Beginning of March Sprint

Happy Friday!

March is definitely going to be a busy month, I'm flying to San Francisco to go to GDC and taking care of a bunch of other things. That being said I still loaded up the schedule as much as I felt comfortable with for this sprint.

The emphasis this month is on getting more content in and hooked up, to put the systems and mechanics to the test and really flesh out the game. Paired with that are some possible work flow improvements, more on that below.

I decided to forgo a build release today, I'm hoping to get a bit more of the work from last months sprint properly integrated into the flow of the game, rather than just working in a test scenario. To give a bit of insight into the construction of the game and keep releases varied and fun, I recorded a short tutorial video. It covers the use of variation "seeds" in the game to produce predictable procedural content. This was largely inspired by No Man's Sky which uses this idea to grow full planets, my hope is to use it to randomize locations and missions. Ideally when you start the game, maps with have their encampments and other such things randomized, so each play through is fresh and interesting, but you can learn the lay of the land in each run of the game. Let me know what you think and if there are any topics you would like to see in future videos.

As I am trying to flesh out more content for this sprint, I have started developing the shoreline map beyond just a placeholder space so it can continue to develop in parallel to the main wilderness map. In order to give future maps a jump start I have started trying out using World Machine to generate new layouts. From world machine I can easily shift large landmass around and try out many options quickly, generating all of the height maps to sculpt and paint the landscape. I don't intend to push this all the way to completion, just to get a loose sense of the space and layout. It will also make it easier to generate similar maps that are meant to share the same biome. I have yet to try the Open World tools developed for the UE4 Kite demo but am hoping I can use those as the next stage to "grow" various foliage around the map. Hopefully these tools together will allow me to get new maps up and running faster so I can focus on the local gameplay and storytelling.

Shoreline map, with loose red circles to indicate possible encampment locations

 Next on the agenda is character art! One of the large missing components of the current pipeline is character art and animation. So I'm trying to rectify that by starting to model the main character and look into rigging options. The render below is the VERY early hi-poly model, which has most of the larger elements blocked out but little to no detail and still missing a lot of the connecting pieces. The most developed elements are the cloth torso and arms, I altered a base model and draped it using Marvelous Designer, pulling it back into Zbrush for cleanup. Still a long way to go on it but happy with it as a start.

WIP "Medical Frame"

Have a great weekend everyone! - CJ

Friday Wrapup! Sprint finish Edition

Hey all!

Its been quite a successful month of progress on the game, lots of work has been added and exciting progress made (at least for me.) The dev diaries have also been going quite well and I've enjoyed making them. Especially as work progresses I will most likely dive into specifics on some, doing tutorials or other content to break things up to keep things fun and interesting.
This week marks the end of the sprint as I mentioned. I tried to get any larger changes done early in the week so I can focus on play testing and fixing any problems with the build to get a nice solid release ready for next week. Work this week included:

• More complete decoration pass on Wilderness map
• Rebuilt resource gathering blueprint functionality to work based on game instance, allows for better persistence and access (previously based in character).
• Initial work on getting controller support for UMG menus (currently only works on main menu, will expand to others once I can better package it up and make it easy to maintain)
• Main map UI and workbench accurately track resources
• Started testing of dynamic menus

The testing of dynamic menus ties into my goals for the next sprint. My high level goals are to focus on the gameplay by A. creating a paper test of the design layer of the game and B. developing the UI framework needed to support the design layer. By focusing on these goals I should be able to get a more complete representation of the intended game as a whole and be able to develop it much more evenly.

Have a great weekend!  - CJ

Start of more "high desert" areas.

Small lighting and decorative revisions

Friday Wrapup! Sweet Cactus Edition

Happy Friday!

I uploaded a new build to IndieDB earlier today and just finished uploading one to Mega (links in "The Game" tab). I'm thinking that I may simply switch over to posting one just on IndieDB as uploading two sets each release is pretty cumbersome. Alternatively I may switch from Mega to another platform that IndieDB supports transfers from, alleviating the uploading that way.

Dev Diary #2 is online and linked below, take a look and leave comments. I'm always open to suggestions and would love to hear about if there is anything particular you all would like to see in these moving forward.

This week was a little on the slow side due to moving but I'm still pretty happy with the progress I've made. Doing well working towards the monthly goals I've set so I'm feeling good about the sprint tasks I've set.

What I did manage to get done this week was some player feedback loop improvements and quick optimizations. For the player feedback I implemented some more UI checks so whether you are interacting with consoles in the base or heading out on missions, you know what is going to happen when you press the button instead of blind triggers.

Optimization is never a glorious thing to talk about but is definitely necessary and can be quite satisfying. I learned a bit about some of the lighting and post process systems, pulling both back pretty significantly which helped performance quite a bit. I probably won't focus on this for a while, just picking where I can but have a few ideas I haven't tried yet that should speed things up significantly. As a related question, do you prefer 32-bit or 64-bit builds? I've been distributing 32-bit as not to alienate anyone but 64-bit would definitely be better for performance. Let me know what you think about that in the comments.

