PDA

View Full Version : Feature Poll Part Deux



hrgiger
04-25-2003, 10:14 PM
Well, since the first poll has been so popular, here is part two.

Now before anyone says anything, I know there are some who will say we don't need all this junk, we just need Lightwave to be integrated, or be more stable or the workflow needs improved. There's no argument. I think eveyone can agree that they would all like to see more streamlined code and better integration. This is purely a candy list and it also gives people who voted on the first poll another chance to vote on something else they would like to see in Lightwave 8. Not to mention, Chuck has mentioned that Newtek has been watching the first feature poll with interest so it doesn't seem pointless at all to post this.

Here's another list of features that some might be interested in.

Hiraghm
04-25-2003, 10:49 PM
Before I vote, I have to show my ignorance and ask... what is edge weighting, and what are ngons / adaptive SubDs?

hrgiger
04-25-2003, 11:00 PM
To get a look at what edge weighting can do, take a look at edge weighting in Cinema 4D:

http://www.maxon.net/pages/products/c4d/highlights_c4d/movie_modeling.html

And the ngons means it can support more then 3 or 4 point polygons that Lightwave currently supports for it's subdivision surfaces.

hrgiger
04-25-2003, 11:05 PM
I should add that edge weighting is different then the subpatch weight we currently have in Lightwave. If you look at that video file above, you see when the put the holes in the plug and sharpen the edeges, the holes stay round but tighten in shape. If you tried to do the same thing in Lightwave with weighting, you would get a hexagonal type of shape instead of a circle. Edge weighting only sharpens the edges, while Lightwave shapens the angle that the edges enter and leave the vetex, making angular, polygonal shapes.

papou
04-26-2003, 07:38 AM
why only 1 vote...pff...so fast rendering.

Flat
04-26-2003, 08:52 AM
Instancing.
While all the others things are desirable, this single feature would allow entirely new scenes to be made.

After all, even without edge weighting people can make truly amazing work.
SubSurface scattering is readily available via plugins.
Soft shadows everywhere would be nice, but there are easy workarounds.
Fast as hell rendering... but wait a minute, renders are already fast as hell ! Just take a look back 2 years before. :)

Without instancing youre screwed if you want to do a forest, an asteroid field, a crowd or whatever comes in numbers.

So there, instancing. :D


Phil, instance number 42

cavalos
04-26-2003, 11:02 AM
HR Giger: How can you say rendering is fast? Please! GI is SLOW as hell and rendering (scanline/raytrace) needs to be improved a lot.
For me instancing is a toy for playing from time to time but at last you ALWAYS have to render your picture or animation so IMHO rendering speed is extremely important.
LW has been considered for many years a top package in first place because it's renderer quality and speed, now the picture is different and there are other high end solutions, so LW is now behind them.
In the best scenario I'd love to see all of the avobe implemented (with top quality) in the next version...that's is highly improbable but to me image is all so my vote is for rendering.

Best
Christian

garv
04-26-2003, 01:04 PM
I'm going with faster rendering......maybe even an option for one pass anti_aliasing..........forgive me I don't know the technical term for this but Lightwaves AA seems to find the edges of the image and AA them....the end result is great but slow. A quicker option to AA would be appreciated...

Also ray tracing is slow....I'm rendering a project right now and anything with ray trace reflection and refraction is slooooowwwww......8 min per frame......I would love to have this sped up in 8.

Man!! that edge weighting in cinima 4D looks cool though!!! I want both
faster rendering and edge weighting.:D

CB_3D
04-26-2003, 04:47 PM
Voted for fast rendering, as this makes possible a few nice tricks with spinning lights amd MB. But its not really a feature;

Id go for full edge weighting now, if i could vote again.

hrgiger
04-26-2003, 07:01 PM
Originally posted by cavalos
HR Giger: How can you say rendering is fast? Please! GI is SLOW as hell and rendering (scanline/raytrace) needs to be improved a lot.


