next »

[epermyakov]

EVERGLORY: An Indie Real Time Strategy Game



Check out the Free Demo on Steam!


I want to share a project I've been working on for a while. It's a strategy game in the vein of Age of Empires or Warcraft. This will be a purely single-player game and focus on a rich single-player campaign with varied missions. The game will offer the ability for anyone to create their own maps and script mission logic with Python. I've laid the first bricks of this project over 2 years ago and I intend to pour all my energy into it to release my game around early 2021.

Some Teaser Screenshots/GIFs










Teaser Trailer





Devlog

I'm starting a YouTube devlog about the project. I'll be posting updates about the progress and mostly talking about the technical aspects of the software. I plan to give an overview of the pathfinding system, integration with Python, and more. The first episode is out now:





Source Code

Did I mention that the engine for the game is fully open-source? Even today. At this point it stands at ~34k SLOC of C. Currently, I am working on it by myself and am not looking for contributors. When my game is released along with the 1.0 milestone of the engine, I hope it will live on as a worthy open-source project.

The code is here on GitHub: https://github.com/eduard-permyakov/permafrost-engine

Next Steps

I'm working heavily on the project. In the coming weeks I'll be ramping up asset production, working to release a trailer and put up a Steam page. The current assets are largely placeholders. In parallel, I'll continue developing the source code. I have some fixes and enhancements queued up in the pipeline. The next big feature will be a Fog-of-War system. You can bet I will make a post and a devlog about it.

Cheers,
Eduard

[epermyakov]

The next episode of the devlog is out. I talk about my approach to implementing a system for saving and restoring the game. It's a universal system that captures the state of the embedded Python interpreter (along with other engine-defined state) and serializes it to a file. The point is that I can have all kinds of user-defined classes, custom UI, and custom event handler functions defined in Python and I never have to write specialized serialize/deserialize logic. The system is able to just take a "snapshot" of any engine session.

A showcase and some more details are here:

[zombieonion]

This looks like a great start! Are you planning to make it more like Warcraft3 with fewer units and more abilities, or the Age of Empires route of massive armies?

[epermyakov]

This looks like a great start! Are you planning to make it more like Warcraft3 with fewer units and more abilities, or the Age of Empires route of massive armies?

In one of the screenshots I have 512 units battling it out on my laptop, and the code is able to scale further. I am targeting a maximum of 200-400 units being controlled by the player. Definitely more similar in scale to AoE than WC3. However, I am going to do more with the hero units than just a buffed up knight. I also want to not overdo the spells, but there will be more than just a heal and a convert. So ultimately it will be something in between, taking the scale of AoE and incorporating some of the RPG mechanics of WC3. Once scene I envision is using around 200 soldiers to take down a massive monster, or having a Helm's Deep style scenario with 500 units at play.

[epermyakov]

Have been spending a lot of time learning Blender. Still doing things badly (not properly managing edge loops and resorting to ad-hoc face surgery), but am getting the hang of it more and more. Here's a preview of some of the first assets for the game.

[epermyakov]

Here's another model for the game. This is the peasant (resource gatherer/builder) unit. Looks less awkward than my last one - has shoulders and non-noodle arms that bend at more or less the right place!

Also, new devlog is out on YouTube here:




In this one I talk go more in-depth about the tech for the movement/pathfinding in the game. This took the most time of all the features in the engine, with many fixes and incremental improvements over more than a year (with breaks and detours to work on other parts of the code).

[epermyakov]

Got my Steam page up. Feel like I've cleared a hefty milestone on this journey.

I had to more or less get my look established for the "reveal" so I've been focusing on asset production over the past couple of weeks. Now back to some more coding to move along the features of the game.

I've updated the OP with some more up-to-date images of the game. Dropped a teaser trailer on YouTube as well:




Would be happy to hear what first impressions are.

