Re: Reddit Thread with Carl Bass, CEO of Autodesk, on 3D printers...

[Alan Fregtman]

lol, he avoided the Softimage questions like the plague.


On Fri Dec 05 2014 at 06:15:21 Simon Reeves  wrote:

> interesting  !
>
>
>
> Simon Reeves
> London, UK
> *si...@simonreeves.com *
> *www.simonreeves.com *
> *www.analogstudio.co.uk *
>
> On 5 December 2014 at 00:56, Eric Turman  wrote:
>
>>
>> http://www.reddit.com/r/IAmA/comments/2oaei2/carl_bass_here_ceo_of_autodesk_designer/
>>
>> Looks to be a currently active thread.
>>
>> Of course there are comments about Softimage, Max, and Maya; some
>> humorous ones if you skim.
>>
>>
>> --
>>
>>
>>
>>
>> -=T=-
>>
>
>

realistic hair shading SI-MR

[Matt Lind]

@Peter

MR hair geometry is flat ribbon, but the normal is computed from a virtual 
cylinder like a normal map with full control over shaft radius and taper 
along the length.  For most practical purposes it should create the desired 
look.  Using the utility shaders (math/lookup nodes) available in the 
rendertree, you can read hair UV coords and surface normal yourself if you 
want to shade them differently than the provided hair shaders.  Mental ray 
stores hair information in the state which can be accessed by any shader 
which cares to dig into that information.


The reason for flat ribbon is to allow hair to be represented as physical 
geometry which can be styled and groomed, respond to dynamics, and use 
material/texture shaders for shading while keeping memory consumption 
low/reasonable (representing hair as geometry is hugely expensive).  Mental 
ray does offer true cylinder hair, but to get cylinder hair would require 
the hair shader to be implemented as a volume shader (which would have it's 
own set of issues).  I believe that was how the original softimage hair 
shader from 10 years ago was implemented.  It rendered hair convincingly for 
the basics such as highlights, but most people didn't like it because as a 
volume shader it did not permit styling and grooming (or at least not much 
control over it), and rendered quite slow with frequent crashing as mental 
ray volumes can be a bit finicky.


I experimented with hair rendering in mental ray for Barnyard all those 
years ago.  From a purely technical point of view, writing shaders for hair 
isn't too hard - it's just standard material/texture shader with additional 
metadata for the hair.  The key to getting results is being smart with your 
render settings.  You must be stringent on ray depth, recursion, shadow 
type, memory limits, and so on.  Setting them too generously will make your 
render times go through the roof as you're telling mental ray to wander and 
find things to do which probably aren't necessary.  the default render 
settings in Softimage are probably too generous for hair.



Matt

Re: Zbrush and vertex IDs

[Sebastien Sterling]

Thx Guys, will try all these. it's nice knowing it's not an isolated
problem.

On 6 December 2014 at 20:31, Nic Sievers  wrote:

> For the UV's make sure in Zbrush in the Export options in the Tool Panel
> you have 'Txr' turned on.  It won't export UV's with that turned off.  Also
> I normally have 'Mrg' on too or the UV's will be separated
>
> On Sat, Dec 6, 2014 at 10:35 AM, Orlando Esponda <
> orlando.espo...@gmail.com> wrote:
>
>> Are you sure is the vertex order that is changing? or just the samples
>> order?
>>
>> Andy's description and solution seems to be the best one. I would try to
>> export from softimage as obj and import again to softimage. The export this
>> new mesh and replace the base mesh in Zbrush.
>>
>> If this is too much, maybe you could try to fix the UVs manually
>> http://vimeo.com/67783125
>>
>>
>> On Sat, Dec 6, 2014 at 10:35 AM, Sebastien Sterling <
>> sebastien.sterl...@gmail.com> wrote:
>>
>>> Might it be a problem with the app link ?
>>>
>>> i notice it to happen with layers as well.
>>>
>>> On 6 December 2014 at 14:41, Andy Goehler 
>>> wrote:
>>>
 I’ve noticed this behavior post Softimage 2010 releases. Where with the
 OBJ exports the vertex order is not preserved. It caused us many headaches
 and countless numbers of redoing UVs. Like you I have nothing reproduce
 able, however your chances of the bug increases when you mirrored the mesh
 before. It’s odd — I’ve advised my team to export to .OBJ and reimport
 before rigging and UVing. It’s a Softimage bug, not ZBrush.

