next »

[tchassin]

Update 75
Hey there!

I finally fixed a visual glitch that had been bothering me for quite a long time. No more random seams between tiles!

But more importantly, I started greyboxing the entire storyline beginning with the Aiguemont Castle. It will greatly help us to iterate on both the story and the level design.
Nothing fancy to show for now, but have a screenshot anyway:

Once we're happy with the level design, we'll begin to work on each level palette.

Pierre spent some time designing new characters and came up with some really nice concepts! Once the artworks are scanned and cleaned, we'll be sure to post them here.

See you next week!

Cheers! 

[tchassin]

Update 76
Hey there!

Since we'll be iterating a lot on the level design, I spent some time improving our pipeline. After switching to Unity's tile system, we got a massive performance boost, allowing us to build much larger maps. On the other hand, the previous maps were too large and felt empty, so we decided to take a different approach. As a result, most of the maps present in the demo were merged into one single map:

This represents about half of the first dungeon. Don't mind the seams, they only appear in the editor.

I then worked on the maps for the rest of the week and we now have a complete skeleton for the whole game world!
This made me realize that some parts of the screenplay didn't make sense from a spatial point of view, so we'll work on that in the next few days.

See you next week!

Cheers! 

[ProgramGamer]

Looking forward to further updates.

[tchassin]

Looking forward to further updates.

Thanks!

Update 77
Hey there!

Sorry for the lack of updates last week, I experienced some technical difficulties.

I added one-way paths to the game since, last time. The main purpose is to add shortcuts (and maybe some secrets). It will need some visual improvements but it works, so we can use it when designing maps.

Additionally, it is now possible to make some enemies unique, meaning that once defeated, they will never appear again. This will be useful both for bosses and optional challenges.

I also made many other little improvements such as saving the player character direction, and displaying the current map name after loading.

I'm also working on adding merchants to the game and implemented the trade system. So far two screens out of three are functional but they use placeholders so I'll show them later.

I had a meeting with Pierre this weekend and we settled most of the story issues. We also discussed the world and level design and found a number of issues and possible improvements. Also some bugs such as these accidental Lost Woods:

Next: more UI for the merchants and more level design. I'll also add some color blocks to make the whole thing more readable and make the screenshots a little more palatable.

See you next week!

[Mochnant]

Just stumbled across this and looks good.  It's nice to see such a long dedication to the game.  Keep up the hard work!

[tchassin]

Just stumbled across this and looks good.  It's nice to see such a long dedication to the game.  Keep up the hard work!

Thanks for the kind words, it really helps to stay motivated!

Update 78
Hey there!

Last update was quite recent, but let's try to get back on schedule!

I finished implementing a first draft of the merchant system. After finding and fixing a ton of UI bugs, the various screens are all functional!
Good news, we also plan on adding a few lines of dialogue to each merchant in the game.
Bad news, it will have to wait the soon-to-come overhaul of the dialogue system.

Level design is also progressing nicely. I made fixes and adjustments here and there and also added basic color to the maps: it's now way easier to understand what's going on.

Pierre also made a brand new artwork! Meet Rosalie, Victoire's little sister:

I didn't write anything about her in this thread, but she's a very important character! We hope you will like her!

See you next week!

Cheers! 

[tchassin]

Update 79
Hey there!

I decided to take a break from UI and level design to work on pathfinding, so today's update is going to be a bit more technical than usual.

Right now, enemies simply move in a straight line toward their goal without taking their surroundings into account, so it's easy for them to get stuck. We made the demo with this constraint in mind and placed them in wide open spaces. But we clearly need to improve on this.

A quick solution would be to check if the player is visible using a raycast before chasing them. However, it's only a temporary solution. We need something more robust if we want to introduce NPCs with a routine, patrolling enemies or use pathfinding in cutscenes.

