Looking Ahead to K-3D 0.9

[Timothy M. Shead]

Gang:

Things always take about twice as long as one would like, even when you
consistently double your first estimate.  We've spent a tremendous
amount of time on the geometric data structures for K-3D 0.7.  That's
not a bad thing - they needed the work, and we have some unique new
capabilities, but we've also been neglecting other important work.

Since the team can really only handle one large project at a time, I say
"let's embrace that way of working" ... I call everyone's attention to
the Roadmap:

http://www.k-3d.org/wiki/Roadmap

... in a nutshell, I have it in mind to give the UI the same level of
focus in 0.9 that the data structures have been getting in 0.7.  In
particular, there's the tremendous amount of work done by Joe to
carry-out and document UI studies.  I want to see all those suggestions
implemented.  That's one of the big reasons for ripping-out the old
tutorial code (already done), because it was a big barrier to UI work.

More than that, however, I want us to look at "big ideas" - new methods
of interaction (the NURBS sketching that we looked at awhile back comes
to mind), a completely new API for writing interactive tools, etc.  The
bigger, the better.  As a personal starting-point, I'm going to be
picking-up a 3D mouse and a tablet, so I can start understanding how to
handle input from nonstandard devices.  I encourage everyone to start
thinking strategically about what it would mean to make the UI as
kick-ass as K-3D's internals.  Then start doing something to make it happen.

Cheers,
Tim

[tshead.vcf]

---

begin:vcard
fn:Timothy Shead
n:Shead;Timothy
org:www.k-3d.org
email;internet:tshead@...
title:Founder
x-mozilla-html:FALSE
url:www.k-3d.org
version:2.1
end:vcard

---

------------------------------------------------------------------------------
OpenSolaris 2009.06 is a cutting edge operating system for enterprises
looking to deploy the next generation of Solaris that includes the latest
innovations from Sun and the OpenSource community. Download a copy and
enjoy capabilities such as Networking, Storage and Virtualization.
Go to: http://p.sf.net/sfu/opensolaris-get
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[Joe Crawford]

On Wed, Jun 3, 2009 at 11:14 PM, Timothy M. Shead <tshead@...> wrote:

> Gang:
>
> Things always take about twice as long as one would like, even when you
> consistently double your first estimate.  We've spent a tremendous
> amount of time on the geometric data structures for K-3D 0.7.  That's
> not a bad thing - they needed the work, and we have some unique new
> capabilities, but we've also been neglecting other important work.
>
> Since the team can really only handle one large project at a time, I say
> "let's embrace that way of working" ... I call everyone's attention to
> the Roadmap:
>
> http://www.k-3d.org/wiki/Roadmap
>
> ... in a nutshell, I have it in mind to give the UI the same level of
> focus in 0.9 that the data structures have been getting in 0.7.

Wow.  That would be rad!  I'm psyched!

Here is my main thought on the next release version:

Make it work well for core, *fundamental features*, with an easy UI.
There a killer 900HP engine under the hood. Let's put a great steering
wheel on it.

In particular, scene management tasks and the missing pieces of the
modeling features that could make K3D very *practical* for a large
number of users.


> In
> particular, there's the tremendous amount of work done by Joe to
> carry-out and document UI studies.  I want to see all those suggestions
> implemented.  That's one of the big reasons for ripping-out the old
> tutorial code (already done), because it was a big barrier to UI work.

Understandable.

> More than that, however, I want us to look at "big ideas" - new methods
> of interaction (the NURBS sketching that we looked at awhile back comes
> to mind), a completely new API for writing interactive tools, etc.  The
> bigger, the better.  As a personal starting-point, I'm going to be
> picking-up a 3D mouse and a tablet, so I can start understanding how to
> handle input from nonstandard devices.  I encourage everyone to start
> thinking strategically about what it would mean to make the UI as
> kick-ass as K-3D's internals.  Then start doing something to make it happen.

All very cool.  I would like to make a quick note that I think we need
to fix up a little bit of the fundamental stuff before we go off
making anything wild.  I will attempt to test the newest version and
go back to my previous tests, finding where users got frustrated, and
also noting some of the things that I think are fairly obvious needs
of a typical artist.

Thanks Tim!

------------------------------------------------------------------------------
OpenSolaris 2009.06 is a cutting edge operating system for enterprises
looking to deploy the next generation of Solaris that includes the latest
innovations from Sun and the OpenSource community. Download a copy and
enjoy capabilities such as Networking, Storage and Virtualization.
Go to: http://p.sf.net/sfu/opensolaris-get
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[Timothy M. Shead]

Joe Crawford wrote:

> All very cool.  I would like to make a quick note that I think we need
> to fix up a little bit of the fundamental stuff before we go off
> making anything wild.  I will attempt to test the newest version and
> go back to my previous tests, finding where users got frustrated, and
> also noting some of the things that I think are fairly obvious needs
> of a typical artist.

To be honest I'm not sure much of anything has changed UI-wise since you
did that work, so don't be discouraged - what might be useful is if you
could reorganize / consolidate / distill the existing work - my
recollection is that there were multiple user studies and writeups
floating around.  I'm at a point where I'm starting to speculate about
what needs to change at a technical level.  Having a well-organized list
of problems would be a good place to start.