 Nowadays, I just import both meshes into Houdini and use the Match
 topology SOP. Bingo.

 Andy


 On Dec 06, 2014, at 11:28, Sebastien Sterling <
 sebastien.sterl...@gmail.com> wrote:

 Ive noticed now a couple of times when sending a mesh back into
 sofrimage, that

 the uv's get erased or that the vertex order, or ID is different, often
 this seems to

 happen regardless of what degree of manipulation i want to effect. I
 can't seem to

 find a reocuring reproducible source for this, with the possible
 exception of dividing

 the mesh, adding a sub div level, something you can do in mudbox
 without any issues.

 *Is there some guard, some box I can tick to tell zBrush to preserve
 vertex order ? or at the very least will prompt me if a perticular
 manipulation is going to break it ?*


 Many thanks and a good weekend to you !

 - Seb



>>>
>>
>

Re: realistic hair shading SI-MR

[Anto Matkovic]

Custom Rman shaders available around, they are only 'core' part. Even you can 
apply them as 3dl property, you can't get proper point-based GI in most cases, 
especially not path traced GI, MIS an so on.Custom color mapping, coming from 
SI, won't work too, of course.All of free Marschner Rman shaders are single 
scatter model ( single scatter = Muh Hair). Implementation of fancy dual 
scattering model, this seems to belong to additional coding in RMan and 
3delight, not a plain RSL - so, unless Pixar will have a mercy, don't believe 
you'll be able to get the dual thing in upcoming Renderman, too.  Maya or SI, 
doesn't matter.

Perhaps the shortest path for one man band and single machine, it is to just 
hack the RSL of 3dl emulation of SI hair shader, add code for shifted 
highlight, map the parameter to something unused in PPG ( 'flat light' for 
example), use two or three opaque shaders in render tree, pass the result 
through constant shader for transparency and GI, at the end. 3delight is quite 
good in optimizing the structure of this kind. But this is really a hack.

Regarding Mr, old Muh Hair definitively won't work, but there is a chance to 
get working version of p_HairTK, somewhere. Or, maybe, working implementation 
of built in 'basehair' in latest Mental Ray 3.12. Both are single scatter 
model, but,  still more than out of the box solution.

  From: Leendert A. Hartog 
 To: softimage@listproc.autodesk.com 
 Sent: Saturday, December 6, 2014 1:55 PM
 Subject: Re: realistic hair shading SI-MR
   
Any chance of achieving better results with 3delight (as it's so "budget 
friendly"), possibly with custom shaders?



Greetz
Leendert

-- 

Leendert A. Hartog AKA Hirazi Blue
Administrator NOT the owner of si-community.com



  

Re: Zbrush and vertex IDs

[Nic Sievers]

For the UV's make sure in Zbrush in the Export options in the Tool Panel
you have 'Txr' turned on.  It won't export UV's with that turned off.  Also
I normally have 'Mrg' on too or the UV's will be separated

On Sat, Dec 6, 2014 at 10:35 AM, Orlando Esponda 
wrote:

> Are you sure is the vertex order that is changing? or just the samples
> order?
>
> Andy's description and solution seems to be the best one. I would try to
> export from softimage as obj and import again to softimage. The export this
> new mesh and replace the base mesh in Zbrush.
>
> If this is too much, maybe you could try to fix the UVs manually
> http://vimeo.com/67783125
>
>
> On Sat, Dec 6, 2014 at 10:35 AM, Sebastien Sterling <
> sebastien.sterl...@gmail.com> wrote:
>
>> Might it be a problem with the app link ?
>>
>> i notice it to happen with layers as well.
>>
>> On 6 December 2014 at 14:41, Andy Goehler 
>> wrote:
>>
>>> I’ve noticed this behavior post Softimage 2010 releases. Where with the
>>> OBJ exports the vertex order is not preserved. It caused us many headaches
>>> and countless numbers of redoing UVs. Like you I have nothing reproduce
>>> able, however your chances of the bug increases when you mirrored the mesh
>>> before. It’s odd — I’ve advised my team to export to .OBJ and reimport
>>> before rigging and UVing. It’s a Softimage bug, not ZBrush.
>>>
>>> Nowadays, I just import both meshes into Houdini and use the Match
>>> topology SOP. Bingo.
>>>
>>> Andy
>>>
>>>
>>> On Dec 06, 2014, at 11:28, Sebastien Sterling <
>>> sebastien.sterl...@gmail.com> wrote:
>>>
>>> Ive noticed now a couple of times when sending a mesh back into
>>> sofrimage, that
>>>
>>> the uv's get erased or that the vertex order, or ID is different, often
>>> this seems to
>>>
>>> happen regardless of what degree of manipulation i want to effect. I
>>> can't seem to
>>>
>>> find a reocuring reproducible source for this, with the possible
>>> exception of dividing
>>>
>>> the mesh, adding a sub div level, something you can do in mudbox without
>>> any issues.
>>>
>>> *Is there some guard, some box I can tick to tell zBrush to preserve
>>> vertex order ? or at the very least will prompt me if a perticular
>>> manipulation is going to break it ?*
>>>
>>>
>>> Many thanks and a good weekend to you !
>>>
>>> - Seb
>>>
>>>
>>>
>>
>

RE: ICE: Speed, Velocity and circular motion

[gareth bell]

Of course!

Thanks chaps

From: pete...@skynet.be
To: softimage@listproc.autodesk.com
Subject: Re: ICE: Speed, Velocity and circular motion
Date: Sat, 6 Dec 2014 20:35:27 +0100







This is normal behaviour.
the pointtangent is not giving a circular force/vector (that doesn’t exist) 
– it’s a straight vector, that is tangent to the circle. So at each simulation 
step, the particle gets moved a bit, away from the circle in a tangent 
direction. At the next step it gets the new tangent vector, which pulls it 
largely back in line and makes it move “roughly” in the circle.
 
Depending on the speed and the amount of steps in the simulation, the 
particle will describe the circle very precisely (and thus pretty much staying 
on it) or less precisely (and thus each time getting off the circle and ending 
up describing a larger circle.)
With a slow speed it’s a non-issue. With a higher speed you can try more 
steps in the simulation.
 
But you’re probably better off adding a force pulling towards the center of 
the circle to compensate for the radius getting larger. (vector between self 
pointposition and the center of the circle, multiply it by scalar and add to 
the 
pointtangent and then set velocity)
now tweak the scalar in “multiply by scalar” to find the amount needed to 
remain on the circle exactly.


 

From: gareth bell 
Sent: Saturday, December 06, 2014 3:42 PM
To: softimage@listproc.autodesk.com 

Subject: ICE: Speed, Velocity and circular motion
 

Hi guys,

Having 
an issue getting my head around this. 

I think I want to use the formula 
for circular motion on this one. 

I've got some particles moving around 
in a circular motion with their velocity determined by a circular curve. Using 
circle>closest location>pointtangent>set velocity.

This works 
great however it's not moving the buggers fast enough. 

if I multiply it 
by a scalar it speeds up but the circle radius increases which is correct 
according to the maths, however, I don;t want this to happen. I want to keep 
closely to the circle radius but simple speed up the particles moving around 
it.

Any ideas?

G

  

Re: ICE: Speed, Velocity and circular motion

[peter_b]

This is normal behaviour.
the pointtangent is not giving a circular force/vector (that doesn’t exist) – 
it’s a straight vector, that is tangent to the circle. So at each simulation 
step, the particle gets moved a bit, away from the circle in a tangent 
direction. At the next step it gets the new tangent vector, which pulls it 
largely back in line and makes it move “roughly” in the circle.

Depending on the speed and the amount of steps in the simulation, the particle 
will describe the circle very precisely (and thus pretty much staying on it) or 
less precisely (and thus each time getting off the circle and ending up 
describing a larger circle.)
With a slow speed it’s a non-issue. With a higher speed you can try more steps 
in the simulation.

