LightWave 3D 9.5 was just released. An excellent update to the current LightWave cycle, which I am extremely pleased with.
With the promised hotfix/service release (9.5.1 maybe?), that should arrive shortly, fixing the minor things that didn’t make it to the Siggraph release (which I hope includes fixes for the FBX bone export and OBJ parts export/import) it will be an awesome production horse pumping out projects here.
But with every new release, the eager part of me can’t help thinking of what I want the next version of LightWave to be like. So here’s my top ten countdown of what love I hope LightWave 10 will receive.
- Micropolygons/Buckets. The recent updates to LightWave’s renderer have made it top notch for fast, high quality GI renderings. And with 64-bit computing and tons of gigs of RAM I can easily squeeze millions and millions of polygons without LightWave even thinking of breaking out in sweat. But as it’s pretty common using displacement maps and highly subdivided meshes today it’s quite an uneffective use of resources keeping it all in memory instead of rendering the scene in buckets. And even with a mesh subdivided to millions and millions of polygons I can’t seem to get the same level of details in LightWave coming through that I for instance can get in Modo when using micropolygon displacements. So this is something I really, really hope will be implemented for LightWave 10.
- Render Pass Manager. This is something that can be solved with 3rd party tools today, but they are all workarounds and implemented as good as the current architecture can allow. In a modern rendering application this is something that should be native in my world. Preferably a node based workflow where buffers and savers can be combined to my liking. I want to have a list of passes and buffers and set it up with different combinations of shaders/lights/objects and just press render and get all the images outputed at once and jump to my compositing package. Currently it’s pretty much a pain to work with render passes in LightWave, and it’s such an important part of a modern workflow so this is definitely something that I really think must make it into LightWave 10.
- Improved Rigging. Character animation in LightWave works pretty well today I think, I don’t really share the opinion that LW is no good for char animation. And with the new pole vector and IK features in LW9.5 things got a bit better again. But the rigging! That’s more or less a pain in the a.. at the moment! It just takes way to much time and it’s incredible incovenient jumping back and forth between modeler and layout during the process. A unified environment will probably solve some of the pain automatically. But what I really want is to be able to edit my weight maps while having my bones deform to immediately see the effect. It’s just to much guesswork and editing in the dark at the moment. And a setup mode would be excellent, like it works in pmG:messiah. I want to be able to jump back to my setup pose and edit any bone or make adjustments without the rig breaking or the animation going bonkers.
- New Dynamics Engine. The dynamics engine in LightWave hasn’t seen much progress for some time and starts to show it’s age. I especially hope the cloth dynamics will get some much needed improvements. A mode tailored for hair simulations would be excellent, but just get the ordinary cloth solver up to date – Closer to realtime calculations and with better precision I’d be very happy.
- Hair tools. With FiberFX making it’s first appearance in LightWave 9.5, I think it’s an excellent first iteration. I am very happy with this new addition to LightWave. For LightWave 10 I hope the progress continues, especially the styling tools, as I think the hair renderer is very good already (not that I don’t hope that it will be even better the next time). But the styling tools would be excellent if the FiberModeler could be ditched and instead having the fibers available in the ordinary modeler view, and having the styling tools as ordinary modeling tools so they can be used and combined with other tools. And with a background constrainer (See 6. Topology tools), the fiber strands could easily be prevented from penetrating the underlying geometry.
- Topology tools. As sculpting packages gets used more and more, some topology tools would be really useful to have to be able to retopologize directly in LightWave. A good constraining system where you can constrain to a background object and build new geometry on top of it would be oh so useful. All tools should respect the constraint and it would be so easy to quickly make lowres meshes of hires sculpts and such.
- More Nodes. The node editor for surfacing that came with LW9 is wonderful. I just love a node based workflow, as it’s such a powerful and intuitive way to work. XSI ICE is a great example of the power of nodes. I do hope the node system in LW9 will be extended in LW10 to work everywhere. Nodes for particles, dynamics, motions, renderbuffers and so on… More nodes to the people!
- Instancing. Instancing is long overdue in LightWave and can only be solved with 3rd party tools at the moment. HD instance is a good rendering instance plugin. But this is something that should work out of the box in a modern application. And when it gets implemented I really hope it won’t be just rendering instancing but also modeling instances. Modeling instancing is where I personally would need it the most.
- Unification. The separate application system (Modeler/Layout) has played out it’s role. Now it’s time to bring all modeler tools inside Layout and just call it LightWave instead. Modo is an excellent example of how one unified environment still can separate different tasks in an elegant way. Also many modeler tools can be unified into one tool (No need for 5 different bevel tools as an example).
- Updated UV tools. The UV tools definitely needs a more modern touch. Currently they are usable because of excellent 3rd party tools like PLG UV unwrap. In LightWave 10 I hope we get a native UV Unwrapper, just as good as PLG or preferably better. Other goodies would be that all modeler tools should work in the UV editor, UV overlap checking and a decent UV to image exporter. The current EPS exporter doesn’t work very well with subpatch UV’s and so on.
One can dream, right? :) To me, if these things made it into LW10, LightWave would be an amazing state of the art 3D application again. The 9.x cycle has been ongoing since 2006 with 9.5 probably being the end of the line for 9.x, so I believe the first signs of LightWave 10 development will surface during late Q4 2008. I hope so anyway. So it will be exciting to see where we are heading then. I really hope as much as possible of the list above is under consideration for LW10 and will make it to the release. Oh, would that make me drool rivers and take a flight to NewTek HQ to crawl in joy before the dev team.
Still if just number 1 and 2 (Renderpass/Bucket render with Micropolygon displacement) makes it I will be thrilled with excitement.