Uh, I didn't. This is a feature request poll.

cavalos
04-26-2003, 07:11 PM
Excuse me Giger, that was Flat not you so I stand corrected on that...Back to my original post I Hope someday we can see LW renderer engine in the very top and

As fast as Vray
As configurable as Mental Ray
As vanguardist as Brazil
As easy to use as Everyday LW rendering engine :)

Best
Christian

harhar
04-26-2003, 07:14 PM
there is no n-gon subdivision. Is not even possible to create n-gon subdivision. Only 3 point subdision, 4 point subdivision, 5 point subdivision ect.

hrgiger
04-26-2003, 07:26 PM
How's that?

I was under the impression that an NGON is anything poly with more then 4 edges or sides. So, if you have five point subdivision, you have NGONS.

Valter
04-26-2003, 07:43 PM
My vote goes to...

NGons/Adaptive Sub-D's
Soft Shadows for all Light types
Just Fast as hell Rendering Please
True Subsurface Scattering
Edge weighting (LW8 should support to EDGES than support POLYGONS and VERTICES) with all possibles edges tools atached

what's mean with Instancing of Geometry??

later

harhar
04-26-2003, 07:44 PM
Ngon, where n is a variable = all integers

no one has invented any >4 point poly subdivision yet. True N-point poly subdivsion means after subdivision, all polys remain N-sided.

To subdivide >4 point polys you simply divide the poly into triangles(any polygon can be divied into triangles) then do a loop scheme subdivision, or if possible, divide into quads and do a a catmull-clark(pixar) subdivision.

hrgiger
04-26-2003, 08:06 PM
Valter,

Instancing of geometry will let you render many thousands of an object yet it uses a minimum amount of memory. For example, Sasquatch instances the hair fibers so that you can render out millions of hair fibers at a minimum of system resources.

Valter
04-26-2003, 09:10 PM
ok thanks

Will be very cool too.

DaveW
04-28-2003, 03:23 PM
Fine, give us fake ngon subpatch support. Who cares if internally the algorithm is converting the polys to 3 or 4 point polys? Being able to convert polys with any number of sides to sub-d's like other software would be really nice. Ngon support, as cool as it is, really isn't all that important to me though, but adaptive sub-d's would seriously kick ***. But I voted for edges, I think that's more important. Faster rendering is very important too, it was a tough choice; but I can always add more cpu's to the renderfarm, buy a faster computer, or use tricks to speed up renders. Instancing can already be done with plugins, so can sss and soft shadows.

unknown limit
04-28-2003, 03:24 PM
i would say all of them but as i have to go for i went for....

ngon subpatch support.

Will

harhar
04-28-2003, 04:11 PM
you are missing the point. Fake ngon subd would be first divide all non-triangle polygons into triangle and do a loop subdivision, which is what lightwave already does. It's what the TRIPLE command is for. So fake ngon subD would simply automatically convert all polys to triangles and do the loop subD without asking you if you want to convert all polys to triangles or not.

mouselock
04-28-2003, 09:36 PM
Originally posted by harhar
you are missing the point. Fake ngon subd would be first divide all non-triangle polygons into triangle and do a loop subdivision, which is what lightwave already does.

No, I believe you're missing the point.

If I draw a pentagon and hit tab, I get an error and a pentagon, non-subdivided, back.

What folks are saying is that even if the program has to algorithmically break an N-gon into triangles/quads to do Sub-D, it should do so transparent of them having to go in and fiddle with it. I know this is fairly trivial for 5-gons, 6-gons, etc, and very nontrivial for 256-a-gons (mainly because you're almost guaranteed to not get your contours correct through auto-triangulation when you make a high vertex polygon nonplanar in multiple ways).

Nevertheless, that's a modeling issue. Most modelers know not to use huge vertex count polygons, but not having to rework a model for a couple of penta or hexagons would sure be a timesaver for a lot of folks. (Especially looking at some of the creative expansion/contraction joints folks develop to selectively increase polycounts in quad only meshes.)