I considered using Unity's built-in navigation system, namely the NavMesh. Unfortunately, it was designed specifically for 3D games and we can't rely on this solution. Furthermore, navigation meshes assume agents can move freely, but in our case almost all characters move only using 8 directions. Sticking to the path would make then change direction very frequently, which would feel quite glitchy:


(in black the wanted path and in red the actual character's movement)

That's why I thought about using a grid-based graph to represent the level geometry. It makes grid-based movement a lot better and can be used for free movement by using path smoothing. All that's left is finding an algorithm with a bias toward long straight lines so that grid-based character don't change direction too often. Luckily, there is a grid-based A* variant called jump point search (JPS) that does just that, while also offering a nice performance boost. I'll give it a try later, once the level build is more advanced and we have different types of maps to test it.

TL;DR: pathfinding is hard.

And now, back to that horizontal slice. See you next week!

Cheers!

[freank]

I like it! Great style! and I love the isometric view in a battle! it's so rare to find!

[tchassin]

I like it! Great style! and I love the isometric view in a battle! it's so rare to find!

Thank you very much! We felt it was underrepresented in RPGs, glad to see we're not the only ones.

Update 80
Hey everyone, happy New Year!

Pierre and I took a break from the game to spend some time with friends and family, hence the lack of updates.

Let's get a bit technical again and talk a bit about connexions between maps. This post will be a lot longer than usual, but I hope you'll enjoy it!

Each map is contained in a single Unity scene. To go from one map to another, I have to load the containing scene, so I have to reference it by name in the current scene. I used to store the target scene name directly in the exit objects, but it had lots of downsides:

• Replacing, renaming or deleting a scene required opening all adjacent scenes and manually update the exits.
• There were no checks on the scene names, so I had to launch the game and test each exit manually to make sure things worked.
• A simple typo would result in a game crash.

I thus decided to create a new type of data asset  simply named entry point containing:

• A map name
• A target position

Each exit references an entry point, which is not contained in the scene and thus can be modified without opening scenes. Moreover, entry points can be shared between exits to avoid redundancy. I also created a custom editor for entry points, which gives me a lot more control over the way data is displayed in Unity.

The custom map entry editor

The custom editor performs lots of checks such as making sure the target scene is contained in the current build. It also allow moving between adjacent scenes quickly thanks to the 'Open scene' button. I also added handles to vizualize and move around the target position, which is so much better than typing a number in the editor then start the game and check the player position after changing the scene.

The target position handle

This definitely improved my workflow and it should prove extremely useful when iterating on the game maps.

Aside from that, I also uploaded the demo build to Indiexpo a few weeks ago and the game got a lot more attention than on itch.io, which is really really cool. If you have an account, please rate our game and help spread the word out there!



(Shameless promotion is shameless  )

See you next week!

Cheers! 

[tchassin]

Update 81
Hey there !

This week is about level design again, and it'll probably be our focus for the next few months, so I hope you're all looking forward to that :p

I've been working on the roads and towns all week and ran into several unexpected issues.

First, I started working on the level design by editing one map after another. It soon proved to be difficult and a bit erratic, so Pierre and I went back to the overworld to get a better sense of structuer. Here's a sneak peek at the current layout of the game (WIP):

We also noticed our characters spend a long time on roads, which can feel repetitive and linear. Now linearity is not necessarily a bad thing but it's important to get it right. It's important to add variety so that the levels don't feel too linear, by adding branching paths or breaking straight lines for example.

Aside from that, we had difficulties to make each town feel unique, large cities in particular. A lot of towns in classic JRPGs can be described as "5 buildings facing south in an open space" but they rarely feel same-ish. It's due to a combination of macro (making important buildings larger, moving them apart from the others) and micro level design (color variations, remarkable details or small decorative objects). Larger cities, such as the main city of the game, raise another issue: how does one make a town feel large without useless padding? We don't want to add dozens of similar-looking empty houses just for the sake of it or make the player feel frustrated by blocking off large portions of the town. For now, we decided we'd begin with small towns with no padding and see along the way how to make them bigger: density is more important than volume.

Since we were unsatisfied with a number of our drafts, we decided to go back and study the classics.

