Game Production

Week 1

Introduction to 3Ds Max

Today I was introduced to the 3D program of sculpting, modelling, rendering etc. - 3ds Max. I was taught the fundamental shortcuts necessary for utilization of the software, such as movement (W), rotate (E), and scale; this was communicated along with the user interface. This is the software we will mostly be using for modelling the foundations of 3D assets, which we can add onto by sculpting with programs such as ZBrush and Mudbox, finally putting it into a scene with Unreal Engine.

Table Model Tutorial:

I will demonstrate my working process for creating a table here step-by-step with the inclusion of annotation, tool explanations, and shortcuts for future reference.

To begin, I opened a new file and navigated to the topmost bar and clicked the "Customize" dropdown menu, selecting "Units Setup".
From here, as the image shows, I checked "Metric" and selected centimeters for my model to be scaled in - this is due to Unreal Engine's use of centimeters as the default measurement as we will frequently be transferring models to and between software.

Once this was checked, I continued with the model creation. This is a task imperative to using 3ds Max as you must check this each time you open it, with it in my regular routine it will become habit to do this.

Following this, I navigated to the Command Panel on the far right of the interface and clicked the plus sign to create a new model. Selecting the box in the standard primitives, I had a new base to work with. I input the given measurements in the keyboard entry and parameters; finally, I had a physical model on the grid to start adjusting.

These are the measurements used for my base model, the keyboard entry section refers to the general magnitude of the shape with the coordinates preceding that. The segments in parameters refer to the grid lines within the model.

The measurements previously listed created this model - to edit this model properly, I right clicked on it and entered the dropdown menu for "Convert to" and selected "Editable Poly". This would be my base.

Then, I pressed 4 on the keyboard to use the face editor shortcut, in which I held the Ctrl key and selected four faces to bring out. I then scrolled down the modify panel on the far right to the face editor tools, in which I used extrude at 40cm to create table legs.

Next, I used "R" as the shortcut for scale to cinch the ends of the table legs and give them a curved effect.

Finally, I used the "Selection Centre" option under the "Centre Flyout" menu of the top toolbar and scaled outwards to flare the table legs out and demonstrate weighted balance.

Dev Log Evaluation:

Today I was taught the fundamentals of 3ds Max and created the model of a table, I believe this was a good start to using the software as I have relevant knowledge from using Maya in college and I'm following steps accordingly; I look forward to using these tools for builds of higher complexities.

I believe I struggled in following the steps at a quick pace but once I become familiar with the program, I will improve in this aspect.

Experimenting in 3Ds Max

Given the fundamentals I learned on the 30th, I applied my knowledge here in an independent task of creating an asset of our choice - I chose to make a 3-tiered fountain with a heart on top. For this task I primarily used the technique of scale and extrude to create the model with the addition of the connect tool and other methods.

Fountain Model Tutorial:

To begin, I entered the "[+]" icon in the top left of the viewport, going through the dropdown menu to "Viewports Configuration". From here, I entered the "Statistics" menu and ticked triangle count, total + selection, and show statistics in active view. This shows me how many triangles I'm using in my model so I don't exceed the budget limitation and cause performance issues and potential crashes, losing work.

This setting is temporary so it's necessary that I build the habit of adjusting this every time I open 3ds Max.

Next, I selected a cylinder in the "Create" menu under the Command Panel with these measurements, particularly with 10 sides to ensure a low poly model. This is how I implemented the model, with a short stature to establish the thickness.

I right clicked the model and entered the "Convert to:" menu and selected "Editable Poly" so I could begin modelling this asset.

Next, I pressed 4 to enter face mode and selected the top plane. I planned to use the scale + extrude method to model most of this asset, as I can hold the Shift key while I scale/move/rotate to extrude any selection without pressing the extrude tool and inputting measurements, the latter method is better for precision. This image shows me extruding the top plane and scaling it inwards to prepare the bowl of the fountain.