[Ramos]

Respect!
RTS games are hard to make but on the bright side, they have a hardcore player type fan base.
Just one thing, as an artist, the game would look a lot better if you use low poly polished 3D or pixel art, the current type of visuals are not so attractive to old school generation(retro fans) nor to current age generation(realistic 3D or low poly) but there might be some small exceptions

Good luck

[epermyakov]

Thanks for the feedback. My personal thought is that the low-poly style has been somewhat overdone in the indie space. I feel it's easy for a game to end up looking super generic. I mean how many games have this exact look? That's why I attempted to step away from the low-poly/flat-shading look into the direction of more gritty/realistic art. Will keep working on improving the execution.

[Ramos]

Thanks for the feedback. My personal thought is that the low-poly style has been somewhat overdone in the indie space. I feel it's easy for a game to end up looking super generic. I mean how many games have this exact look? That's why I attempted to step away from the low-poly/flat-shading look into the direction of more gritty/realistic art. Will keep working on improving the execution.

The example you gave is not very polished but still better than current visuals (at least in my personal hardcore RTS fan point of view), I recommend you check Overland for example on low poly done well in a tactical game.
Or at least do a mockup of low poly version and current version and present to a test audience to see which they like more but the bottom of the line is that this game is your dream and you should make it however you want, that is the advantage of an indie game developer.

Follow your dreams

[epermyakov]

Fog-of-War system is in the works. Here's an early version.

[epermyakov]

Finished the Fog-of-War implementation. Here's it in the works:



Basically just added some blending in the shader to make the edges smooth. And made sure not to render the objects that are in Fog or Unexplored.

Another Devlog is out where I go into all the details:

[epermyakov]

New devlog is out. Spent the last while trying to improve the OpenGL performance of the engine. Documented my findings in the devlog. The results are modest, but it was an interesting learning experience. Performance is hard!

(The gif captures the kind of stuff I was debugging when working on this)

[epermyakov]

New episode of the devlog is out. This is another technical one, where I talk about the implementation of the job/fiber system in the engine. Yes, yes, I know. Spending forever doing the behind the scenes stuff and no actual visible progress on the gameplay. I believe (think? hope?) it will all be worth it in the end. Need to lay a strong foundation, then the gameplay code will be smooth and easy to write later on. Going to focus on some "visible" progress for my next devlog. Anyways, here's the video:

[Ramos]

Nice progress, I can see the gameplay becoming very solid

I am curious, how many you raised on Steam wishlist so far?

[epermyakov]

Nice progress, I can see the gameplay becoming very solid

I am curious, how many you raised on Steam wishlist so far?

Thanks. I'm at about 120 wishlists currently. Getting a slow but fairly consistent trickle, however. So I agree with the wisdom of announcing your game as early as possible. Gives you more time to accumulate views, wishlists, etc.

[epermyakov]

Here's a preview of EVERGLORY's main menu:



Went for a "Slavic Embroidery" look for the UI. Will tweak it a bit still, but overall happy with the look and feel of it. A bunch of work went into getting the main menu up: allowing scripting of the camera to place it in a fixed position, the tech to actually load a mission (i.e. "switch scenes", loading a different script), and more.

[epermyakov]

Spending more time on UI work. Have made a full-on file browser inside of my engine. It's a lot of work, but ultimately worthwhile for a good user experience. And honestly, something 8x worse (more primitive) takes 1/2 the effort. But yeah, this is why people tell you not to make your own engines 

[epermyakov]

New episode of the Devlog is out now. Wrapped up skinning/styling the UI and getting the main menu logic up. Also implemented loading of different missions now. Happy to say that that I'm officially out of the "one demo scene where a bunch of stuff is rendered" forest with my engine now. At the point where I can actually script gameplay, though I'm going to finish off a couple of core mechanics (base building, resource gathering) first.

Here's a quick sneak peak:



And here's the full devlog where I break everything down:

[epermyakov]

Working on the systems to enable gameplay programming in the engine. Here's a Pong demo that I made to test it out - this standalone "mission" is about 100 lines of Python!



Make no mistake, I'm making an RTS, not Pong I just want to be able to be able to create a lot of different, varied missions with numerous scripted interactions. This is just one example to test how achievable that is.

Will be publishing the script on GitHub and dropping a new devlog soon ~~!