Object: Skin Sample (12)

[tgialdini]

:D

yummy...

nice work.... but why do I wanna go play hacky-sack?

[victor throe]

am i the only one whos gonads hurt when they see that pic?


really nice skin though


stick some hairs on it

[brittnell]

Thanks Victor...

Yeah, I'm not sure of any way to put a skin sample on a ball object without illiciting that feeling though.

Funny thing is that I didn't have any problem looking at it for hours while working on it, but my wife refuses to look at it.

;)

As for the hairs, they're in there, they're just really small (and the DOF kills a lot of 'em). Here's a close-up:

http://cinematico.com/images/Sample12cu.jpg

Maybe Sample 13 will be a little hairier...

Also, here's a wire, in case anyone is interested (I know it's not as interesting as a Ferrari to look at) :rolleyes: .

http://cinematico.com/images/Sample12wire.gif

Cheers,

[/b]

[bix]

One crit - The bump map looks a little too reptilian, too regular of a pattern and to much contrast between the cells. Maybe it's just zoomed in so far on the texture. Looks really good other than that, love the pushpin tag.

[brittnell]

Hey bix,

Yep...

That was actually one of the things I was figuring on changing for Sample 13.

The textures are a mix of hand (Wacom) painted, mixed with some procedurals.

The "reptilian" bit was done by hand, based off of the pore pattern on the edge of my hand ("blade" edge, near the pinky).
I just got a bit lazy, and made the tiling of it a bit too small, so the repetition is pretty apparent. I should also mask it a bit as well, so it isn't so consistent over the surface.

As for scale, yeah, it is supposed to be zoomed in quite a bit. Hence the scale of the tag/pin, and the amount of DOF. I was figuring it's about the size of an earlobe. That said, I think the scale confusion is due to the vein scale/density to pore scale/density ratio... I think it would help clarify if the veins were a bit bigger.

Speakin of... the veins are hand painted as well, and work both in the diffuse, bump, displacement, and 3S channels (so they are denser than the surrounding tissue when lit).

I've actually got an animated lighting test going today, and should have it up tomorrow.

Now I'm going to play around a bit with GhostPainter 2.0 (which just showed up at the door while writing this!)

:D

Thanks again for the feedback!

[/b]

[CuTnPaste]

wow

try to applicate it to a arm or a face or leg....

:)

good!!

:bounce:

cutnpaste

[clusterfuq]

wow, incredible - you have to put it on something. hard to tell scale at the mo. the hairs could be more visible.

could you write a tutorial on this please? are the hairs done using a plug-in?

really amazing work. well done :applause:

[brittnell]

Thanks again guys...

Cutnpaste, yes, I think that's a good idea. My wife just got some pics from a friend of her broken wrist with pins set in it (right through the skin)... should be interesting in 3D.

:eek:

clusterfuq, hmmm... tutorial... I'll definitely keep it in mind. Shouldn't be too difficult to explain, just a matter of finding the time to do it (you kow how it goes).

The short story is mix and mask materials... quite a few of 'em.

As for the hairs, no plugin. I created one very narrow pyramid primitive with 4 height sections. Then I applied both bend and twist modifiers to it. I then took this object and used scatter to, uh, scatter it over the flesh surface (adjusting scale/local rotation/etc. 'til it was good). Then I collapsed it and removed the non-visible faces. The material for the hair is a semi-transparent anistropic, with a very high specular. This helps the little hairs catch the light a bit (visible in the full-size image, where the hairs sit on the lit/shadowed boundry).

Cool?

Thanks again.

[/b]

[g0tenks0]

WOW:bounce:

[Renzsu]

as mentioned the bumpmap seems off, and by looking at my own hands I think it'll help if you make the bits where those lines meet deeper than the actual lines (so that the lines go from deep, to shallow, to deep again). The deep parts are where the hairs grow..

[brittnell]

Yep... That's exactly what I had in mind.
The old pores are on the right, new pores on the left...
http://cinematico.com/images/Sample12pores.gif
(image scaled)
Thanks!
[/b]

edit:
Any ideas on how to implement the one-hair-per-pore concept?
I was thinking about using a variation of the pore map (pores only, no wrinkle) as a displacement (on a clone of the main "ball"), volume selecting only the points of the now-displaced pores, deleting the rest, and using that as a base for my hair scatter.

I'm sure you can use maps to drive hair density in something like Shag:Hair, but I only have a demo of that, and it doesn't seem to work well with GI.

[Renzsu]

yeah, looks much more convincing, though I'm not sure whether those lines should be so wrinkly.. could be it depends from person to person though.

[brittnell]

Here's that lighting loop I mentioned.

Sample12.avi (452kb/divx) (http://cinematico.com/images/Sample12.avi)

There are still the things missing mentioned earlier, but this is until I can collect sample 13.

Thanks again,

[/b]

[clam_mechanic]

do you use actual camera samples for your dof? or through render effects?
Sam

[SimonReeves]

heyyy nice job that looks really cool,

the only thing that i'd crit is that the bump looks too even etc.
but you've clearly looked at that with the new bump

:thumbsup:

[brittnell]

Originally posted by clam_mechanic
do you use actual camera samples for your dof? or through render effects?
Sam Actually, I'm rendering out .rla files, and doing the dof in post... close enough for government work. ;)

Thanks again all for the feedback.