With the bowl created, I held the Ctrl key and selected one face, and double clicked the next which selected the whole ring of faces - that way I could continue working on the bowl.

With the previous extrusion, I selected the bottom of the bowl's plane and held shift to extrude with the movement tool selected - that way I could deepen the bowl. Following this, I scaled the bowl inwards again to begin creating the central pole.

[I used the movement tool and brought up the pole to an ideal height before modelling the rest of the fountain to understand how far I wanted it to go.]

Here you can see how I repeated the same process to create 2 more bowls of exponentially decreasing size, with a blocked out shape on top as the foundation of a heart. This base model is now ready for detail implementation and mesh error/bug fixing.

Approaching and Fixing a Mesh Error:

Viewed from the top, it occurred to me that I made a mistake in scaling the first bowl inwards; it is apparent that the bottom bowl's exterior was scaled with the interior triangle that applies all axes for an equal transformation. However, I believe I only scaled the interior vertically which led to this elongated effect that carried through to the rest of the model.

I approached fixing this issue by viewing the model from the top and selecting the faces of the bottom bowl's exterior and adjusting them to the others as this saves time compared to fixing 80% of the model instead.

I've demonstrated here how I fixed the issue by scaling the selected faces on the X-axis to thin it out and become parallel with the rest of the model and therefore resemble a wider fountain over an equidistant one.

Before and After

You can see the apparent improvement through this adjustment compared to the previously inaccurate and distorted model.

On top of the heart's foundation, I implemented vertex connections so I could sculpt it into a heart - this was also done to prevent mesh errors as having flat faces with many edges can cause bugs (known as "ngons"), so I must add these lines for that purpose too. I connected the top three to the bottom three using the connect tool within "Edit Vertices" under the modify panel.

Finally, in edge mode, I selected the middlemost line atop the heart's foundation and brought it down with the movement tool which created the illusion of a heart and successfully confirming the model's completion. Yay!!

Dev Log Evaluation:

Today I applied the knowledge I gained in the 3ds Max fundamentals lesson to independently create a 3-tiered fountain.

I believe this task was a success due to my previous knowledge of creating assets similar to this in Maya, but that can't be discredited as this is an unfamiliar program; therefore I am proud to have done this with no necessary feedback. My approach to the inaccurate model issue was efficient and saved time, so I believe I've improved in making executive decisions independently.

Although, what I could do to improve this mesh is by garnering feedback from peers or tutors by considering suggestions for extra details or understanding where my design could be lacking in visual interest. This would test my mindset confronting criticism and build opportunities to adapt and improve my work, strengthening my portfolio.

Week 2

Creating a Chest

This week, we followed a tutorial of creating a generic treasure chest to understand the fundamentals of its construction and the tools necessary to do so. This way, we can model our own chest with heightened confidence.

I began with a box primitive, adding to the width and length segments to attain enough faces for an effective foundation to build off of. I followed this up by converting it into an editably poly so it'd be ready for modelling.

Following this, I entered vertex mode and scaled the bottom row inward for a slight curved effect, along with scaling the middle row outward to replicate the shape of a chest.

Next, I selected the top faces in face mode, using the inset tool at around 2cm to establish the area where the cavity will be. I then held shift and dragged the selected area downwards to extrude it, finally pressing delete to achieve the hole.

Next, I added extra dimension and slight curvature by selecting each corner's edges in edge mode and using the chamfer tool at 2 segments for this effect.

In vertex mode, I selected the two vertices from the chamfer and the vertex diagonal to it, pressing connect to remove the possibility of ngons that would cause mesh errors; I repeated this process on each corner.

I selected the four vertices of each inner corner and used the weld tool to bring the corners to simplify the mesh.

Once again adding extra curvature to the chest's rim through border selection and the chamfer tool.

Finally, I selected the faces of each side of the chest and used the inset tool to add a rim to each side where metal plating would be, the insides featuring wood.

Selecting certain edges and pressing the connect tool repeatedly led to a rough shape defining the lock.

