Sep 1, 2023

Dev Log #1: Lost Eidolons - Veil of the Witch

Lost Eidolons: Veil of the Witch - ODS_ThorThor

Hello everyone!

It's been a while since our last major Lost Eidolons update, when we added New Game+ mode to the PC edition.

Since the game's release, we've spent a lot of time collecting feedback and having internal discussions about the game's future. As promised, we delivered LE on Xbox and Playstation this month, with the help of our porting partner, PID Games.

Now it's time to decide where the world of Artemesia goes next.
There were a lot of paths available to us in making this decision. Should we do DLC? Big or small? Or should we just jump straight into Lost Eidolons 2?

In the end, we've decided to take a different path, and try something a bit unexpected.

Today I'm happy to share what we've been working on: an exciting new spin-off game, titled "LOST EIDOLONS: VEIL OF THE WITCH"!

We refer to the game as a "spin-off" (rather than a true sequel), as it shares a lot of Lost Eidolons' core identity, and even a few characters. But the gameplay and general presentation is a major departure.
Now let's dive into some questions you may be curious about.

What kind of game is "Veil of the Witch"?

Veil of the Witch is a turn-based tactical RPG with rogue-lite elements. It takes place in the same world as Lost Eidolons, and is set shortly after the events of the first game. But it's an entirely standalone experience you don't need any prior knowledge to play.

Fans of the first game will find Veil of the Witch to be a tighter, more focused game, that offers more of what they love with less friction in between.

New players will find a rich world and characters, a fast and gritty combat system, and an addictive, looping structure that sets the game apart from the rest of the genre.

I'm confident the game will speak for itself.

What's changed since Lost Eidolons?

In one word: speed.
Lost Eidolons was a linear, story-driven RPG that unfolds gradually. Veil of the Witch, on the other hand, is fast, gameplay-driven, and hits the ground running immediately. One of the most immediate changes is that we've done away with sprawling class trees that take 30 hours to get to the cool stuff.

In Veil of the Witch, things are much more stripped-down. Each character has a single class, with fixed weapons and armor, and a very specific playstyle. There's still progression and choices to be made (skills to unlock, gear to be upgraded, and so on) but generally it's all a lot tighter and more focused now.

We've also made a number of changes to the combat system itself, and the game's overall structure is radically different from Lost Eidolons. I'll save the details for another day, but for now I'll simply say that I'm a huge fan of games like Hades and Slay the Spire, and they've had a huge influence on Veil of the Witch.

The result of all this is a leaner, meaner game with more excitement, greater strategic depth, and the freedom to approach battles how you decide.

But this doesn't mean the story experience isn't important to us. Like much of the rest of the game, our approach to narrative this time around is: "Less is more". We've got a great setting, a tight cast of characters people will love, and a new approach to story presentation that (while sparse) we're hoping will be a lot more effective.

Oh, and we've taken the art direction for characters and UI in a completely new direction. I personally love how this game looks, and I can't wait to hear your thoughts on it soon.

Why make a spin-off instead of Lost Eidolons 2?

While it took a lot of internal discussion to reach this decision, in the end, our reasons are pretty simple.
It won't surprise anyone to say that, while we're proud of Lost Eidolons, there are some elements that didn't quite come together as well as we would have hoped, and many of you felt the same.
When we dove into player feedback for the game, two big points came up over and over again.

First, players said they liked Lost Eidolons' story and characters, but found the actual execution to be sometimes lacking, and often inconsistent.

Second, most players loved the combat system, but felt progression lacked meaningful choices, and encounter design could be rigid and unforgiving.

After weighing this feedback and discussing it internally, we decided to try something new that would let us build on the engine we've created already, but take a big swing in a new direction, with a storytelling approach that was lighter and simpler.

Also: I've just always wanted to make a game that brought rogue-lite stuff into the turn-based tactical genre! So we decided that before jumping into a sequel, we'd try a spin-off, and see what happens.

But I want to be very clear: this isn't some little pilot project or cash-grab pseudo-DLC. We're very serious about making Veil of the Witch an excellent RPG that stands entirely on its own two feet, and offers a ton of depth and replayability.