As an aside, shouldn't any n-gon which tiles a plane be able to be the basis of a subdivision scheme (hence, hexagonal subdivision would be possible)?

harhar
04-28-2003, 10:14 PM
If I draw a pentagon and hit tab, I get an error and a pentagon, non-subdivided, back.

Use the damn TRIPLE command. Do it transprently would be auto triple the polygon. Are you really so lazy you can't use the triple command? Just don't select any polygon, and the triple command automatically convert all non-triangle polys to triangle.



it should do so transparent of them having to go in and fiddle with it.

Is pressing a triple command so difficult.


but not having to rework a model for a couple of penta or hexagons would sure be a timesaver for a lot of folks.

You don't have to, use the damn Triple command.


As an aside, shouldn't any n-gon which tiles a plane be able to be the basis of a subdivision scheme (hence, hexagonal subdivision would be possible)?

yes, if you can come up with an algorithm that divides them into quads.

harhar
04-28-2003, 10:36 PM
here's a video showing how to subdivide n-point polygon.

www.duperworld.com/subpatcher.rar

Anttij77
04-28-2003, 11:26 PM
Ok, i asked a friend of mine to subdivide a cylinder with 5 sides ( that's 4 quads and 2 pentagons ), just to see how Maya does it...

And here's the image...
As you can see, there are no triangles after the subdivision!

Valter
04-29-2003, 12:52 AM
well.

I would be satisfy if lw don't drop or desapear with subpatched polygons when I drawing edges on subpatched polygons (edgeloop, knife, split), but modeler don't suport subpatch polys > 4 vertices.

ali
04-29-2003, 06:24 AM
instancing geometries...because i work on architectural visualisation and wish to see that, when i change size of a column, hundreds of others change also...and rendering capabilities of course...the others are also very important but i am not interested in organic modelling and animation....

Matt
04-29-2003, 07:05 AM
Ali - ditto!

Flat
04-29-2003, 11:39 AM
Instancing can already be done with plugins, so can sss and soft shadows.

Dave, what plugin (besides Sas) does instancing exactly ?
We're not talking about instancing fibers of hair, it's about instancing LW objects, even characters.

Say you want to render a forest in LW right now, starting with a tree.
You've got 2 solutions AFAIK : load from scene ad nauseam, but then your scene will take N times more memory if N is the number of trees.
Or you can point-clone+ your single tree to death, with about the same result.
You also have to composite a few layers because neither method allows for enough geometry in a single scene.

Instancing can cure that, and much more. It can allow to render crowds of animated characters without compositing, it can add thousands of nurnies, pipes and weird stuff to a big model without going through the memory roof, it can allow for incredibly complex environments, etc.
If a plugin does it I'd love to know, but given LW's present architecture I seriously doubt it.


Phil, who agrees a "quadruple" command would be nice, though :p

DaveW
04-29-2003, 12:58 PM
flat: hd_instance will do instancing of objects. http://www.happy-digital.com/instance.asp

harhar: the problem with tripling is that it's an extra step, and if I have a few 5-sided polys I wouldn't want them tripled, I would want them split into a quad and a tri. And when I'm using tools like knife or edgetools, I don't want my subpatch object turning into a mess. And on top of that, there was a thread a while back where someone demonstrated that simply tripling (and even merging back to quads) was not the same thing as turning ngon's to sub-d's in other packages. Anttij77's post demonstrates this as well, but doesn't go into as much detail as the old thread. I can't remember which forum it's on but if I find it I'll link to it.

I'm fully aware of the triple command and how to effectively deal with 5+ sided polys, but I'd rather LW took care of that for me transparently like other apps. It would save time. That's what new features are supposed to do.

harhar
04-29-2003, 02:35 PM
Ok, i asked a friend of mine to subdivide a cylinder with 5 sides ( that's 4 quads and 2 pentagons ), just to see how Maya does it...

yes, it seems after subdivision, all polygons become quads. So maya's algorithm is that instead of divide all polygons into triangle like lightwave, divide them all into quads, then do the subdivision.


I would want them split into a quad and a tri.