To prevent ngons, I used the target weld tool and connected the middle vertex at the bottom to the left corner. I repeated this with the other side too.

Once again, I used the connect tool to prevent ngons at the top here.

This is the result of the lock.

With these faces selected, I extruded and scaled inwards for extra dimension and depth.

Clicking edge constraint and selecting full edge loops, I clicked the X button in the "Edit Geometry" tab to normalise this loop for X, which straightens it out.

This is an example of how normalizing the edge loop for X straightens it out.

-------->

This is the model with the edges straightened out, looking much neater.

For the lid, I began with a cylinder primitive turned on its side, removing the bottom and the sides.

Selecting the side's edges, I used the bridge tool to bring back a surface. Following this, I used target weld to cover it completely.

With the sides created, the line down the middle on each side creates an opportunity to connect the bottom. To begin, I used border selection and the cap tool to fill out the underside. With the previously mentioned central lines down each side, I entered vertex mode and selected each vertex at the end of the line, using the connect tool to draw a line down the underside. I next used the cut tool to add horizontal lines corresponding to the segments atop the cylinder, preventing ngons.

Selecting the corners in edge mode, I used the chamfer tool at around 2cm with 1 segment for some extra dimension.

I then repeated the connect method of creating the lock on the chest here, followed by selecting all the faces within the top and using the inset tool to define the outer rim.

Applying the FFD 4x4x4 modifier to the lid and with "control points" selected, I was able to give the lid a more dynamic and curvy shape.

This is the lid after using the FFD 4x4x4 modifier.

Finally, I selected all of the faces on the underside and used the inset tool to establish a hole in the lid. I followed this up by extruding it upwards for hollowness and scaling it inwards for the final detail.

This is the end result of making the chest. Yay!

Dev Log Evaluation:

There was a lot to take in and remember for this tutorial and I actually missed the lesson where we built the chest. The amount of information was overwhelming and difficult to keep up with, but for the most part I followed along with it quite well; I'm happy with the end result and I'm also proud of the fact that I did the bottom half completely independently. When the session's playback was uploaded, I took notes and did it by myself after the next class and I carried it out efficiently with no problems at all.

My only criticism is my ability to follow along with tutorials as if your mind wanders for a second, you'll be lost immediately; although I did my best to concentrate and follow along. The final struggle I faced in following the tutorial was just getting lost in the interface as we got introduced to many new tools, however I believe this issue will naturally solve itself as I familiarise myself with 3Ds Max's array of tools.

Other than that, I'm pretty happy with how this went and I'll take what I learned here into the next session where we create our own chests. I want to challenge myself and completely start from scratch, deviating from this model. It'll be hard, but I think the knowledge I gained today will aid me thoroughly.

With the tutorial of a generic treasure chest complete, I began some research to consider the theme I want to portray on my custom chest. I achieved this through a mood board demonstrating two concepts I was drawn to, compiling relevant images such as artwork, materials, and 3D models from artstation with a respective colour swatch and annotation, using analyses of real game asset examples.

Mood boards are a great way to begin a project thanks to the prompting of idea generation and visualisation of ideas through images - this gave me a great idea of how to paint my textures and what colours to choose.

Week 3

Starting My Chest

In this session, we began working on our custom chest models. Thanks to the mood board I created and the knowledge I gained from previous tutorials, I had a good idea of where to take this project; before modelling, I made an initial sketch to visualise my chest in my head so I wouldn't be clueless at the beginning.

To begin, I created the base using the same process as before; implementing a box primitive and adding 3 length segments, 1 width segment and 2 height segments for a proper opportunity to alter my model. This screenshot from before is essentially what my base looked like before I made my own alterations.

Finally, I selected this edge loop and scaled it outwards to get an overall more curved, organic shape for my chest.

For the lid, I repeated the same process as previously demonstrated up until this stage. From here, I made my own customisations.

I then right clicked the FFD button and clicked "collapse all" to finish the use of this modifier.