Cheers,
Tim

[tshead.vcf]

---

begin:vcard
fn:Timothy Shead
n:Shead;Timothy
org:www.k-3d.org
email;internet:tshead@...
title:Founder
x-mozilla-html:FALSE
url:www.k-3d.org
version:2.1
end:vcard

---

------------------------------------------------------------------------------
OpenSolaris 2009.06 is a cutting edge operating system for enterprises
looking to deploy the next generation of Solaris that includes the latest
innovations from Sun and the OpenSource community. Download a copy and
enjoy capabilities such as Networking, Storage and Virtualization.
Go to: http://p.sf.net/sfu/opensolaris-get
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[Joe Crawford]

On Thu, Jun 4, 2009 at 12:00 AM, Timothy M. Shead <tshead@...> wrote:

> Joe Crawford wrote:
>
>> All very cool.  I would like to make a quick note that I think we need
>> to fix up a little bit of the fundamental stuff before we go off
>> making anything wild.  I will attempt to test the newest version and
>> go back to my previous tests, finding where users got frustrated, and
>> also noting some of the things that I think are fairly obvious needs
>> of a typical artist.
>
> To be honest I'm not sure much of anything has changed UI-wise since you
> did that work, so don't be discouraged - what might be useful is if you
> could reorganize / consolidate / distill the existing work

Ok.  I'm not sure how much of that stuff I'll be able to track down,
but I'll take a look.  Also, I could test the last released version
with some new users.

There's a few things that are so obvious and fundamental that I can
simply list them.  Other things will take more discussion and back and
forth testing with users.

> - my
> recollection is that there were multiple user studies and writeups
> floating around.  I'm at a point where I'm starting to speculate about
> what needs to change at a technical level.  Having a well-organized list
> of problems would be a good place to start.

Yes, agreed.  I will try to track that stuff down, or re-post it, and
put it in a reorganized wiki page.

I'm excited!


Sincerely,

Joe Crawford
___________________________________
Owner - Celestine Studios   and   Joetainment Enterprises
Cell: 604-866-3050
Email: joetainment@...
Web: http://celestinestudios.com

------------------------------------------------------------------------------
OpenSolaris 2009.06 is a cutting edge operating system for enterprises
looking to deploy the next generation of Solaris that includes the latest
innovations from Sun and the OpenSource community. Download a copy and
enjoy capabilities such as Networking, Storage and Virtualization.
Go to: http://p.sf.net/sfu/opensolaris-get
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[Daniel Scott Matthews]

On Thu, Jun 4, 2009 at 5:00 PM, Timothy M. Shead <tshead@...> wrote:

> Joe Crawford wrote:
>
>> All very cool.  I would like to make a quick note that I think we need
>> to fix up a little bit of the fundamental stuff before we go off
>> making anything wild.  I will attempt to test the newest version and
>> go back to my previous tests, finding where users got frustrated, and
>> also noting some of the things that I think are fairly obvious needs
>> of a typical artist.
>
> To be honest I'm not sure much of anything has changed UI-wise since you
> did that work, so don't be discouraged - what might be useful is if you
> could reorganize / consolidate / distill the existing work - my
> recollection is that there were multiple user studies and writeups
> floating around.  I'm at a point where I'm starting to speculate about
> what needs to change at a technical level.  Having a well-organized list
> of problems would be a good place to start.
>

Joe, did you have a break down of the differences between the needs of
3D character animation artists and more engineering oriented people?
The reason for asking is that for character work you use an
approximate mesh and a detailed displacement map, whereas a person
working on buildings or technology models (even for the same scene)
requires a whole set of different tools.

Can we start with a set of sser types and then compile sets of shared
and unique tool requirements?

------------------------------------------------------------------------------
OpenSolaris 2009.06 is a cutting edge operating system for enterprises
looking to deploy the next generation of Solaris that includes the latest
innovations from Sun and the OpenSource community. Download a copy and
enjoy capabilities such as Networking, Storage and Virtualization.
Go to: http://p.sf.net/sfu/opensolaris-get
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[Daniel Scott Matthews]

correction:
sser = user


On Thu, Jun 4, 2009 at 5:09 PM, Daniel Scott Matthews
<dsmatthews@...> wrote:

> On Thu, Jun 4, 2009 at 5:00 PM, Timothy M. Shead <tshead@...> wrote:
>> Joe Crawford wrote:
>>
>>> All very cool.  I would like to make a quick note that I think we need
>>> to fix up a little bit of the fundamental stuff before we go off
>>> making anything wild.  I will attempt to test the newest version and
>>> go back to my previous tests, finding where users got frustrated, and
>>> also noting some of the things that I think are fairly obvious needs
>>> of a typical artist.
>>
>> To be honest I'm not sure much of anything has changed UI-wise since you
>> did that work, so don't be discouraged - what might be useful is if you
>> could reorganize / consolidate / distill the existing work - my
>> recollection is that there were multiple user studies and writeups
>> floating around.  I'm at a point where I'm starting to speculate about
>> what needs to change at a technical level.  Having a well-organized list
>> of problems would be a good place to start.
>>
>
> Joe, did you have a break down of the differences between the needs of
> 3D character animation artists and more engineering oriented people?
> The reason for asking is that for character work you use an
> approximate mesh and a detailed displacement map, whereas a person
> working on buildings or technology models (even for the same scene)
> requires a whole set of different tools.
>
> Can we start with a set of sser types and then compile sets of shared
> and unique tool requirements?
>