So when can I play it?

This time around, we're going to be pursuing an Early Access model, and do actual open game development with ongoing community involvement.

One of my biggest regrets about Lost Eidolons was that we didn't do public playtests until after much of the game's design had already been locked-in, and was difficult (or in some cases impossible) to change.

One of the main reasons we did that was because it was a linear, story-driven game, and we consistently got feedback that players didn't want to play an incomplete version of that kind of game. After all, no one wants to read a messy first draft of a novel that spoils the plot.

But as Veil of the Witch is a more gameplay-driven title, I feel a lot more confident sharing early builds, so we can gather and integrate community feedback. If everything goes well, I'd like to launch in Early Access sometime in 2024, but it's likely we'll run some beta tests before then.

Also, as I'm really hoping to release the game on Switch, our team has put a lot of effort into lowering the overall hardware requirements. By the time we hit full release, we're hoping to make the game available on all consoles. (No promises on simultaneous launch, though.)

I'm very excited to finally share our second Lost Eidolons project! We're planning to share more details in regular dev blogs down the line (starting with this one).

I hope you'll join our player community on Discord, where you can chat with us about the game. And don’t forget to wishlist Veil of the Witch on Steam!

See you soon,
Jin Sang Kim
Creative Director, Lost Eidolons

DISCORD | TWITTER | FACEBOOK | INSTAGRAM | YOUTUBE | TWITCH

https://store.steampowered.com/app/2530490/Lost_Eidolons_Veil_of_the_Witch/

Sep 1, 2023

Devlog #32: August 2023

Astride - Raidho Games

It's been a month since the team returned from their summer breaks. We are currently working on the upcoming update and we are happy to announce that we have a new programmer on the team! We believe they are a great match and will be an amazing addition to the team. Ouroboros will be joining the next month's devlog, so stay tuned.

Horses, horses, horses

By Maja - CEO, Concept Artist

It’s been amazing to be back at work after the summer break. I’ve been focusing on NPCs, which I love to draw the most except for horses. We have made a list of NPCs and there are many to draw still in addition to others that need an update. A lot of time goes into creating the NPCs only by text. Who they are, what they do, and what their story is. Some NPCs are more important than others for the gameplay and story. On the side of all these NPCs, we are preparing for another government funding application. Therefore, I am also working on the art bible and updating the game design document.

The first NPC is a fancy guy with glasses. The idea for him is to be a beauty salon worker, he is who you visit to change your appearance. But that may change. He is a person who loves the 1800s clothing style and wears a slightly modernized version of that. He is tall and skinny with sand-colored hair. I of course had to give him a mustache. Even though he has a very cool outfit, his color palette is not the most vibrant. The main focus is his yellow vest. The vest is inspired by the flowery pattern on the “bunad-vests” (traditional clothing from Norway). His pants are gray and his leather shoes are brown. I really felt like glasses fit him perfectly. He has an eye for detail, but his vision is not the best.

I have made plans for NPCs of different ages since I don't want everyone to be a young person except Einar. This is Elin and she is in her early 50s. I wanted her hair to be graying. She is wealthy and cares a bit about fashion and appearance. Her favorite accessories are her pearl necklace and earrings. Dressage is often associated with black and white. Her riding boots are made of black leather and her overall appearance is clean and calm. And what is her role in Astride, you ask? Elin will be the dressage instructor!

Our beloved cat Oskar has been well known by the community ever since Andrea was first announced. However, Oskar turned out to be a more important character than we imagined. So of course he needed to get his very own concept art. Oskar is a Norwegian Forest cat, and I studied the breed closely to make it correct. They are so cute and fluffy! I did not know I would have this much fun drawing a cat. I had not been drawing many cats before Oskar was created. And you know what? I LOVED IT. And to think that at first, he was just a fast sketch on the side of Andrea.