Before and After Feedback:

Before I began my unwrap, I received feedback for my model so I could improve upon it. The primary critique was to add complexity to the topology, I believe hearing there would be a budget for the tri-count intimidated me into simplifying the mesh to its detriment. My tutor added to the face of the chest, as the bottom half was generally fine; he also got rid of ngons I missed. He brought the nose out and defined the brow so it'd resemble a face a little stronger.

In regards to the horns, he advised that I detach them so they're their own separate asset and add jagged floating shapes to act as the horn tips broken off, for extra visual merit.

Dev Log Evaluation:

I believe the start of this project has been strong and I've got a good custom model to work with; using 3Ds Max independently increased my confidence using it as I problem solved on my own, learning a lot in the process.

I can't think of any complaints for this session, just that I need to keep working hard at a good pace to complete this project to a good standard.

Week 4

Finalising the Model and Unwrapping:

I was supposed to unwrap this week but I realised but I realised the 3D model didn't save correctly and my progress was set back to before my feedback was applied. This situation sucks due to the progress lost, but it was important to bounce back and continue to work hard. While this saving issue could've just been a fault of 3Ds Max, it also could have been me; I have to take accountability for not double checking it saved properly before going back home off campus and in the future I'll be extra careful with this. A tough but necessary lesson.

Since I was back home, I had to catch my model up on my own in my own time which was a difficult task as I didn't know entirely what changes my tutor made to the model, but I used the screenshot I luckily captured and worked off of that and the notes I took.

Seeing as the side of my model was wonky, I selected the edges and, with edge constraint toggled, normalised it for x which straightened these edges for a cleaner model.

I wanted the bottom half to match the top in detail so I used the swift loop tool in the modelling ribbon dropdown for extra faces to work with.

For the final step of the bottom half, I used the FFD 4x4x4 modifier to increase the asymmetry and jagged nature of the chest by flaring certain areas out and dragging some inwards. I scaled the base inward for an overall curved silhouette, resembling an organic form like the rock I aim to imitate.

Next, I used the chamfer tool for extra dimension on the lid's edges, followed by inset on the sides extruded inward slightly for depth.

Finally, I adjusted the horns by adding more edge loops with swift loop and the cut tool to add more faces so the curvature is slightly more emphasised.

With these horns done, I was finished with catching my chest up again and could progress properly once again.

I created a new plane primitive for the lava inside the chest with moderate segments for less rigid sculpting, along with the FFD 4x4x4 modifier with control points selected. I used this modifier to create variation in the vertices' height, making it flowy like lava.

Plinth and Props:

With the model done, it's time to unwrap.

Chest Unwrap Components:

Select least visible edges and convert to seams, then unwrap for shapes like this.

When unwrapping, I split the components (Lid, Chest, Horns) into the quadrants for organisation purposes, ready to pack.

Packed and ready for texturing!

Unwrapping the plinth was easy as it was basically flat; I repeated the same edge selection to seam process seen on the horns on the rocks. I also realised I could stack the unwrap for each rock as it's the same mesh, just differently sized. I did this through entering vertex mode and enabling "snaps toggle" to connect vertex to vertex. With all six rocks stacked into one, this would save me a lot of time texturing.

Dev Log Evaluation:

I've come out of this session feeling much happier about unwrapping and UV mapping than I previously was; unwrapping the chest was a slight challenge but it was much easier to do compared to the tutorial which introduced us to it with a generic chest model and figuring out the rocks on the plinth map could be stacked to save time felt so satisfying. I believe I'm getting the hang of 3D production and will continue to learn more as I go.

Hand-Painted Texture Blog Task

For this blog task, I am being asked to research for and analyse a 3D model in which the hand painting technique was used to implement textures and detail, exploring the colour choice, values, edge dressing and painted factors. I aim to leave this with an improved understanding of the hand-painting technique's fundamentals, able to apply them to my own model effectively.

