Jeer blocks, metal hooks, and rigging that connect the anchor stock loops to the catheads Improve the collar cleats on the bowspirt Model the gammoning cleats on the bowspirt Add the saddle for the jib boom onto the bowspirt Wrap the bowspirt woolding with rope Add 2 additional rope loops and thimbles to the tip of the jib boom Decide between knight's heads or a flying forecastle to lock the bowspirt in place

Finally showing some long overdue love to the stern: The stern lantern has been completed, including mounting brackets. The taff rail has been modelled. Newel posts along the back have been placed (still missing supports along the sides, though I want these to look different) Corbel-style relief carvings (baroque ornamentation) added beneath the davits. Floral scrollwork has been added to visually separate the gallery planking from undercarriage, as the angle between these is not overly pronounced. Red painted wood texture has been enhanced to show random areas of scuffing (similar to the yellow wood texture)

A few definitions to start things off. This is a public forum, and in practice many experienced CG artists will use these terms interchangeably, which is fine because we can usually infer the correct terminology from context. I want give precise definitions upfront so any replies to this topic don't inadvertently confuse someone who is trying to learn for the first time. And if you are new to this, DON'T WORRY - it's not as complex as it may initially seem. I promise. Texturing: The art of overlaying materials onto the mesh of your model so that it looks like it's actually made of real-world substances Material: A 3D simulation of a substance that exists in real life such as wood, metal, fabric, etc. In a 3d application such as blender, the term 'Material encompasses the entire node set-up including textures, operators, and shaders. Physics Based Rendering (PBR) Materials: A Physics Based Rendering, or 'PBR' material is generally understood to be any material that is contains more than just color data. The foundation for any PBR material is a related set of textures. Screenshot #1: A set of PBR Textures from the textures.com website. Texture: a 2-dimensional image file (.jpg, png, .tga) that contains some sort of light-based physics information. Most texture files begin as actual photographs taken with a camera. Multiple copies of the photo are made, and each is edited differently to serve a unique purpose in the 3D application. Once imported into the 3D software they are once again combined and processed through a shader, at which point the rendering engine can interpret the data to simulate color, ambient occlusion, roughness, specular, displacement, transparency, translucency, emission, and much more. The thing that all of those have in common is they are all real-world properties/behaviors of light. Shader: A 'hub' node which allows the input of various textures and/or values of certain properties, then combines and calculates the results of the overall material Screenshot #2: The Principled Shader in Blender. Note all the different properties that can be edited. We can set these values manually or import and hookup our PBR textures to the shader. In many cases, the name of the texture file will match its corresponding input on the shader (compare screenshot #1 to screenshot #2). For example, the 'normal' texture file maps to 'normal' input on the shader, and same goes for the 'roughness' and 'metallic' files. If the name doesn't match, for example, the 'albedo' or 'height' files - there's usually a good reason, such as they require an operator between the texture and the shader, or we have a choice on how we use those files (For the experienced folks - I'm talking about displacement vs. bump mapping here) Node: A visualization of any component of a material (i.e. texture, operator, shader) and the relations to other components inputs and outputs. Screenshot #2: Any 'box' in the screenshot below is considered a node. Okay! With then definitions out of the way, let's create some materials for our ship...

Topic Opener: My favorite Blender tutorial vlogger, Andrew Price, once said [paraphrased] "Google the words 'Blender + Car' and I bet you'll find over 1,000 videos on how to model a 3D car, but not one video on how to present that model." Now, I recognize and respect that my audience for this topic may be pretty niche - after all, to the vast majority of members here, the only function of 3D software in ship modelling is drafting usable plans or exporting datasets for 3D printing parts. I'd wager that the notion of a 3D ship model as the final output instead of an input is a foreign or uninteresting concept to many, but I have gotten a few requests to cover texturing and materials, so here we are. With that in mind, I'd like to formally introduce this topic as a way to help one another breathe life into our digital ship models and to be able to display them as evocative artwork in still renders, movie clips or video games. Here's my agenda and I hope to do one section/week. I'm also happy to focus more on certain topics or skip over others based upon your feedback: 1. Understanding PBR (and why it matters) 2. To Plank or not to Plank - Which technique is really more accurate? 3. UV magic: It's all about that scale, 4. Avoiding the Pitfalls: reverse color grading 5. Avoiding the Pitfalls: use of color palettes 6. Cliche's to Live By 7. Special Effects and Overlays We'll dive in soon gang, -Nate

@Lieste Thanks for the feedback. So I'm not sure I complete follow what you are saying, but I was careful to shape each sail from a real-world template (HMS Pandora). It is possible, however, that the wind simulation I used didn't affect the sails the same way it would in real life. Here's a screenshot of the actual cut of the sail canvass with no wind in them Let me know if it still seems off to you and I'll take it under consideration. Thanks!