I'll leave you to your weekends with this small model addition, inspired by very cool cactus trees. Thanks! - CJ

Friday Wrapup! Moving Edition

Hey all!

Keeping things short today as I am in the middle of moving. This week was a good one as it started with the release of the playable build and was followed shortly by the associated inaugural dev diary. I'm making good progress on this month's sprint tasks which mainly comprised are mainly comprised of getting quick functional models into both the Base and Wilderness maps. Hopefully I will have more finished and a few gameplay improvements for next week.

Thanks! - CJ

Slight Delay

Sorry for the slight delay, I have a build all ready and packaged up but am running into some intermittent internet issues that are making the uploading a bit difficult. I will keep you posted when it is ready and will try posting it also to IndieDB as an alternative to anyone that is having issues with Mega.

Thanks for the patience! - CJ

Friday Wrapup! Foliage and awkward bugs edition

Hey all!

This week I attempted to focus on gameplay which was a lot of fun and made some good improvements, albeit somewhat stymied by a persistent bug.

The gameplay was mainly focused on providing good user feedback and making the combat a bit more "snappy". I was looking quite a bit at the isometric Tomb Raiders as reference for this, even implementing a version of the trigger-able mine seen in those games. Also touched up the effects and feel of the primary weapon which I'm pretty happy with for now.

Unfortunately I have been stuck with a bug where the player mesh is lost every time I leave the editor so until I sort that out I can't really make a build. Frustrating but I will keep on it, I'll even rebuild the main character from scratch if I need to.

Due to the bug and migrating the project to the newest version of Unreal, I also spent a little time adding some more foliage. I'm excited about the new broad leafed plant as it was my first test in migrating the foliage over to gradient map texturing. The plant uses a height map to define the shader driven gradient mapping so I can easily swap out colors or stack them using indexes. My hope is that this will make them easier to use over a variety of environments and easy to balance the color in the scene. I will migrate the grass and other plants over soon as I start tightening up the environment more.

Hope everyone has a good weekend, and this is the last post prior to the holidays so there won't be any post next week.

 Happy holidays! - CJ

Friday Wrapup! Workflow and UE4 stream edition

Happy Friday!

This week was a fun one for me as I had a bit more time to delve into the game than usual. The two things I chose to focus on were artistic process and management design.

Art

I've always been a bit fan of workflow improvement and optimization and was really inspired by Michael Pavlovich's GDC talk "Blurring the Separation Between Concept and Production." This led me to try out a new twist on my modelling process with the intent of getting things in engine faster. In general it is:

1. Greyblock model - Quickly test something a very block, ugly version in engine to test overall scale and sillouhette
2. High poly - Still focused on large shapes and not on details, using tools like Zbrush's Zremesher to get sculptable topology.
3. "Auto" low poly and unwrap - Using tools like decimation master and unwrap in zbrush to get quick representative models
4. Batch texture - Pipe low poly and high poly into Substance designer or Quixel suite to both bake and texture.
5. Import into engine.

There are still a lot of steps I would like to streamline and automate more but I really like this process because it means in a very small time I can go from high poly to having something in engine that I can actually look at. Where this really shines is the fact that I can go back to any step in the process, refine something, and everything will update with very little to no effort.

Using the above process I made this crate, medical tub, and partition curtain. I'd like for everything to feel almost like a handpainted miniature, very tangible when I'm done so they all were piped through the same slightly dirty clay substance for now. I will be doing more of this work to flesh out the other models in the base before refining anymore textures too much so the area feels more lived in and less empty.

Design

I also made some headway on the management portion of the game. I wanted to get some of the managment gameplay in and working as soon as possible to balance and give context for the more action oriented parts. I know that this isn't quite as exciting but I think will be very important for the overall game feel and balance. Right now I'm developing it as its own module so I can play through the days very quickly but will hook it up into the main game as soon as I feel it is more complete and balanced.

Right now it is loosely functional, you can cycle through the day's with a placeholder section that gives you a random amount of resources that you would normally fight for in the missions. I'm hooking up the UI so you can assign personnel and choose destinations now, with base upgrades to follow.

Misc

Finally, I also sat in on the weekly Unreal Engine twitch stream. This week they talked about developing the characters for Paragon and rendering challenges. Lots of fun to watch and interesting to hear them talk about some of the more technical aspects and developments. New upcoming tech includes:

• New hair shading solution - Anisotropic highlights on masked hair with supporting techniques that get rid of most hard edge penetration issues.
• New eye shading solution - Similar to skin shader but allows for more convincing wet surface. Coupled with material techniques for recreation of the lens, iris and other physical phenomena associated with the eye.
• New cloth shading solution - Based on research and work done by Ready at Dawn for the Order, adds "fuzziness" which enables the recreation of a wide range of cloth types without fresnel fakery.

Hope you all have a wonderful weekend!     - CJ

Friday Wrapup! Locker edition

Happy Friday everyone!

