V0.05 Progress Update!

Hey everyone!

It's been awhile since there was a legitimate progress update, so the following is my effort to hit the important points of the past month or so without making you read a novel about it.

EDIT: Just kidding, it turned out to be a novel anyways.

Spring Cleaning Complete!

Well, it's summer technically, but as I think I mentioned in the previous post I was making a huge effort for the first few weeks after we put out the last patch to put optimizations ideas into effect that had been stacking up for awhile. This is especially important since TK was coming on board full time and it'd only make sense to reorganize things before teaching him processes and such.

• We now have complete cloud integration, meaning TK and I share all of the art related files and 3D libraries and receive updates to each other's work in (nearly) real time. It took over a week to upload and sync all of the necessary files.
• We've moved to a new chat client for development. We're all a bit meh about it, but overall it's better for working with a team than what we had previously.
• I've personally updated Photoshop from CS 5.1 to CC so TK and I are running on the newest version (and so that I could get the last 6 years of updates). Definitely worth it.
• I heavily upgraded my internet connection to remedy my super slow upload speeds (would've been nice to have during the V0.04 patching debacle... and for the aforementioned cloud integration).
• I redid my laptop's OS install entirely (a nightmare by the way since I "downgraded" Win 8.1 to 7 and the laptop has a really weird hybrid SSD hard drive configuration, meaning I had to HIGHLY customize the Win 7 install in order to fit the OS on the SSD partition), with the result being that I now have two rendering rigs that basically share the same file system due to the aforementioned cloud integration.
• I redid my entire desktop workstation's layout, bought a bigger desk to mount the laptop on, recycled a bunch of old PCs / hardware / sold music equipment and generally improved my cramped living conditions (I live and work in a 9'x13' room, basically). I also added a hefty UPS battery backup so I don’t have to worry about severe weather or hardware damage / data loss from power outages.
• TK and I found a solution for customizing CPU priority in our rendering engine of choice. This means we can actually work on other stuff while things are rendering instead of the system being completely locked down by an overzealous rendering engine hogging resources.
• I did at least a week's worth of massive cleaning and much needed reorganization of all the game development files (nearly 200 GB worth).
• We added a private forum for Inner Circle members so that there can be actual discussion and feedback on topics instead of one way presentation of content.
• TK introduced me to Pinterest, which we will now be using for easily stashing and visually referring to links to licensable 3D assets, reference art, and relevant game development techniques/tutorials. The idea is that over time we will accumulate a shared library of potential assets / concepts that we can quickly refer back to.
• I upgraded my phone all the way from iOS 6 to iOS 9 so that I can actually be productive when away from my workstation. Apps now work, and it's pretty glorious.
• I now have Maya (a popular 3D development program), and now I need to learn to use it… this leads to the next topic.

TK's Map Editor

When TK came on full time he proposed a map editor that should drastically increase our efficiency when it comes to making the overhead retro JRPG style maps.

Our current method is to make 2D tilesets, build the maps in RPG Maker’s editor using these tilesets, and then export images of these maps to be edited in Photoshop. This is highly time consuming and labor intensive since a lot of elements such as shadows have to be placed by hand.

TK’s idea is to use a heavily customized UI in Maya with scripts designed for our workflow needs to remove the need to make 2D tilesets entirely. Instead, the entire tilemap can be built in Maya, and in theory we should also be able to make the maps at a higher quality and seem less repetitive. The results should look comparable to the sideview maps and battle backgrounds, which are done entirely in 3D with relatively easy 2D post in Photoshop.

Some of the major hurdles he’s overcome so far include learning the process for map creation so as to improve upon it, finding and implementing plugins and write additionally required UI scripts for Maya, and replicating our 3D shaders for the rendering engine we plan to use with Maya (shaders are programs used to get the basis for the unique MATM aesthetic when rendering the 3D models).

Assuming it all works out (which at this point it looks like it should), the amount of time saved over the course of the project will be huge, and we should end up with higher quality maps to boot. Naturally, I haven’t used Maya before (I’m a Blender guy… not really by choice, but it’s what I had available and thus stuck with), so TK also will be teaching me the basics of navigating Maya so I can actually use the new map editor.

Here’s a small gallery showcasing his work in replicating the MATM look in the new rendering engine: http://imgur.com/a/cL8SA

AltairPL's Map Overhaul

TK’s map editor is an idea that comes at a good time, as AltairPL has been progressing steadily with his map overhaul, the goal of which is to not only enhance performance but also simplify the map and event making process. He’s written up a super high level description of some of the stuff he's already accomplished. He also included a really impressive summary of the FPS gains he's managed so far.

• Basic optimizations of maps and characters [performance]
• Added basic (for now) customization of map events, which allows easier creation of more advanced and dynamic events [functionality]
• Allowed for RPGMaker’s tilemap structure (which is no longer even necessary for our maps) to be removed entirely, freeing up a ton of processing power on overhead maps [performance]
• Removed a lot of the map-oriented 3rd part scripts and have begun replacing their functionality with more case appropriate code [performance].
• Changed how map events are handled, which cuts number of needed events considerably [performance]
• Added map data preprocessing, which greatly improves access time to map data in game [performance]
• Changed how map data is created and edited, which greatly improves map configuration time and minimizes the chance of an error (invisible wall, walking in the air, etc.) [functionality]

Sample of map performance change for some of the Access Tunnels maps (in FPS using AltairPL’s PC for testing) - not everything is implemented, so these values may change (both ways):

• B5 Hub: 56 => 78
• B5 West 1: 45 => 63
• B6 Hub: 55 => 84
• B6 Puzzle: 60 => 93

Neon's 2nd Outfit

The boobs are loose!

I just spent the last few days working on armor damage for Neon’s new outfit.  The idea was to do it entirely in 3D if possible so that the very time consuming manual painting wouldn't be required like it is on Malise's.  As I've mentioned before however, getting the hentai-esque details we want is a really tough task in 3D since the models still have to remain highly poseable.  Suffice to say, it’s been accomplished and looks great.

I’m also roughly halfway through the general combat poses for the outfit. Once finished there will be about 45 total variations on the 9 base poses, since armor damage on this outfit happens in such a way that a few more variations are required.

Here’s a small preview, including a before and after GIF that shows the new post processing I’ve begun using that features more vibrant colors and improved shading: http://imgur.com/a/KKYD2



That's about it for now!