But you’re probably better off adding a force pulling towards the center of the 
circle to compensate for the radius getting larger. (vector between self 
pointposition and the center of the circle, multiply it by scalar and add to 
the pointtangent and then set velocity)
now tweak the scalar in “multiply by scalar” to find the amount needed to 
remain on the circle exactly.

From: gareth bell 
Sent: Saturday, December 06, 2014 3:42 PM
To: softimage@listproc.autodesk.com 
Subject: ICE: Speed, Velocity and circular motion

Hi guys,

Having an issue getting my head around this. 

I think I want to use the formula for circular motion on this one. 

I've got some particles moving around in a circular motion with their velocity 
determined by a circular curve. Using circle>closest location>pointtangent>set 
velocity.

This works great however it's not moving the buggers fast enough. 

if I multiply it by a scalar it speeds up but the circle radius increases which 
is correct according to the maths, however, I don;t want this to happen. I want 
to keep closely to the circle radius but simple speed up the particles moving 
around it.

Any ideas?

G

Re: ICE: Speed, Velocity and circular motion

[Orlando Esponda]

Angular Velocity. Take a look at this video
https://www.youtube.com/watch?v=zAx61CO5mDw

On Sat, Dec 6, 2014 at 8:42 AM, gareth bell  wrote:

> Hi guys,
>
> Having an issue getting my head around this.
>
> I think I want to use the formula for circular motion on this one.
>
> I've got some particles moving around in a circular motion with their
> velocity determined by a circular curve. Using circle>closest
> location>pointtangent>set velocity.
>
> This works great however it's not moving the buggers fast enough.
>
> if I multiply it by a scalar it speeds up but the circle radius increases
> which is correct according to the maths, however, I don;t want this to
> happen. I want to keep closely to the circle radius but simple speed up the
> particles moving around it.
>
> Any ideas?
>
> G
>
>
>

Re: Zbrush and vertex IDs

[Orlando Esponda]

Are you sure is the vertex order that is changing? or just the samples
order?

Andy's description and solution seems to be the best one. I would try to
export from softimage as obj and import again to softimage. The export this
new mesh and replace the base mesh in Zbrush.

If this is too much, maybe you could try to fix the UVs manually
http://vimeo.com/67783125


On Sat, Dec 6, 2014 at 10:35 AM, Sebastien Sterling <
sebastien.sterl...@gmail.com> wrote:

> Might it be a problem with the app link ?
>
> i notice it to happen with layers as well.
>
> On 6 December 2014 at 14:41, Andy Goehler 
> wrote:
>
>> I’ve noticed this behavior post Softimage 2010 releases. Where with the
>> OBJ exports the vertex order is not preserved. It caused us many headaches
>> and countless numbers of redoing UVs. Like you I have nothing reproduce
>> able, however your chances of the bug increases when you mirrored the mesh
>> before. It’s odd — I’ve advised my team to export to .OBJ and reimport
>> before rigging and UVing. It’s a Softimage bug, not ZBrush.
>>
>> Nowadays, I just import both meshes into Houdini and use the Match
>> topology SOP. Bingo.
>>
>> Andy
>>
>>
>> On Dec 06, 2014, at 11:28, Sebastien Sterling <
>> sebastien.sterl...@gmail.com> wrote:
>>
>> Ive noticed now a couple of times when sending a mesh back into
>> sofrimage, that
>>
>> the uv's get erased or that the vertex order, or ID is different, often
>> this seems to
>>
>> happen regardless of what degree of manipulation i want to effect. I
>> can't seem to
>>
>> find a reocuring reproducible source for this, with the possible
>> exception of dividing
>>
>> the mesh, adding a sub div level, something you can do in mudbox without
>> any issues.
>>
>> *Is there some guard, some box I can tick to tell zBrush to preserve
>> vertex order ? or at the very least will prompt me if a perticular
>> manipulation is going to break it ?*
>>
>>
>> Many thanks and a good weekend to you !
>>
>> - Seb
>>
>>
>>
>

Re: realistic hair shading SI-MR

[Angus Davidson]

The trick is you don¹t get the Data to Maya. You get it to a solid
dependable format that Maya (or whatever can read) In theory you can send
it to blender which is actually pretty decent for rendering and
compositing.