------------------------------------------------------------------------------
OpenSolaris 2009.06 is a cutting edge operating system for enterprises
looking to deploy the next generation of Solaris that includes the latest
innovations from Sun and the OpenSource community. Download a copy and
enjoy capabilities such as Networking, Storage and Virtualization.
Go to: http://p.sf.net/sfu/opensolaris-get
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[bART Janssens-2]

On Thu, Jun 4, 2009 at 8:14 AM, Timothy M. Shead <tshead@...> wrote:
> More than that, however, I want us to look at "big ideas" - new methods
> of interaction (the NURBS sketching that we looked at awhile back comes
> to mind), a completely new API for writing interactive tools, etc.  The
> bigger, the better.  As a personal starting-point, I'm going to be
> picking-up a 3D mouse and a tablet, so I can start understanding how to
> handle input from nonstandard devices.  I encourage everyone to start
> thinking strategically about what it would mean to make the UI as
> kick-ass as K-3D's internals.  Then start doing something to make it happen.

One important aspect of UI work should be the ability for plugins to
create custom constrols in the viewport, much like they can create
custom controls in the panels right now. As I am slowly moving towards
the construction of a house, I'd like to focus on getting the
"architectural tools" idea that has been proposed for SoC off the
ground. This should fit in nicely with UI work, as I guess many of
these tools will require more UI interaction than other plugins. I do
fear the house may be built before the tools are implemented in K-3D,
but that's life ;)

Cheers,

--
Bart

------------------------------------------------------------------------------
OpenSolaris 2009.06 is a cutting edge operating system for enterprises
looking to deploy the next generation of Solaris that includes the latest
innovations from Sun and the OpenSource community. Download a copy and
enjoy capabilities such as Networking, Storage and Virtualization.
Go to: http://p.sf.net/sfu/opensolaris-get
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[Daniel Scott Matthews]

On Fri, Jun 5, 2009 at 8:07 AM, Bart Janssens<bart.janssens@...> wrote:

> On Thu, Jun 4, 2009 at 8:14 AM, Timothy M. Shead <tshead@...> wrote:
>> More than that, however, I want us to look at "big ideas" - new methods
>> of interaction (the NURBS sketching that we looked at awhile back comes
>> to mind), a completely new API for writing interactive tools, etc.  The
>> bigger, the better.  As a personal starting-point, I'm going to be
>> picking-up a 3D mouse and a tablet, so I can start understanding how to
>> handle input from nonstandard devices.  I encourage everyone to start
>> thinking strategically about what it would mean to make the UI as
>> kick-ass as K-3D's internals.  Then start doing something to make it happen.
>
> One important aspect of UI work should be the ability for plugins to
> create custom constrols in the viewport, much like they can create
> custom controls in the panels right now. As I am slowly moving towards
> the construction of a house, I'd like to focus on getting the
> "architectural tools" idea that has been proposed for SoC off the
> ground. This should fit in nicely with UI work, as I guess many of
> these tools will require more UI interaction than other plugins. I do
> fear the house may be built before the tools are implemented in K-3D,
> but that's life ;)
>

I also have a personal interest in using K-3D as an architectural
design and simulation tool.
My wish list includes output to the Indigo render,
http://www.indigorenderer.com/  and interaction with
http://apps1.eere.energy.gov/buildings/energyplus/ as well UI tools
with the same level of fluidity and expressiveness as SketchUp.

The above wish list would benefit from a reworking of the materials
part of the application so that I can define a "substance" e.g.
concrete-123-xyz and that substance has physical properties that
include the optical (material) properties, but also thinks such as
mass, R value etc. so that one model can be used in Indigo for
presentation and visualization and also for LEEDs analysis and
documentation.

With the geometry associated with "substances" (phycal materials) K-3D
could also interface with architectural engineering tools to calculate
loads within the structure. e.g. http://www.csc.fi/english/pages/elmer
 or http://impact.sourceforge.net/ and this capability would also
benefit animation production requiring physics simulation.

If I had to pick one, of the above, I'd choose output to Indigo, and
the associated improvements in handling materials such as texture
baking so that Renderman SL can be used to define maps for output to
renderers that don't support it natively. i.e. have an internal
Renderman renderer and use it as just another way of defining a
texture map.

------------------------------------------------------------------------------
OpenSolaris 2009.06 is a cutting edge operating system for enterprises
looking to deploy the next generation of Solaris that includes the latest
innovations from Sun and the OpenSource community. Download a copy and
enjoy capabilities such as Networking, Storage and Virtualization.
Go to: http://p.sf.net/sfu/opensolaris-get
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[Timothy M. Shead]

>> Joe Crawford wrote:

> Ok.  I'm not sure how much of that stuff I'll be able to track down,
> but I'll take a look.  Also, I could test the last released version
> with some new users.
>
> There's a few things that are so obvious and fundamental that I can
> simply list them.  Other things will take more discussion and back and
> forth testing with users.
>
>> - my
>> recollection is that there were multiple user studies and writeups
>> floating around.  I'm at a point where I'm starting to speculate about
>> what needs to change at a technical level.  Having a well-organized list
>> of problems would be a good place to start.
>
> Yes, agreed.  I will try to track that stuff down, or re-post it, and
> put it in a reorganized wiki page.