A good example of making a path feel less linear without altering its structure too much from Sword of Mana.

If you have good examples or pieces of advice, feel free to share it with us!

Next week, I'll keep working on roads and towns. We'll also be studying the amazing tile system used in the Mana games, especially for cliffs: if we succeed, this should prove tremendously useful to improve the game feel and visuals.

See you next week!

Cheers! 

[tchassin]

Update 82
Hey there!

Another week, another devlog about level design.

After settling on the overworld design I showed you last week, we had to modify a good number of maps to fit the new layout of the game.

First, we split maps in three categories:

• Roads (large but simple)
• Towns (small but complex)
• Dungeon (large and complex)

Since roads are the easiest category of maps to work on, we decided to start working on those first.

Pierre and I then spent some time finding strong map concepts to guide us when working on the details. It can as simple as "A vertical river splits the town in two" or "many cliffs and small caves". Here's an example:

Trying to create this in-game brought two issues to light:

• It's not possible to pass under and above a bridge at the moment
• Tile size was too large and had to be reduced from 32 pixels to 16 pixels

The first issue will require some programming but is pretty easy to fix. It requires placing triggers on slopes that changer characters layer (both for sprite and the collider) : whether the character is moving up or down can be determined using its direction.

Changing the tile size is simple and does not require any additional programing but it's a pain in the neck since I'll have to resize each map and place more tiles. Well, it happens I guess.

As shown in the sketch above, some areas in the game will have a lot of cliffs, so Pierre spent a lot of time studying cliff tiles from classics such as Final Fantasy V and Sword of Mana. He found some interesting tile patterns to break the grid, so we started working on a template mimicking this:

We now have to test in game to adjust it : it needs to be both simple to use and complex enough to be visually interesting. Iterations in sight, I guess.

See you next week!

Cheers! 

[tchassin]

Update 83
Hey there!

I spent a lot of time working on building the cliffside map from last week. I was not familiar with the layer system so moving the character between layers took longer than expected, but our custom collision system now supports layer changes! I took some time to remove old unused layers and reorganize the remaining ones, and also fixed some minor bugs and glitches I found while working on collisions.

Making that complex map took quite some time and I feel like I'm not proficient enough with Unity's tile-oriented tools, so I may take some time researching them in the next few days.

In the meantime, Pierre came up with level design concepts for each of the various towns and villages featured in the game. Progress is slow but steady and I think it's coming along pretty nicely.

See you next week!

Cheers! 

[tchassin]

Update 84
Hey there!

Pierre and I were both busy with personal things this week so today's entry is a bit on the short side.

I started the week by searching for in-depth tutorials about Unity tilemap system. Alas, I couldn't find anything satisfying
I may have to create some utility tools at some point because I feel some features are missing or too limited.

Level design is progressing nicely. I think I'm really getting the hang of it and hope to finish the current batch of maps in about two weeks, perhaps a little more.

Pierre is still focused on map sketches and town layouts. Here's a sneak peek of Réamont, the second largest town in the kingdom of Verdance:

Next week should be a lot more productive. Since Pierre won't have a lot of time to work on the game this week, I might take a break from making maps and play Monster Hunter write a first draft of the dialogues.

See you next week!

Cheers! 

[tchassin]

Update 85
Hey there!

After finishing building the levels Pierre sketched last week on Monday, I decided to start working on dialogues in order to have them in-game. To do that, the dialogues must be written in translation files, which are plain text files I edit with Notepad++.

I started writing dialogues with Open Office then copying them in translation files, which quickly turned out to be pretty cumbersome. I thus created a specific syntax highlighting for our dialogue system and added comments and empty lines support. It is now much easier to edit translation files directly. Currently, this is what an open dialogue file looks like:

The scenario and dialogues are entirely written in French for now.

Since it took me one day to write the prologue and the first chapter, I think a first draft of all story-related dialogues should take a week or so.

I won't be able to work on the game next week so see you in two weeks!

Cheers! 

[tchassin]

Update 86
Hey there! It's been a while!

