Sculpting the armour
Production:
Creating the armour:
I started by taking the helmet design that I had created in my experimenting session (#2) as I was very pleased with the helmet design I was able to create. To note, I took the 4th helmet design from my list earlier, since the design was very closely reflecting the reference of the 15th century Italian design I have been following. The only change I have made is the size of the visor, to which I have reduced the size of the facial slit for the eyes of the character to better reflect the real design.
After implementing the helmet into my new Zbrush doucment, I decided to create a basic body for the character. This would not be seen, but I felt that it would help me create the armour so that I could essentially 'layer' the individual armour pieces across the body. This way, I would achieve realistic proportions and fittings for the character's armour. I started by getting a relatively realistic image of a male character as I intend for my character to be male. In hindsight writing this blog post, I figure a better image would've sufficed as the focal angle of the reference image I had used was not so useful and a bit misleading. I also should have gathered reference imagery from the side and top. This is a mistake I should have avoided, but proceeded nonetheless.
To create the body, I used the ZSpheres tool to create a basic A-pose of the character, matching the proportions to that of the reference image. To use the reference image effectively, I used a texture overlay option in Zbrush which allowed me to have the reference image as a transparrent overlay in my viewport, making the process much easier than working on an image to the side of my screen for instance. Once I had created a character using ZSpheres I was happy with, I proceeded to use the 'Adaptive Skin' option which turns the ZSpheres into a mesh. From here, I was able to use the basic tools within ZBrush (Standard, Clay and Move) to create a very basic body shape for the character that I was happy with.
From here, I started working on new pieces of armour, paying close attention to my reference. I started from the chest since it is the biggest single piece, I wanted to work center -> outwards as I figured it would be easier to layer the armour pieces this way and not the other way around. I began with the basic shape. I achieved this by inserting a new primitive shape (a cube) and moving the basic shape with the Move tool.
In order to achieve the sockets for the arms and neck, I used a boolean operation to create this. The process involves using the Dynamesh tool within Zbrush to enable the effect, which also provides a refinement and more resolution to the geometry of the mesh. I did this for the chestplate mesh and a new primitive shape (cylinders) which pertruded from where the sockets for the arms and neck would go. I then set the cylinders to Negative in the subtools panel, and merged the subtool with the chestplate subtool. After simply dragging with the mouse to remove the 'mask', the boolean operation was complete. This left some holes in the mesh from the inside and outside, but this was easily fixed by re-applying the dynamesh option. I then proceeded to smooth the result and move it into a shape which fit the body nicely.
From here, I started adding plates for the arm. This included a shoulder plate complete with a stop-rib (The plate pertruding outwards, mentioned in my success criteria), Rerebrace and Vambrace (upper and lower arm plates) and a Couter (at the elbow of the arm). I continued my technique from earlier, and used the boolean operation too. I also used the Pinch and Flatten tools to get seams and finer edges/finishes to the plating to make it seem more realistic and solid material.
Though, I am personally not keen on my design for the Couter and found it the most difficult to design due to its complex shape. I decided to continue for now and come back to it at a later date.
To complete the arm, I added a Gauntlet to the hand. I tried my best to replicate the design used in the reference I was following. Though the design features small plates connected to the fingers of the glove of the wearer, and I had not included fingers in my basic body shape I had created earlier, I figured gauntlets of the Middle Ages used gauntlets where the fingers were all one plate and the thumb was separate, but I was mistaken. This is something I can easily ammend later in the project. For now, I will continue on refining the plates and continue applying them to the rest of the character.
I started to add some basic details and definitions to the armour; Most notably the seam central to the chestplate as well as the Fauld/Tassets section (below the waist, above the legs seen in the image). The Fauld is usually made up of lots of individual plates to offer more movement for the wearer, but I did this using the Pinch tool to basically pinch lots of the geometry of the mesh at a point to give it the appearance of multiple parts. However, I do feel like the look is mismatching from the other plates which are evidently separate subtools/layers, so I will come back and separate this later on in the project. I felt that delaying some of these tasks might be better than refining one small area, especially if I decide at a later date that I wish to further change these areas or undo my changes.
I also started to add some other layers in the model to give it more depth: The belt around the chestplate. This is how the armour would be worn and kept together in the real world, so it will be necessary to include such aspects in my design. I am happy with the result, and works well like the strap on the helmet design from before.
This was the work I was able to achieve before Term 2 begun. I am pleased I was able to make some progress, especially when ironing out the basics.