On 2014/12/06, 6:54 PM, "Leendert A. Hartog"  wrote:

>Oh, it's definitively worth looking into once it finally arrives.
>As a renderer I'll surely take it for a spin.
>Even if that means my return to Maya.
>But I somehow don't like having to get data from Softimage to Maya and
>vice versa.
>To most of you it must be second nature. To me it still has an touch of
>voodoo.
>
>Greetz
>Leendert
>
>-- 
>
>Leendert A. Hartog AKA Hirazi Blue
>Administrator NOT the owner of si-community.com
>

 

This communication is 
intended for the addressee only. It is confidential. If you have received this 
communication in error, please notify us immediately and destroy the original 
message. You may not copy or disseminate this communication without the 
permission of the University. Only authorised signatories are competent to 
enter into agreements on behalf of the University and recipients are thus 
advised that the content of this message may not be legally binding on the 
University and may contain the personal views and opinions of the author, which 
are not necessarily the views and opinions of The University of the 
Witwatersrand, Johannesburg. All agreements between the University and 
outsiders are subject to South African Law unless the University agrees in 
writing to the contrary. 

Re: realistic hair shading SI-MR

[Leendert A. Hartog]

Oh, it's definitively worth looking into once it finally arrives.
As a renderer I'll surely take it for a spin.
Even if that means my return to Maya.
But I somehow don't like having to get data from Softimage to Maya and 
vice versa.
To most of you it must be second nature. To me it still has an touch of 
voodoo.


Greetz
Leendert

--

Leendert A. Hartog AKA Hirazi Blue
Administrator NOT the owner of si-community.com

Re: realistic hair shading SI-MR

[Angus Davidson]

True, But I am assuming that you can use softimage, for pretty much
everything, export to alembic, apply render man mats and render in Maya.

Part of the reason for the Renderman NC 19 is so that they can polish the
docs and examples for first time users. So worth having a look at.

You can do the same pipeline and render in the free version of Houdini as
well. For testing of course.

Kind regards

Angus



On 2014/12/06, 6:38 PM, "Leendert A. Hartog"  wrote:

>Worth pointing out in turn: that's Maya-only IIRC. :D
>
>Greetz
>Leendert
>
>-- 
>
>Leendert A. Hartog AKA Hirazi Blue
>Administrator NOT the owner of si-community.com
>

 

This communication is 
intended for the addressee only. It is confidential. If you have received this 
communication in error, please notify us immediately and destroy the original 
message. You may not copy or disseminate this communication without the 
permission of the University. Only authorised signatories are competent to 
enter into agreements on behalf of the University and recipients are thus 
advised that the content of this message may not be legally binding on the 
University and may contain the personal views and opinions of the author, which 
are not necessarily the views and opinions of The University of the 
Witwatersrand, Johannesburg. All agreements between the University and 
outsiders are subject to South African Law unless the University agrees in 
writing to the contrary. 

Re: realistic hair shading SI-MR

[Leendert A. Hartog]

Worth pointing out in turn: that's Maya-only IIRC. :D

Greetz
Leendert

--

Leendert A. Hartog AKA Hirazi Blue
Administrator NOT the owner of si-community.com

Re: Zbrush and vertex IDs

[Sebastien Sterling]

Might it be a problem with the app link ?

i notice it to happen with layers as well.

On 6 December 2014 at 14:41, Andy Goehler 
wrote:

> I’ve noticed this behavior post Softimage 2010 releases. Where with the
> OBJ exports the vertex order is not preserved. It caused us many headaches
> and countless numbers of redoing UVs. Like you I have nothing reproduce
> able, however your chances of the bug increases when you mirrored the mesh
> before. It’s odd — I’ve advised my team to export to .OBJ and reimport
> before rigging and UVing. It’s a Softimage bug, not ZBrush.
>
> Nowadays, I just import both meshes into Houdini and use the Match
> topology SOP. Bingo.
>
> Andy
>
>
> On Dec 06, 2014, at 11:28, Sebastien Sterling <
> sebastien.sterl...@gmail.com> wrote:
>
> Ive noticed now a couple of times when sending a mesh back into sofrimage,
> that
>
> the uv's get erased or that the vertex order, or ID is different, often
> this seems to
>
> happen regardless of what degree of manipulation i want to effect. I can't
> seem to
>
> find a reocuring reproducible source for this, with the possible exception
> of dividing
>
> the mesh, adding a sub div level, something you can do in mudbox without
> any issues.
>
> *Is there some guard, some box I can tick to tell zBrush to preserve
> vertex order ? or at the very least will prompt me if a perticular
> manipulation is going to break it ?*
>
>
> Many thanks and a good weekend to you !
>
> - Seb
>
>
>