Zanga_Games_Art_Studio (2022) Hand-painted 3D Building Model - Venomous Pool. Available at: https://www.artstation.com/artwork/o2zO9L (Accessed October 2024)

My Chosen Example:

I chose this 3D model by Zanga Games Art Studio through searching "Hand-painted 3D model" on artstation which gave me relevant results by professionals demonstrating their knowledge which would provide accurate colour theory, light dynamics, etc for me to analyse. I also clicked the "Hide projects with AI" quick filter under the search bar; while this doesn't guarantee the full removal of AI art, it's preventative to a degree. It is important that I still put in effort to confirm whether something is AI generated or not.

Colour Task:

Considering the asset's context as a venom pool in an ancient viper temple, the colours were chosen by the artist to represent this concept. The primary colour is a cool, desaturated blue that leans towards grey to embody the texture of stone which would be used to construct the statue. I believe this blue undertone was applied to support the idea of an analogous colour palette and reflect the theme of danger in this piece - the primary colour is a desaturated blue which is the colour least prominent in nature, conveying the unfamiliar setting and risky ventures within the environment. The secondary colour of gold is on the opposite side of this transition, as the warmest hue providing variety. The use of gold shows variation in material through its metallic irradiance, also portraying a visual narrative with its expensive nature, perhaps suggesting opulence regarding whoever constructed this statue. Green acting as the accent colour sits in the middle of this analogous colour palette, tying the other two together for a cohesive appearance. It's also thematically relevant, as snakeskin is often associated with the colour green despite their varying tones, and venom is portrayed as green in media and in this asset to be reminiscent of poisonous afflictions, as snake venom is actually white.

Value Task:
Column 1 - Highlights

Column 2 - Mid-tone/Local Value

Column 3 - Shadow

The asset itself is relatively muted to fit a more detailed and gritty stylised direction in art - this is demonstrated by the lacking in distinction in local values, which causes difficulty in viewing this piece in grayscale as the tones are muddied. This is a weak point in this piece and the artist could improve the values by making the gold lighter to generally make each material distinctive. However, I believe the artist relied on their colour choices and use of vibrancy to make the asset effective as the pool of venom doesn't stand out here, but the green used is the most vibrant tone in the piece and could be regarded as the focal point due to it's vivid nature drawing the viewer's eyes to it, almost as if it were magical, in contrast to the muted surrounding colour which embody natural materials like metal and stone.

Edge Dressing Task:

On materials such as the stone and metal, the artist has used lighter colours just above the main highlight on the edges to hide the seams which improves material read and conceals seams in the unwrap. Masking the seams with lighter colours and variation is known as "edge dressing".

With the stone, the edges have a lighter colour with loose lines to portray an aged effect and corrosion on the corners where it would've been sharp at creation. The stone also has cracks and scratches for detail which is also purely painted on using edge dressing. The metal is a little cleaner, but still uses the same method.

Finally, towards the bottom of the model, the artist added clumps of sand to hide the seams further down; plausibility is implemented through bringing the sand down and following the cracks of the stone so users don't immediately recognise it to be edge dressing. It's extra effective as the ground the model sits on is likely to be sand, providing environmental narrative.

Hand-painted Detail Task:

In the first example, you can see how the wireframe is simply a square with two segments - however the artist took this opportunity to see it as a canvas and implement the most detailed part of the asset here, painting details as if they were precisely sculpted.

Below this example is a simpler version of this, showing a rectangle with another inside it, extruded inwards as demonstrated by the edge's shadows. This could've been done using the inset and extrude tool, however using this method allows the artist more budget in the 3D model.

Finally, the snake's torso was constructed with squares and painted over with directional lines, cracks in the stones, and venom leaking from the core - this could've been portrayed through adding insets to each V-shaped line and having their own unwrap on the UV map but the artist simply kept the torso as squares and painted over them, saving more budget.

Conclusion:

With each of these factors analysed using a professional-made model, I've come out of this task enlightened about many aspects of rendering my model's textures and will be applying each method to my model.

