Archive for 2013

I’ve worked a little more on it today, I finished the strap harness, (the strap ends are still missing, I know) and made a filter.

For those interested it’s in my past broadcasts on twitch, I didn’t feel like announcing that I was streaming today, I just wanted it recorded.

The wasp, is this a good name? I like it, at least for now, it’s never too late to change it… calling it the psycho II is probably not going to happen, that was back when I thought gas masks where weird instead of cool.

This modeling session went very well, better than I hoped. I almost expected that I was going to start over several time over the course of the day, but what do you know, I might be getting better after all. I know someone is gonna ask for this to be lockable, I’m not sure how to make it happen (from a visual standpoint) but I will think about it.

It’s this time again, i will be streaming most of the day on my new project: A new gasmask 🙂

[livestream type=”twitch” channel=”kyrah3D”]

I’m sorry for the lack of news, i tried to relax the past week, but i think i still need a couple more days before i am fully “recharged”.

I believe the next project will be a new gas mask 🙂

One month in the making, finally completed 🙂

They feature my “classic” leather, in 8 colors and 8 stitching, so basically 64 possibilities, you can also change the hoof color to match the boot body (should you want to) and it has a  “smooth” and a “textured” version.

The heel can be either color coordinated or use a contrasting varnished wood finish. The laces and metal can bear any colors you may want to use (RGB selection). Finally, they are mod, so you can tweak around. The single mesh structure makes them also extremely easy to adjust to your avatar.

Check it out at the KDC main shop or on the Secondlife Marketplace.

I’m sorry for the long silence, I didn’t have anything new to show really.

Tuesday was all about getting my new displays calibrated (it takes a couple of hours per display). Today, as I was assembling the textures I realised that there was overlaps that I could never fix so I had to scrap a bunch of the work to rearrange the UV clusters so they won’t be this tightly packed.

Which means I’m rendering textures, again.

On the flip side I added a few wooden materials for the heel spine, but I still need to recover this wasted time.

This is going to be a long night -_-.

The SL model is pretty much final, I won’t touch it anymore. Texture wise however, I think the black wood is too “black” and things could use a bit of contrast, but maybe that’s just me.

The metal will be tintable, the usual way, same for the laces and the hoof, if i make a light version of the texture. There is still a couple of texture glitches here and there, but nothing major that I can see.

It’s using more triangles than what I was hoping to remain under, but there is only so much I can do with these laces, short of doing it all in texture (which is the proper way to do this by the way) but you guys would probably not like that.

What’s left:

• Getting all the parts UVWmapped.
• Making all the textures.
• Some scripting, but considering these do not lock or have any bondage features, coloring HUD and maybe a walking sound.

Another work in progress, I deleted a bunch more “unnecessary” edges on the boot itself and added the LPM horseshoes. I’m sure “someone” will complain that I’ve been cheap on the polygon count, by the look of it, I won’t be able to maintain the boot under 2000 triangles, and 3000 triangles is not far considering I’m still missing the zipper pull, lacing, hooks and eyelets. I have no doubt that it will remain under 10k triangles but beyond that?

The devil is in the details…

Fun fact: a sculpted prim is 1024 triangles.

Now that the high polygon version is ready, it’s time to work on the low polygon version, the one that will be specially optimized for real-time rendering and obviously, secondlife.

I never talked about these things, but when it comes to low polygon modeling, it’s all a balancing act. Triangles take pretty much the same time to render regardless of their size, so if you want your model to be “kind” to the user’s computer, you have to balance between quality and performances.

I could write an article about it, i have a folder full of extremely bad examples of  models that have no place in SL, yet are your typical “top seller” shoes.

As of today, SL doesn’t support normal mapping (yet), so we cannot use fancy tricks. Which means that regardless of your ability to optimize, you have to shoot for a little higher polygon count than you normally would because of the tendency SL residents have to move their camera around and zoom all the way in on everything, down to the smallest piece of jewelry.

But even with normal and specular mapping, I fear that those unreasonable expectations for high fidelity rendering will only get worse.