Crest cover
Crest screenshot
Linux PC Mac Steam
Genre: Simulator, Strategy, Adventure, Indie

Crest

Dev Blog #153 - Retrospection



It's been an eventful week: On Tuesday everyone stopped what they're doing to talk about the future of Crest and on Thursday we were left without internet at the office due to heavy rain and storms. Nevertheless, a lot of progress has been made this week.


Community



Like mentioned above, on Tuesday we had a day long discussion about the current problems of Crest and how we'd like to fix them in the future. What came out of it can be read in yesterday's post: The Future of Crest.

Two important dates to take away from the lengthy post:

  • The Community Module will launch at the end of November
  • The "final" release for Crest has been pushed back to Q1 2018 for now

This October actually marked the start of the 5th year of Crest's development. You can find out about everything that happened on this long road and the hardships we went through in our post The Development History of Crest.


Art



Martin
I've been working on the new trailer, most of the time has been spent on making a storyboard. It makes sense to create images outside of your head if you collaborate with others on a project, but the added bonus is that the ideas you have in your head sometimes don't look that good when you see them with your eyes. So it's always a back and forth between what you think and what can be made. Storyboards are great!



Emma
I've been working on figuring out technical solutions for our new website and working on some new UI. From our long discussion about how to improve the game, an overview of cities was born. At the moment, it's kinda hard to get a clear view of how each city is doing if you don't go through every city pop-up and go over every tab.

This new UI will provide the player with a summary of every city, easily readable with not too much information. At the moment, we need to figure out how to display this information without making it looking too "information-heavy", as it shouldn't scare the player away. Below are two of the mock-ups!



Code



Johannes
I've not worked much on the game this week, I've mostly done company stuff.
I have however made a tool to find mismatched calls to Profiler.BeginSample() and Profiler.EndSample(), which helped us find a couple of mistakes in the code.

I'm now back at focusing on improving the island generation, to make the islands more varied and interesting.

Johannes W
This week I fixed some minor bugs with the camera movement. There was some old code that kind of worked against the new camera movement which caused the camera to sometimes behave weirdly.

I've also been working with the story builder for the new end screen. For example, a new feature for the purpose of development is the ability write sentences that refers to methods in the script that is responsible for the story generation. So the string "Your era came to an end when {FollowerName} died of old age." results in a sentence that has replaced "{FollowerName}" with the follower's name.

For randomization, "{Random(proud, noble, grand, glorious)}" results in a randomized word from the arguments "proud", "noble" etc. The purpose for this is to enable us to create dynamic and interesting stories for our players! If you are a programmer reading this, the library Reflection in C# was used to accomplish this.

Tomas
I have finished the new Follower word discovery speech/thinking bubble I talked about last week.

I am almost done with making it so that followers with high faith never do actions that the player tells them to not do. I was almost done but after some code reviews I have chosen to re-do the feature a bit.

I also looked into a major change to how Followers choose targets when they build their orders. When I am done, this should make Cannibals, Killing Warriors and Hunters behave as expected. Previously there was a possibility that they performed the actions on unwanted targets. Like for example if the player had written the commandment "... eat old followers." and there were no old followers, they would still eat people... just not their favored target.



Jens
This week I implemented the playback for all the new sound effects. This means that we can start testing them during play, seeing if they work or what potential adjustments need to be made. For example, the notifications will now play different sounds depending on the severity of the message. Monuments and expertise buildings now play rewarding sounds when built or upgraded. All in all, more than 20 sound effects have been added with this batch.

Otherwise I've been working through some issues we have had with the disposition system (relationships between cities). Currently the trading is getting a makeover as the cities trade too often, with too many resources, essentially making them bounce resources between each other. We want to balance the pacing of the game from the point of the second city being founded. Right now all the different aspects of the disposition system, trading, diplomacy, alliances, and war kicks in with the second city, which can be overwhelming in its current state.