I don't think any subdivision algorithm can handle a polygons made of a combination of quads and tris, it's either all triangle, or all quads.

harhar
04-29-2003, 02:53 PM
:I

DaveW
05-01-2003, 02:40 PM
So what's going on in LW when I have some quads and some tris in the same model? It doesn't appear to be converting the tris to quads since I end up with tris in the end unlike metaform, where tris turn to quads (and merging the tris into quads afterwards leaves some tris behind) and it's certainly not turning the quads into tris; if I use the triple command the object subdivides a lot differently than if I didn't triple them.

edit:
Just want to add that I really don't care what the underlying math is, LW should be able to do what other programs can and that is cleanly sub-d polygons with more than 4 sides. Tripling the polys does not do this. Tripling just the polys with more than 4 sides does not do this. There is no simple way for LW to get the same results with a 5 sided cylinder as Maya.

harhar
05-01-2003, 06:22 PM
subdivided a 5 sided cylinder in LW in under a minute, didn't seem very difficult to me. Notice all the polygons turned into quads like in maya.

http://www.duperworld.com/subd.jpg

DaveW
05-01-2003, 06:43 PM
I get the same result you got, and that's exactly the problem. It looks awful. It turned into a lumpy blob. Take a closer look at the Maya results, it looks a lot better.

harhar
05-01-2003, 08:56 PM
.

wacom
06-07-2003, 06:36 PM
I voted for more presets for a couple of reasons.

First: They are easy to make for developers.

Second: Look at half the posts on the LW site- if there were a few more preset settings for each option people wouldn't need to post as much.

Third: When you're in a pinch and need
to get something done yesterday it's always nice to fall back on some tweakable presets.

Fourth: It would help "sell" lightwave to people better- they could sit down with the demo in a day and come out seeing just how powerful the program is- instead of finding out six months later...

I guess I'm using "Presets" to define more than just textures- and going into all the other options in Layout (like several gravity presets, presets for partical effects, presets for lights, preset expressions (or lscripts) that are built in for things like a moving wheel on a car etc.)

sailor
06-24-2003, 03:02 AM
ok about the subdivision prob here its is full LW...just use metaform plus and it will do exactly the same

DaveW
06-25-2003, 04:15 PM
OK, metaform plus will work if I want to freeze the object. But the whole debate started because we wanted this kind of subdivision to work with subpatches. If I'm modeling a character and I need a 5 sided patch somewhere then metaform plus doesn't help me. Or if I'm modeling a hydraulic part that needs to have variable resolution then metaform plus doesn't help me.

sailor
06-26-2003, 01:54 AM
yes u are right...
this said the number of iterations in Metaform plus could have been tweakable in the layout at rendering time but i dunno if this is possible...remember that in Maya the Polysmooth operation does exactly this and the object is like "freezed" the only differnece is that with the history and the LOD manager u can change the number of iterations at rendering time.
AFAIK u can not do this with metaform plus the only possibility is to use the very old trick of LOD management and keep ur base object and make differnt objects with differnt iterations as object replacements....but i understand your need for this and agree with you eventhough i actually never have used subpatches for their capability of LOD management (lucky me) what i mean is that i have never encounter the case where i start far from the object and zoom in very fast until a total close up and never had the use for a multi iteration object..

in conclusion it will be nice to have N-gon subpatches support even if most of the time i end up just tripling the ngons...i dunno why there is so much buzz about the triangles in subpatches actually the most messy layout of polys is a tri fan (that u can convert in quads) or the 5 edges point problem wich happens mostly with quads (wich proves that quads are not a miracle with subpatches)

i would prefer to have construction history or hierarchical subdivs that will be kick *** IMHO :)

Bigboy
07-01-2003, 01:09 PM
Actually... I wish you'd added bezier patches in there :-)

Lots of graphics hardware now supports this, and it would be really handy yo have...

