Stuff from LW modeller you'd like to see in Blender?

Hi folks, following the discussions about Lightwave, I thought about starting a thread here on what tools / features from Lightwave modeller we would like which we think are missing from Blenders toolset.

I want to state up front, this is not a question of “which is best”, nor do I want the thread to spiral into a “make blender a LW / MAX / XSI clone”. Rather, speaking as an ex-Lightwaver, there are certain features I still miss from LW which I think would be advantageous for workflow & modelling if they existed within Blender.

I know there are quite a few “ex-wavers” out there so it would be good to get the input on how Blenders toolset could be further improved.

For myself, though I acknowledge Miratools has gone some way to providing interactive bend/taper and the existence of the simple deform addon, within LW, the taper/bend/twist tool had an iteractive “widget” which allowed direct manipulation to determine both where the falloff occurred and provide control over the type of curve as shown in the screen shot. The sliders under the “preset” button control the shape of the falloff curve, depending on which viewport you used depended on the resulting axis being affected. Here, I think a hybrid would be better as Blender has a great x/y/z locking system or shift+x/y/z for excluding an axis. To be able to set the falloff then use the axis constraint I think would be superior to clicking in multiple views.

I also think some of the functionality of LWCAD would be really useful in Blender. I know there are CAD style addons already out there, but how do people think they stack up against LWCAD functionality, which workflow do you think better suits Blender, or indeed would a hybrid of the workflows be better?

For arch viz for example, whilst we can construct a floor plan using edge extrudes, the way LWCAD handled building wall layouts was very efficient. The “profiler / engraver” tool was excellent in being able to create clean angled coving and skirtings, it was a tool I used a great deal, fast and intuitive.

Do you agree with the above observations, do you have any thoughts on where you feel there are gaps in the workflow / toolset when it comes to modelling?

I want this to be a constructive thread, where developers can look and see if there is any consensus over features and then they can decide if these are do-able and whether they would consider adding them to the software.
I am not shooting for the moon here, just looking for a few tools which may make modelling faster / smarter.

Attachments


I really need these tools:

  1. Fast Viewport (more fps for heavy objects, faster selection\deselection, entering in edit mode for heavy objects) - (will be in 2.8 I gess)
  2. layers - (will be in 2.8 I gess)
  3. Edit object normals (YAVNE comes to the rescue)
  4. UV tools
  5. Good UV Packer
  6. AA for Cycles backing

This a quote from the soon to be released Lightwave 2018.

Cel Shader and Enhanced Edge Rendering: Offers flexible non photoreal render control over material shading and allows gradient-based cel shading, while Edge Rendering uses any material available in the Surface Editor to shade any line.

What is important to note, is that all of this is on screen, ‘realtime updates’ not a Post Process. Blender 2.8 is slated to drop the Internal Render, without replacement, which currently is the mainstay of NPR work for Blender, and no plans to replace the Freestyle line rendering. Both of these conditions make Blender feel like it is losing ground in the NPR world, of anime, animation, illustration, arch-viz.
I mean, take a look at the video link below, which was done in Lightwave 2018 Beta, and imagine being able to tweak calligraphic attributes of your lines, on screen, real-time, and then work with hand-drawn textures, updating with each change on the shader.

May I happily inform you that LW devs went to Japan to collect wishes for the new Cel interactive shader amongst others. And for you the cherry on the top is Batman-Ninja, movie soon tbr’ed by Warner Bros all done in LW2018 beta by Kamakeze Douga. It ‘s Kurosawa runnin’ with with DC. Be amazed, I was, all my Xmas luck to Douga on release! https://www.youtube.com/watch?v=CwPFxcefpdU

I know that work is going into Grease Pencil for the ‘Hero’ project, but all of those lines are drawn, not rendered from models, useful in select cases, but not the same as NPR rendering solutions.
This is the one area that I find Lightwave towers above Blender 2.8 development, wish it was otherwise. I would gladly welcome to be wrong, but since I use both tools, I am telling it as I know, but I am well aware of my limitations.

Thanks for starting the thread. As an ex-waver and current Blender developer, I’m watching with interest.

1 Like

I do like the falloff modes in Modeler as well. I also like Muiltishift. Anyone can make that into a Modifier would be perfect.

Some better procedural textures would also be welcome.

Regarding other stuff, I think it would be fair to wait until after 2.8. I think LightWave performance as of 2018 is going to smoke Blender though. Just from some tests I did with similar polycounts to Lino’s tests.

Not LW’s Modeler related but I would love to see Layout’s camera mouse controls in Blender! :eyebrowlift:

While I like Falloff modes in Lightwave, they are not the target for my tastes. Modo has a step ahead, for me, and the Falloff choices in Modo for modeling are excellent, a little more controllable and ‘live’ than Blender or Lightwave.
On the second most important area of need, UV tool development, and for that I do most of my UV work in Modo. Lotsa options, great control, I just get UV work done and get back to something more interesting and creative faster.

But this is a LightWave to Blender thread not a Modo to Blender thread. Right?