Design



Emelie
This week we had a workshop where we went through things that we want to fix with the game, it was very beneficial and added to my work as a designer. When we start implementing these things I hope that you will notice the difference and get a nicer time playing our game. Other than that I tested a part of the storybuilder, now that it's testable in a scene that Johannes W has made, I noticed some things that needed changing but it's coming along nicely.

Production



Marcus
Work on the Community Module is getting completed at a steady pace and most of the art assets are done, which means that we can have one of our artists, Emma, begin working on the updates and features we have scheduled for our next big update. These updates are based on community feedback and internal playtesting, which have identified the biggest issues with Crest, and I have begun scoping and planning these iterations.

_____________________________________________________________

Have a nice weekend!

The Future of Crest



With the release of the Community Module behind us, we started talking about what’s next for Crest. For a whole day the company divided up into groups and discussed what the current problems are and how we could fix them. The following points are to be taken with a grain of salt and are not set in stone.


Pacing/Difficulty



The problem: Crest is overwhelming for new players. There are a couple of changes we have in mind to lower the difficulty curve and to let you have more control over it.

  • Word tablet: Have empty spaces to show that there are words that you have not discovered yet. We’ll keep the word discovery tab but we feel it should be more visible to let you know that you can discover words. The words will also always stay in the same spot and not switch around when you discover a new word like they do now.

  • Tech tree: We’re thinking of adding some sort of tech tree to the word discovery to give you microgoals and have more control over how and when you progress.

  • Building of new cities: We’ll change it so your followers don’t build new cities on their own anymore, only when you write a migrate commandment. There’s a huge increase in difficulty when your followers build a new city and we want you to feel comfortable with the one(s) you have first. This will also give you more control in the beginning on where your followers expand, but make it more difficult the more cities you have. In addition there will be greater consequences for overpopulation, making it not sustainable in the long-run.

  • Limiting actions: Words like “trade” or “socialise” are pretty much done by the followers on their own and make you as the player feel obsolete. We’re going to make adjustments to those, and also tie them in the aforementioned tech tree to not suddenly overwhelm you with another layer of complexity which doesn’t even consider you.


Follower AI (Free Will)



The problem: Followers are too clever without commandments and too stupid with them.

Tweaks: Followers will take more immediate action upon your commandments, eradicating unnecessary waiting time.

Self-preservation: Followers will be more self-conscious about their own mortality, i.e. they won’t blindly walk into an area full of hippos anymore. Hunger will get adjusted so the priority for collecting food will get raised exponentially until they eventually stop their current task to take care of themselves.

Associations: They’ll be more biased towards similar words to appear less random. Right now each word has three or more words that make sense (e.g. Coast - Ocean, Fish, Food) connected to it, where one of them gets randomly chosen when an association is formed.


UI/Visual Feedback



The problem: Lots of information is hidden or spread out in too many places.

  • City needs window: While the spiderweb might look nice, it’s not really useful. We’ll change it to provide a quick overview of the city and summarize Needs, Faith & Doctrine.

  • Bubble over city when a commandment has been written: We’ll change it to make it show more specifically why a commandment has been liked or disliked.

  • World overview: Every city will be shown in a quick overview to make mid-/endgame navigation easier. The city window will therefore be cleaned up and have more info delivered in the world overview.

  • Visual feedback for water supplies: Did you know that mountains are the main source for water distribution on the island, and that water in general has an influence on basically everything? We discussed things like showing snow on top of the mountain, depending on whether it’s full or dried up, as well as animating the rivers that can eventually dry up, but there will probably also be a general overview.

  • Free will feedback: When a follower is acting on his own will, it currently states “I’m doing xyz because I want to.”
    We’ll change the information in the follower window to also show according to what need/craving they’re doing an action. For example: “We’re farming food because we’re starving.”


Animals



The problem: They don’t really matter or in some cases have too much influence.

The animal world will be more dynamic with more interaction between each other and have more drastic consequences for the world.

  • Animals will die more frequently: Starving, killed by predators, hunters or warriors, you name it. Currently too many animals get tolerated within the city borders without any consequences for them or the followers.

  • More migrating: It doesn’t make sense for antelopes to hang around next to lions without panicking or hyenas chilling where there’s no food for them. Not only will them migrating more create a more dynamic world, but also more tension between animals and followers.

