Hello everybody,
I finally started making the Medieval Collection Series I developed for the Unreal Marketplace.
I will continue adding everything. I hope I can complete this
My Personal Portfolio : ArtStation - S.Firat Demir
Hello everybody,
I finally started making the Medieval Collection Series I developed for the Unreal Marketplace.
I will continue adding everything. I hope I can complete this
My Personal Portfolio : ArtStation - S.Firat Demir
Well, first I decided to do the environmental design and simple lighting.
I’m going to use HDR instead of using Direct Light. In this way, I think I will get a more realistic background.
The first thing that is necessary for me is to have a medieval atmosphere. Then I will create all the colors and color profiles accordingly. I’m going to go a little beyond the usual color palette logic.
I also simply needed a floor material. You know, to make it easier to scale, I wanted it to have something of my own.
I needed some specular effect and roughness effect. I’ve made a few Value entries for now, maybe we can edit custom maps in the future.
I am very impressed by your portfolio and look forward to see the Medieval Collection come to life.
Should you have any interest in exploring shaders, I would be pleased to offer my assistance. Please feel free to contact me at Elias Wick.
Hey Elias, I’m glad you’re here. This is a great idea.
Also thank you for your nice comments.
Hey @SkyeEden Thank you for the arrangement. You’re amazing, it’s nice to be with you.
First I’ll need a few references for proper scaling. That’s why I decided to make a character.
I had prepared my own 3D face model before. I thought it was suitable for this He just needed clothes. A very dear friend of mine helped me with this and created some outfits. I improved them a little and prepared their coloring.
I wanted the character to be detailed, so it consists of 4096 x6 texture indexes. At the same time, I prepared it as a low polygon to optimize. (This is why I removed the eyelashes etc. I actually tried the Opacity map, but I canceled it because I thought it wouldn’t matter if we wore helmets anyway.)
I created 1 pose using a standard T pose and simple RIG and now it’s ready in the unreal scene.
We will take the time to develop it further. Seems to be manageable for now.
I have a simple question. I’ve never been able to create shadows in Chainmail. If anyone knows how to do this, help would be greatly appreciated.
I immediately thought that we should prepare the boards and started looking for the most optimized solution for this.
As a result, the use of Trim Sheet will work for us. I want the whole collection to use a single material. In this way, I can make the optimization better.
For this, I prepared and put together some boards using substance designer and zbrush.
This part was a bit boring… But then I tried to calculate the boards. I’m stuck on pinterest. I have identified the pieces of wood we will need. But if new parts are needed in the future, it won’t be difficult at all. All that’s left is to UV sit the new part on the trim and send it to the unreal engine side.
I noticed that the trim sheet is having problems with the edges and needs to be reset properly. After fixing that too, everything seemed to be ready.
At this very point, the incredible problem that I had overlooked appeared.
Lightmap … Fail - I realized that we need to use UV Channel 0 and UV channel 1. In this case, the problem was solved. But you know, this is incredibly boring. lol.
Look at this, red colored things can never be a good sign. But that’s okay, because that’s just UV channel 0.
Here is UV channel 1 and residual light calculations problem solved.
I also needed a material to generate randomization.
You know, every board should have looked different even if it used the same map. There are different approaches for this. But the best thing for me was to set 3 different masks according to RGB and fix it.
I used my concrete material on the boards according to these masks, you know this is actually a trick.lol
But this is not understood, who cares. It’s enough for me to make the boards look more worn out.
Each board now behaves differently. In this way, when we bring the same ones side by side, we can get a good view for the beginning. We will return in the future to take this topic to a higher level.
My Material Instance settings seem a bit complicated, I know. But I wanted to group it in a way that I can understand. It’s a good start right now, but I’ll definitely improve it. We’ll also be able to use it as a blueprint for almost anything on stage.
I wanted to streamline the material hierarchy a bit. I’ve been this successful at most I’m not very good at making this kind of schemes regularly.
Thanks to Mert Tastan, he described some updates to the atmospheric height fog settings and it looks much better now.
Maybe we can consider using cubemap later.
We will also need iron materials. Barrels, tables, whistles, etc. It will fix a lot of problems.
In order not to have high polygons, I applied the same cooking operations and now they use a single map.
In fact, it didn’t need to have incredible details. Because these will be the metals we will use in simple works.
It seems more correct to make separate studies for inlaid irons.
Unfortunately the lightmap calculations did not go well. So it was necessary to create a new lightmap compute map for all board surfaces.
For this, I produced a randomization and UV build script in max. It’s working fine now.
3Ds Max and Unreal Direct link system work very well. Very good for sending material, model and light at the same time. But I’m not in favor of sending lights on huge stages. Instead, I favor the use of emission maps over the mesh, which most of us are familiar with. In this way, we can provide optimization much easier.
Generally, disconnections occur. But if you’re good with file naming, you don’t experience this. It’s best to reimport occasionally and keep the link up to date.
Try not to complicate the materials too much in 3Ds max, just make simple connections and use vray.
I’ve noticed that Arnold materials have problems with the transfer so vray or scanline looks better. We’re just making simple connections anyway, and we’re doing the rest in unreal.
Since it will not combine mesh structures, it is much more correct to proceed with attaching in general. But for this reason, you won’t have many objects that move independently in unreal.
It’s always up to you to make adjustments for varying uses. But don’t forget to keep the optimization.
I’m trying to get the best result, my current settings are like this, you can see the problem I’m having in the picture below. I’m trying to come up with a solution to fix modular structures and lightmap problems.
I think I’ll try to fix it after this stage by making changes with the material. This may be a contrast blend issue. I will do some experiments for this.
Here is the node I used to fix the problem. With the low contrast blend settings I’m going to make here, I can correct these blackouts. But I’m not entirely sure and I can’t understand why this is happening.
This will be a temporary solution I did, yes it will prevent blackouts. But these are not actually blackouts. It provides a parameter for the second material and its aging. It’s supposed to do light-colored wear, but it causes blackouts on many parts.
Anyway, I will temporarily solve this wooden surface problem and continue from the iron prop set. I thought I used a lot of damage, dirt and rust in the previous version. So I decided to make simpler irons and simplified the texture structure. Maybe a few usage parameters.
But other than that, I decided to continue using this structure for anchors.
I want to try a few. Because it is the best way to determine if everything is enough.
Let’s try to make some simple models with a few references. But the important thing here is to get the best quality as quickly as possible. That’s why it’s important to me that we move forward using what we have.
In fact, the general plan is to do these last, and that’s how teams usually work. But for now I’m on my own and who cares about the overall planning lol
It’s better to do this when I’m really bored. I will not use these models in the future, but it is really pleasing that good results are obtained and can be made according to references.
Here is a video below. It can be fun to watch.
Manufacturing a torch using irons and it looks nice now. I guess I’ll have to produce some more parts. At least wall torches etc.
Yes I designed a torch for the walls. No extra parts were required to make this torch. So the parts I planned still seem sufficient.
I only make it from wood without adding metals. I’ll add the metals later. I started making the cranes.
I prepared some variations of metal probes. In this way, I will continue to create differences in the image using the same mesh structures. It’s such a small detail, but it really pays off.
Simultaneously, I started preparing bridges. The current system works really fast. Maybe we can try to create a procedural structure later.