Now in term 2, I continued progress on the armour. In the above screenshot, you can see I have added an additional smaller Tasset plate which is included in the reference armour.
I found it difficult to model the Sabaton since I did not find much reference imagery of the design for the 15th century Italian armour design I am following. I may decide to improvise the design here later on to ensure realism over accuracy of the reference armour.
Here is an image to show the inside region of the leg. As you can see, the Fan-Plate only appears on one side. It is worth noting that any gaps of the armour such as between the Cuisse and Greave will be covered with chainmail, which I will achieve by using some simple geometry with what will most likely be a chainmail texture applied to it.
During the modelling of my character, I was beginning to notice a problem in the proportions, specifically the leg of the character which will be shown below. The image above is one of the other various references I was using to design my character; This is a 3D model of a suit of armour which is seen here. I found this source to be highly useful due to the 3D view of the armour, as well as assessing how it was made to potentially influence my workflow. On this particular occasion, I used it to compare my model to notice the obvious flaw in the legs of my character, which were way too short and unrealistic, even when accounting for the difference in focal angle of the camera view in this source.
BEFORE AFTER
Here is a before and after shot of the proportions of my character. In the left image you can clearly see that the proportions are flawed, and the knee-joint is far too low. This was unfortunate as it meant manually adjusting all the armour pieces from below the leg region, as well as adjusting the mesh to fit and moving the geometry of the body underneath the armour. However, I am glad I spotted this problem now rather than later in the production, which could have been disaterous. This shows how important it is to get reference from all 3 orthographic views when developing proportions, not just the front as I had done before. Thankfully, I was able to adjust the model appropriately to result in the right-most image shown above, which I think have much better proportions. Sadly this did impede on my progress, but was necessary.
I started refining the arm from before. In essence, I went to fix the elbow reigion of the armour, the Couter. I started by adding a few invidiaul plates like I had done for the knee to give some well needed depth. I then completely redesigned the Couter, as I was not happy with the original design. I did this in invidual pieces - again like the knee from before - which the depth certainly helped the design to appear more realistic.
Here is the Couter design from the front/top. The Couter looks quite close to a lot of Couter designs in other armours too. However, I still wasn't absolutely keen on this design as it still felt quite clunky and big, despite this part of the armour naturally being a larger piece.
Instead, I decided to polish the gauntlet of the armour by adding fingers to the model. I did this by removing the 'placeholder hand' from the basic body shape I had initially created, and made a separate subtool for the hand to serve as a Glove to be worn with the plates attached/sewn into it. This appears to work quite well and I am happy with the basic design and shape here.
I still wasn't happy with the Couter design, so I proceeded to re-design it again. I understand and acknowledge that perhaps this was wasted time, especially since it took me some time to create these designs each time being a more complex shape. However, I was personally sold on this design since it felt more flexible and less clunky.
Here is the design from the front. I also added a strap to the design which I felt really help to sell its believability by adding that extra layer of depth. I believe that little details such as this can really help to sell something as more believable and realistic. The design is more flexible too, and should offer more freedom when moving the arm, which I believe might be more useful when wielding a large two-handed sword that my character will possess, though I must admit my knowledge is limited on this so I cannot say for certain. I decided to keep this design for now, leaving a copy of the previous design as an older save so that I could come back and re-use the Couter from before if I change my mind, or if it is recieved more positively from any feedback I might recieve in the coming weeks.
During one of my mentoring sessions, I had an opportunity to show my character off to someome in the industry with a lot of VFX and CG practice and knowledge, with the intention of getting highly critical and useful advise. I have made note of all of his suggestions and recommendations, but the most immediate one was regarding my design for the Stop-Rib in the above image. He pointed out that the Stop-Rib would most likely be a separate plate attached to the Pauldron (Shoulder), rather than being a part of the same piece as I had done before. Upon checking my reference images, sure enough this was the case too. I had overlooked this small detail. I decided to take this advice and quickly added the Stop-Rib as a separate subtool, removing the previous one.
___________________________________________________________________________________
Experimentation - Self session:
During production, I was having some concerns about my intended workflow when modelling. Initially, I was thinking that I would create all of the details in Zbrush and then retopologise it in Maya to then be UV mapped, textured, and so on... However, my discussion with one of my mentors raised some valid concerns with my approach for the workflow. Whilst I see myself as good when it comes to retopology, naturally you will be losing a lot of detail that would have been achieved in Zbrush whilst sculpting the details very finely and at a high poly-resolution.