FWIW, I've tried to consistently keep

http://www.k-3d.org/wiki/Developer_Documentation

focused on documenting the way things are, while parking new features and
ideas on

http://www.k-3d.org/wiki/Future_Work

so you should start there.

Cheers,
Tim






------------------------------------------------------------------------------
OpenSolaris 2009.06 is a cutting edge operating system for enterprises
looking to deploy the next generation of Solaris that includes the latest
innovations from Sun and the OpenSource community. Download a copy and
enjoy capabilities such as Networking, Storage and Virtualization.
Go to: http://p.sf.net/sfu/opensolaris-get
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[Timothy M. Shead]

> On Thu, Jun 4, 2009 at 8:14 AM, Timothy M. Shead <tshead@...> wrote:
>> More than that, however, I want us to look at "big ideas" - new methods
>> of interaction (the NURBS sketching that we looked at awhile back comes
>> to mind), a completely new API for writing interactive tools, etc.  The
>> bigger, the better.  As a personal starting-point, I'm going to be
>> picking-up a 3D mouse and a tablet, so I can start understanding how to
>> handle input from nonstandard devices.  I encourage everyone to start
>> thinking strategically about what it would mean to make the UI as
>> kick-ass as K-3D's internals.  Then start doing something to make it
>> happen.
>
> One important aspect of UI work should be the ability for plugins to
> create custom constrols in the viewport, much like they can create
> custom controls in the panels right now. As I am slowly moving towards
> the construction of a house, I'd like to focus on getting the
> "architectural tools" idea that has been proposed for SoC off the
> ground. This should fit in nicely with UI work, as I guess many of
> these tools will require more UI interaction than other plugins. I do
> fear the house may be built before the tools are implemented in K-3D,
> but that's life ;)

This is a pretty good example of what I mean when I talk about interaction
- at this point we have a great pipeline, great data structures, and lots
of experience / understanding about how to solve problems within that
framework.  However our interactive tools are pretty crude and (more
importantly, from my perspective), the implementations are extremely
difficult to follow.  I see the big, strategic technical goal as one of
developing a framework for tools that makes it as easy to write a new tool
as it is to write a reader.

Cheers,
Tim



------------------------------------------------------------------------------
OpenSolaris 2009.06 is a cutting edge operating system for enterprises
looking to deploy the next generation of Solaris that includes the latest
innovations from Sun and the OpenSource community. Download a copy and
enjoy capabilities such as Networking, Storage and Virtualization.
Go to: http://p.sf.net/sfu/opensolaris-get
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[Timothy M. Shead]

Daniel Scott Matthews wrote:

> I also have a personal interest in using K-3D as an architectural
> design and simulation tool.
> My wish list includes output to the Indigo render,
> http://www.indigorenderer.com/  and interaction with
> http://apps1.eere.energy.gov/buildings/energyplus/ as well UI tools
> with the same level of fluidity and expressiveness as SketchUp.
>
> The above wish list would benefit from a reworking of the materials
> part of the application so that I can define a "substance" e.g.
> concrete-123-xyz and that substance has physical properties that
> include the optical (material) properties, but also thinks such as
> mass, R value etc. so that one model can be used in Indigo for
> presentation and visualization and also for LEEDs analysis and
> documentation.
>
> With the geometry associated with "substances" (phycal materials) K-3D
> could also interface with architectural engineering tools to calculate
> loads within the structure. e.g. http://www.csc.fi/english/pages/elmer
>  or http://impact.sourceforge.net/ and this capability would also
> benefit animation production requiring physics simulation.
>
> If I had to pick one, of the above, I'd choose output to Indigo, and
> the associated improvements in handling materials such as texture
> baking so that Renderman SL can be used to define maps for output to
> renderers that don't support it natively. i.e. have an internal
> Renderman renderer and use it as just another way of defining a
> texture map.

All this is already doable today - creating a new material type is
trivial.  Suppose I created a couple of plugins to get you started -
would you be able to take the ball and keep it going?

Cheers,
Tim

[tshead.vcf]

---

begin:vcard
fn:Timothy Shead
n:Shead;Timothy
org:www.k-3d.org
email;internet:tshead@...
title:Founder
x-mozilla-html:FALSE
url:www.k-3d.org
version:2.1
end:vcard

---

------------------------------------------------------------------------------
OpenSolaris 2009.06 is a cutting edge operating system for enterprises
looking to deploy the next generation of Solaris that includes the latest
innovations from Sun and the OpenSource community. Download a copy and
enjoy capabilities such as Networking, Storage and Virtualization.
Go to: http://p.sf.net/sfu/opensolaris-get
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[Daniel Scott Matthews]

On Fri, Jun 5, 2009 at 8:29 AM, Daniel Scott
Matthews<dsmatthews@...> wrote:

> On Fri, Jun 5, 2009 at 8:07 AM, Bart Janssens<bart.janssens@...> wrote:
>> On Thu, Jun 4, 2009 at 8:14 AM, Timothy M. Shead <tshead@...> wrote:
>>> More than that, however, I want us to look at "big ideas" - new methods
>>> of interaction (the NURBS sketching that we looked at awhile back comes
>>> to mind), a completely new API for writing interactive tools, etc.  The
>>> bigger, the better.  As a personal starting-point, I'm going to be
>>> picking-up a 3D mouse and a tablet, so I can start understanding how to
>>> handle input from nonstandard devices.  I encourage everyone to start
>>> thinking strategically about what it would mean to make the UI as
>>> kick-ass as K-3D's internals.  Then start doing something to make it happen.
>>
>> One important aspect of UI work should be the ability for plugins to
>> create custom constrols in the viewport, much like they can create
>> custom controls in the panels right now. As I am slowly moving towards
>> the construction of a house, I'd like to focus on getting the
>> "architectural tools" idea that has been proposed for SoC off the
>> ground. This should fit in nicely with UI work, as I guess many of
>> these tools will require more UI interaction than other plugins. I do
>> fear the house may be built before the tools are implemented in K-3D,
>> but that's life ;)
>>
>
> I also have a personal interest in using K-3D as an architectural
> design and simulation tool.
> My wish list includes output to the Indigo render,
> http://www.indigorenderer.com/  and interaction with
> http://apps1.eere.energy.gov/buildings/energyplus/ as well UI tools
> with the same level of fluidity and expressiveness as SketchUp.
>
> The above wish list would benefit from a reworking of the materials
> part of the application so that I can define a "substance" e.g.
> concrete-123-xyz and that substance has physical properties that
> include the optical (material) properties, but also thinks such as
> mass, R value etc. so that one model can be used in Indigo for
> presentation and visualization and also for LEEDs analysis and
> documentation.
>
> With the geometry associated with "substances" (phycal materials) K-3D
> could also interface with architectural engineering tools to calculate
> loads within the structure. e.g. http://www.csc.fi/english/pages/elmer
>  or http://impact.sourceforge.net/ and this capability would also
> benefit animation production requiring physics simulation.
>
> If I had to pick one, of the above, I'd choose output to Indigo, and
> the associated improvements in handling materials such as texture
> baking so that Renderman SL can be used to define maps for output to
> renderers that don't support it natively. i.e. have an internal
> Renderman renderer and use it as just another way of defining a
> texture map.
>

Another program like energy plus is,
http://www.esru.strath.ac.uk/Programs/ESP-r.htm , and it is also open
source. N.B. it's material and other databases,
http://www.esru.strath.ac.uk/Programs/ESP-r_capabilities/databases.html

For more on LEED, see
http://en.wikipedia.org/wiki/Leadership_in_Energy_and_Environmental_Design

------------------------------------------------------------------------------
OpenSolaris 2009.06 is a cutting edge operating system for enterprises
looking to deploy the next generation of Solaris that includes the latest
innovations from Sun and the OpenSource community. Download a copy and
enjoy capabilities such as Networking, Storage and Virtualization.
Go to: http://p.sf.net/sfu/opensolaris-get
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[Timothy M. Shead]

Daniel Scott Matthews wrote:
> On Sat, Jun 6, 2009 at 1:02 PM, Timothy M. Shead<tshead@...> wrote:
>> Daniel Scott Matthews wrote:

>>> If I had to pick one, of the above, I'd choose output to Indigo, and
>>> the associated improvements in handling materials such as texture
>>> baking so that Renderman SL can be used to define maps for output to
>>> renderers that don't support it natively. i.e. have an internal
>>> Renderman renderer and use it as just another way of defining a
>>> texture map.
>> All this is already doable today - creating a new material type is
>> trivial.  Suppose I created a couple of plugins to get you started -
>> would you be able to take the ball and keep it going?
>>
>
> I can have a go and see if I can make it useful.
>
> So any piece of geometry could have any property and we could have a
> way of managing them?
> There are a lot of possibilities with the possibility of some or all
> of the visual characteristics of a material actually being parametric,
> calculated from other "physical" properties.
> http://www.google.com/squared/search?q=material+properties#

Yup, this is all doable today (and another good candidate for better
UI).  I've committed the beginnings of an Indigo module that includes an
Indigo render-engine, and Indigo material plugins (currently unused).
It's a work-in-progress, but my intent is to get it to a point where you
can populate a scene with Indigo materials and do some simple rendering,
then hand it off to you for feedback.