My model will have a similar amount of colours included to this asset, with a primary and secondary colour featured. I'll apply the same parts of colour theory to it, using an analogous palette to include cohesivity in it with a warm colour scheme instead - this is due to the volcanic theme of mine. My stone/rock texture will lean towards a purple hue with red undertones to complement the lava seeping from it.

For my values, I plan to have my rock significantly darker than the lava, with the lava relatively in the middle regarding value scales. This is because it's just the local value, and I'll be building up the highlights heavily, as I aim for the lava to glow to advocator for its temperature visually. The rock's local value will be moderately dark as the edge dressing will be lighter, along with the protruding faces for a good example of range in its values.

I will be using the edge dressing technique heavily for the rock, as the corners of my chest are relatively simple and edge dressing will add more complexity. This goes for the overall texture, too - I was initially going to mix up the topology and make it wonky, using inset for most faces and extruding them to implement the rock texture, but with my teacher's feedback, I learned that it'd be heavily costly budget-wise.

The previous point also applies to the hand-painted detail factor, due to it elevating the rock's texture. To save budget, I'll be doing the rock texture entirely hand painted. I'll add small air bubbles to the lava like this, too. I plan to add cracks and corrosion to show age through hand painting too.

Week 5

Textures in Photoshop

This week, I would begin texturing my chest as the unwrap was complete and ready to translate to Photoshop. I was admittedly nervous as the rocky texture is inconsistent and would be hard to align regarding the seams, so edge dressing would be a difficult task - but I had to be brave and give it a go.

In the tools section of the UV editor, I selected "Render UVW Template", creating one for a wireframe and one for a mask with these settings respectively. On the mask, I deselected the "show overlap" option as it changed any stacked objects to bright red which would hurt your eyes working on and disrupt your ability to discern values when painting. Once I clicked render, I saved each file as a png respectively named as "wireframe" and "mask" for effective file organisation.

With a new photoshop file created in 2048x2048 dimensions at 300dpi, I imported my wireframe and mask files ready to paint.

(Practicing layer discipline through labelling each layer accordingly.)

(Relevant references kept on a separate layer to aid me.)

On a new "base value" layer clipped to the mask, I changed the values of each component. The chest would be relatively dark like most rock and the lava would be bright for solid distinction between the two. Atop this layer, still clipped, I created a sketch layer where I marked the basic details I'll be painting.

Rock Rendering Process:

Simplified Process:

Before and After:

I realised I made the values too dark on the top compared to the rest of the chest's rock textures; to fix this, I selected it with the lasso tool and went to Image -> Adjustments -> Brightness/Contrast and turned the brightness up slightly and the contrast down slightly which matched the other areas better.

Lava Process:

Here, I Checked my textures looked decent so far by importing them into 3Ds Max before I add gradient maps. I opened the Material Editor with M and dragged in a physical material, followed by a png version of the textures which I attached to base colour. Selecting the chest, right clicking the material and clicking "Assign Material to Selected" allowed me to see the textures on my model and approve it for the colouring stage.

When applying gradient maps, I went to Image -> Adjustments -> Gradient Map and chose a suitable preset to customise. From then, I'd desaturate it slightly and add my own colours to it like this ashy indigo I chose for my chest. This is to demonstrate the muted nature of rocks, with a hint of violet warmth to show cohesivity in the volcanic/lava theme which is scorching hot.

I repeated the same process for the lava and then proceeded to add finishing touches such as dark outlines and filled in areas around the edges of the rocks to hide the seams for effective presentation, known as edge dressing, and soft airbrushed circles around the lava done with an orange hue and a multiply layer. Finally, I flattened this image and saved it as a targa file, downsized to 1024x1024.

The plinth would be a mixture of ash and solid dirt, so I chose to use a noise brush and impressionist blend brush to capture the grain of ash and the clumpy roughness of dirt. For the gradient map, I chose a light maroon to complement the indigo of the chest as these colours resemble the complementary pair of blue and orange, only shifted closer together on the colour wheel for a cohesive overlay.