Jimzip
07-21-2003, 08:03 AM
Just model in quads. Geez, it's not that hard to select four points and press 'p', and unless your model is really screwed it shouldn't take long.
Also, it's all in the workflow. People are creating some pretty amazing stuff just with 3 and 4 sided polys, it's kind of efficient.. I'm not against implementing this kind of tool, but you can achieve the same outcome with the same amount of key strokes.
(Also, why subpatch a cylinder, why not just stretch a sphere?)

Jimzip :D

DaveW
07-23-2003, 04:27 PM
You *can't* get the same outcome with the same amount of keystrokes, that is what sparked the whole dabate. And certainly not with point-by-point modeling, talk about a slow and painful way to model. The 5 sided cylinder was simply to illustrate that tripling a 5+ sided polygon does not give the same results. I don't think anybody here has a hard time modeling a stretched sphere.

Obviously people have gotten away with using only 3 and 4 sided polys for quite some time, but there are almost always situations where support for 5+ sided polys would speed up workflow. Last time I checked, faster workflow was a good thing. If people didn't run into 5+ sided polys so much, there wouldn't be so many tutorials/discussions about how to avoid them and how to deal with them when they can't be avoided.

jeremyhardin
07-29-2003, 02:51 PM
hrgiger- by motion tracking, you mean put in movie or image sequence, LW tracks camera and/or object motion?

Is that a realistic request? I mean, if that could happen this new version, I'd vote for that. But i don't want to waste my vote on pipe dreams.

Dodgy
08-03-2003, 05:03 AM
I went for faster rendering, cause all the rest can be worked around, or you can get a plugin for (HD instance from www.happy-digital.com is amazing !) so faster rendering is the one for me :)

A Mejias
08-09-2003, 10:28 AM
Originally posted by hrgiger
To get a look at what edge weighting can do, take a look at edge weighting in Cinema 4D:

http://www.maxon.net/pages/products/c4d/highlights_c4d/movie_modeling.html

And the ngons means it can support more then 3 or 4 point polygons that Lightwave currently supports for it's subdivision surfaces.


Yes yes YES! I want that!

papou
08-09-2003, 03:36 PM
yeah, we want that :)
tried to make the same thing with LW. WeightSlider plugin helps little but u have to triple, subdivide, mergetrigonX, make rounder, add faces...hmm..I understand now, what are edges selection.
Ngon looks fantastic in C4d.

but i still need a quicker renderer.

evenflcw
08-09-2003, 04:23 PM
C4D doesn't have ngons(?)! Although it has this neat feature of automatically splitting subDs into tris or quads whenever a patch become an nGons. If we can't have real nGons in LW, I'd love an _option_ to activate a simular function. Atleast that would take care of subpatches reverting back to polygons whenever points go beyond 4, which is a real workflow spoiler.

Akelig
08-13-2003, 07:21 AM
I voted motiontracking...
perfect 3d handheld camera matchmoving.....jummi:D

ren_at_home
08-13-2003, 11:21 AM
who couldn't vote for "just fast as hell rendering"!?

how fast is hell anyway...
are you basing this on the average time it takes to burn in hell?
thinking about this I'm having second thoughts as usually
"burning in hell" is for an eternity...


but yes - faster GI, faster rendering with a reduction in global warming caused by all that burning would be most welcome.

Akelig
08-14-2003, 12:30 AM
Build yourself a hell of a renderfarm ;)

ren_at_home
08-14-2003, 04:47 AM
buy yourself a decent tripod...
;)

jorbedo
08-24-2003, 08:30 PM
Global Illumination - RADIOSITY

Come on akelig, you don't get it, With LW 5.6c and Lightscape solution, I'm having rendering times from 1 to 2 minutes at 720x480 on Athlon XP1700 with shadows, raytracing, enhanced antialias, soon to be upgraded to a P4 3.2GHz.

And you suggest a render farm, I think that you are joking, right?.