Horses, horses, horses

By Mathilde - Community Manager, 3D-artist

Before we went on our summer break, I worked a bit on fur/hair planes with the Norwegian Fjord and the Norwegian Warmblood horse. This was mostly to test if I would have these planes blend into the main mesh without creating hard edges. It turned out to be a bit finicky, but I believe I came up with a good temporary solution. The shape and the placement of the hair planes are definitely not final, I think I will have to do some more research on hair and find better solutions at a later time but it’s a start.

As I mentioned in the work-in-progress post over at our Discord server, I'm thinking of having three different sets of feathering. Sparse, medium, and full. Sparse is what you would typically see on Icelandic horses and Northlands horses during the summer, the medium is what you would expect on a Norwegian Fjord or a Friesian (not a planned breed, just an example), and the full feathering would be what you'd expect from a Norwegian Dole or a Jutland. We'll have to figure out a way for the horse's coat to affect the color of the feathering, but that's for later and with the help of our freelance technical artist.

When we finally came back from our break, the first week was mostly about the team getting back into the office and our routines. We had a lot of meetings for planning but we were also in the final stages of the hiring process of our new programmer. In between this, I got back into the groove with 3D by making a new wheelbarrow for the stables. It still needs a high-detail map and some fresh colors but it is ready nonetheless. Personally, I am so excited for an additional team member and I’m looking forward to what progress we can make in the future. Our new programmer, Ouroboros, joins the team on September 1st and will most likely be a part of the next devlog here on Steam.

Something we all know and hate is the lack of sound that is currently going on in Astride. We are working on having sound added, and I hope we can have this issue fixed sooner rather than later. We have music ready, it just needs to be implemented. As for sound effects, we had a sound designer join us for a day in the stable to record audio and he is currently working on these files. Hopefully, we got a lot of useful sounds that we can implement into Astride. We will have to do at least one more session of sound recording with the sound designer as he wasn’t completely happy with some of the recordings. Especially for the hoof beat sounds. For the next session, we will be seeing if individual mics on all four horse legs are a good idea. The reason we want to record single hoof beats is for the purpose of making dynamic hoof sounds in the game. Meaning that they will be played whenever a hoof touches the ground, not just a static sound when the horse is doing X gait or X animation.

Part of my tasks for the dressage development phase is to create different stages of muscle in the horse meshes with a change in the high-detail map to accompany it. There will be differences between a well-conditioned horse and a horse that hasn’t had a lot of conditioning. For this, I made a new shape for the Norwegian Warmblood where the muscles were larger and more defined to illustrate a well-conditioned horse. After the new shape was done, I brought this into Substance Painter to add further details to the muscles. The thought is that there will be a natural blend between these two types depending on how your horse is conditioned. I’m excited to see my theory in action at a later stage.

Moving on, I’ve been mildly annoyed by how poorly I did the fleabitten textures that are currently in the game. They were previously done in a rush and it shows. I like to redo and quality-check textures from time to time, and this time it was the fleabitten's turn. I also experimented with how a bloody shoulder could appear in-game. Bloody shoulders are very interesting in the way the pattern works and I hope I was able to make something that resembles it. I'll try to add these textures and other variations into the engine right away so that they will be available from the next update.

A couple of weeks ago, I created a poll to help us decide which breed I would work on next. I was excited to see the Percheron in the lead but I was also nervous since this would not be an easy task. The Percheron horses are known for originating in France but they are, to my knowledge, most popular in the United States. Essentially, there are two types of Percheron horses; The French type and the American type. They are the same breed, but their conformation is vastly different. The American type is sporty and long-legged, while the French type is much heavier built and with shorter legs. This kind of brings me back to the Norwegian Fjord horse as my initial thought was to represent the older, draft-type Fjord and the sportier, leaner-type Fjord. Essentially, I hope to do the same for the Percheron and represent both types of builds.

I’ve already created the first draft of the low-detailed mesh for the American type Percheron and added this to my sculpting software, Zbrush, to decimate it and work on the high-detailed mesh for adding more details and definition. The high-detailed mesh is done and next I will bake it with the low-detailed mesh (projecting the high-detail mesh on top of the low-detail mesh) and later go over the low-detailed mesh again to finalize the shape. When I am happy with the American type, I will work on the French type as they will share the same high-detail texture. I’m excited to share more progress of this breed later.

