Cheetah3D 7.3 Beta 3

#1
Cheetah3D 7.3 Beta 3

Hi,
it's quite some time ago since I've released the last Beta and although Beta 3 might not look too spectacular it's actually quite a big one. I've mainly worked in two areas. The first is the transition from libstdcpp to libc++. That was something which was overdue since quite some time and I'm happy that I'm done with that work now.
The second big project I've started it preparing Cheetah3D for the Metal API. To do that I'm in the process of moving most of the OpenGL code behind some wrapper classes. That will make the transition to Metal much easier once I perform the final step. So please have a special look if the 3D view still works as before.
Last but not least there is also a new feature in Beta 3. The GLTF/GLB exporter now supports skeletal animations.:smile:

Next I want to start work on supporting the dark mode in upcoming macOS Mojave.

New features in Cheetah3D 7.3 Beta 3:
Code:
-added skeletal animation export to GLTF/GLB exporter
-transitioned from libstdcpp to libc++
   -upgraded Bullet physics engine (v2.79 -> v2.87)
   -upgraded OpenEXR library (v2.0 -> v2.2.1)
   -upgraded Carve CSG library to latest version
   -updated FBX SDK library (libstdcpp -> libc++)
   -removed dependency to BOOST library
-repaced lots of OpenGL code (in preparation for Metal)
-fixed bug in glass shader (appearing in Falcon with very low roughness values)
-fixed bug in particle mesh (particle alignment problem)
-fixed bug in SVG import
-fixed bug in gadget tool (problem with algning too (0,0,0))
-fixed crashing bug in scalpel tool
New features in Cheetah3D 7.3:
Code:
-added PBR shader
   -based on Disneys Principled shader
   -added support for GGX BRDF
-added glTF exporter
   -supports glTF and glb (binary glTF) output
   -supports the following meshe attributes
      -vertex
      -normal
      -two uv sets
      -vertex color
  -supports PBR materials
      -color texture
      -AO texture
      -metalness texture
      -roughness texture
      -emissive texture
      -AO/roughness/metalness texture merging to embedded texure export
      -embedded textures
  -supports animation export
      -Takes
      -object animation
      -skeletal animations
   support camera export
-added support for refractive caustics to Falcon renderer
-added "R", "G", "B" and "A" output channels to Image material node
-added support for DirectX style normal maps
-transitioned from libstdcpp to libc++
   -upgraded Carve CSG library to latest version
   -upgraded Bullet physics engine (v2.79 -> v2.87)
   -upgraded OpenEXR library (v2.0 -> v2.2.1)
   -updated FBX SDK library (libstdcpp -> libc++)
   -removed dependency to BOOST library
-removed all OpenCL code
-repaced lots of OpenGL code (in preparation for Metal)
-changed specular color of default material to black
-improved node editor (copies all possible parameter values to new shader)
-fixed bug in node editor (fixed node position when changing shader node)
-fixed redraw bug in node editor
-fixed .h export problem
-fixed bug in material preview of state node
-fixed crashing bug in material import (failed to properly connect nodes)
-fixed bug in photon mapping
-fixed object selection bug
-fixed bug in node editor (displayed broken connections)
-fixed bug in "Add node" menu
-fixed material delete bug
-fixed bug in glass shader (invisible area lights)
-fixed bug in .jas importer (problem with v6.x skylights)
-fixed bug in .jas importer (converting v6.x point lights with radius to sphereical areas lights could fail)
-fixed bug in shadow mat (blocked glossy reflections)
-fixed bug when copy&pasting objects (take manager didn't' update)
-fixed bug in glass shader (appearing in Falcon with very low roughness values)
-fixed bug in particle mesh (particle alignment problem)
-fixed bug in SVG import
-fixed bug in gadget tool (problem with algning too (0,0,0))
-fixed crashing bug in scalpel tool

Attention:
PLEASE BE AWARE THAT THIS IS BETA SOFTWARE SO DON'T USE IT ON IMPORTANT DATA.