Re: realistic hair shading SI-MR

[Angus Davidson]

Worth pointing out that Renderman r19 will be free from early Jan for non
commercial work. I would suggest waiting a little bit for that .

Kind regards

Angus

On 2014/12/06, 5:33 PM, "Leendert A. Hartog"  wrote:

>Interesting. Will have a look. Thanks.
>
>Greetz
>Leendert
>
>-- 
>
>Leendert A. Hartog AKA Hirazi Blue
>Administrator NOT the owner of si-community.com
>

 

This communication is 
intended for the addressee only. It is confidential. If you have received this 
communication in error, please notify us immediately and destroy the original 
message. You may not copy or disseminate this communication without the 
permission of the University. Only authorised signatories are competent to 
enter into agreements on behalf of the University and recipients are thus 
advised that the content of this message may not be legally binding on the 
University and may contain the personal views and opinions of the author, which 
are not necessarily the views and opinions of The University of the 
Witwatersrand, Johannesburg. All agreements between the University and 
outsiders are subject to South African Law unless the University agrees in 
writing to the contrary. 

Re: realistic hair shading SI-MR

[Leendert A. Hartog]

Interesting. Will have a look. Thanks.

Greetz
Leendert

--

Leendert A. Hartog AKA Hirazi Blue
Administrator NOT the owner of si-community.com

Re: realistic hair shading SI-MR

[Andy Goehler]

Hey Leendert, look here:

http://www.3delight.com/en/modules/PunBB/viewtopic.php?id=2030 


Oldie but goodie. 


> On Dec 06, 2014, at 14:08, Leendert A. Hartog  wrote:
> 
> Thanks, I was just coming to the same conclusion... ;)
> In 3delight would the converted Softimage hair shaders suffice
> or would still have to look for some custom shader?
> 
> Greetz
> Leendert
> 
> -- 
> 
> Leendert A. Hartog AKA Hirazi Blue
> Administrator NOT the owner of si-community.com
> 

ICE: Speed, Velocity and circular motion

[gareth bell]

Hi guys,

Having an issue getting my head around this. 

I think I want to use the formula for circular motion on this one. 

I've got some particles moving around in a circular motion with their velocity 
determined by a circular curve. Using circle>closest location>pointtangent>set 
velocity.

This works great however it's not moving the buggers fast enough. 

if I multiply it by a scalar it speeds up but the circle radius increases which 
is correct according to the maths, however, I don;t want this to happen. I want 
to keep closely to the circle radius but simple speed up the particles moving 
around it.

Any ideas?

G

  

Re: Zbrush and vertex IDs

[Andy Goehler]

I’ve noticed this behavior post Softimage 2010 releases. Where with the OBJ 
exports the vertex order is not preserved. It caused us many headaches and 
countless numbers of redoing UVs. Like you I have nothing reproduce able, 
however your chances of the bug increases when you mirrored the mesh before. 
It’s odd — I’ve advised my team to export to .OBJ and reimport before rigging 
and UVing. It’s a Softimage bug, not ZBrush.

Nowadays, I just import both meshes into Houdini and use the Match topology 
SOP. Bingo.