  • Nerf animal stats (I’m looking at you, hippo!): We’ll change how much they affect the environment and also how aggressive they behave against followers.


These are all the changes we want to apply to Crest before the final release. Things like polishing, balancing, bug fixing and technical optimization are not taken into account here since we’re constantly applying them and before release we’ll additionally reserve at least one month for it.

This does not mean that there won’t be any new features coming to Crest after the Community Module. We feel like it’s more important to fix what we have now before adding even more things. We’ll have to wait and see how fast we get through the tasks before we can talk about planning production for new features.

You might have already guessed that we won’t have Crest finished as planned by the end of this year. Our new goal is to have it finished by the end of Q1 2018. We’ve already been in Early Access for a long time, it just wouldn’t make sense to rush it now and release something half-baked. Thanks to all your feedback in the recent months, we have a much clearer idea now of what’s left to fix, where the problems lie, and we want to take action on it.

What do you think of all the mentioned changes? Are we on the right track? We’d also like to know what you are currently missing in Crest. Let’s pick up that discussion in the Gameplay/Balancing Suggestions thread.

Dev Blog #152 - Bubbles



Bubbles…bubbles…the character on The Wire? The 1982 arcade video game? Michael Jackson’s chimpanzee? The skull enemies from The Legend of Zelda? The song by System of a Down? Find out what kind of bubbles we’re talking about below.

Community



We posted a F.A.Q. today to answer some of your most frequent questions. Feel free to comment and ask away.

Our wikia keeps growing every day. Let us know if you’d like something to be explained more deeply and we’ll prioritize it.


Art



Martin
I have spent the majority of the week with doing preparation for the new trailer. I'm trying something new, in the film industry you usually do this thing called a ripomatic, which is a fake trailer which takes music and clips from other sources. They show this to the film distributors and producers to get greenlit, but it rarely gets outside of that setting, mostly because it's not really fair use.

Since as I've said it's not really common to share I've censored the screenshot below, so mysterious, right? But I'm confident that the visuals that we will replace it with will look neat, you'll find out soon enough! ;)




Code



Johannes W
I have been working on generating stories this week. The task has been bigger than expected, resulted in not being fully completed yet. I started off this week by transferring texts for different stories to the Unity project. Generating the texts was pretty straightforward. The difficult part has been giving tools for non-programmers to change the story generation. Localization has also been in mind meaning that the stories has to be told in multiple languages.

The current solution for generating the texts in different languages has been to use Unity's ScriptableObject for storing texts. Because human languages can vary a lot in structure, sentences in the generated story is built like formatted strings with numbers that gets replaced in code, for example "{0} effort but it wasn't {1} this time" where "{0}" and "{1}" get replaced by specific random words. For more complex words, like the name of a city or a follower in-game, the numbers in a formatted string are replaced by specific keys. These keys is then recognized in code and adds its corresponding value, like the last name of a city that survived in the game or the animal that ate the last follower...

This is only the surface of the story builder and I am working hard to get it to work properly. We feel like this feature is gonna be an interesting part of Crest!

Jens
A lot of progress was made with the expertise buildings and resource notifications. The buildings now use a better algorithm for placement around the city. They also make use of the new floating notifications to show when an expertise level has been increased.

Otherwise the notifications will be seen for followers that are either spending resources (such as building materials for constructions), or gaining resources (mining, farming, etc.).



Tomas
I have made several changes to the word discovery feature.
Words now have individual progression rates, which makes some words easier to discover than others. These rates are randomized when a new
game is started. There will be a discovery balloon that pops up above followers heads when they progress the discovery of a word. The balloon is finished except for the implementation of new graphics for it.

I have also made Warrior followers stronger against animals, giving the Destroy animal actions a buff. Previously, Destroy animals (Warrior follower) was the exact same action as Consume animals (Hunting follower) except that you get no food from Destroying animals, but you do from Hunting them.



