Eromancer posted
March Progress Update!
Hey everyone!
We’re less than three weeks out from the Battle Test 2 release, so this update is not be quite as lengthy as our usual ones. We still need up to a week or so to get Battle Test 3 (BT3) ready, so the plan is to release it for the $3 tier at the beginning of April and we’ll finally be back to posting monthly updates at the beginning of the month. The good news is Ubercharge completed the battle/portrait lighting rig that we intend to use throughout the game, and I’ll be using it to light the portraits for BT3. The results are better than our previous ones (you’ll find an example below), and it takes far less time to configure. Some other big news is that we’ve updated all of Malise’s base files (we had a pretty huge list of fixes piling up) and in doing so took some time to improve upon lingering cosmetic issues / act on feedback we’ve received. The result is a new face, a slew of body fixes, material edits, and new hair. Here’s the list of topics for this post (I’ve updated these sections since the last IC update):
- Malise Updates Part 1
- Malise Updates Part 2
- Ubercharge’s Battle / Portrait Lighting System
- H Artwork / New Processing Test
- AltairPL’s Coding Progress Update
- Ubercharge’s Map Progress
- To Do List
Malise Updates Part 1
TK and I went through probably three different variations of Malise after Battle Test 2, and it’s been both very frustrating and a solid learning experience in how to manage stylization with the new art style. Our most realistic version, which we put three days into,
You must be registered to see the links
but simply didn’t mesh with the body, hair, or other characters. While that sucked, it did help me to learn what features work with our style and what doesn’t. For the past few weeks I’ve been studying artists like
You must be registered to see the links
and
You must be registered to see the links
(League of Legends), as I think segments of their work are approximately the level of stylization we need for faces/bodies. At any rate, our work resulted in quite a few changes.
You must be registered to see the links
Below you can check out the new face model, as well as the more realistic one we passed on. The most obvious change is probably the
much better eye shape. You can also expect better facial expressions once we’re done.
There were also a host of changes/fixes to the body. The biggest difference however is probably the hair, which I know some people felt was too plain. One especially good thing about the new hair is that it should be much easier to pose.
Lastly
You must be registered to see the links
showing the (almost) finished face. The hair wasn’t done yet when these were captured. Please let me know what you think!
Malise Updates Part 2
Since we revealed the above changes last week to the Inner Circle crowd we’ve done some additional work, and I’ve rendered a new set of images to illustrate some of these as well as show some expressions on the new face.
You must be registered to see the links
(no processing, these are straight from the viewport).
And here’s the list of edits:
- I’ve finalized the hair material edits.
- We’ve further sculpted the hair and bangs (sharpened the points and tried to make the bangs look more natural and mesh better with the new ponytail by messing em up a bit). Check it out in the images below.
- Finished a set of fixes for her mouth and teeth position, which were making expressions with an open mouth generally weird.
- We’ve completed other miscellaneous fixes such as
You must be registered to see the links
and
You must be registered to see the links
that I’ve expertly hidden until now 
(I think).
- TK re-rigged the holsters, which now work perfectly (
You must be registered to see the links
).
- Improved skin glossiness and sub-surface scattering
- TK has completed the transfer of suit morphs/configurations to the new body shape.
- Forgot to mention this previously, but the suit’s highlights will be subtler and less distracting with its new material, but still be obviously shiny.
Ubercharge’s Battle / Portrait Lighting System
The goal of system is to simplify the battle lighting process so we don’t need to move lights around for every environment (something really time consuming as you saw from our work on Battle Release 2). To accomplish this, the vast majority of lighting for portraits and battle artwork will be done in post.
You must be registered to see the links
.
The system works by using
You must be registered to see the links
. When I render out a battle portrait, I simply have to render each light group individually, and then I’m done with that portrait for good. This will also work for enemies. Then, using Uber’s black magic, we can actually blend those light groups’ intensities and color in post to create lighting for any environment we may come up with in the future. All three of the very different lighting examples in the image linked above were generated with the same set of renders. Eventually we may even be able to do this in-game in real-time, which will cut down on the hard drive space required for portraits by a ton.
You must be registered to see the links
, which will use the new lighting rig as an opportunity for me to get some practice with it.
H Artwork / New Processing Test
I’ve started working on some H artwork and took a couple tests to completion. I’ll likely be putting these together intermittently over the next few weeks and hopefully by the time we need them I’ll have a full H skill chain or two ready for one of the releases.
I’ve also been testing out some new processing effects, including depth of field and chromatic aberration, both of which are present in the above images. Uber taught me how to add depth of field entirely in post, which is a pretty great benefit of the 32-bit workflow. I’m using the chromatic aberration (color bleed around the edges) effect largely to add to the retro film vibe. I’m considering carrying this
You must be registered to see the links
and elsewhere now that we can blend effects better.
AltairPL’s Coding Progress Update
If you saw the Inner Circle updates this month, skip to the “***” symbol for the new stuff.