I was originally planning to mimic the workflow I had carried out during my Experimentation session #2, since that had gone well. However, my mentor was able to give me very valuable critique due to his experience in the industry that had not been raised by me nor peers prior. The finer details I had included in the helmet - besides the strap at the back - were all done in Substance Painter, via texturing. This was something that was critiqued since texturing doesn't give the same level of detail when it comes to sculpting the detail with geometry. My thought process was further limited due to the reference of the 3D armour model I have displayed earlier in my blog, which was very reliant on texture work.
It was explained to me that this sort of process is better applied in game design, where textures are used for efficiency in the real-time rendering of the game engine. In CG/VFX work, you go for as much detail as possible to get the best quality output possible. In essence, I should be creating all of these details (Within reason) in the geometry of the model. The textures should be used for areas of 'High frequency' - so for areas very zoomed in and close up which will benefit from high resolution texture work.
From this, I was able to take away that my prior workflow was limiting and constricted, as I would not be getting a substantial amount of detail through just texturing, especially not for my project and scope of work. As a result, I needed to rethink my workflow from the modelling process to the UV mapping stage, since work beyond this point would not need to be changed (thankfully).
I decided to experiment with a method of workflow which might be more suitable, and see what results I could get. My plan was to adapt my workflow so that I am creating the base geometry of my characer in Zbrush, which would be labelled Stage 1. Once this has been done, I will retopologise the armour so that I get nice and clean topology in the model, and straight edges. I have been struggling to get nice and flat edges in my model so far, and have been heavily reliant on the Pinch tool which essentially pinches the geometry together at a point, which I have been using on the edges of the armour plating to give it a more realistic thickness, especially as this was a point raised during my first experimentation session. This retopology pass would comprise as Stage 2. So far, the workflow is the same as what would have occured but does not account for the finer sculpted details.
For what I am calling Stage 3 of the modelling process, I will be re-importing this now-retopologised mesh back into Zbrush so that I can add finer details onto the model. The reason for this is so that I can clean up the topology of my model and give it straighter and cleaner edges. Moveover, during my second experimentation session, I had tried to import all 4 helmet designs I had created. However, I found that the geometry of atleast 2 of the helmets were very glitchy, and had lots of visual errors such as backface planes appearing across the model. This was a result of using the Pinch tool, and unfortunately this did not show in the Zbrush viewport, but did in Maya. I do not want to be in a similar situation whereby I have worked very hard on my model for the visual result to be glitchy and visually flawed. This is why retopology is essential for my workflow, but needs to happen before the sculpted details are applied as to preserve those finer sculpted details.
Another issue I have had is dealing with resolution in Zbrush. I have been using the Dynamesh tool which aims to improve the shape of what you are creating, and is useful when using certain processes such as the Boolean technique I had used in production so far. Unfortunately, if you wish to add more detail to a subtool using Dynamesh, you must increase the resolution. Doing so will reset the Dynamesh tool, which loses lots of detail in the mesh you apply it to, and the resolution seems to dynamically change depending on the scale of the object, making it hard to maintain a sense of continuity in the geometry. Additionally, the geometry appears to be more suited for organic modelling, the edges and overall shape tend to seem more curved and organic, but this could be wrong and is based on my own take from using the tool. Finally, when applying Dynamesh to a particularily thin mesh, the mesh will develop lots of holes since the tool must be assuming the topology is wrong or incorrect, and is trying to compensate. Though this has been only causing me problems, and has set me back a couple times already. This is another set of reasons for the change in workflow.
To summarise, I would start by creating the basic shapes and layers of the armour (Stage 1). Next, I would retopologise all of the individual layers and shapes appropriately in Maya (Stage 2). Finally, I will reimport the now-retopologised model back into Zbrush, where I will be able to add lots of finer details. This would be exported back into Maya to be UV mapped, which will end (Stage 3) and finish the modelling process for my character. It is a lot of back and forth between programs, but on paper it sounds like a solid system. I decided to test this using the helmet I had created and retopologised in my experimentation session #2:
After importing the helmet into Zbrush, I had to clear its applied textures (I had used a UV Checkermap for diagnosing the scaling of UV Shells) before proceeding. The process worked smoothly, all individual layers remained and the geometry worked as intended.
To get more detail without using Dynamesh, this time I was simply able to Divide the mesh, which subdivides the geometry to give more polygons and thus more detail. I did this a couple times and tested some alpha brushes and other default brushes to see what sort of details I couple create. I was able to do some nice details and the process seemed to be working well so far.