Cheers,
Tim


[tshead.vcf]

---

begin:vcard
fn:Timothy Shead
n:Shead;Timothy
org:www.k-3d.org
email;internet:tshead@...
title:Founder
x-mozilla-html:FALSE
url:www.k-3d.org
version:2.1
end:vcard

---

------------------------------------------------------------------------------
OpenSolaris 2009.06 is a cutting edge operating system for enterprises
looking to deploy the next generation of Solaris that includes the latest
innovations from Sun and the OpenSource community. Download a copy and
enjoy capabilities such as Networking, Storage and Virtualization.
Go to: http://p.sf.net/sfu/opensolaris-get
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[Daniel Scott Matthews]

On Mon, Jun 8, 2009 at 1:04 AM, Timothy M. Shead<tshead@...> wrote:

> Daniel Scott Matthews wrote:
>> On Sat, Jun 6, 2009 at 1:02 PM, Timothy M. Shead<tshead@...> wrote:
>>> Daniel Scott Matthews wrote:
>
>>>> If I had to pick one, of the above, I'd choose output to Indigo, and
>>>> the associated improvements in handling materials such as texture
>>>> baking so that Renderman SL can be used to define maps for output to
>>>> renderers that don't support it natively. i.e. have an internal
>>>> Renderman renderer and use it as just another way of defining a
>>>> texture map.
>>> All this is already doable today - creating a new material type is
>>> trivial.  Suppose I created a couple of plugins to get you started -
>>> would you be able to take the ball and keep it going?
>>>
>>
>> I can have a go and see if I can make it useful.
>>
>> So any piece of geometry could have any property and we could have a
>> way of managing them?
>> There are a lot of possibilities with the possibility of some or all
>> of the visual characteristics of a material actually being parametric,
>> calculated from other "physical" properties.
>> http://www.google.com/squared/search?q=material+properties#
>
> Yup, this is all doable today (and another good candidate for better
> UI).  I've committed the beginnings of an Indigo module that includes an
> Indigo render-engine, and Indigo material plugins (currently unused).
> It's a work-in-progress, but my intent is to get it to a point where you
> can populate a scene with Indigo materials and do some simple rendering,
> then hand it off to you for feedback.
>

Thanks Tim, I look forward to testing it out. It will definitely be a
bragging point if K-3D can, from a single scene, produce separate
outputs that will satisfy both artists and scientists. Once we have a
useful Indigo output plug-in I will start marketing K-3D to all of the
existing Indigo users in the hope that they produce material for our
gallery, help with debugging and perhaps get interested in developing
that aspect of K-3D further. With luck you have opened the door to a
whole new network of people and Energy Plus (or similar) functionality
will do the same.

At the moment I am looking for anything that relates to an
"Architectural Ontology" or an "Architectural Ontology of Materials"
so that any library constructed has good coverage and is compatible
with existing systems and practices. Once I have that sorted out I can
then add the Physical Properties data so that the materials are
meaningful to any simulators that the scene will be used as input for.
I am looking at existing product catalogs and material archives in the
hope of finding a pattern we can adopt but hope also to find a more
formal ontology. If anyone know of one please let me know.

On the point of simulators, there is an issue of what to do with the
data returned from them as it can take several forms, tabular data
that may have rows that relate directly to scene geometry or zones, 2D
maps (heat, light, etc.) that relate to a 2D plane cutting through the
scene, or as 3D data points such as voxels or a sparser set that needs
to be represented only as 2D maps on the faces of the scene geometry.
The rest of the output may be in report form, but still may contain
references to the scene that would benefit from linking. And all of
the above can change over time. e.g. a Thermal or Light level
simulation of a structure that changes throughout the day and the
year.

On the point of time, K-3D needs to know about dates and this needs to
feed into a Sun model, which may as well borrow from the astronomical
code in one or more of the existing FOSS projects that cover the area.
i.e. the user should be able to place geometry by Scale, Latitude,
Longitude and Altitude, for any celestial body (usually Earth, but
NASA and Game developers would want any location to be possible).
Obviously Lat. Lon. Alt for a given celestial body changes, in terms
of absolute coordinates, over time and this raises the question of
where 0,0,0,0 (yes 4D) actually is. I will leave that open for debate,
but we do need to decide. One all that is working then input/output
with GIS systems such as GRASS are possible, as well as Google Earth
(Universe), allowing for the automatic generation of environments for
scenes so that the user can focus on there model but know that when it
is rendered it's will be done in context. e.g. a texture map for and
environment sphere is rendered based on the models location
parameters.

Then there is the question of weather and atmospherics... but I'll
leave that for another day. ;-)

If all of the above is OMG! inducing do keep in mind that I am talking
firstly about ensuring that the K-3D application architecture design
decisions ensure that such options are possible in the future and that
the necessary fundamental properties and methods are in the API to
allow such functions to be built.

------------------------------------------------------------------------------
OpenSolaris 2009.06 is a cutting edge operating system for enterprises
looking to deploy the next generation of Solaris that includes the latest
innovations from Sun and the OpenSource community. Download a copy and
enjoy capabilities such as Networking, Storage and Virtualization.
Go to: http://p.sf.net/sfu/opensolaris-get
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[Timothy M. Shead]

Daniel Scott Matthews wrote:

> On Mon, Jun 8, 2009 at 1:04 AM, Timothy M. Shead<tshead@...> wrote:
>
>> Yup, this is all doable today (and another good candidate for better
>> UI).  I've committed the beginnings of an Indigo module that includes an
>> Indigo render-engine, and Indigo material plugins (currently unused).
>> It's a work-in-progress, but my intent is to get it to a point where you
>> can populate a scene with Indigo materials and do some simple rendering,
>> then hand it off to you for feedback.
>>
>
> Thanks Tim, I look forward to testing it out. It will definitely be a
> bragging point if K-3D can, from a single scene, produce separate
> outputs that will satisfy both artists and scientists. Once we have a
> useful Indigo output plug-in I will start marketing K-3D to all of the
> existing Indigo users in the hope that they produce material for our
> gallery, help with debugging and perhaps get interested in developing
> that aspect of K-3D further. With luck you have opened the door to a
> whole new network of people and Energy Plus (or similar) functionality
> will do the same.