Download link of Cheetah3D 7.3b3:
http://www.cheetah3d.com/download/Beta/Cheetah3D7.3b3.dmg

I hope you like the new features in Cheetah3D 7.3.:smile:

Kind regards
Martin
 

frank beckmann

Well-known member
#2
Thanks for the SVG fixes+improvements - most welcome. :icon_thumbup: :p
(Copy&pasting from Intaglio ignores Cheetah3d´s preference setting though - no big deal - have to save the file first.)

Cheers
Frank
 
#3
Ops, I've just released that I forgot to completely enable the scalpel tool after fixing a bug.:redface: Re-uploaded Beta 3 with the working scalpel tool. So please download it again. Sorry for the inconveniences.
 

uncle808us

Active member
#5
That will make the transition to Metal much easier
What will the transition mean for us older Mac OS system users? Should we be transitioning to other 3D software. will this be the end of updates for our beloved (learned) Cheetah 3D?:frown:
 
#6
What will the transition mean for us older Mac OS system users? Should we be transitioning to other 3D software. will this be the end of updates for our beloved (learned) Cheetah 3D?:frown:
You have a MacBook Pro Retina which supports Metal so you have nothing to worry since your Mac supports Metal.

I want to delay the transition to Metal as long as it makes sense. I just want to be prepared for the case that Apple gives up OpenGL faster than expected.
 
#7
Thanks for the SVG fixes+improvements - most welcome. :icon_thumbup: :p
(Copy&pasting from Intaglio ignores Cheetah3d´s preference setting though - no big deal - have to save the file first.)
There are no preferences for Copy & Pasting paths. Which preferences do you mean?
 

frank beckmann

Well-known member
#8
There are no preferences for Copy & Pasting paths. Which preferences do you mean?
I just recall that copy&pasting will maintain "pt" as unit. So 50mm=141,732 Pt=1.4173 units in Cheetah3d.
And to keep 50mm at some reasonable scale a factor of 0,003528 comes in as 0.5 units which let´s me easy do true to scale (1:10) work. :icon_thumbup:
Freu freu. :p

Cheers
Frank
 

uncle808us

Active member
#9
You have a MacBook Pro Retina which supports Metal so you have nothing to worry since your Mac supports Metal.

I want to delay the transition to Metal as long as it makes sense. I just want to be prepared for the case that Apple gives up OpenGL faster than expected.
My other Mac is Mid 2009 both running El Cap. I don't plan on changing to a newer OS on either computer so that would mean the end of C3D updates correct?
I may be nearing the end of my run with Apple. ( since March 2 Macintosh II Mac II January 15, 1990)
The new Apple is not what it used to be (to me anyway.)
Then a lot of things aren't.
I still drive a 1993 Jeep Wrangler.
That's right I'm an old fart stuck in my ways.
 
Last edited:
#10
I am curious about what transition to Metal means as well. I too am on El Capitan with no plans to upgrade, but only because I am on a nicely running Hackintosh and do not currently want to go through involved process of upgrading. A new Hackintosh with latest Mac OS would be easier If I got a more current mother board and processor but I am actually waiting for next year to see what the redesigned Mac Pro holds in store. I am hoping I can dump Hackintosh all together if Apple returns to a form factor more like the towers used to be. Also most of my 3d work is now done on a PC. My Hackintosh is primarily used for Cheetah and a few other apps (I still do prefer the Mac OS and probably always will though).
 
#11
That questions is actually pretty difficult to answer since I also don't know what the future brings. The future of OpenGL on the Mac is unknown and Apple keeps back any information if I ask them directly. So being PREPARED for faster than expected end of OpenGL is not the worst idea.

The other side I already planed to update the whole OpenGL code to a more recent version. I actually wanted to use OpenGL 3.0 pretty soon. Unluckily that makes no sense anymore. Every single line of OpenGL 3.0 code would be a total waste of time since OpenGL has no future anymore.

So if I want to support newer graphic card features I have to go directly Metal.