Here is the view from the side. I was using some preset alpha brushes to create details, which would've worked better had I used more subdivisions. This raises a question regarding performance, whether using this many polygons in a model with many layers will run on my machine or not, despite my machine being a high specification. This is a downside of COVID-19 as I will be unable to access the University computers for some time if it were to be necessary, this is pure speculation at this point. If this would be the case, I would hope to be able to apply for exentuating circumstances.
Here is a view from a side-angle of the helmet, using the default grey material. So far, the results seem promising.
Dynamesh: OFF Dynamesh: ON
Using the SHIFT+F function, I was able to display a wireframe view of the model. I decided to use this as an opportunity to test my theory regarding the use of Dynamesh. You can see the difference in the above screenshots when Dynamesh is applied. Detail is naturally lost, and this is to be expected. What I found interesting was that the topology of the mesh is almost completely rotated, as if to match the world axis rather than the original topology. A result of which causes the horrible bumpy edges at the bottom, as polygons are being stretched diagonally from the outside to the inside of the plate. There are also some Tris present in the mesh, but this shouldn't be a problem here since this mesh would not deform when animated. All in all, it seems advisable to avoid the use of Dynamesh in 'Stage 3'.
At this point, I was reassured that this method would serve as a suitable workflow that I could use. I am currently thinking that certain embossed areas might be better to be done using masks or perhaps done in Maya, but I am unsure. As a result, I have decided to await an upcoming feedback session this week to see what my tutors may advise/recommend.
___________________________________________________________________________________
Resuming Production:
Recapping from before my experimenting session with extra details, I had previously talked to a mentor with industry experience; He had suggested to create the stop-rib of the shoulder as a separate model to indicate realism, as the stop-rip would be a component separate from the shoulder. Having this separation would help me to create a more defined shape, one where the viewer can easily see the plates are separately made rather than all one piece and thus would achieve more realism and be more detailled.
I decided to use this approach for the rest of my design, as prior to this I had been using the Pinch tool in Z-Brush to create the appearance of indents where plates might be separated. In the image above, I have refined the tassets of the armour so that they are now separate meshes and no longer one whole mesh. This helps them look more realistic as they have more defined edges where they would realistically be separated. Furthermore, separating the mesh like this allows for better potential with the rigging of the character, as these could be animated to have independant or more flexible movement.
I continued this approach to appropaite areas of my armour. The image above demonstrates how I have done the same for the Sabaton (protective foot-wear) of the character in which I have created them using separate 'pieces'. Again, the result is more detailled with more defined edges to the mesh.
I wanted to add extra bits of detail to the model. I went ahead and added a strap for both knees which would theoretically hold the knee plating components in place around the knee of the wearer, again adding realism to the design. I added this similarily for the elbow designed prior.
I refined the mesh of the Hand Cover, adding a curved edge to the back of the cover which connects to the sides. The idea being that - when the cover is bent at the clenching of the fist for instance - this curved edge covers what would otherwise be an exposed gap in the armour plating. This helps to refine the model better so there are no large gaps between places where unneeded.
Whilst revisiting my main source of inspiration/reference, I noticed that the back of the upper leg plates were not fully enclosed cylindrical shapes as I had modelled already. Instead, there is a gap in the plates which are connected by straps, which - presumably - allow the wearer to attach them onto his/her leg. I had failed to realise that the shape I had currently used would be impossible to wear. I decided to revisit my design to better reflect what is seen in the above image.
I have made some very large improvements to the back of the leg to better reflect a more realistic and functional design. I have removed a large section from the plate, and connected the gap with straps. These hold the large place in place. I believe the refined design is much improved, not just for functionality but I would argue the design is much more interesting and varied to add some visual appeal to the design. All in all, this was a largely successful modification to the upper leg that I ultimately stuck with.
Here is a view of the revisited upper leg from the front to better show how it will look from a front-perspective. The straps can easily be seen from the front and thus the details should not be missed, whilst also not being too intrusive.
I still wasn't happy with my stop-rib design. Despite it now being a separate plate, I felt that it was way too large. This is something I had confirmed when looking at reference images which typically used a shorter stop-rib (length-wise). This makes sense from a functional standpoint, as a sword would hit a stop-rib even if the length was not covering the entire length of the shoulder plate.
I revisited my design for the stop-rib and shortened it significantly. The stop-rib is now a very simple shape covering mostly the front region of the armour where blows from opposing forces would be most likely to strike.
