Archive for 2016
So my grand plan for texture recycling ended up not really working that well, so I will do some regular diffuse baking instead.
Now as you can see that’s some seriously compact UVmapping :P, save for that empty bit in the top. Don’t you worry there will be stuff in there eventually.
I haven’t unwrapped those back straps yet because I really want to use the same texture for the tied/untied versions (and even the same texture for all four straps) and that will require some planning.
But the rest looks okay, save for the glitches which will obviously be fixed.
Now I really have to finish those sleeves…
Alright, most of the gaps are now closed, I even capped the openings. I also finished the “buckled” version of the straps. This is hovering around 6000+ triangles. It’s not making me happy but I needed the strap geometry to be “texture recycling friendly”. You can’t have it both ways I suppose.
The sleeves are still on symmetry, and I need to make the low poly version of the buckles (this is going to suck).
Some more cleanup might be required before I can call this good.
It’s getting there, I’m making a bunch of mistakes that I thought I wasn’t gonna be making anymore… like NOT COMPLETING THE MODEL before going into detailing and optimization.
Oh well, what’s done is done.
Left upper back is just a clusterfuck that I’ll probably end up trashing.
Blocked out the arm so I could work on the hands and then I’ll fumble something together something for the elbows. The shoulders should work okay with this geometry I hope.
Since texturing is a bit pointless if I can’t test the baking quality, I’ve started the low polygon model. It’s really rough in places and not all that great in general. I initially started in mirror mode, but because all the creasing was done out if symmetry mode, there is not much point in doing that.
Initially I tried to follow the main peaks and valleys of the creases, but I ended up switching to “peaks only” to keep the polygon density low. You can see in picture 2 and 3 the “improvements” in the topology (yeaaah… not there yet)
Oh yeah I also added the mandatory collar trim.
I’ve done a rough unwrapping of all the pieces and straightened what needed to be. I’ve also reused some of the maps I made for the KDC Institutional straitjacket.
Not entirely sure how I want to proceed, unless I want to blow up the texture budget, there is a pretty big chance that the weave of the fabric will be invisible.
Alternatively, sl does allow to use different repeat values for the diffuse and normal maps so I could use very small but very efficient repeating textures for the diffuse and do most of the detailing in the normal map.
I might also try to come up with different fabric weaves that don’t have to be so fine.
Enough! Well almost. I might tweak the sleeves a little more later. I’ve detailed the chest and curved straps inward a little too (because tension!), I think it looks “just right”.
Thinking about it, I cannot remember a recent time when I did crease modeling. The RevosuitX was creased mainly through bump mapping, and I suspect it might go back as far as the Delilah corset.
The front mid-section creasing will be minimal because the fabric should be fairly tight there anyway and because I need to adapt this for flat chests too.
I should kill myself for such a terrible pun.
1st attempt at creasing the lower back, the right side looks better in my opinion. Front seems like it should be. It will be a little trickier to get the pulling force in the right direction around the collar.
Because I sure feel cured from any strap making urge, at least for a little while.
On straitjacket anyway, the back is usually fairly interesting detail-wise.
If I’ve done my job well, the right side is clearly going over the left side. A few wild creases appear and I kinda like how they turned out.
I’ve also pulled the back further up, it’s starting to look a bit off-balance but some generous creasing should help explaining why it is riding so high up.
I was gonna start on the buckling straps but I need to make a reference for the friction buckles first.