@Martes - So I got a rather interesting answer regarding our mystery barrels: Turns out they are a fitting for whaling ships and were filled with fire suppressant powder (sand could be used in a pinch). Similar to grease fires, water has little to no effect in putting out fires fueled by whale oil/gasses. Obviously, I'm not doing a whaling ship here, so I'm removing them, but I thought I'd share as its interesting.

Thanks @Martes Okay, so based upon your initial feedback: 1. I have removed the secondary jib boom, which reduces the overall length of the bowspirt quite a bit. 2. I will also re-work the spanker boom and gaffe (make them longer) because that much I can do without ripping the whole thing apart. I'll updated images when finished. Regarding your second post: 1. I'm honesty not sure about the barrels. I saw a photo of a model of the Niagara someone had done where in order, it went: main mast (and rails), grate, barrel storage (they are small barrels, and I would say powder kegs if that weren't so inherently dangerous), then capstan, then the binnacle, then the choice of either: a. hatch for the aft staircase followed by the tiller OR b. the Ship's wheel (for which the gooseneck and pulleys will fit below deck if you forego the lazarette space. Otherwise, you're correct, it has to be a tiller.) 2. Fair enough - this was a guess on my part. I've never been able to see what's in between the hammock cranes and mesh netting. I thought the name was coincidence, or it referred to how the netting was attached. I didn't know the actual hammocks were stored there... wouldn't they get soaked? 3. Already covered this, but using certain configurations, yes, brigs could have steering wheels in lieu of tillers, though I'm not sure how common this was. 4. This is another configuration choice. There are many dependencies when it comes to configuration choices, and I like to think of it like this: If there are deckhouses, a quarter gallery is required because the houses need to be positioned wide enough so as not to impede the swing of the tiller. Deckhouses also preclude the addition of the bridge deck because there is nowhere to climb up. So on my model there is a ship's wheel. The wheel means no lazarette to accommodate the gooseneck. No lazarette means the captains quarters can be on the birthing deck, which eliminates the need for deckhouses (small flag lockers still exist they just don't need their own room). Finally, no deck houses means we can have a bridge deck, for which I still need to model the stairs. 5. Hmm... I'm not sure I agree on that point. Flying Forecastles on brigs, especially those of the late 18th/ early 19th century, were not all that common. 6. Yep, you're right - but it's a lot of work and I'm not looking forward to it lol. 7. Yep, you're right - so how do you feel about Scotland Yard instead (get it?)

A few quick updates on the ship: 1. Gun port and sweep ports texture replaced to a smoother red painted look (compare to the anchor stock texture). 2. bevel added to the front edge of the keel and cutwater. Note: The decorative 'curl' from the upper headrail does extend beyond the cutwater bounds slightly, but any noticeable deformity should be covered up when I attach the figurehead 3. Detailed algae overlay added to the bottom strakes of the tarred planking just above the copper plating. This is only on the hull strakes right now, but I will be extending it around the cutwater on the bow, the stern post at the stern, and onto the rudder eventually. Algae does bleed onto the copper plating as well, but it interacts and visually looks quite different on metal than wood (pinkish streaks or an oxidized patina look), and I have yet to figure out how to handle that detail. 4. Parrels added around the spanker boom and spanker gaff, although I'm not too happy with the current grain. I also need to add the leather wrapping that prevents scuffing and chipping of the mast. That's al for now!

Thanks Martes. So I was definitely playing around with a weird focal length on that particular render, so it probably was the camera angle to a certain extent (my fault, see updated screenshot below). With regards to it being 'undermasted', this is a curious observation I'd like to pick your brain on... If anything, it should have the opposite problem - here's why: The masts, while modeled on the fore mast of the HMS Pandora, were scaled to the exact specs of the Niagara (120 ft above the waterline for the main mast and approx. 84 ft for the fore mast). The Alert calls for a 100 ft length 'deck', which I extrapolated to be around 105 ft from cutwater to taff rail. Niagara, by comparison, is 118 ft in the same dimension. Thus, the masts should be slightly oversized, not undersized. Take a look at the updated screenshot and tell me if it still appears that way to you, I'd be curious to know. Finally, thx for the info on the Sophie. Unfortunately, bc of the time spent modelling the rigging, I can't make any drastic replacements to the masts or yards on this project, but it'll be a good lesson-learned for any future brig I do. -Nate