Here is the refined stop-rib design from the side. I believe the design is much improved as it looks much more tidy and efficient from an aesthetic point of view. This is also a similar size that I had seen in many reference images, suggesting that - despite being smaller than the previous iteration - that the functionality would still be retained and significant. I decided to use this improved design.
Upon my research, I discovered that helmets - specifically Armets (Armet being the type of helmet I am designing) would typically feature a Rondel on the back of the helmet, a protective circular metal disc. Oddly, despite researching into the purpose of this component, I was unable to find any accurately cited functional purpose of a Rondel, though it is suggested that it may protect the strap which holds the helmet in place. Since this is a common occurance in Medieval armour designs, I decided to include it to add a sense of realism to the design as well as potential functionality. I did have to modify the strap of the helmet and raise it a little to incorporate the Rondel and the shaft connecting it to the helmet.
In other reference imagery, large plates around the neck can be seen. I cannot explain why the designs differ, though I would argue that the addition of these neck plates would help to protect the neck of the wearer better, much more so than that of the previous reference image. I decided to incorporate these plates in my design to improve my armours functionality (and thus its realism).
Here is my implementation of the plates around the neck of the wearer. I believe I have done a good job at recreating this neck plating, using two separate meshes to potentially offer better movement when rigging the character as it could help the character to look down/move their neck better.
Next, I added some basic straps around the tassets of the armour. The purpose of these straps were to hold the tasset plates in place, as I had mistakenly overlooked this detail and - without these straps - these large plates would be essentially floating. These were using similar designs to other straps used across the model, such as the knees and elbows seen prior.
From this point, I was pretty happy with my design. I decided to mirror the armour so I could preview it and see it better before moving on to the modelling of the sword.
Front view:
Back view:
Front/Side view:
Arial view:
Overall, I was pretty pleased with the design. Once mirrored the armour felt much more protective and proper, giving me a better overview of its design. My decision to create separate plates as separated meshes showed its effectiveness, as you can really tell that all the parts of the armour are separate and pieced together to make a suit of armour, which I believe adds a huge amount of realism to the design.
However, I still wanted to improve the design. I felt that it was a little too uniform, and needed a bit of variety to help lift the design from being too symmetrical.
After revisitng my reference images I noticed that one of the shoulders varied from the other. Upon some light research, I understood that the left side of the knight is more plated and protected as the right side of the knight would wield the weapon and thus require the most mobility and freedom to move. As a result, the shoulder on the left-hand side covers the entirety of the armpit reigon, a part of the body which - from my research - I know is one of the most vulnerable spots of a knight.
In addition, a part I had failed to implement as of yet were pieces of chainmail. Most notably, covering the pelvis region and the back of the neck. I had been giving consideration as to how I might go about implementing these, in consideration to the rigging/animating of the character. I had considered creating individual chain links manually for incredibly realistic shape and movement, but ultimately dismissed this due to both time to create and render/animate. Instead I decided that using a plane with an appropiate texture would serve my needs. I had experimented with nCloth in Maya in my Pre-production section of this blog, so I knew that I would be able to attach a plane to an object and have it deform automatically via simulation, rather than having to animate it by hand which could be very time consuming to get it looking right. Ultimately I wasn't sure if this was the right decision, but this would be the most reliable due to having some knowledge in using nCloth and that this could be feasible via cloth physics.
After this deliberation, I went ahead and refined parts of my model. I started by modifying the left-hand shoulder of the character, allowing it to cover a larger region of the character and thus be more protective for the wearer. This also added some variety in the armour design.
Afterwards, I added a plane just below the tassets, covering the pelvis region of the character. The plane would connect to the bottom of the chestplate and hang using a constraint in future. This would serve as added protection for the wearer too, protecting the pelvis region which was otherwise previously rather exposed. This helped to bring my armour design more in line with what I was aiming for.
Furthermore, I did the same for chainmail covering the back of the neck. This would be connected to the bottom of the helmet and would rest at the top of the chestplate. By comparison, this piece of chainmail would not be seen as often nor require too much deformation. This, again, improved the functionality and protective nature of my armour design.
Scabbard - Design and discussion on functionality:
My main reference of armour design features a scabbard around hips
For my character, I wanted to include a Scabbard in the design, in which "the scabbard is an item used for storing a blade" (Angus, n.d.). These were very often carried to hold the weapon of the wearer conveniently. The inclusion of a scabbard would add some believability to my design, whilst also adding some variety as scabbards "were primarily composed of wood or leather in the medieval period" (Ibid.), thus the material would differ from the otherwise repetitive metal materials which will be used.