Jorbedo

silverlw
10-06-2003, 05:11 AM
Something got to be done with Lightwaves renderengine but im a user not a programmer. I love Lightwave but the renderengine is almost the same since i first started with lightwave on the Amiga, except for 128bit framebuffer and Montecarlo Rad. I could give my balls if i was able to use Brazil together with Lightwave. I hate Max gui and workflow but i have to switch to Max or Newtek got to radically improve the renderengine. Multibounce radiosity above 3 bounces stands still in Lightwave but in Brazil when i do indoor radiosity i can often use up to 8 bounces and still survive the rendertimes. i can continue babeling about renderfeatures but i hope Newtek takes this seriously.
This is no photo it's a character rendered with Brazil,subsurface scattering and Radiosity probably used.
http://www.splutterfish.com/sf/ArtistSpot/yama/mamegal_jp_face.jpg

WizCraker
10-06-2003, 05:40 AM
you are making us choose when everthing on that list should be included. I voted for Ngons, but more presets would be nice for at least MD, however with the new soft/rigid system it looks like they have already fixed that, and of course faster rendering is always needed but it is not the most important part and true instancing would be nice.

hesido
10-16-2003, 04:08 PM
Ppl. Look at this, most of you have seen this already, but lightwave deserves such an update, just to let you remember.

Hierarchical BSSRDF Evaluation (http://graphics.ucsd.edu/~henrik/papers/fast_bssrdf/)

Notice the light bleeding thru translucent material, and download the papers for more superb renders, look at the ultra short render times of the teapot example. I mean , come on! ;) This rocks. And the method is deterministic, so it doesn't flicker when animated, the method can be used also for scanline renderings. (I may not need color bleeding from translucent material but SSS is way to cool :) ) Anybody who can fake a glass of milk as good without SSS??

I guess it is too late to put it in LWave8. But Lightwave needs such an important update.

Also: Instancing is real important

Aaand: I should be able to group the objects (instanced, duplicated, or anything) so when I have 150 objects attached to a particle system, I don't have to browse thru 150 objects to select the lwave cow in the scene. I hope they bring that into LW8. Hell, I'd like to group'em just for fun!

Raytraced non-area/non-linear lights leave buggy shadows, everybody knows it.

There is tearing artefacts when you morph an object with triangles. Maybe I am just inable.

tasmanian
10-17-2003, 03:57 AM
What do you mean choosing one of these features ? I wan't them all ! Preferably now ! :D

Lude
10-18-2003, 11:18 AM
can Lightscape still be bought in a form that will work with Lightwave?

I've been trying to find it on the AutoDesk website but it just talks about using it within there products.

Has it been bought up and discontinued?

lude

erikals
01-01-2004, 11:05 AM
Originally posted by DaveW
....flat: hd_instance will do instancing of objects. http://www.happy-digital.com/instance.asp .....

...but this doesn't really support crowd generation, which I guess would kind of be taking away a major feature. (In the examples shown the crowds have the same basic movement). Or am I wrong, maybe it would support crowd generation if LW had a crowd generation feature?

pixelinfected
01-04-2004, 08:35 PM
Instancing of Geometry
very useful to build a lot of complex situation, in many genre of images and animation, from vfx to medical, documentaristic, technical, visual design and more.
i try hd istance and i cannot understand how i can live without it, fast, working with reflection, refraction and more.

True Subsurface Scattering
i bought g2, which gave me that and many more, but is not a very urgent needing, if is not fast like g2 do.

Soft Shadows for all Light types
yes, immediately, true or not, i think a soft and diffuse shadow must be implemented as soon as possible, realism can grow fast with it.

NGons/Adaptive Sub-D's.
for know a simple ngons sudiv, like c4 do, are simple, myself i wrote two years ago a beta of script which work when lw must convert poly to subdiv, before to convert check if there are >4 point poly and if exist convert in 4 and 3 point poly, with if possible, a good poly distribution.
adaptive subdiv are a bit complex, be cause they must rewrite a subdiv engine of layout, which is based on original metaform research.

More presets (surfaces, MD, Hypervoxels, etc...)
very useful and lost cost way to show more power of lw, to do simpler the life of lw user, and help to learn lw parameters.

