eromancer
patreon
Inner Circle Update for August 16 2018!
🕑 Added 2018-08-17 05:01:18 +0000 UTCHey guys!
Not as much to show off visually this week, but we’ve made steady progress nonetheless. I’ve learned a lot about scripting Maya, which I’ll bore you with below :D. TK has redone the face rig. Mr. Kittyhawk and Ubercharge have been working on clothing assets, and Limbo has started on a new character design. Check it out :0!
Ero’s Maya Conversion Progress
I’ve been knee deep in Maya scripting all week, tackling some short-term material solutions as well as opening up cans of worms to get the lay of the land for longer-term improvements. The first major success is that I have the material transfer script I had started at the end of last week working. I’ve spent more time than I wanted to here, as I’ve tried to flesh this script out and futureproof it as much as possible because I have something larger in mind for it in the long-term. Here’s what we can now do that wasn’t possible last week:
- Transfer materials between any two similar objects in one click. Shape can be different but polygon number and structure must be the same… ideal for character bodies, since all of our human characters now use Genesis 3.
- The same for entire hierarchies of similar objects (two rigged characters for instance, with all their clothes, accessories, and such).
- Quickly make polygon selections by material. If a coat is all one mesh, is leather, and has metal buttons, I can now select just the buttons in 2 clicks.
If this stuff sounds like basic functionality, it’s because it is in my opinion; all the more reason why not having it during the conversion process has been a pain in the ass. I haven’t been doing flashy character renders recently because our character rigs are constantly getting updated, and without a way to quickly transfer materials to the new models I’m left with gray blobs :D.
Regarding that “long-term” idea: look closer and what I’ve created so far is basically the groundwork for a material preset system. I used material presets a whole lot in my previous Daz workflow, and need it now because of the degree of detail we’re going for with stuff like material-based wetness. In Daz I could save a material loadout for a character and reload it later from a library. Maya doesn’t support this out of the box, and we haven’t found third party software that addresses it. What I’ve worked on this week has gone a long way in giving me perspective on what we’ll need to do to make that possible down the road.
On that earlier note of character rigs getting updated constantly, I was hoping to have a small library of facial expressions to show off, but TK’s been working exclusively on the face rig this week, so it’d be pointless since I’d have to redo it. While Genesis 3’s facial expressions have been brought over from Daz, I think some kind of suck, and having easy access to the face rig should allow for some improvement – especially with the eyebrows and mouth. So, in place of getting that done, I’ve been making posing more efficient. TK taught me how to make marking menus in Maya; essentially radial quick menus like you'd see in a game. I’m currently making scripts for things I’ll use regularly during posing; stuff like resetting the face rig, toggling joint limits on/off, and so forth.
TK’s Rigging Update
The face and eye control setup script got a nice round of rewrite and polish. The previous version would move the controls to the joints (with an offset) and go straight to work on setting constraints for animation. The problem that Ero had was that they can get visibly lost underneath the geometry while posing, and also that the initial offsets would be a problem on other face shapes. Say a character with a bigger nose might not even have a visible control. Clearly not good for daily use. Repositioning these controls once they're set is both technical, time consuming, and a big mood killer.
The idea is simple: the initial import and setup only places the controls in roughly the correct place. This lets us make whatever changes, before moving on to shiny new button in the importer UI for attachment.
Problem was, I was super-duper "smart" about how I wrote the first version. So smart that I realized 3/4ths through writing it that it would need a big redo! I had written it to position and attach everything in one big logical go, but separating those two ideas out of it totally broke the whole routine. So, this new version is actually a smart idea, and it's something I can easily make changes to later if I need to (like an option to detach again). It's working great and I even found some bugs and improvements over the first version.
Clothing Asset Progress!
Limbo, TK and I worked on some basic 3D shapes to mock-up Exarch Nine’s visor. I wanted Limbo to do some extra sketches of it, and it evolved into 3D mock-ups since differences in the shape are subtle but change the vibe a lot. I decided we should probably move on until after I finalize the character’s face and hair, since going further at this stage would be kind of moot.
Mr. Kittyhawk finalized the topology and created a new non-procedural material for the syndicate guys’ coat. We decided my procedural version wasn’t a great idea considering we’ll likely need to animate the coat, and it would likely produce strange results while moving due to all the folds in it. Here are the final results!
Last but not least, Ubercharge has mostly finished up the model for one of the Splicer masks, which you can check out some pictures of here.
AltairPL’s URGE Engine Update
Well, I think I might have exaggerated a bit when I said lately that I would be long done with the archive stuff if not for my family medical emergency last month. Even though I'm making pretty good progress lately, I still see a lot of work on the horizon, like finishing implementing versioning stuff and a few other more or less small things.
I consider all of this version stuff cursed, since every time I start working on it, something entirely different is taking me away from it. But between all those interruptions, I already made solid progress on it, both planning/designing and actual implementation. During designing phase I started thinking about something that was bothering me for some time now. Apparently, when some people see something like '0.05' they automatically treat it as a number equal to 5%, which never was our intention. I was able to convince Ero to adopt popular [MAJOR].[MINOR].[PATCH] version scheme, which should work much better for our needs and make my life a little bit easier. I really hope that changing version notation from '0.0563' to '0.5.63' won't cause too much confusion, but Ero is a little bit less optimistic about it. Previously, we have planned to skip few versions and tag next big release as 0.10 instead of 0.06 (0.10.0 looks less confusing in new notation than 0.6.0), but I'm not sure we should make such a huge version leap for Battle Tests that will come before the next big release.
Anyway, all the stuff that was distracting me from version stuff would have to be done sooner rather than later, so I'm not fretting about it. Most of it was done in preparations for the inevitable first URGE-only release, and was related to getting rid of some old and decrepit RM compatibility stuff, implementing functional changes that veer a lot from old RM approach, and to utilize URGE capabilities better. One of such things was implementing a dedicated game ini for URGE, which fits our needs much better and makes configuration much easier. Another thing wasn't really planned, but after seeing how good the game archive performs, I decided to get rid of the mini URGE archive and move its contents, mainly shader source files, to the game archive(s). This allowed me to get rid of some extra code required by this mini archive, not to mention that shaders modification will be now much easier.
Once in a while, when testing stuff, I would notice a small problem related to my implementation of rendering queue when resetting the game. I was always forgetting to even make a note about it, so when I noticed it again recently, I decided to fix it once and for all... and I did.
Ruby scripts finally got some love too - another batch of workarounds were removed or conditioned, which will make their later removal much easier. Scripts preprocessing was also updated, which will streamline making release builds for both RM (mainly for testing) and URGE. I also made a lot of preparations for changes that will be required for new version of version stuff ^^.
Hopefully now I'll be able to concentrate on the version stuff and everything related to it instead of jumping all over the code. I would really like to finish everything archive-related and put it all behind me.
I get the feeling I forgot about something important, but there's always another update, right? ;)
Meet Mr. Black!
Mr. Black is yet another new character that you’ll encounter in the next major version. Limbo and I have been working on a bunch of concepts for him this week. We’re not quite there yet, but hopefully these will give you an idea of what we’re going for.
Let us know what you think!
Comments
MB
If this was a democracy, I would vote for the change to a [MAJOR].[MINOR].[PATCH] version scheme.
MangoFishSocks
All this coding stuff has my head aching. Can't imagine yours. Best of luck as always. Mr. Black looks very ominous but not in an antagonist way. Just unfortunate, worn and beaten down. Think Patches from Dark Souls, maybe mixed with some Fallout's Mysterious Stranger. You don't trust him but he's always there to help you in his own way.