Andy


> On Dec 06, 2014, at 11:28, Sebastien Sterling  
> wrote:
> 
> Ive noticed now a couple of times when sending a mesh back into sofrimage, 
> that 
> 
> the uv's get erased or that the vertex order, or ID is different, often this 
> seems to 
> 
> happen regardless of what degree of manipulation i want to effect. I can't 
> seem to 
> 
> find a reocuring reproducible source for this, with the possible exception of 
> dividing 
> 
> the mesh, adding a sub div level, something you can do in mudbox without any 
> issues.
> 
> Is there some guard, some box I can tick to tell zBrush to preserve vertex 
> order ? or at the very least will prompt me if a perticular manipulation is 
> going to break it ?
> 
> 
> Many thanks and a good weekend to you !
> 
> - Seb

Re: realistic hair shading SI-MR

[Leendert A. Hartog]

Thanks, I was just coming to the same conclusion... ;)
In 3delight would the converted Softimage hair shaders suffice
or would still have to look for some custom shader?

Greetz
Leendert

--

Leendert A. Hartog AKA Hirazi Blue
Administrator NOT the owner of si-community.com

Re: realistic hair shading SI-MR

[peter_b]

well, if it's a financial issue - get 3delight. You can use it for free with 
some number-of-cores limitation, that still makes it awesome at a lot of 
things were MR isnt. Hair in particular being one of those.



For me, a huge issue with hair in XSI/MR is the normals. The hair primitive 
is done as a flat ribbon, for decent light interaction it should be a 
cylinder. (which is what 3delight does - or is it a cone? anyway not a flat 
ribbon)
You can have cylinder like normals with a custom shader (if you write it 
yourself it's 'free', if you have someone write it for you this could end up 
more expensive that a renderlicense - so that's not very feasible).


You can make things a bit better (just a bit) with a gradient on the hair 
(black white black - crosswise) as a bump. This will give you somewhat 
better normals. But you will need very decent sampling for this to really 
matter - and decent sampling with hair in MR is going to be painful.
There's a couple of (out of the box) things that can improve hair rendering 
in MR - shadowmapped lights and  the rasterizer in particular jump to mind. 
But don’t expect miracles either.

...


-Original Message- 
From: Leendert A. Hartog

Sent: Saturday, December 06, 2014 1:16 PM
To: softimage@listproc.autodesk.com
Subject: Re: realistic hair shading SI-MR

Okay, this was intended as a thread with a serious question about hair
shading in MR,
but it's turned into a Redshift infomercial. ;)
If I had the funds at my disposal, I already would have bought it.
Ah, well, a bald human head definitely is an artistic statement too.

Greetz
Leendert

--

Leendert A. Hartog AKA Hirazi Blue
Administrator NOT the owner of si-community.com 

Re: realistic hair shading SI-MR

[Leendert A. Hartog]

Any chance of achieving better results with 3delight (as it's so "budget 
friendly"), possibly with custom shaders?


Greetz
Leendert

--

Leendert A. Hartog AKA Hirazi Blue
Administrator NOT the owner of si-community.com

Re: realistic hair shading SI-MR

[Rob Chapman]

yeah sorry Hirazi I clearly saw, its just the latest and greatest theories
of shading hair are including in the redshift hair shader and rendering
tens of millions in 20 - 30 seconds wheras mental ray for softimage...
 when was the last time a decent new technology researched shader come out
for it.. and for SI? and be able to render in any kind of reasonable
timeframe. depending on   which version of mr are you using? the latest?
there may be some new shader tech with the mila shaders am unaware off with
sss but in my humble opinion if it does not do this dual scattering stuff
(which is in RS) it is not realistic enough...

http://www.cemyuksel.com/research/dualscattering/



On 6 December 2014 at 12:16, Leendert A. Hartog  wrote:

> Okay, this was intended as a thread with a serious question about hair
> shading in MR,
> but it's turned into a Redshift infomercial. ;)
> If I had the funds at my disposal, I already would have bought it.
> Ah, well, a bald human head definitely is an artistic statement too.
>
>
> Greetz
> Leendert
>
> --
>
> Leendert A. Hartog AKA Hirazi Blue
> Administrator NOT the owner of si-community.com
>
>

Re: realistic hair shading SI-MR

[Leendert A. Hartog]

Okay, this was intended as a thread with a serious question about hair 
shading in MR,

but it's turned into a Redshift infomercial. ;)
If I had the funds at my disposal, I already would have bought it.
Ah, well, a bald human head definitely is an artistic statement too.