Finishing the environment and starting planning

By Tirna - Project Manager, 3D-artist

It's good to be back in the office again! After we were done sharing our plans and tasks with the new programmer we hired, I could finally pick up the environment again. I've been working on the environment for a while but now it's finally coming to a point that is ready for implementation. Firstly, I had to do some optimization work on the riding arena. I repaired a few textures that were broken and didn't show up correctly in the engine. While I was working on these fixes, I also fixed a few issues that were previously bothering me, such as the rotations of some cobblestones and other details in the riding arena textures.

As mentioned previously, I spent some time polishing up the environment. I adjusted a lot of areas on the map that were a bit empty and filled them with flora. My goal is for the nature and environment to feel natural and beautiful. I have also taken some time to work on other improvements such as ensuring proper lighting inside the riding arena and fixing the issue we had with reflections in the riding arena windows.

Now that we have the environment closer to how we want it, I shifted over to updating and refreshing our saddles to match better with the new environment. I started with the dressage saddle and gave the leather some TLC after remaking the entire mesh for it to match the new mesh of the show jumping saddle. After the dressage saddle was done, I started working on an upgrade for the show jumping saddles as well. So now both of the saddles are matching in quality.

Other than this, I have been spending a lot of time working on the Gantt diagram for our new application. This means that I've been planning the next phase that we are estimated to last from December 2023 till December 2024. I have also been helping out Maja with the game design document and to get everything in order.

Shoulder contortions and character movement

By Marius - Animator

Before we went on break for the summer, the last animation work-in-progress showed the beginning of adding blend shapes to the shoulders. In addition to making the blend shapes, I started making a system to control the blend shapes and future-proof that system to deform correctly according to arm movement. This turned out to be a bigger hurdle than I first anticipated… It is easy to just imagine and say that when the arm points forward and is twisted, use this blend shape and add the twisting blend shape.

The problem is quaternions, the equation that is most commonly used for calculating rotation in 3D is not that intuitive. While normally when working in 3D we are used to thinking in three dimensions X (left/right), Y(height), and Z(depth), quaternion rotation adds another dimension called W. “The W dimension is a scalar that stores rotation around the vector” according to the unity manual, personally I like to think of it as incomprehensible horrors you wouldn’t understand. On a serious note, it is complicated math that I personally don’t have the capacity to manipulate through equations. The gist of the main problem from my perspective is that the 0 point for the W changes according to the rotation of the other 3 dimensions, so trying to use the W value as a reference for an equation becomes unintuitive.

After trying to comprehend the incomprehensible, I decided to forego that venture as it was becoming a time sink. But the work wasn’t for nothing as it got me to think of a way to get around the problem in ways that I normally would not think of. Finally, I got a system that worked for deforming automatically.

The first test of this system is shown above. At this point, the equations for both the left and right arm used the same arm for value references, so you can see on the right arm (your left) that there are weird deformations. This is because that arm does not move while it activates the same blend shapes as the moving left arm. I decided to leave it in this work-in-progress as it shows what is happening with only the blend shapes and might help to give an understanding of what blend shapes are.

I would like to add that when making and testing these blend shapes, I mostly put them in as extreme positions as I can. For example, an overarm would never actually rotate over 180° (90° each way from the neutral position), especially without moving the shoulders to accommodate the twisting motion. You can test this yourself by bending your elbow (to not misinterpret forearm rotation with overarm rotations) and rotating your arm as far as you can using only the overarm. This rig should in theory be able to twist the arm almost up to 360° (≈180° each way), though the more extreme the twist the worse it looks as there is a limit to how much you can patch up the twist on a small area.

When the system was functional I went into updating the value references of the right shoulder to the correct arm, so I could move them independently and still get the correct blend shapes. While doing this I did some further tweaking and added more minor blend shapes. Before the next WIP though, there had to be blend-shapes on the clothes as well.