These are the texture sheets applied to materials in Unreal Engine; the RGB values are attached to base colour and I added a new value of 1 attached to roughness so they'd be relatively matte like the textures would be in real life.

Dev Log Evaluation:

I worked hard on the textures for this chest to achieve a look crossed between realism and stylisation which you can see in the moderate detail of the lava, but the cartoonish flow of it found in the linework. I believe I did a decent job but I'm not entirely happy with it; I think some implied lighting on the texture sheet could've improved the dynamic factor in the textures but I believe that lighting in Unreal Engine will help the textures pop and not look so out of place. My previous experience with UE5 aided me in the process of creating materials and placing everything, but I'm nervous for alphas as I've never tackled that before.

Week 6

Importing my Scene to Unreal Engine

In this session, I imported my assets from 3Ds Max to Unreal Engine where I would add lighting, walls and a floor, and my alphas. I had to export the chest and plinth separately and as autodesk FBX files for correct compatibility with Unreal Engine. I ensured the scale was set to centimetres, as that's what UE5 works in, for a correctly translated model.

With the materials made before applied to the meshes imported into the scene, I could start on the lighting. Removing the skybox light and the floor beneath the model on a new level, I began adding lighting for my chest. The reason for the skybox removal was for a subtle lighting carefully accenting the model and flattering it, without overexposing it and harshly lighting it. I added a rect light in front of the model with a slightly warm but bright tone for main exposure. At the back, I put a warmer red one to imply the blazing environment you'd find this chest in, and two cool (very subtle) rect lights to the side just to define its silhouette in the dark surroundings.

Using two cubes in the place actors tab, I implemented them and scaled them out width and length-wise to create a floor and walls for the chest to sit on, just so it's not floating in an ominous void.

With the walls created, I created a material to go with it. I simply chose a solid deep maroon colour to correspond with the fiery concept I chose without drawing much attention.

Finally, this was where my chest was at before alphas. I added a final rect light with a deep red hue at a high intensity but low radius to support the chest's silhouette without lighting up the floors beneath its position.

Dev Log Evaluation:

With my chest almost complete, I can say I feel happier about its appearance now that it's in Unreal Engine with flattering lighting. It wasn't too challenging to do this with my previous knowledge of Unreal's fundamentals and I quite enjoyed experimenting with lighting, I just have to hope the alphas process goes well as the finishing touches.

Week 7

Alphas

Alphas are 2D images pasted onto a scene as additional decoration that can elevate a model's visual presentation. These are created through alpha channels in art programs and ported into 3D software with masks applied. With my chest scene complete, I chose to add alphas as a final touch which would be a challenge as something completely new but I was willing to try to boost its presentation. I aimed to create smoke trailing upwards to help with environmental storytelling of the volcanic environment of this chest.

Smoke Alpha Photoshop Process:

This is how the alpha looks as a material in 3Ds Max.

I added plane primitives for my alpha placements (top photo) and applied the material to them to check they worked (bottom photo).

High Res Captures:

Following the same process as the previous materials created in Unreal, I added the alphas (ensuring the masked option was toggled) for the final touches. With that, my chest was finally done! Yay!

Dev Log Evaluation:

I'm feeling really glad that I did include alphas as they really add to my scene and give it extra flare; I'm proud of myself for pushing through this learning curve and doing the complicated stuff mostly independently due to lesson absences.

I feel as though the smoke render could've been more effectively done if I'd planned its appearance more, such as finding references for stylised smoke and marking the flow of it. I also am disappointed I didn't get time to add the extra alphas from my stretch goal which were floating sparks but they may have crowded the scene anyway.

Overall, I'm happy with this project completion and for my drawbacks, I'll be sure to always have references on hand and commit to time management better; this block had many ups and downs, a lot out of my control, but I think I had a great comeback and still got everything done to a decent standard.