I am still working on the dialogues. The first draft is roughly 50% done. Things are taking a little longer than I thought because I keep finding little things that need to be fixed in the scenario (and sometimes in the world design). I hope I can finish this week.

Pierre has almost finished sketching all the road maps. Once he's done, I'll start building them in game while he works on the town and village map sketches.

See you next week!

Cheers! 

[tchassin]

Update 87
Hey there!

I couldn't finish the first draft of the dialogues this week, but it's almost done. It shouldn't take very long since all left inconsistencies have been taken care of. I think I should be done by Tuesday.

Since I miss coding these days, I decided to work on the new battle system starting tomorrow. Of course, I'll keep working on dialogues and level build in parallel, but I think this will boost my motivation.

See you next week!

Cheers!  

[tchassin]

Update 88
Hey there!

I got quite a lot done this week. First, I completed the first draft of the dialogues, just as planned. Hurray! The next step is for Pierre to review everything.

I then worked on improving the battle system. Many of those who played the last demo told us the game lacked a bit of tactical depth, so we decided to add rows to the battle system.

Much like Etrian Odyssey, there are two rows. Types of damages are affected by the distance between the caster and its target:

• Melee attacks are weaker when the target is far away.
• Ranged attacks are stronger when the target is far away.
• Magic spells are not affected by distance.

We also added an option to target a row instead of either a single fighter or a whole group. Additionally, some skills can only target a specific row: each fighter has a role for each row, which determines which skills are available. Roles are attributed based on character personality and backstory, so players won't be able to change them. Here is how the roles are dispatched between characters:

• Victoire is a melee fighter on the front row and a buffer on the back row.
• Latharn is a black mage on the front row and a healer on the back row.
• Erik is a debuffer on the front row and a ranged fighter on the back row.

Those aren't absolutes, and some skills can be used on both rows.

You can change rows while in battle, either with the "Switch row" option during your turn or by using dedicated skills as shown below:

You can define your default formation in the tactics menu (early WIP):

Right now, the role system is still a work in progress and the targeting system needs some work for row-based attacks to work, which I'll start tomorrow.

See you next week!

Cheers!

[Ikkir_]

Oh, nice. The latest battle aesthetic really reminds me of Breath of Fire for some reason.
On turn order: will it be possible for characters/monsters with high speed to have multiple attacks, or have attacks that cause the next turn to be faster/longer?

Edit: also, that demo link on update 80 is broken, I think?

[tchassin]

Oh, nice. The latest battle aesthetic really reminds me of Breath of Fire for some reason.

Thanks! Breath of Fire is one of our main visual influences and the isometric battle perspective is an homage to the series!

On turn order: will it be possible for characters/monsters with high speed to have multiple attacks, or have attacks that cause the next turn to be faster/longer?

We used to have an ATB system to make speed an important part of the game, but we felt it didn't make the combat system much more interesting. With our more traditional turn-based system, speed is less important and is mostly used to determine the original turn order. Allowing the characters to play twice if they have 50+ speed or something similar would introduce thresholds in our system, something Pierre and I are not very fond of.

Instead, we decided to have ranged skills rely both on strength and speed to deal damages. We might also introduce some other speed-based mechanics.

Eventually, skills that can modify the turn order are indeed something we would like to have in the game, we are trying to come up with rules can be simply explained and displayed in the UI while preventing the player from indefinitely delaying an enemy's turn.

Edit: also, that demo link on update 80 is broken, I think?

Yes it was, thanks for letting us know! I should have double-checked that.

[tchassin]

Update 89
Hey there!

It seems like I underestimated the amount of work required by the targeting system. The previous one could only target a single character or a whole party, so I had to rewrite a lot of code to get it working with the new row system. There was also a number of problems with the UI. Since it's a very important part of the game, I decided to take some time to improve the UI navigation system.

Much better now!

I also wrote most of the code required for skills to be usable only from certain rows. All that's left to do is displaying the compatible rows in the UI and decide which skill can be used from which rows.

See you next week!

Cheers!