After adding and copying the blend shape system to the clothes I wanted to test the shoulder blend shapes in an animation and not just spinning the arms around semi-randomly. And while we didn’t have movement animations on the new rig excluding on horseback, those types of animations would not show or test shoulder movement to the extent I wanted. So I made an interaction animation of scratching the horse’s neck while idle, the timing for scratching was deceptively hard to get down, so I had to ask for references of scratching from the team members with horses. And even with the reference I had to adjust the timing to look good animated. That said: the scratching animation will get worked on more as I feel it is still far from finished, but as of now it is not a prioritized animation.

Moving to more high-priority animations: We want to implement the character soon and with it make it possible to not be mounted. To do that we need movement animations. “But you already have that” I hear you say, and it is true that at one point we had them. But those animations are old enough to have been made on a completely different rig, and the animations don’t transfer automatically. Unless they are similar enough that they use the same naming conventions and the same approximate base rotations, at which point there would not necessarily be a need for a new rig. Above is the jog animation, which I am fairly happy with for the time invested in it. I would like to adjust the hands to not be as stiff and maybe generally tweak the upper body motion a little.

With the movement system Green/Red Horse is working on it will be possible to crouch and for that, we need a crouching animation. Crouching was actually a little weird to plan because you have to think and agree with the others on how far down you want the character to crouch. For example: in the first iteration the crouching animation was much lower, as I personally almost sit on my heels when crouching.

There’s also the fact that I usually stay away from using other animation as a reference, as that would mean I am making an imitation of an imitation of the original movement, so why not just reference the original? I like to think of it as a game of telephone; for each time it is referenced some information is lost and that loss will add up after a while, and you don’t necessarily know at what level of reference telephone the animation you are referencing is.

Finding a good reference turned out to be more difficult than I first anticipated though and decided to just film myself crouching forward. I had to adjust a couple of times to get a height the rest of the team was happy with but there was still a lot I had to do differently from my own movement. You can learn a lot of things you never thought about, by filming yourself move and then analyzing it. I say this because one of the things I had to fix is that apparently when I crouch walk, my natural position is very roguelike with one hand in front and the other further back and to the side. It gave me a slight laugh but we aren’t making a stealth-action game so that would not really fit Astride. So the hands were adjusted to move more like when you walk upright, right arm forward when the left foot moves forward.

Tidying up

By Green Horse - Programmer

Inviting friends over is a good way to motivate oneself to clean the house. Similarly, hiring a new programmer has provided a perfect opportunity for me to clean up the project.

Before I began tidying it up, the project was a jumbled mess of prototype code, abandoned features, and workarounds. Now, it's much more organized, with separate packages for each part of the codebase, logging, asset pipeline documentation, a structured folder hierarchy, and standardized naming conventions for all assets.

Since the last devlog, my focus hasn't solely been on cleaning up the project, though that was the main objective. I've also accomplished other tasks:

Coded a Garry's Mod-inspired character controller.
Downloaded and implemented a development console.
Developed an editor window for internal use.
Programmed map loading and streaming.
Implemented terrain collision.
Prototyped mount/dismount functionality for horses.
Added even more settings.
Began documenting the asset pipeline for the team.

The upcoming game update will likely be the last purely "horizontal" one, where we feel we're laying down the game's framework. With this robust foundation in place, we can begin to "scale vertically" with our new gameplay programmer and introduce more gameplay features. For instance, a short-term goal after this update is to introduce the ability to pick up or push/pull physical objects. However, it's important that we align ourselves with the roadmap we've made and finish the show jumping and dressage game modes first. But some tiny, fun, low-effort digressions are allowed, I say. I will add a soccer ball. Horse soccer anyone?

I'm eager to explore gameplay ideas with the community and I'm looking forward to seeing how we can enable players to roleplay in future updates. More livestreams will come, but we just have to finish this next update first before we can play together on the stream again. More information will come when we're ready to start the Patreon testing phase. See you around!

Please be aware that the content featured in these devlogs represents an ongoing work in progress and may be subject to further changes. We are committed to continuously developing and refining our game. To stay informed about the latest updates and connect with the rest of the Astride community, we invite you to join our dedicated Discord server community here! For additional information about Astride, please visit our official website.