There's a lack of love for Tube Tops with the only tube-top shirt being Synonymous's RE3 Jill Valentine outfit top, Unfortunately that's not hue-shiftable.
and if you want it to be a quality mod, breast slider support would also be nice.
~just decompiling synon's mod and putting on the hue shift would be easy, doing the breast slider would actually require drawing a few extra frames
Alternatively, you can designate anchor points (one on the back, 5-6 along the perimeter of the breast) and then use the physics system to generate an appropriate neckline.
The physics system doesn't care about layer boundaries, so it will stretch the entire neckline ("back" section, "chest" section, and "breast" section) into a single continuous curve. This curve will accommodate the breast slider, temporary breast transformations (e.g. heavy breathing and jiggle physics), and breast reshaping (e.g. via the superbreast mod). The curvature adaptation is demonstrated in the animation above (which also includes a physics-based waistline).
It requires a bit more work to assign an actual fill to the underside of the neckline (and the topside of the waistline), but it's doable. RGB support is straightforward.
Problem: I don't actually like the Jill Valentine top. There are too few visual details and I don't get a sense of physicality by looking at it. I don't want to want to work on it.
Proposal: I'd like to hijack this thread and ask for tubetop reference images. Once I see an interesting design, I'll apply the WIP physics stuff and create a usable SDT mod.
Guidelines
reference images must be clean side views with a reasonable resolution
reference images should include a reasonable amount of visual detail (e.g. shading, highlights, stretch lines in the fabric). I'm not an artist, so it's difficult for me to improvise such details.
If you submit a flat-colour sketch then I'll probably ignore it.
If your image has too much visual detail then I'll probably just omit half of it while tracing.
I can work with a semi-rigid garment (e.g. bustier) or a soft-but-elastic garment (e.g. breast band) or a soft-and-loose garment (e.g. cropped blouse).
However, I'd like to adhere to Gerve's original intent, so let's focus on realistic and/or reusable clothing designs. Something which could conceivably be mix-and-matched into many different outfits.
Fully-rigid garments (e.g. armor plates) wouldn't fit the Request. Similarly, anything with sleeves or shoulder straps would be inappropriate (unless these elements can obviously be omitted without ruining the design).
The reference image should show medium-sized or large-sized breasts, because it's easier to scale a design down than to scale it up.
Also: the physics system doesn't work very well with flat chests, so they're a low priority for me.
Coverage is negotiable. I'd prefer a somewhat-modest design which covers part of the midriff, but I could be persuaded to create something skimpier (e.g. underboob bandeau) -- if you can find a cool design with a good reference image.
I'd like to avoidhorizontal stripes. It requires a LOT of additional effort to make them look correct when they cross the inframammary fold.
The source material is unimportant. I'll accept original designs, concept artwork, video game characters, anime framegrabs, etc. But they must be colourful cel-shaded images. I'm not going to accept photorealistic stuff because I lack the artistic skill needed to adapt it to the SDT art style.
This one could work, although I don't know what it's supposed to be. The construction (lacing + rivets) suggests a stiff corset or even a breastplate. Yet the front of the garment appears to be squashed inwards by light pressure from the thigh. And the visible muscle definition suggests that it's made of lycra (or - more probably - bodypaint).
Out of scope. I don't want to include a skirt, because then my OCD kicks in and I need to physics-animate the whole skirt (so that it swishes back and forth when the girl moves, and automatically wraps around her legs and/or puddles onto the floor when she's in a kneeling position).
Also - most of these designs show off too much breast. As a "physics assignment" it's more interesting to work on a design which covers most of the upper breast while exposing a moderate amount of cleavage.
Out of scope for OCD reasons. I already have a leotard project on-the-go. If I can get the layers of the [buttocks, hips, thighs, pelvis, abdomen] to work together nicely, then I'll immediately apply that technique to the Angela Balzac project. Afterwards, we can talk about one-piece swimsuit designs.
Strapless Leotard (There's no playboy bunny leotard with breast scaling or hue-shift either)
Same problem as above - I'm not willing to tackle this costume design until I've sorted out the layers - so that fabric can wrap over and under the hips properly.
The garment seems to be made of shiny translucent material (e.g. latex?). SDT can't produce a convincing depiction of such materials - we'd need a lighting engine which could alter the gloss and reflection patterns in realtime.
The image doesn't show the full garment, and the squashing of the breasts adds complexity (because the modder would need to guess at the "unsquashed" shape, instead of tracing the image as-is).