Design



Emelie
In the beginning of the week I started writing the messages that I wanted in the tutorial but it proved to be tedious work and not very inspiring because I couldn't picture how it would be in game. I talked to Emma and came up with a much better way for me at least. I took screenshots and edited in information to see the flow and how I could fit the text in these smaller bubbles. "What is the most vital points that the player needs in order to play our game?" was the focus for the text.

For the remainder of the week I made 60 images about all that is in the tutorial right now, not counting most of the new features. I thought I was done, but then Martin gave me a link to this article and it made me rethink some of my decisions, which is where I am at now, changing the texts and some of the features I thought the tutorial needed but maybe it doesn't. I can't wait for the new tutorial to be implemented in the game and tested!



Production



Marcus
This week I have continued with my behind-the-scenes work that probably won't be noticed in the game but is kind of vital to the company, which I unfortunately can't talk that much about.

I have also been playing the game a lot in order to get a better grasp of where we need to focus our energy in the continued development of Crest.

_____________________________________________________________

Have a nice weekend!

Dev Blog #151 - Brushing Up



Most of the work this week has been around improving usability and user experience based on all the feedback we got from you in the recent weeks.


Community



We started a Wikia for Crest to make it easier to understand all the systems in Crest and in the future we'll eventually integrate it into the game itself. It's bare bones right now but we're continuously adding text and would appreciate your help in making it look more pleasing:

Crest Wikia

This week we also managed to go viral on imgur and made it to the frontpage with the following gif. We accidentally applied a shader that mixes vertex colours and diffuse to a model without UV coordinates. The shader chose to mix the colours with the camera instead et voilà: "Crest - Other Dimensions" was born:




Art



Martin

This week has been eventful, I've redesigned the faith interface as you can see below. We didn't think it reflected clearly enough why you get influence. So we've streamlined the feedback and in the process removed the upper influence limit. Instead you get an allowance calculated from how many followers are happy or dissatisfied with you. This was how it worked generally before, but was more obtuse and confusing. Hopefully you'll like this better!



Emma

This week I've been working on new UI and rework of the current UI. The Faith System is being changed slightly, which affects both the UI for Influence and adds a new tab of Faith. The layout of the Diplomacy tab was also changed, to provide more visibility for each city. With this, it'll be much easier to keep track of how your followers feel about you and how the cities feel about each other!




Code



Johannes

Some more tweaking was done on the loading screen, and it is now done in our internal build. The experience of starting the game is roughly 1.2 million times better now. I've now started doing some research related to making the islands a bit more varied.

Tomas

I have implemented a new city symbol, based on the Acacia trees and shrubs.
I have polished up the monuments feature and added monument ruins using Jens’ glorious remnant system.

Also, I have changed cannibalism so that followers no longer eat each other immediately. Instead, they build up a motivation to do so over time before actually doing it. This should give the player a chance to intervene before cannibalism happens... if they want to intervene of course. :)



Jens

This week has been split into two major tasks. I have been working with our audio designer, iterating through new sound effects that are going into the game later and I have also finished the new Builder follower, adding all the behaviour it needs.

Most of the UI for the expertise buildings is now working, though it will need more iterations later since I used some placeholder items for now. Next step will be to get a floating feedback for when the followers spend resources.



Johannes W

This week I've done a lot of refactoring and adjustments for the in-game underworld. The new game-over screen we are working on includes the chronicler showing the history events spiral. To this point, the underworld scene was coupled to the main world scene in way that it was impossible to use the same assets and scripts for the game-over screen. It has now become more modular, but there is still some more work left to do.

I also started to work on generating unique stories when a game has ended, just to take a little break from the previous task. Emelie wrote different combinations of sentences and words depending on what happened in the game. It should be implemented in our internal build next week, at least.

Design



Emelie
I have finally started work on the tutorial. I have spent time pondering on what is vital for the player to know at the very beginning and focused on those. Many messages were cut and the ones left will need to be reworded. After I’d decided on what to keep in the basic tutorial, with help from the team, I made a flowchart for everyone to see. Now I am reworking the messages that will guide the player through our tutorial, wish me luck!