OK, time to test!  I've written-up some very brief instructions at

http://www.k-3d.org/wiki/IndigoEngine

along with a sample image.  So the first step is to get it built and see
if you can get an image out of it.  Once we've got that functioning, I'm
going to point-out some places where I'd like to have you start to
"fill-in the blanks" functionality-wise.

Cheers,
Tim

[tshead.vcf]

---

begin:vcard
fn:Timothy Shead
n:Shead;Timothy
org:www.k-3d.org
email;internet:tshead@...
title:Founder
x-mozilla-html:FALSE
url:www.k-3d.org
version:2.1
end:vcard

---

------------------------------------------------------------------------------
OpenSolaris 2009.06 is a cutting edge operating system for enterprises
looking to deploy the next generation of Solaris that includes the latest
innovations from Sun and the OpenSource community. Download a copy and
enjoy capabilities such as Networking, Storage and Virtualization.
Go to: http://p.sf.net/sfu/opensolaris-get
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[Daniel Scott Matthews]

On Mon, Jun 8, 2009 at 3:25 PM, Timothy M. Shead<tshead@...> wrote:

> Daniel Scott Matthews wrote:
>> On Mon, Jun 8, 2009 at 1:04 AM, Timothy M. Shead<tshead@...> wrote:
>>
>>> Yup, this is all doable today (and another good candidate for better
>>> UI).  I've committed the beginnings of an Indigo module that includes an
>>> Indigo render-engine, and Indigo material plugins (currently unused).
>>> It's a work-in-progress, but my intent is to get it to a point where you
>>> can populate a scene with Indigo materials and do some simple rendering,
>>> then hand it off to you for feedback.
>>>
>>
>> Thanks Tim, I look forward to testing it out. It will definitely be a
>> bragging point if K-3D can, from a single scene, produce separate
>> outputs that will satisfy both artists and scientists. Once we have a
>> useful Indigo output plug-in I will start marketing K-3D to all of the
>> existing Indigo users in the hope that they produce material for our
>> gallery, help with debugging and perhaps get interested in developing
>> that aspect of K-3D further. With luck you have opened the door to a
>> whole new network of people and Energy Plus (or similar) functionality
>> will do the same.
>
> OK, time to test!  I've written-up some very brief instructions at
>
> http://www.k-3d.org/wiki/IndigoEngine
>
> along with a sample image.  So the first step is to get it built and see
> if you can get an image out of it.  Once we've got that functioning, I'm
> going to point-out some places where I'd like to have you start to
> "fill-in the blanks" functionality-wise.
>

Builds from svn fine, but there are a few issues with coordinate spaces.
the camera is looking from the opposite side to the K-3D viewport and
the light object is rotated 90 deg.

Tim, I sent you a test scene and render frame that shows the problem.

I'd also add a few more options to the IndigoEngine including the choice
of using the -n m switch which is very handy for people with LANs and
idle CPUs. :-)
Basically everything under, renderer_settings section of the "Indigo
Manual.pdf" should be in the IndigoEngine properties.

------------------------------------------------------------------------------
Crystal Reports - New Free Runtime and 30 Day Trial
Check out the new simplified licensing option that enables unlimited
royalty-free distribution of the report engine for externally facing
server and web deployment.
http://p.sf.net/sfu/businessobjects
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[Timothy M. Shead]

Daniel Scott Matthews wrote:

> On Mon, Jun 8, 2009 at 3:25 PM, Timothy M. Shead<tshead@...> wrote:
>> http://www.k-3d.org/wiki/IndigoEngine
>>
>> along with a sample image.  So the first step is to get it built and see
>> if you can get an image out of it.  Once we've got that functioning, I'm
>> going to point-out some places where I'd like to have you start to
>> "fill-in the blanks" functionality-wise.
>>
>
> Builds from svn fine, but there are a few issues with coordinate spaces.
> the camera is looking from the opposite side to the K-3D viewport and
> the light object is rotated 90 deg.
>
> Tim, I sent you a test scene and render frame that shows the problem.

Yup, got it here ...

> I'd also add a few more options to the IndigoEngine including the choice
> of using the -n m switch which is very handy for people with LANs and
> idle CPUs. :-)
> Basically everything under, renderer_settings section of the "Indigo
> Manual.pdf" should be in the IndigoEngine properties.

I'll take a look at the transformation issues, FWIW these things always
crop-up when we're working with another tool, so this isn't entirely
unexpected.

As far as the parameters go, this is where I'd like you to lend a hand
... if you look at modules/indigo/render_engine.cpp line 529, you'll see
where the current set of properties are defined.  Then take a look at
line 92 to see where they're initialized.  Finally, look around line 372
to see them in-use.  You should be able to add to the list pretty
easily, let me know if you have questions.