Things were pretty slow this week as I am still waiting for news on whether my harddrive can be rebuilt or not. So unfortunately that means that there is no new build today. However I have tried to keep things moving forward in other areas, working in parallel to avoid overlap should I be able to integrate some of the lost progress from my old drive.

So this week was mainly comprised of:

• Writing and Design - Wrote up a blog post with a synopsis of the game's story and overall design intent. Check it out here if you haven't already!
• Learning - For a couple different reasons I have migrated to using Modo as an upgrade for my horribly outdated version of 3ds max. It is slow going but am getting a bit more used to it as I work.
• Learning - I'm always on the lookout for pipeline and workflow improvements and decided to give a method of normal map baking by Nick Quackenbush for Substance designer a try. I tweaked it a bit by creating two unique bakers in Substance so I didn't have to save the files out. Pretty cool so far, definitely useful for the hard surface stuff I'm starting to work on!
• Production - In an effort to work through the aforementioned "education" I decided to work on a couple very simple designs for the base, namely lockers. I wanted to learn using something relatively simple so I could work out the kinks while still moving forward. I will probably end up exposing a couple controls in Substance and saving out a few variations for the different areas but I'm pretty happy with how they look so far, even in an unpolished state.

Have an excellent weekend all! - CJ

Friday Wrapup! Spooky edition

Happy Halloween everybody!

This week didn't quite go as planned but still managed to make some progress here and there. The barrier of course was my storage hard drive dying on me, but fortunately I already have a replacement installed and a couple week old backup of the data restored. I did end up sending out the dead drive for data recover, we will see how that goes. So in the meantime I am switching over to parallel tasks that won't require ugly reintegration should I get the data back in a usable form.

So this week was mainly:

• Fighting with my hard drive
• New text game build - More in depth design work, specifically on the text game test and its role in character abilities and upgrades.
• Writing some basic lore and game premise info (will be shared soon, needs editing)
• Made a material orb to be used as an standard primitive when authoring smart materials. More info below.

Please take a look at the text game and leave feedback and thoughts. The game is a lot cleaner now, and though it still doesn't have much in the way of player choice, it is getting closer to a rough "alpha" state.

The one art task that I focused on this week was to make a material orb for use when authoring textures. I've wanted to make one for a while so I could standardize and make it easier to craft "smart" materials more readily.

 If you aren't familiar with the concept of smart materials, they are a relatively new way of texturing models. By supplying the material with a small set of support textures that are baked for each asset, the material is able to adapt the visual detail to the contours and shapes of that specific asset. This means that if you make a "smart" metal material and then apply it to two different models, it will generate textures for each with the rust accumulating in all the right cracks and oil residue dripping down the surface in the right direction, etc. This coupled with physically based rendering makes it much easier to achieve good looking art that doesn't need adjusting in different lighting scenarios.

 A material orb simply makes this process a bit simpler and more streamlined as you can test the materials against the same model, and because it is built with this in mind, will look good on a variety of surfaces. By decoupling the material authoring from the unique model you can make a handful of materials that can texture an entire environment, and using newer previs techniques, use the materials to inform both the concept art and final models without any extra work.

Happy Halloween all!

- CJ

Rough textures testing out new material orb. 3 different lighting environments.

Friday Wrapup! first edition

Hey all!

This week was an exciting one for me, to finally have the game shared with the public is something I've wanted to do for a long time. I will try to do these weekly wrap ups to share any progress from the week, and it will also help me to stick to a bi-weekly build release. Look forward to a new build next week!

This week mainly consisted of:

• General - Getting the game posted and setting up a few main threads to share it (Polycount, IndieDB)
• Art - Greyblock modelling for base architecture and some initial texture and material work.
• Design - Minor work on Design Test text adventure. New version posted on the game page.
• Design - Cleanup and rework of design doc to prepare for next steps.

The main focus this week was Art, specifically for the game's base of operations. The location is intended to be a semi-derelict medical facility/bunker. I greyblocked (rough un-textured proxy) models for the various areas and started doing test decoration of the spaces. The current idea is to have four main sections; Clinic, Fabrication/Engineering, Lab, Bunks/Dorm. The layout is still quite temporary, it still needs to be informed by the design and narrative needs of the space but that being said I'm still happy with it as a start. I continue to enjoy making "smart" art using UE4's blueprints, using procedural rules to make it faster and easier for me to decorate.

For example, a simple blueprint that stacks a randomized number of crates on top of each other, randomly rotates each in 90 degree increments to break up the faces, then offsets and rotates by a couple of degrees for a more naturalistic final look. A pretty simple example but hopefully gets the idea across.

If you haven't already, please download and play the main game from the game page, or if you are short on time, take a look at the design text adventure.

Have a great weekend!

- CJ

Some of the greyblock models created for the base.

Slightly revised clinic with vertex painted sand and a few models slapped in.

Updated base with vertex paint and a few of the new models. Still needs actual decoration pass.