Greetz
Leendert

--

Leendert A. Hartog AKA Hirazi Blue
Administrator NOT the owner of si-community.com

Re: realistic hair shading SI-MR

[Mirko Jankovic]

Just have to add redshift is god send and life saver esspecialy for
freelabce work :)
On Dec 6, 2014 1:09 PM, "Mirko Jankovic"  wrote:

> Some time ago i spent couple months trying to render hair i. Mray. In the
> end all the time wastted and all the issues and speed wise when you count
> it is more expensive to do it with mray then to buy single redshift lic and
> solve or your rendering problems if you have any decend nvidia card
> On Dec 6, 2014 12:15 PM, "Leendert A. Hartog"  wrote:
>
>> Sadly, like I stated, that is not an option, financially (very poor
>> hobbyist).
>> But there must surely have been something people did before the advent of
>> usable renderers?
>>
>> Greetz
>> Leendert
>>
>> --
>>
>> Leendert A. Hartog AKA Hirazi Blue
>> Administrator NOT the owner of si-community.com
>>
>>

Re: realistic hair shading SI-MR

[Mirko Jankovic]

Some time ago i spent couple months trying to render hair i. Mray. In the
end all the time wastted and all the issues and speed wise when you count
it is more expensive to do it with mray then to buy single redshift lic and
solve or your rendering problems if you have any decend nvidia card
On Dec 6, 2014 12:15 PM, "Leendert A. Hartog"  wrote:

> Sadly, like I stated, that is not an option, financially (very poor
> hobbyist).
> But there must surely have been something people did before the advent of
> usable renderers?
>
> Greetz
> Leendert
>
> --
>
> Leendert A. Hartog AKA Hirazi Blue
> Administrator NOT the owner of si-community.com
>
>

Re: realistic hair shading SI-MR

[Leendert A. Hartog]

Sadly, like I stated, that is not an option, financially (very poor 
hobbyist).
But there must surely have been something people did before the advent 
of usable renderers?


Greetz
Leendert

--

Leendert A. Hartog AKA Hirazi Blue
Administrator NOT the owner of si-community.com

Re: realistic hair shading SI-MR

[Rob Chapman]

If you can expand your rendering choices I can highly recommend redshift
for quality of shader & speed
On 6 Dec 2014 10:25, "Leendert A. Hartog"  wrote:

> What would be the "best" shading solution for rendering (more or less)
> realistic hair in Softimage using Mental Ray (sadly: financial
> constraints)? Note: shading/rendering hair, not "creating" it. Are there
> still any custom solutions "floating around", like the ancient muhhair and
> such?
> Any other pointers for getting the best possible result would be more than
> welcome. Thanks in advance.
>
> (Inspired by a question on the German xsiforum.de BTW)
>
> Greetz
> Leendert
>
>
> --
>
> Leendert A. Hartog AKA Hirazi Blue
> Administrator NOT the owner of si-community.com
>
>

Zbrush and vertex IDs

[Sebastien Sterling]

Ive noticed now a couple of times when sending a mesh back into sofrimage,
that

the uv's get erased or that the vertex order, or ID is different, often
this seems to

happen regardless of what degree of manipulation i want to effect. I can't
seem to

find a reocuring reproducible source for this, with the possible exception
of dividing

the mesh, adding a sub div level, something you can do in mudbox without
any issues.

*Is there some guard, some box I can tick to tell zBrush to preserve vertex
order ? or at the very least will prompt me if a perticular manipulation is
going to break it ?*


Many thanks and a good weekend to you !

- Seb

realistic hair shading SI-MR

[Leendert A. Hartog]

What would be the "best" shading solution for rendering (more or less) 
realistic hair in Softimage using Mental Ray (sadly: financial 
constraints)? Note: shading/rendering hair, not "creating" it. Are there 
still any custom solutions "floating around", like the ancient muhhair 
and such?
Any other pointers for getting the best possible result would be more 
than welcome. Thanks in advance.


(Inspired by a question on the German xsiforum.de BTW)

Greetz
Leendert


--

Leendert A. Hartog AKA Hirazi Blue
Administrator NOT the owner of si-community.com