At the end I'm in a pretty clueless state at the moment. I wished Apple would be a little bit more open on that subject.
 
#13
Martin, thank you for the work you've put in on Cheetah, it certainly is an amazing piece of software.
That said, Macintosh computers are lagging behind. They're very expensive, and the new AMD processors are looking quite appealing; 12 Cores and 24 threads for reasonable prices. But moving to an AMD system means leaving Mac OSX behind. Some have managed to create AMD Hackintoshes, but supposedly it is not easy, nor are the results stable.
Are you completely dedicated to the Mac OS? Any chance of even a Linux port?
Thanks.
 

uncle808us

Active member
#14
That questions is actually pretty difficult to answer since I also don't know what the future brings. The future of OpenGL on the Mac is unknown and Apple keeps back any information if I ask them directly. So being PREPARED for faster than expected end of OpenGL is not the worst idea.

The other side I already planed to update the whole OpenGL code to a more recent version. I actually wanted to use OpenGL 3.0 pretty soon. Unluckily that makes no sense anymore. Every single line of OpenGL 3.0 code would be a total waste of time since OpenGL has no future anymore.

So if I want to support newer graphic card features I have to go directly Metal.

At the end I'm in a pretty clueless state at the moment. I wished Apple would be a little bit more open on that subject.
Thanks for the reply Martin. I understand your thinking and I agree with you if I were in your place I would do the same. (probably ):smile: :icon_thumbup:
 
#15
Falcon is so much lagging behind the competition regarding render speed, it badly needs GPU support IMHO.
Connecting a box with decent graphics is also cheaper than a new Mac with more CPU cores.
I don't know if a Ryzen CPU really renders faster than Intel but with the Metal transition ahead I can't really imagine another one on top of that.
 
#16
That questions is actually pretty difficult to answer since I also don't know what the future brings. The future of OpenGL on the Mac is unknown and Apple keeps back any information if I ask them directly. So being PREPARED for faster than expected end of OpenGL is not the worst idea.

The other side I already planed to update the whole OpenGL code to a more recent version. I actually wanted to use OpenGL 3.0 pretty soon. Unluckily that makes no sense anymore. Every single line of OpenGL 3.0 code would be a total waste of time since OpenGL has no future anymore.

So if I want to support newer graphic card features I have to go directly Metal.

At the end I'm in a pretty clueless state at the moment. I wished Apple would be a little bit more open on that subject.
Martin, do you think you will implement SSS and volumetrics before Cheetah becomes Metal-only ?

Thank you.
 
#18
Martin, do you think you will implement SSS and volumetrics before Cheetah becomes Metal-only ?
Probably. :wink: My goal was to have SSS by the end of this year (for the 15th Anniversary of Cheetah3D). But WWDC 2018 crossed that plan and I spent almost my complete time since then to make Cheetah3D future proof. Although I still have quite some work to do the dust definitely starts settling.

P.S. The Mojave Dark mode will be available in Beta 4.:smile:
 

Attachments

Last edited:
#19
Probably. :wink: My goal was to have SSS by the end of this year (for the 15th Anniversary of Cheetah3D). But WWDC 2018 crossed that plan and I spent almost my complete time since then to make Cheetah3D future prove. Although I still have quite some work to do the dust definitely starts settling.

P.S. The Mojave Dark mode will be available in Beta 4.:smile:
Thanks Martin. Looking good.
 
#20
At the risk of sounding a bit controversial, I really don't have a problem if Cheetah doesn't run on computers made older than 3-5 years ago. Fact of the matter is, a lot of professional 3D modeling software in many cases costs as much as a new Mac every year as a subscription, thanks to the new (dreaded) software-as-a-service economy these days. Cheetah provides pretty much everything that a casual and professional-ish user could hope for in a very reasonably priced package. I'd hate to see it held back by not taking the fullest advantage of modern systems. I think you're striking a good, if perhaps a little conservative, balance to keep everyone happy.

Also, the dark mode concept looks amazing!
 
Top