Motion tracking
naah, there are application build only to do that, i prefer a specialized application instead of a rought implemetation in a 3d software.

Just Fast as hell Rendering Please(Especially GI)
this area must be enhanced, actual lw render, if you want to have all quality of lw render, can be very slow, especially with refraction and reflection enable. i hope optimization of render, memory and quality.
actul modern render like renderman, mental ray, and other render like vray, final render offer quality and speed with low time, and this mean that lw must speed up its render to be competitive, i know that i have all render node free, but if i need to buy 10 times cpu than other render i lose the advange of lw render...

Dodgy
01-05-2004, 04:19 AM
Originally posted by erikals
...but this doesn't really support crowd generation, which I guess would kind of be taking away a major feature. (In the examples shown the crowds have the same basic movement). Or am I wrong, maybe it would support crowd generation if LW had a crowd generation feature?

The lastest version of HD instance allows you to pick multiple objects for your instanced objects, scattering them over your surface, as well as allowing you to tint those objects randomly. If you had say a dozen objects, with one offset motion mixer motion, you could get a fairly good crowd...

Dick Nelson
01-22-2004, 09:46 PM
What the world really needs is "Lightwave for Dummies." Some of us have need for software that will produce relatively simple 3D animated graphics, but aren't doing the next sequel of "Star Wars." I don't have a clue what "specularity" and "radiosity" and "volumetrics" are. For instance, I just want to EASILY and QUICKLY take a 2D image (gif, jpg, etc.), such as the one at left, and turn it into 3D and animate it. Or produce a 30-second animated news open. I've done the latter, but it took way more time than I could really afford to devote to it. I've consulted a couple of tutorials on making 3D logos, but they were unintelligble to me. I am generally able to figure out how to use software, even without consulting manuals (which are generally useless anyway), but Lightwave is way beyond me. There must be others who could really use "Lightwave Lite."

marlo.steed
01-25-2004, 12:29 AM
Right, Lightwave is not that user friendly. Many of the neatest features are buried in dialogue boxes that are nonintuitive. It needs to take a look at Strata Studio for a lesson on ease of use. I used Strata for quite a while but found it too limiting but it was much more intuitive than Lightwave.

I vote for more intuitive interfaces so that anyone can figure out the powerful features.

3Demon
02-18-2004, 11:52 AM
Hey Guys,

Hijacking, cross-posted spam has been removed by Moderator.

hesido
02-18-2004, 12:09 PM
3Demon, thanks for the tip, but just what on earth is this related to a lightwave feature request?

BTW- We need SSS, realbad.

the moderator was faster than me

colkai
02-19-2004, 03:36 AM
I've said it afore, and I'l say it again.

Edge weighting - we need edge weighting and we need it now (;)). When it comes to building Sub-D objects, like cars etc.. this would cut the amount of geometry down dramatically.
Not to mention removing the need for a whole lot of knifing and bandsawing so consequently would also be a huge workflow improvement. Allowing modellers to really speed up their work.

Please..please..please let this be in a 8.X release!

Lus Santos
04-12-2004, 12:32 AM
Subsurface scattering please...

Based on my experience:

Lightwave surface thickness:

I think we all like this name, but the results...hmm...

Ska plugin:

Slow, extremely buggy and unpredictable.

G2 plugin:

http://www.3ddmd.com/images/G2/SSS_default.jpg

I think this image tells all. I guess we dont want to fake it or making tricks to achieve a sss look, we want the real thing!

----------------------------------------------------------------------------------

I want this! ;)

http://www.lowpoly.com/sculpture/marble-ear-sss.jpg

thekho
04-12-2004, 02:18 PM
I'd love to have all but i voted for NGons cos they is important for my character models.



Originally posted by sailor
ok about the subdivision prob here its is full LW...just use metaform plus and it will do exactly the same

Where did you get metaform plus? I can't find it on my Lightwave 7.5.

thekho
04-12-2004, 05:06 PM
Oh...I found metaform plus from Legacy Plugins....it's cool!