A bit more cleared things now :)
But let me be stupid now once and learn instead of not asking and
staying stupid.
So if I understood correctly.. for example tool that someone create
for rigging and animation for example, should be, in theory possible
to work in all supported applications via Splice.
So you can use that tool, rig in Maya, animate in SI or vice verse or
completely in one and use in another?
Just trying to figure out if I got a grasp of idea Thanks!
On Wed, Mar 5, 2014 at 7:18 PM, Paul Doyle <[email protected]
<mailto:[email protected]>> wrote:
(cross-posted to SI-Community and to our mailing list)
Hi everyone,
We’ve been mentioned in a few places recently in relation to what
has happened to Softimage, and particularly in discussions
regarding ‘what now?’. I’m going to try and cover everything
that’s come up in various emails and forum threads. If I missed
anything, let me know and I’ll do my best to get back to you. You
can reach me directly using [email protected]
<mailto:[email protected]>, and I will track responses in the
places this is posted.
Before getting into anything else, you need to know that *FABRIC
IS FREE* to individuals. Freelancers, hobbyists, students,
professionals - we don't care. We strongly believe that the way
for Fabric to become successful is for people to have easy access
to it. We make money from studios adopting the technology, not
from people trying to use it.
“What the &^%$ is Fabric Engine?”
Fabric Engine is a platform for building custom tools and
applications. It’s designed to be open and extensible. That
doesn’t really mean much though, so let me explain a bit about the
major components.
*Fabric Core - *this is a high-performance multi-threading engine.
It takes the code you write in KL, and makes sure that it runs as
fast as possible on the hardware you’re running it on.
*Kernel Language (KL)* – this is the high-level language used for
writing this code. KL has been designed to be accessible to
someone that is most comfortable writing Python code – it behaves
as a dynamic language, so you can iterate quickly. However, it is
as fast as statically compiled code. Soon we will be able to
execute KL on your graphics card, without needing to change a line
of code - GPU compute for free, using a language that anyone can
learn. You can learn a lot more about the KL language here:
http://fabricengine.com/splice/kernel-language-kl/
*Extensions - *we ship a range of libraries that Fabric can use.
Alembic, FBX, Bullet, Open Image IO, hardware devices and so on.
We also ship the EDK to make your own extensions, and all the
source code to our extensions, and tools to make it as easy
possible. This means that you can use these libraries inside of
your existing DCCs.
*Splice API - *this API makes it possible for you to use the
Fabric Core and KL within other applications. That might be a
commercial DCC - we currently support Maya, Softimage, Arnold and
Nuke (Max and Houdini coming soon) - or it could be a custom
framework. This allows you to use Fabric to build tools that are
DCC-agnostic - you can easily move them between applications,
which can be very powerful. There is a full demo here:
https://vimeo.com/76325922
*Get Started*
So if you're a Softimage user, this is where I suggest you start -
you'll be working within your familiar application environment,
and you'll get a good sense of what is possible with the Fabric Core:
1) Get Fabric: http://fabricengine.com/get-fabric/
2) Get Splice: http://dist.fabric-engine.com/FabricSplice/1.11.0/
3) Follow the tutorials:
http://documentation.fabric-engine.com/Splice/latest/HTML/Workshops/klworkshop.html
*Scene Graph - *We also have a standalone Python/Qt framework.
However - we are currently working on 2.0 of this (due early
summer). My recommendation is that you take a look at the version
2.0 outline presentation: https://vimeo.com/84300368 but that you
limit your testing to the Fabric in Softimage stuff linked above.
*Managed transition - *Everything that you do in other
applications with Fabric and the Splice API will be portable to
the 2.0 scenegraph. This gives you a path to the future without
abandoning your current toolset - you might decide to move to Maya
down the line, or another application, or to the Fabric standalone
framework. It gives you options and most importantly, it allows
you a managed transition.
*Real-Time Renderer (RTR) - *we have a powerful real-time renderer
that is targeted at production. It's written completely in KL,
which is testament to the power of the framework - it also means
it's completely open. You must remember that it is a platform for
building rendering applications - so far we have customers using
Fabric and the RTR for deep image compositing, scene assembly with
Arnold integration, animation preview, projection mapping and
asset preview. You can see some additional information here:
http://fabricengine.com/splice-2/fabric-engine/rendering/
*“What does it do out of the box?” - *right now, not much. We ship
a range of demo applications, but right now if you aren't
comfortable writing python or similar then you're going to
struggle. However, Fabric 2.0 will have much more for you by way
of visual programming and 'instant utility' tools that you can
just use. Over time we will have more and more functionality like
this, but we elected to start with the platform capabilities first
- it's taken us a while but it's starting to pay off. This will
change over the next year as we see the community pushing
functionality themselves, and hopefully we'll see a few
collaborative projects come to fruition soon. We're also
supporting 3rd party developers like Eric Mootz, so we're looking
forward to seeing more artist-centric tools coming through this year.
*“What’s the long term roadmap?” - *watch the Fabric 2.0 video
(linked above - https://vimeo.com/84300368) for an inkling. We are
thinking in broad strokes about: scene assembly, rigging and
animation, virtual production and visual programming. We see these
as areas that are ripe for some innovation and we have a platform
that offers all of the building blocks to do this. However - we
aren't beholden to investors or shareholders, so our roadmap
sometimes changes according to what our customers are telling us
they need.
*"Are you in control of the company?" - *Yes. We have investors
but Phil, Peter and I run the company. Nothing can happen without
us agreeing to it.
*We can't do it alone* - if you want to see change happen, you
have to get involved. Small companies like Fabric need your
support, we need you to kick things around and tell us what you
think. We need to know what you need. It's immensely frustrating
to be told "this is cool, if only it did X" and then we do X and
the response is "now if it did Y then I'd take a look". Get
involved - it's free!
*Let's get creative - *we are open to creating a consortium and
finding ways to open-source work done there. Obviously there are
hooks into Fabric and the concern will be around vendor dependency
- however, a lot of that can be addressed in the design of a
particular project. We have done deals that give source code
access to customers after a certain number of years, and we will
work with studios to give that kind of security. We see this as
something where we would not be controlling anything, but working
on a partnership basis with the studios that want to do this. It
has to be driven by studios that want to see some control over
their destiny, with companies like Fabric getting involved to
support and drive innovation. We are a platform, so for us this is
the way to success - providing high-performance, dependable
components that can be used to build production-specific tools,
which are best built by you. If you are interested in becoming a
part of this working group then please email me
([email protected] <mailto:[email protected]>) - right
now I'm just gauging interest with the hope that we can do
something amazing together.
Thanks,
Paul and the Fabric team