Just before I decided to take a break from the battle engine overhaul, which resulted in URGE’s conception, I made 2 serious mistakes. I left the targeting system untouched AND I made no notes or comments about it. Both came back to haunt me when we started releasing Battle Tests... the 1st Battle Test had some serious issues with invalid actions being performed, and the 2nd one was riddled with similar unwanted behavior, mostly caused or made apparent by fixes for issues from Battle Test 1. What a mess...
The only thing I could do in a situation like this was to finally rewrite the targeting system, but after starting this, I've learned that all action progression was depending on more or less vanilla RPG Maker behavior and some other wrong assumptions. So, the only choice I had was to redesign and recode the whole action progression with new target determination in mind.
The new design was relatively simple on paper, but actually coding its prototype required some pretty nifty work. I must admit I'm very happy with both the speed and the results. Suffice to say, just this one thing eradicated most of the current problems with in-battle actions. The new system is not only less buggy and more stable, but also allows for some additional action features to be implemented, including one or two already done or planned. Also, while implementing it, I've made a lot of notes about things that may require some more attention sooner or later - possible death-traps (including stuff actually related to character/enemy death), things to optimize, good places for planned features and so on.
A lot of work, but well worth the effort and time. It still needs a lot more work and testing, but after a while I got fed up with it, and since Ero required assistance with something entirely different, I've changed gears for a while.
***
What Ero needed from me was some answers about how well the engine handles diagonal stairs. As some of you remember, the last time those were used in a game was in the “Old Town” maps from v0.02 (area with Splicers and Stigmata), which was a long time ago. In the Old Town maps, diagonal stairs were handled by a 3rd party script, which did its job, but was extremely limiting and a bit clunky. On top of that, it's nowhere near being compatible with my implementation of map levels/layers, so I needed to code it more or less from scratch. Problem here was that stairs are pretty much a part of movement code, which was in shambles and needed to be dealt with before I even started adding to it. Fortunately, I made some conceptual design of stairs mechanics pretty quickly and gave Ero his answers, so he and Uber could continue working on maps with stairs while I revamped the code needed to handle it. Suffice to say, all movement code was updated, some hard-coded things were replaced with more dynamic counterparts, a lot of stuff got cleaned up and better organized and diagonal stairs are now fully functional.
After finishing the stairs/movement stuff, I needed a short break from coding. Since some things related to battle still require some kind of overhaul, I've spent a few days on writing a short novel detailing what and how they need to be done – it should help a lot when the time to actually overhaul them comes.
With this out of the way, I turned my attention back to the battle stuff. The first thing after getting back to coding was to implement the sprite layering system required for bloom and HDR processing - gonna spare you the gory details. Portraits should look much better in Battle Test 3, with hopefully not too big a performance price to pay.
The next thing on the agenda is the aforementioned "notes about things that may require some more attention sooner or later". It's my current work-in-progress, but I'm already very glad I started doing it now, because it helped me to eradicate one critical issue I thought was limited only to holds, but as it happens, could occur pretty much for every action. I really hope this will finally make the battle engine rock-solid.
Ubercharge’s Map Progress
Eromancer here again. I helped Uber out with a block-out diagram for his new map, for which he intends to use his procedural shaders that I showed off during the monthly update. He just got to the point where he is able to test them out, but so far it looks like they will perform great. Once he got it working
You must be registered to see the links
-- no texturing or UV mapping required.
Here’s a set of images showing his progress from the initial block-out stage to the mostly finished geometry (still needs to add props).
-
You must be registered to see the links
-
You must be registered to see the links
-
You must be registered to see the links
-
You must be registered to see the links
To Do List
Here is a broad version of our to-do list for the next few weeks. I’ll probably add one of these more often from here on out. I’ve checked off stuff we’ve completed since the last Inner Circle update.
Eromancer
- COMPLETE: Some remaining Malise hair edits
- IN PROGRESS: Update Malise battle portraits with her new version
- IN PROGRESS: Create upper body exposed portrait versions once TK has the suit ready
- COMPLETE: Work with Uber to recreate his battle/portrait lighting rig in Daz
- Continue experimenting with Police H skill artwork
- Calibrate Daz lights for HDR processing (currently doing this in post and it’s resulting in degraded dynamic range)
- IN PROGRESS: HDR processing on Neon/Malise’s new portraits. I learned batch processing in the program we’re using for this and completed Neon’s portraits, but may do them over with Uber’s battle rig lighting (shouldn’t take long since it’s all procedural).
TK
- COMPLETE: Update Malise’s Maya scenes to reflect the new body edits
- COMPLETE: Transfer updates to alternate suit configurations (zipper open, etc)
AltaiPL
- COMPLETE: Implement sprite layering system required for bloom and HDR processing
- IN PROGRESS: Continue work on battle engine
Ubercharge
Post BT3
- Paint Malise armor damage concepts for TK
- TK and Uber to begin sculpting / implementation for Malise’s armor damage
- Enty launch
- Update Neon and her base scene files / prepare Neon's armor damage
- Help Ubercharge with Splicer overhaul
- Start Ven overhaul(?)
You must be registered to see the links