Quick Comparative shot to self-test my skills. I'd be happy to know what you guys think. Truly - I'm not just fishing for complements here, any constructive criticism is very much welcome at this stage as I approach the final touchup phase and prepare her for use in animation, video games, and still renders. A couple things to note: - My model is actually based upon the hull from the HMS Alert (The 1779 twin-masted Brig version just to clarify because there were quite a few ships named 'HMS Alert' over the years), which is about 10 feet shorter and 34 years older than the US Brig Niagara of 1813. - The Masts, Yards, Bowspirt, Spanker Gaff and Boom were all drafted from HMS Pandora's plans. The main mast and yards on my model were actually from the fore mast of the Pandora (whose dimensions closely matches that of the Niagara's mains. Beyond that, I could find no documentation on Niagara's masts and yards other than total height, length, etc.. - The rigging is bastardized from multiple sources, but I tried to follow the Brig Niagara's where possible. - I chose to use the old-style head rails and cutwater from the HMS Alert. - I followed the classic Royal Navy's color palette. While the Niagara is also black and yellow with red trim, the Niagara was way too 'banana' yellow and saturated for my taste, so I went more towards the HMS victory's look instead. I then played with on overlay intended to scuff up the paint, which gives it a more muted look in certain lighting. Thanks! 3D Model: Brig Niagara:

Just to lend credence to the curve to mesh and bridging method... It's tedious to be sure, but it's hard to argue with the results:

Hi Srenner, So the good news is, to my mind at least, this is a really easy issue to solve. I would simply start by converting the curves to meshes. Go to the object menu -> convert -> mesh. Do this for all of them. Now each curve is its own 2d object. Next, position them in line per the plans (I think you already did this), select all of them and press ctrl+j to join them into a single object. Finally, provided each curve has the same number of vertices, you can select 2 at a time in edit mode and bridge the edge loops. Let me know if this is accomplishes what you were trying to do. Best, -Nate

Thanks for the kind words Kurtis... And yeah, I totally have a love-hate relationship with Boolean, because it always involves retopology after its used. That said, if used before you solidify and/or cut the individual planks, the clean-up is harmless. Regarding your modifier stack, I'm interested to learn how/why you used the geo nodes and the weighted normals? Perhaps you know something I don't, and I'm always eager to learn new tricks... lol Finally, on a slightly different topic, I thought I'd share a screenshot of the last figurehead I did. I think on my current project i'm going to attempt to replicate a wooden sculpture but below is an example of a worn bronze patina look w/ oxidation streaks. Note: which ever route you go, it may behoove you to model the generic character first, and use an armature to pose it on the front of the ship... you'll have way more flexibility this way Best, -Nate

Ok, so I lied about the last post being the final step lol. This really is the conclusion however (I promise) Step 10. Apply all modifiers on all the plane objects, then use ctrl+J to join them into a single object. Tab into edit mode, select two adjacent edges at a time a bridge them (make sure the 'merge' option is selected) To do this go to the 'Edge' menu -> 'Bridge Edge Loops' After bridging: Step 10b. Once the edges have been bridged and merged, we actually want to dissolve the resulting edge. This may seem redundant, but this is necessary because the program has merged the points at the shortest distance and has not smoothed out the connection. Press 'delete', and select 'dissolve edges' from the resulting pop-up menu. Step 10c. Finally, there will be excess mesh on the outermost edges that doesn't follow the curve of the hull. Simply select these edges and choose 'delete vertices' as seen here: Ta Da!!!!!!!! So this is where I'll leave you. From here, I usually just boolean cut the front with the keel object itself (which is always the next thing i model) to ensure a perfect fit, and then select alternate rows from the mesh and separate and solidify to start the planking process. There is always some amount of tweaking involved, but I've found through many failures that doing hulls this way sets me up for success, as its easy from here to create the frames, internal planking, and external planking all from one 'shell' so-to-speak. Hope this helps, -Nate