I've got a bunch of stuff to finish up before I can go out hunting for Sample13, but I promise that the new "chunk" will be even "livelier" (and possibly a bit more dynamic as well).

Cheers,

[/b]

[Renzsu]

question, how do I deal with those RLA files? As Photohop can't read them?

[brittnell]

After Effects is what I've always used for this, although I think most motion graphic programs can open them (eg. Combustion).

It's the only way I know of getting at the depth buffer info (which you can then use as a mask to control blurs/etc.).

Cheers,

[/b]

[Renzsu]

Ah, thanks for clearing that up.. seems weird though don't you think that Pshop can't read RLA files..

[victor throe]

rla question

when rendering out from max5

the channels in the rla are not antialiased

why is this?

[brittnell]

Originally posted by victor throe
<snip>...when rendering out from max5 the channels in the rla are not antialiased
why is this? You know, I always thought that was weird as well...

I think it has something to do with the way the information is genererated in the first place, somethng like a series of clipping planes through your camera's Z space.

Honestly, it is still a bit of a mystery to me...

:shrug:

I think it uses 256 levels of gray, and then uses one level per "clipping plane"... Although, I could be talking out my ass here.

I have one little trick, that I'm not sure even works, but I "feel" like it is doing something (kinda like carrying around a lucky charm); What I do is make sure the environment ranges for my camera are set to just the boundries of what I want my dof to be, before rendering. I think this places the majority of the gray-scale depth information into this area, rather than stretching it over a huge world-space (giving you more lattitude to work with, and minimizing the aliasing issue).

Speaking of aliasing, one thing I usually do is ad a bit of gausian blur and levels adjust to the mask (.rla depth), to get it a bit smoother and tighter across the depth levels.

Definitely feels more like art than science...

Best of luck!

And about P-Shop not opening .rla's... Yeah, it does seem weird. I wonder why it doesn't? Maybe someone has written an import module for it?

Oh, and one more thing... You can also use .rpf files to do the same sort of thing, but you need the latest AE production bundle to do so.

Cheers,

[/b]

[dans]

You can use the 3ds max Render Element Z-Depth (Render Scene Diaolog > Render Elemnts > Add.. ) to get z-depth for compositing. (The values for min and max depths you can figure out using the camera clipping planes)

And if you have GhostPainter you can use psd-manager to get your zdepth render element as a PSD layer into Photoshop :)

Daniel

[brittnell]

Originally posted by dans
<snip>...And if you have GhostPainter you can use psd-manager to get your zdepth render element as a PSD layer into Photoshop :)Oooh... nice tip! Thanks for that one (as noted earlier, I do have GhostPainter, so yay!).

For quick and easy dof (in After Effects), definitely check out Frischluft Lens Care (http://www.frischluft.com). Easy as pie, and well worth the money!

Cheers,

[/b]

[brittnell]

http://www.cinematico.com/images/cineSAMPLE12c.jpg
Just a little update before moving on to 13...
Cheers,
[/b]

[Finkster]

mmmmm.....pube-licious

[brittnell]

Yeah, isn't it funny the things you find when you finally get around to cleaning behind the toilet?
"Oops, wonder who's missing this then... Guess I ought to put it up on the lost-and-found board".
:surprised

[Finkster]

Yep, as Bill Cosby would surely say "you can find inspiration in the darndest places". Possibly.

Looking forward the the meat bag that will be skin sample 13.
Maybe a little more translucency would help - emphasize the earlobe scale you mentioned.

Good stuff.

[kr3ml]

lol... cool...:D

[brittnell]

Thanks!

Just FYI, I just figured out an easy way to break up the repetition of the pore texture a bit (without creating new bitmaps).

I took the original texture, and used it in channel 1 of a mix map.
I then copied this to channel 2 of the mix map as well, and rotated this version (w) an arbitrary amount.
Then, for the mix amount control, I used a simple fractal noise material and adjusted the size/balance to suit.

This worked fairly well, but then I...

Took this mix map, and made it channel 1 of a new mix map (you following?).
I copied this sub-mix-mix map to channel 2 of the "parent" mix map, then went into each of the sub-maps in channel 2 and rotated them a new arbitrary amount.
I then used a new fractal noise to mix between the two sub-mix-mix maps.

Whew!

The end result is much more naturalistic (and chaotic), while still retaining the feel of the original.

I'm not going to post another full-out render, but hopefully I'll be able to illustrate the technique in the next sample.

Cheers, and thanks for the feedback.

[/b]

[Doubt]

Very good results man...

Your changes to the bump bitmap and as well as the "double mix" for randomness certainly help a ton. I'd love to see a tutorial of the whole process as well.

Keep up the good work--

[UnlikelyCorny]

Originally posted by brittnell
(I know it's not as interesting as a Ferrari to look at) :rolleyes: .


[/B]

I get so bored with these matchbox models of toys for boys. This is definately a lot more interesting, please elaborate as much as possible. The animation is beautifull...

[Jango]

;) Your latest render looks a bit like the skin at the balls :wip: (kidding)

I like it. As you mentioned, you should bring in more variation, even to such a "small piece" like that. When I look at my skin I see it looks different after every few centimetres.

But all in all, very nice skinball :-)

[CGTalk Moderation]

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.