But I agree for that it is worth. Modo is way ahead in most areas. Just to be honest.

The built-in Archipack does a pretty good job - walls are as easy as LWCAD to set up (if not easier, if I recall correctly), and they also remain live and editable. Roofs and other objects have quite extensive parameters.

What I do miss from Modeler are the layers in mesh editing mode. I wish we could have independent mesh layers implemented - outside the scene layers, and organize our work in layers specific to a particular mesh. Similar to Modeler.
An add-on ($3) exists, but it is pretty limited in scope. https://blendermarket.com/products/accent-pro-edit-mode-layers

Unlimited scene layers. And labelled scene layers. A proper layer manager panel with options to quickly set view properties for each layer.

Custom wireframe colours per object. Long requested, but not possible.

Another thing that is quite essential is the ability to customize our own tool sets and tool bars in Blender. In Lightwave it is very easy to create you own customized tabs and groups of tools in the sidebar, but in Blender this is not possible yet. I am hoping this option will be added soon in the 2.8x development cycle. We REALLY need to be able to create simple task-oriented workspaces, and remove all the fluff in the toolshelf.

Lastly, more procedural texture generators - like the ones in Lightwave and Cinema4d. I miss those.

Regarding other stuff, I think it would be fair to wait until after 2.8. I think LightWave performance as of 2018 is going to smoke Blender though. Just from some tests I did with similar polycounts to Lino’s tests.

I did some testing with quite heavy objects in Blender 2.8, and 2.8 utterly destroys 2.79 in regards to viewport performance. Many, many times faster. In one particular case v2.79 lagged with one complex object consisting of many layers of objects and rather heavy meshes, and in 2.8 alpha I was able to place 8 or 9 of these and the viewport remained smooth.

It will indeed be interesting to see how both apps’ viewport performances will be improved over the coming year. Based on my testing in an old 2.8 alpha I am looking forward to Blender 2.8.

Finally, I’d like to have an option for custom profiles for the bevel command and the bevel modifier (and other commands). Similar to commands like the taper fallof in Modeler, but integrated better. I guess custom falloffs? The proportional editing falloffs in Blender are okay, but I’d love a custom option added with a custom spline/curve.

Love the idea! Can anyone make these small demonstrative video gifs or time link to youtube video? Walls of text are hard to interpret. I think people use https://getsharex.com/

It’s not a question of where LW will kill Blender, or come to that where Blender kills LW, I am not trying in any way to compare packages, my only aim here, as an ex-waver, is to highlight modelling features we think would be useful in Blender that LW had. There are plenty of threads about comparing software and all of the pitfalls such threads bring with them, here I want to be constructive, simply to benefit the Blender modelling toolset.

I am hoping to get some video up of some examples, but to be honest, at the moment, I have some “real world sutff” I have to put time into, but after the weekend, I should have time.

Understandable, thank you! I am also wondering that since Modo is derivative of LW, perhaps some of the functionality sought has been implemented in better ways there? (E.g falloffs) If so, maybe knowledgeable users should highlight/showcase that as well.

Personally, I’m all for that, as you say, Modo is a LW derivative and if the falloff tools are even more flexible than the LW ones I think it would be good information.

I wasn’t making a comparison thing. I was simply stating two facts. 1 is that some of the things I want from LW will come already in 2.8. 2 When LW 2018 gets here one of the things I would want is the geo performance -assuming it will be as good as tests I have made - Because this is one of the biggest complaints working in Blender on large scenes. And so it would be a feature I would be asking for. That’s all.

ok, a little hesitant to say this, but back in my LW days (long time ago for me now!) I loved the LW lensflares :slight_smile:

Well flares do add realism. Of course they get a bad name for being over-used.

Here is something you might find interesting:

The one thing that makes LW really fast is the ability to invoke a command like bevel over and over again just by right clicking and continuing from where you left off. With Blender’s control scheme, I don’t think that this is possible. Holding ctrl or the MMB to constrain movement along an axis. Right clicking on a primitive during creation/before confirmation of that primitive creates a new one of the same dimensions. I’m sure there are a few more subtleties.

The one thing I don’t like about Blender is the imperial units where it usually measures in inches and then skips to yards, nobody in the States measures in yards commonly, I didn’t even do it when studying civil engineering. But I usually stick with the metric system anyway.

And the statistics panel helps a lot.

As far as the camera navigation above, you can lock camera to view, pretty close in this regard.

I think it is the navigation style difference that is wanted here. The LightWave camera navigation is different than manipulating the viewpoint. When you lock the camera in Blender it follows normal viewport navigation rather than giving you a different set of navigation controls you get with LightWave. When you get used to it, the LightWave camera does make more sense. But I don’t even do that anymore. I prefer camera rigs. That way my shots have that intangible feel of moving through space on a crane in an arch where that applies.

Blender wins here in that regard with included camera rigs. Just for what it is worth. But I can see the need for the LightWave style navigation.

  • subgrid snapping + the abillity to quickly change the grid size without zooming
  • multishift
  • widgets (for example bend twist / interactive primitve creation)