Cheers,
Tim

[tshead.vcf]

---

begin:vcard
fn:Timothy Shead
n:Shead;Timothy
org:www.k-3d.org
email;internet:tshead@...
title:Founder
x-mozilla-html:FALSE
url:www.k-3d.org
version:2.1
end:vcard

---

------------------------------------------------------------------------------
Crystal Reports - New Free Runtime and 30 Day Trial
Check out the new simplified licensing option that enables unlimited
royalty-free distribution of the report engine for externally facing
server and web deployment.
http://p.sf.net/sfu/businessobjects
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[Daniel Scott Matthews]

On Tue, Jun 9, 2009 at 2:51 PM, Timothy M. Shead<tshead@...> wrote:
> As far as the parameters go, this is where I'd like you to lend a hand
> ... if you look at modules/indigo/render_engine.cpp line 529, you'll see
> where the current set of properties are defined.  Then take a look at
> line 92 to see where they're initialized.  Finally, look around line 372
> to see them in-use.  You should be able to add to the list pretty
> easily, let me know if you have questions.
>

I have questions, many many questions.... :-)

I had the render called with the network rendering switch but as that
side of things is in flux I commented it out and got on with looking a
rendering parameters that go into the .igs file.


One thing that came up was how to handle the conditional existence of
parameters in the GUI, e.g. I need to have a drop-down list for things
like tone mapping choices and depending on which one the user selects
the gui will then show the appropriate set of parameters.

at the moment I have just done this so I can test things but you can
see that there are two sets to handle in this case.

      // Setup tonemapping ...
      stream << "<tonemapping>\n";
//      stream << "<linear>\n";
//     stream << "<scale>" << "1.0" << "</scale>\n";
//      stream << "</linear>\n";
      stream << "<reinhard>\n";

      stream << "<pre_scale>" << "6.0" << "</pre_scale>\n";

      stream << "<post_scale>" << "1.0" << "</post_scale>\n";

      stream << "<burn>" << "6.0" << "</burn>\n";

      stream << "</reinhard>\n";
      stream << "</tonemapping>\n";

We could add save_untonemapped_exr true so that there is always that
data available for post production and default to reinhard for the
.png output.


Also, when you do this

      const k3d::int32_t pixel_width = m_pixel_width.pipeline_value();

are you doing this to ensure a whole number for pixels, and if so why
are they not whole elsewhere?

------------------------------------------------------------------------------
Crystal Reports - New Free Runtime and 30 Day Trial
Check out the new simplified licensing option that enables unlimited
royalty-free distribution of the report engine for externally facing
server and web deployment.
http://p.sf.net/sfu/businessobjects
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[Timothy M. Shead]

Daniel Scott Matthews wrote:

> One thing that came up was how to handle the conditional existence of
> parameters in the GUI, e.g. I need to have a drop-down list for things
> like tone mapping choices and depending on which one the user selects
> the gui will then show the appropriate set of parameters.

In the short-term, you should create the union of all the parameters you
need, so:

* An enumeration property to choose between "linear" and "reinhard".
* A set of properties required for "linear".
* A set of properties required for "reinhard".

This will be immediately usable, albeit suboptimal.  Then, should it be
necessary, we can do some custom panel work to address hiding/showing
controls.

> Also, when you do this
>
>       const k3d::int32_t pixel_width = m_pixel_width.pipeline_value();
>
> are you doing this to ensure a whole number for pixels, and if so why
> are they not whole elsewhere?

Not sure if this is answering your question or not, but "m_pixel_width"
and "m_pixel_height" are both integer properties - so this line is just
reflecting the type of value that's already there.  What "elsewhere" are
you looking at?

Cheers,
Tim

[tshead.vcf]

---

begin:vcard
fn:Timothy Shead
n:Shead;Timothy
org:www.k-3d.org
email;internet:tshead@...
title:Founder
x-mozilla-html:FALSE
url:www.k-3d.org
version:2.1
end:vcard

---

------------------------------------------------------------------------------
Crystal Reports - New Free Runtime and 30 Day Trial
Check out the new simplified licensing option that enables unlimited
royalty-free distribution of the report engine for externally facing
server and web deployment.
http://p.sf.net/sfu/businessobjects
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[Timothy M. Shead]

Daniel Scott Matthews wrote:

> Builds from svn fine, but there are a few issues with coordinate spaces.
> the camera is looking from the opposite side to the K-3D viewport and
> the light object is rotated 90 deg.

So, kudos to Daniel, I've just committed his first patch that extends
the Indigo capabilities.  I've fixed some problems with transformations,
so everything should be as expected, save one item:

>From what I can see, you rotated the light because you wanted it that
way, right?  For the life of me, I can't find any documentation on how
to apply a rotation to an Indigo rectangle_light, nor can a find a
sample file that provides a rotated light - that's why the lights always
point along the same direction.  If you can find an example, let me know ...

Cheers,
Tim

[tshead.vcf]

---

begin:vcard
fn:Timothy Shead
n:Shead;Timothy
org:www.k-3d.org
email;internet:tshead@...
title:Founder
x-mozilla-html:FALSE
url:www.k-3d.org
version:2.1
end:vcard

---

------------------------------------------------------------------------------
Crystal Reports - New Free Runtime and 30 Day Trial
Check out the new simplified licensing option that enables unlimited
royalty-free distribution of the report engine for externally facing
server and web deployment.
http://p.sf.net/sfu/businessobjects
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development