Any of these would work. I'd prefer the white corset with the ruffled trim. The trim itself is fairly wide, which will conceal any minor inconsistencies between the static and simulated sections of the garment. The accessories and posture of the girl make it a very good reference image for SDT.
the artist was slightly lazy. He drew a gap along the spine, but the trim is unbroken at top and bottom. The lacing is partially visible, but the grommets are too far apart and the knotting is omitted.
I'm going to emulate his lazy approach. As we switch from the original oblique image to a full profile, the gap will vanish and the lacing will become almost wholly invisible. The trim will appear as two uninterrupted bands.
the image shows some "creased" or "crinkled" details at the edges of the ribcage.
I'll probably omit these (simplifying the shading pattern into 3 radial gradients) because the creases would look silly as static elements, and it isn't worth the trouble of simulating and animating them.
the garment would be setup as a "Bra" import - because you can't really wear a bra under a corset
the garment would be form-fitting but would not actually constrict the ribcage or abdomen (no tightlacing). It will fit closely around the breasts but will not cause a muffin-top (aka push-up bra) effect.
the garment will constrain breast movement slightly -- this is a default effect applied by SDT whenever a bra or shirt is worn (even a loose T-shirt ... the scripting isn't very clever). After testing this version, we might decide to release a variant which heavily suppresses breast motion (for a more "corset"-y effect)
note: this additional scripting would influence only the breast motion. Heaving of the diaphragm and abdomen would be unaffected. It's technically possible to limit breathing, but this Request began with normal clothing so let's not push it into bondage/fetish territory.
RGB slider support is do-able, but I'll initially aim for bright colors only. I may add support for dark color schemes later on.
Progress is kinda slow at the moment. I made a few adjustments to the physics stuff, but the modeling of the neckline (and waistline) can't proceed until the tracing is finished. And that's going to take a while because the ruffled trim is fairly intricate.
If anyone wants to help out with tracing then let me know.
Speculation and Discussion
the neckline shown in the reference image is slightly S-curved.
this may be merely a perspective effect (due to the low "camera angle"), rather than a deliberate statement about the shape of the garment.
I like the effect, because it hints at the 3-dimensional prominence of the breast. It suggests that the garment is partially elastic (i.e. a hybrid between a rigid corset and a form-fitting tubetop), and helps to explain why it conforms to the inframammary fold.
the hybrid appearance may be important/useful if the mod permits a lot of breast bounce -- such bouncing would simply look "wrong" if the viewer believes that the garment is a traditional corset.
meh; I'll do some testing when the art assets are ready. It may be possible to reproduce the S-shaped neckline while maintaining a smooth curve.
the occulted sections of ruffled trim (behind the arm, behind the rose) will be filled in via a lazy "clone stamp" approach
I plan to present the bones (vertical struts) of the corset as embossed details
any bone which wraps around the breasts will need to be modeled as a physics object
the mid and rear bones may also be physics-based ... or I might draw just draw them onto the static elements (to improve runtime performance)
I plan to increase the number of bones
the reference image shows two, but the character therein is much more slender than the SDT girl.
we'd naturally tend to add more bones as we widen the garment to fit a larger frame.
I'd want to add bones anyways - even if the reference image fit the SDT girl perfectly - because I feel that the number two is insufficient to convey the idea of "corset" to the viewer.
we must rely on the bones to give an immediate visual impression of "corset" because the lacing details will be mostly obscured, and the frontside of the garment is form-fitting rather than smooth
RGB1 slider will control the main body of the garment (including the embossed bones); RGB2 slider will control the trim
when the mod is loaded, both sliders will be default to white (to resemble the original reference image)
the RGB fills and shading will be fine-tuned to match the original reference image, and should work reasonably well with bright colors. You'll be able to switch to a dark fill (e.g. crimson, black) but it won't be as convincing.
If you're hoping to achieve a dark-and-shiny effect (e.g. satin, vinyl) then you'll need to edit the FLA.
Thank you for the offer, but I know that you have your own projects to deal with.
I'm hoping to convince (or "guilt trip") some of the newbies or lurkers to contribute. Tracing paths isn't especially difficult, and in most cases it's an unnecessary timesink for modders. The SDTMods community could be much more productive (in terms of "mods released per month" or "amount of time spent waiting before a Request gets fulfilled") if the junior people would handle some or all of the vector work. Experienced modders could then focus on the complex tasks (such as fitting, rigging, shading, scripting, etc) which require special skills and/or software tools.