Production



Marcus
This week I have been busy with getting to know the project better and doing Producer stuff, such as planning, providing feedback and generally helping out where possible, both for the project and for the company as a whole.

_____________________________________________________________

Have a nice weekend!

Dev Blog #150 - Dimensional Growth



Another Unstable Build, a new employee - we keep growing in all dimensions. Besides these big news, many things we’ve been working on for a while got finished this week. Exciting times!

Community



We published the second Unstable Build for the upcoming Community Module today. It introduces remnants of the world as well as monuments that your followers can build. You can read all about the new features here.

We got a new producer on Monday! Say hi to Marcus and hear about his first week here:



I have mostly spent the week getting to know the people and the project in order to effectively do the awesome work of planning pretty much everything we do here. So far it's been really fun and welcoming and I aim to do great things with the team going forward, though probably not too much that actually shows up in the game. Hopefully I'll have some more interesting things to say next week beyond hello!


Art



Martin
I've been focusing on working with the new website for our studio, it's coming along and hopefully will be up in a few weeks! I've also done a mockup for the new loading screen, fancy isn't it?



Emma
I've had a full week of just UI stuff, both general UI fixes and new UI for the so very secret feature you've been hearing about. There's been a need for some small UI changes with the additions of renaming, expertise buildings and monuments. Below is four different versions of a new UI, next to the current version.




Code



Johannes
Apart from some bug fixes for the unstable build, I've worked on the new loading screen, which is just about done, but unfortunately won't make it to the unstable build.

Jens
Unfortunately I can't give any big details about my work this week either since the feature is still not revealed. Though I can tell you it has been an exiting week with a lot of progress. Next week I'll be back to working on the things that are left with the expertise buildings. It is mostly going to be about polishing and fixing the UI since the ground work already has been laid.

Johannes W
This week I've been working on three different things: chronicler end-screen, camera movement, and GUI for the secret feature. The end-screen is starting to take shape but there is still some design stuff that is needed before continuing with the code. The work with camera movement has mostly been about polishing things. Most of my time this week has involved working with GUI.

As a programmer, I've always been curious of learning new programming techniques. The GUI for the secret feature had a lot of visual animations that was created using closures. In short, a closure in C# can encapsulate variables storing state within delegates/lambdas (C#). These closures can then be passed around in the program like any normal type of data. This made it possible to create objects with complex behaviors, like having time varying values, without the need to define new classes that do few specific things or having many adhoc variables scattered around in one class whose only purpose is to keep state. In this case, state where kept in closures instead. If any C# programmer read this, I highly recommend learning about closures! This technique helped me a lot to program behaviors very efficiently.


Design



Emelie
I have continued my work on the story builder, finishing up sentences for a couple of edge cases. I have also focused on the conditions and triggers of the sentences, all that is left now is to see it implemented and watching it work. Of course there will be some finetuning but I'll do that when we get to it. Finally, I have started the work on the dreaded tutorial, we will see how it pans out.

_____________________________________________________________

Have a nice weekend!

Community Module: Unstable Build #2 Now Live!



The second Unstable Build for the upcoming Community Module is now available on the unstable branch. This time we have a few major features instead of many small improvements for you to try out:


New features



  • Monuments - Your followers are now able to build monuments to express their faith in you.
  • Remnants - Instead of disappearing, followers & animals will now leave behind skeletons and cities slowly degrading ruins.
  • New word "Middle aged" - Not young, not old, but middle aged.
  • Bulletin Board tweaks - The notifications won't overlap anymore.
  • Improved camera control - There's now a lot more flexibility, allowing for more beautiful camera angles.


How to access


Right click on Crest in your library
Click on “Properties”
Select the "Betas" tab
Select the dropdown menu and choose "Unstable"
The game will update, please wait (restart Steam if the game is not updating)
Play

Known Issues


There should be no major game breaking bugs, at least none that we know of ourselves.

Have fun building monuments and please let us know how you like the new improvements. This is the last Unstable Build before we release the Community Module in the not too distant future. Here you can read what's left to come.