Hip scabbard example
Most Scabbards which were used during the Middle ages were "fixed at one’s hip" (Ibid.) due to the ease of drawing/re-sheathing a sword from the hip. They were much more effective in terms of combat, since "The sword would be drawn in a defensive position and allowed for extremely versatile maneuvers independent of the location" (Millins, 2019).
Unfortunately, during my research I came across a bit of a hurdle. To recap, I intended to feature a Greatsword as the weapon in my design, which was considered "a massive sword with a broad blade and usually two lethal cutting edges" (Greatsword, n.d.). This was an issue as Greatswords were simply too large to be held in a Scabbard fixed at the hip. Because of this, Greatswords would either be carried over the shoulder or transported to a battle, where it could then be used. It was looking like the inclusion of a Scabbard might be out of the question for my character.
I came across some interesting research on Back Scabbards, a type of Scabbard carried across the back of the wearer instead of beside the hip. These were rarely used since drawing a sword from the shoulder (as opposed to the hip) "limits you to blades typically not more than about 18 inches or so, depending on exactly your arm length" (Barrett, 2020), where "with a blade that short, you can carry it at your hip where it will be much easier to draw" (Ibid.). As mentioned earlier, Greatswords - being too large for a Hip-Scabbard - would usually be carried to the battle, either transported or ovcr one's shoulder.
Back-Scabbard example (Leather Back Scabbard, n.d.)
However, Back-Scabbards could be used to help carry a Greatsword into battle. They would not be used to draw from as they would be largely impractical, but could help a person carrying such a heavy sword in a relatively convenient manner. In fact, the Scottish Claymore (a type of Greatsword) would be carried in a Back-Scabbard due to its large size preventing it from fitting in a traditional Scabbard attached to the hip. A good example can be seen in the image above, which also highlights how these Scabbards different to a traditional Hip-Scabbard; Instead of covering the entirety of the blade, they would only cover the tip of the blade in order to allow for a larger sword to be unsheathed, since the sword could be pulled over the shoulder much sooner - only needing the tip of the blade to be unsheathed before doing so. Whilst this style of Back-Scabbard would offer less protection of the blade, it would be a relatively accurate and feasible means of carrying a sword too large to be fixed to the hip instead.
From my research, I have found that Back-Scabbards would rarely be used to carry a sword, since a Hip-Scabbard would be much more practical for both carrying and drawing from. However, due to the large size of a Greatsword - my intended sword design - a Back-Scabbard could be used to carry a Greatsword for the purpose of convenience. Despite Greatswords more commonly being carried over the shoulder or transported to battle, I decided to implement a Back-Scabbard in my character design purely for adding some asymmetrical and material variety in the design. Although these were uncommon they would still be feasible. In order to create one functional, I referred to the image above in which a Greatsword would rest on two hooks at the top of the scabbard, with the tip of the blade encapsulated in preferrably a wooden sheath at the bottom. This would allow a Greatsword to be pulled out of the Scabbard. By comparison, Back-Scabbards which cover the entirety of the blade are impossible to pull a sword out from, and thus this design should definitely be avoided.
Here is my implementation of a Back-Scabbard. The design is very representative of the one shown prior, being carried over the shoulder with a large wooden sheath at the bottom to hold the tip of the blade in place.
From here, I believe the modelling of the armour of my character is complete. Here is a turntable of the model:
Bibliography:
Angus, H., n.d. Scabbard · Medieval London Objects. [online] Medievallondon.ace.fordham.edu. Available at: <https://medievallondon.ace.fordham.edu/exhibits/show/medieval-london-objects/scabbard> [Accessed 4th Febuary 2021].
Millins, J., 2019. Why are some swords held on the waist, and others on the back?. [online] Quora. Available at: <https://www.quora.com/Why-are-some-swords-held-on-the-waist-and-others-on-the-back> [Accessed 4th Febuary 2021].
Lordsandladies. n.d. Greatsword. [online] Available at: <http://www.lordsandladies.org/greatsword.htm> [Accessed 4th Febuary 2021].
Barrett, M., 2020. Are back scabbards for swords actually viable?. [online] Quora. Available at: <https://www.quora.com/Are-back-scabbards-for-swords-actually-viable> [Accessed 4 February 2021].
Outfit4events. n.d. Leather Back Scabbard. [online] Available at: <https://www.outfit4events.com/eur/product/10574-leather-back-scabbard-baldric-for-claymore-longswords/> [Accessed 4 February 2021].
Comments
Post a Comment