Continued from above: Step 5. Add a small plane. I usually use 6in x 6in but it should be the approximate width of each plank in the hull. (Note: the example in the screenshots is actually a longboat, so I used a 4in plane) Tab into edit mode and rotate the plane along the 'y' axis 90 degrees, then align it so the bottom vertices are at 0 'Z' location. Finally press 'ctrl+r' to cut it vertically down the middle. Step 6. Add an array modifier with the following settings: - Fit type = Fixed Count (Very important - the number you put here corresponds to the number of planks on the ship from the rabbet at the keel up the sides of the hull to the bulwark cap rail. This should be the same for each curve, so count the number of planks up to the lowest edge of the bulwark, i.e. ignore the forecastle, quarter deck, or poop deck planking for now. - Relative Offset Factor: X=0, Y=0, Z=1 - Merge = Checked (default distance value is fine) Step 7. Add a curve modifier with the following settings: - Curve = use the eyedropper to select the curve you want the array to follow. You will do this for all of them so it doesn't matter which you start with - Deformation axis = 'Z' Note: if the tilt is wrong like in the image above, just tab into edit mode and rotate the whole thing on the z axis + or - 90 degrees, until it looks like this: Step 8. Duplicate the plane, and clear out the 'Curve Object' value in the curve modifier. Then use the eye dropper to select the next curve in the series. Do this for all curves (assign 1 plane to each curve) You will notice that on the narrower sections of the hull the same number of planks will overshoot the curves, and on the widest parts of the hull, the planking may not reach all the way up. Step 9. To Fix the height inconsistencies, tab into edit mode and make the individual planks taller or shorter. We must do this BEFORE we apply the modifiers. When done with all of them, it should look like this:

Looking Great Kurtis, Your project is kinda like ghost of Christmas future for me, as I've decided a 1st or 2nd rate ship of the line will be my next project. I didn't become a member till your project was well under way, but for what its worth I have hull creation down to a science that doesn't require hours of smoothing, subdividing, and sculpting. I'm sure you probably did some of these steps, but I'm curious to see another blender artists strategy: 1. Import three copies of the plans (as reference images), scale to blender world scale accordingly, and align the plans in all three dimensions. Tip: The origin should be the top of the rabbet (or top of the keel, failing that). 2. Trace the lines with a line mesh or a curve (We'll be converting these to curves, but I'm better working with meshes, so I prefer to start with a plane a delete two of the vertices so I have a line segment, then just extrude as needed. untill it smoothly follows the plans. 3. Repeat step #2 for all hull 'slices' on the plan. Switch into plan view (X-Orthographic) and align each line or curve segment to its correct position along the 'Y' axis when done, it should look like this: 4. Select either the front or back curves, tab into edit mode, and with the pivot point set to 3d cursor, type 's' to scale, 'x' to lock on the x-axis, and then '-1' to essentially flip everything to one side. Finally, in object mode, select everything, convert to curves (if they aren't already), and go to 'object' menu -> 'apply' -> 'apply all transformations' The apply all transformations part is vital to the FINAL STEP FINAL STEP is continued in next topic reply...

A few minor improvements made over the last week or so: 1. Cross-lashings (supports the course yard jeer blocks) on the fore and main mast heads/caps added 2. Reef tackle lines for the fore and main topsails added 3. Carronades replaced with higher quality versions. They don't currently appear tied to their brackets on the hull, but I will update that a bit later

New screenshot as part of on-going materials enhancements... The Objective in Layman's Terms: Make the sails look as realistic as possible. The Objective in Geek-Speak: This screenshot was part of a ray-tracing test (i.e. Cycles Rendering Engine) for the sail canvass material, and updates to the underlying texture nodes to achieve the desired results, which include: 1. If back lit, I want to reproduce the iconic 'golden' effect of subsurface scattering properties. To accomplish this, I figured out that the sails cannot simply be 2-dimentional planes; they must have an actual thickness to them. However, because the quilting, seams, and ropes fed through the sails only protrude from the front, I needed to invert the colors of the normal, bump, roughness, and ambient occlusion textures on the backsides of each sail while ensuring that the normals themselves continue to point outwards. 2. Also when back lit, the translucent - but not transparent - properties should become obvious. This is accomplished by adding a transmissive element to the material, and using a mask to ensure its only applied to the thinner parts of the sail cloth. 3 When front lit, I want the 'micro details' (wrinkles) cloth brush to be visible first on a high-poly copy of each sail, then on a baked normal map applied to the original copy. I've only done the Fore Topsail like this so far, but if it works, I'll do the same to the others.

Thanks! Unfortunately, all my work is destined to remain digital. I was born with a disability that severely limits the use of my hands, so I'll never be able to construct the outstanding 'real' models the way you guys do... That said, I can do this:

Rigging! AHH! So. Much. Rigging. Lol.