I've put some more thought into the "hybrid" nature of the garment, and I think that I'll handle it via the breast slider
for small breasts, the garment will seem more like a true corset (or breastplate) - its frontside will be fully convex. It will not be form-fitting; a small space at the top will be deliberately left empty.
the entire garment will stretch slightly to accommodate the breathing cycle, but the waistline will stretch much more than the neckline
as the breast size increases, only the lower half of the frontside remains rigid and convex. The upper section wraps around the breasts. In a sense, the design has shited to [underbust corset + shirt] combo, akin to a dirndl. But the artwork (and animation/physics) will attempt to present it as a single continuous surface rather than a composite.
when the breast size exceeds 50%, we'll see that the bones visibly "kink" when they intersect the inframammary fold. The bust section of the garment is bowed outwards towards the viewer.
at the maximum breast size, we observe a significant S-shaped deviation in the neckline. This effect actually occurs throughout the slider range, but it's less noticeable with smaller breasts.
the garment will include an inframammary crease (as can be seen in the reference image). I omitted it from the mockup due to laziness.
the mockup image below shows the garment as semi-transparent. This is a visual aid - to make the bones more obvious and to clearly display the gaps (where the garment does not fit tightly to the skin). It's intended that you'll be able to change the color via RGB sliders. If you actually reduce the Alpha slider from 100% then the garment will look terrible -- there are too many overlapping elements; Alpha support is not feasible.
the garment will rely on non-traditional methods for drawing fills and shading. These techniques will be fine-tuned to deliver best results for medium breasts with limited bounce (i.e. with the "corset" behavior activated).
if used with very large breasts, or with boosted breast-bounce (e.g. via modification of your Settings file), some of the highlights and fills may bounce too far -- they'll deviate from their proper position and may even pass beyond the the outline of the garment. Caveat emptor.
the RGB fills and shading will be fine-tuned to match the original reference image, and should work reasonably well with bright colors. You'll be able to switch to a dark fill (e.g. crimson, black) but it won't be as convincing.
To elaborate on the previous mention of RGB shifting - here are the current vectors. They're incomplete because a few sections were obscured in the reference image; I intend to fill in these gaps by tracing frilly details from another (less detailed) reference image. Possibly this one. The main point is that the primary fill (RGB adjustable) is overlain by highlight and shadow fills, both of which are partially transparent. This approach looks reasonable so long as your custom color is similar to the original color (0xF0FAFA). It works fairly well with pale yellows and pinks.
It works less well with colors which are dark or saturated.
The same coloring and shading approach will be used for the bodice, breast cups, and the bones. Even though you'll have separate sliders for the main corset (vs the trim), the you'll face similar restrictions. If you try to make the corset black, then you'll find that the shadow details get washed out (because Flash can't draw "darker than black") and so the costume will probably look bad.
You're free to test the limits, of course. Pure black won't work, but if you're looking for a classy lingerie effect then you could experiment with dark greys.
Normally, it would be possible to revise the highlight/shading stuff by modifying the FLA project file after the mod is released. Hence, someone could produce a variant of the corset which is fine-tuned to support dark colors. Such revision work would be difficult in this case, because most of the fills will be defined within physics objects rather than static clothing elements.
I'll share the project file, of course; I'm just warning you not to get your hopes up.
here are the current vectors. They're incomplete because a few sections were obscured in the reference image; I intend to fill in these gaps by tracing frilly details from another (less detailed) reference image. Possibly this one.
The frill vectors are done as well as the basic shading for the bodice.
I relied heavily on this image while filling the gaps. The "frilled pillow" image tag ended up providing a lot of good references, because it tends to include upward- and downward-facing frills (with different overlap and shading patterns) as well as corners.
Todo list:
fabric creases on the bodice
lacing along the posterior edge
during this step, we'll push the posterior edge forward. You might be able to see a bit of skin under the laces - I'm not sure yet.
division of the frilled trim into segments
probably 6 for the top and 3 for the bottom
shape optimization
physics assignment for the bones
physics assignment for the bodice fills and shading
testing and fine-tuning of the bones w/r/t breast slider keyframes
ActionScript coding for the corset behaviour (temporary increase in breastFirmness)
Edit: this project is on hold, awaiting feedback from Gerve.
I've been interested in this project, sadly I haven't learned draw or mod for SDT.
I was wondering what people could do in the absence of answer from Gerve. Maybe test the mod if you could provide it ?
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.