Archive for the ‘pet projects’ Category

Here is another attempt at the interrogation chair project, I like it a lot more than my previous attempts which I’m not sure I’ve ever posted here. I was hoping to integrate some kind of knee-spreader, but so far, cannot figure out of a good approach that also fits the overall design of this contraption.

I look so unhappy in this pictures, ^_^ I wonder why!

Blender’s renderer & denoising have really come a long way, and this is nearly at real-time speed on GPU compute. Would you look at that.

As for the Vergilius collar, I finished writing the code yesterday, I was going to do product pictures, manuals, and all those little things but my home server appears to have given up the ghost, which pretty much ate the rest of the afternoon.

I haven’t entirely lost hope but I might have to look for a replacement… and an IDE drive adapter of sort (it is THAT old). Thankfully, it did not contain much of value other than lots of scripts and somewhat useful services and things I figured out over the years. For the time being, I’ve moved the really important stuff (customer mails & sale reports) to a different machine.

Alright we’ve had a few setbacks. Last minute change and I decided to add an occlusion map, which required of me to try a bit harder with my UVs, which also required rebuilding the LOD models that I thought I was done with Monday.

I’ve also decided of a name for this line “Vergilius”; It has a nice ring to it.

There is a problem however, and that is related to the script/behavior. I really have two options with this slightly unusual “side locking” system:

• Option A: Treat both lock points as one, simple to use, but that means they cannot be chained to or from.
• Option B: Treat both points independently, that means the collar would only be RLV-locked if BOTH lock points are fitted with a padlock, both would then be valid chain targets and sources (side collar points currently only exist on the braced collar)

EDIT: You can vote which option you prefer here.

PBR is officially out, right? It’s been for months on the official viewer, and is now on Firestorm, which means that it is out for 99.9% of the user base.

I’ve decided to try my hands at a quick little project and see how well it ends up doing.

In general, I tend to avoid making latex products unless it is something truly special, as there is little point in competing with creators that pump that stuff out all year round. That being said, I think I’ve gotten pretty good at making actual, realistic latex/rubber using PBR materials, and people have been asking for a serviceable rubber set for literally years at this point.

This will be PBR-only (sure, try to change my mind on that one) and should be super easy to recolor and mod. I think I’ll also keep the front pin/post as a separate mesh, this way you can move it in another hole if you don’t want it centered, for some reason.

Ended up going with modeling the LPM apron bib after all. A bit dense but it is probably okay. There really isn’t that much left to model at this point, and I’m going to HAVE to move to the more nerve wracking parts…

I’ve actually gotten the logic for the VRChat version of the stunprod nearly done over the weekend. The behavior is pretty much identical to the one in SL, but in SL you can’t do screen-space shader shenanigans :3

Oh, boy do I love optimization… even if it is slowly killing me.

I ended up redoing the surface of the apron a good 5 or 6 times, I’m still not happy of the geometry it is giving me, but it kind of works. It currently clocks at 27000 triangles, all the pleats add so much geometry… I’ve optimized the corset lacing a bunch, but I need to remove a lot more to make it acceptable. I also modeled the collar.

During the party last weekend I’ve identified a horrible bug on the KDC MixMaster – Bartending System so expect an update for this soon as well as a slight change in how recipes are handled. (if I can wrap my head around it)

It feels like all I’ve been doing is modeling ruffles at this point.

Lots and lots of them. It’s been going okay despite the fact my left hand is giving me some trouble, I’m having some sleep troubles too as I regularly catch myself nodding off while doing these (It is such a slow and boring step!).

It feels like this is a decent chunk of work for the past 3 days. I’m trying to keep the geometry of the apron somewhat close to the skirt underneath to avoid obvious clipping issues later (which will probably happen anyway, no matter what I do)

The apron surface need a few tweaks to connect to the trim properly, and I need to clean up where the ruffles meet the belt, it is a bit messy in there. Like with the underside of the skirt, a back-face might not be necessary here either if I use two-sided rendering.

Once the tea apron is all good I think that I’ll jump to the collar & tie before worrying too much about the apron bib option. In fact, I might put it aside for a time and see if I can get a preliminary weight painting/UV map done, just to see if I can expose a few obvious problems this model is going to “feature”.

I also want to take Robust Weight Transfer for a spin.

The latex maid corset body is complete, I ended up redoing most of it to simplify the geometry (same amount of steps all around it) and anchor it better to the torso and skirt. I filled in the gap in the front of the body, it is a bit… weird? But, works for me.

The back I still open, I’m having trouble thinking up a good edge flow here to connect everything together.

Lots of health related appointments this week and a bad sleep schedule, (oh, and even hail!) that’s my excuse this time, other than… the WRITING(?!)…

I spent… apparently 14 hours over the last 3 days trying to put down ideas and assemble them into some kind of coherent story. I’m not a writer, so of course it is god awful and a waste of time.

(Still more efficient than trying to turn it into a video game…)

I don’t know what to do about it, it’s so far outside my skill set to build a compelling story, other than the torture porn scenes… I guess. Maybe I’ll put the introduction on Patreon, or something.

I was a bit sick again, but I worked on the skirt pleats on and off through the rest of the week, and during the weekend. The junction between the skirt and the trim is about as “clean” as I can make it, they don’t have the same number of edges, and I do not want to overdo it. Unless I really need to increase the density on the skirt… it will do.

It brings me to another little thing: With PBR we now have the ability to decide whether polygons should be rendered single-sided or double-sided, this means that I might be able to get away with not explicitly modeling the underside of the skirt, and just using a double-sided material for that part. Double-sided materials do take a little longer to render, but they are still faster than an actual underside geometry.

The last picture illustrates what I could potentially do: Model only the underside of the trim and join with the (now visible) backside of the skirt.

There is still a lot left to do, I’ll get back to the corset and finish topside tomorrow.

My internet connection decided to die for most of the evening and until now. I got some decent progress done over the past two days, but I have to say that all those pleats are making me lose my mind little by little.

I’ll definitely think again before I do something like this in the future.

It is, somewhat clean, I’m not sure how I’m going to fuse some of those sections together, but I’m sure that I’ll figure it out.

I think this is a good point to wrap up the detailing of the model. I added some discrete creases on the chest and neck, fixed up the dress & corset meeting point at the front busk, tweaked the apron bib here and there. I could spend a hundred more hours on this, but, at least for now, this looks good to me.

I started the low-poly, but there is not much to write home about, next update I’ll have more.