PDA

View Full Version : Discussion Thread for the New Blog Post: Part 2 on the new Unified Mesh System



Pages : [1] 2

robpowers3d
12-01-2015, 03:52 PM
Discuss the LightWave Blog Post: Part 2 on the new Unified Mesh System Here.

hdace
12-01-2015, 03:53 PM
Doesn't have anything much new but does clarify several things. He says various tools to use the new architecture will become available over time. I'm wondering how good the SDK will be so that maybe third parties will develop new tools a little faster.

wyattharris
12-01-2015, 03:53 PM
Well yay and drat.
Sounds really good overall but soooooo few details. Give us more! :D


...the unified mesh system which we have implemented in LightWave Layout is an improved version of the modern mesh system which we developed for ChronoSculpt.
Alright alright, now we're talking. ChronoSculpt is a sweet experience. I'm hopeful now to see more sculptural based tools developed along with the modeler staples.


The unified mesh system does make our development job easier and will allow us to implement workflows involving the mesh system much more easily than we were able to do in the past.
I figured that must be the case. Having to wrestle Historical LW into a headlock so you can try and force it to do something it was never designed for must be a shade more difficult than a modern LW with none of those restrictions. And instead of reading about a new industry standard like Pixar OpenSubdiv and thinking, "How the heck are we going to implement that?" you simply do it in the manner that Pixar intended.

I know as a user I have always thought that way. "Oh wow, OpenSubdiv looks great. Too bad LW will never have it." It would be nice to fix my broken attitude in that regard. I know I'm getting waaayyyy ahead of the curve but I'm excited!

Question Rob: In UGE's current state, what kind of performance are you getting tumbling, deforming etc. on say a million poly mesh? (fps)

hrgiger
12-01-2015, 04:03 PM
Doesn't have anything much new but does clarify several things. .

Did you miss the part about having a deformation stack? That means we can now choose how deformations are ordered in Layout. And that's pretty huge.

wyattharris
12-01-2015, 04:08 PM
Did you miss the part about having a deformation stack? That means we can now choose how deformations are ordered in Layout. And that's pretty huge.
Yeah, this is a huge amount of information dropped in a few short paragraphs. And there's enough there to have us speculating for weeks. :D
For example, if there is a true deformation stack then there must also be a...
Nah, I'll leave that to others.

Chris S. (Fez)
12-01-2015, 04:12 PM
Did you miss the part about having a deformation stack? That means we can now choose how deformations are ordered in Layout. And that's pretty huge.

Awesome: "a true deformation stack that can be reordered, improvements to morph and deformation speed, and much more."


I am very curious about the implementation. I hope we can precisely stack subdivisions and displacements. ie. base + displacement + 1 subdivision + displacement + 2 subdivisions + displacement + 1 subdivision. A nodal editor for stacking deformers and subdivisions might be nice.

mummyman
12-01-2015, 04:16 PM
Yeah, this is a huge amount of information dropped in a few short paragraphs. And there's enough there to have us speculating for weeks. :D
For example, if there is a true deformation stack then there must also be a...
Nah, I'll leave that to others.

lol. Nice. And finally!!! Glad to hear it's based on Chronosculpt... which is pretty sweet when I tested it out last week. Fast as all hell. Good things... good things.

mummyman
12-01-2015, 04:19 PM
Great read, Rob! Thanks! Glad to hear about similarity with Chronosculpt. I got to test that out last week. And was amazed throwing a super dense mesh at it. Very exciting! Can't wait to hear/see more. Thanks. Now, just post another! :)

hrgiger
12-01-2015, 04:19 PM
Rob, this post makes me quite happy. I've said it before and I'll say it again, architectural improvements are where its at in my book. Great work to you and the team on laying what sounds like a great foundation for future development.

Was hoping to hear more about just how improved deformation and/or scene handling is with the new mesh system. Like just how improved are we talking? Hopefully we'll see some demonstrations or comparisons with current LW.

Kuzey
12-01-2015, 04:41 PM
So..these new modeler improvements shown in the previous blog post, are they using the "unified mesh system" or are they hacks for now?

robpowers3d
12-01-2015, 04:41 PM
Rob, this post makes me quite happy. I've said it before and I'll say it again, architectural improvements are where its at in my book. Great work to you and the team on laying what sounds like a great foundation for future development.

Was hoping to hear more about just how improved deformation and/or scene handling is with the new mesh system. Like just how improved are we talking? Hopefully we'll see some demonstrations or comparisons with current LW.

Great to hear that you liked the post HR! Yes we plan to share more stats on improvements in the future and further details on deformation workflow updates as well.

cresshead
12-01-2015, 04:42 PM
would love to see some video on this in the next blog.

nothing get's the point across better than a video 'showing you'.

we've seen video on the renderer, on hair on volumetrics....modeller layout camera view
mesh system please!

If your writing about it...there must be something to see right!
do it...open the box!

erikals
12-01-2015, 04:43 PM
makes me curious to if importing Fluids objects from for example RealFlow an Houdini will fly...

http://erikalstad.com/backup/misc.php_files/question.gif

Verlon
12-01-2015, 04:44 PM
Architectural enhancements are great. They may not be as sexy as new features (non-value added is what I hear in manufacturing), but they are certainly important. I am glad this is coming as it opens up a lot of doors not just for LW3DG, but hopefully plugin developers as well. We are all looking forward to using ChroNevroModeLayout 2016 in early January. :devil:

You said we would not see a full modeling toolset in Layout, but are there tools you are looking to implement there in the next iteration? While I am not averse to a somewhat slow transition, I don't want to find myself having to model in both applications because of specific tools that only exist in one of the other.

hrgiger
12-01-2015, 04:49 PM
So..these new modeler improvements shown in the previous blog post, are they using the "unified mesh system" or are they hacks for now?

Well they're in modeler so no, they're not using new mesh system.

cresshead
12-01-2015, 04:52 PM
so is this

so....is this layout viewport showing sub d mesh...
not triangulated like layout2015??

https://blog.lightwave3d.com/wp-content/uploads/2015/11/Heros_Mesh-1024x549.png

Kuzey
12-01-2015, 04:54 PM
Also, would this unified mesh system fix modelers internal point order or that objects need to be in the world center for tools to work etc

hrgiger
12-01-2015, 04:56 PM
so is this

so....is this layout viewport showing sub d mesh...
not triangulated like layout2015??



Good Question Cress. So how about it Rob? Of course it could just be a render using new edge rendering from 2015.

- - - Updated - - -


Also, would this unified mesh system fix modelers internal point order or that objects need to be in the world center for tools to work etc

Kuzey, the unified mesh engine has nothing to do with modeler.

Kaptive
12-01-2015, 04:57 PM
One thing I take from this (and all of Robs posts) is that we should keep expectations for the next release realistic. Though it IS going to be a significant release to be sure (and I mean, if the new renderer is close to Arnold, but with unlimited nodes... for the upgrade price, that alone is actually pretty huge imho), but the real magic is going to be from here on in.

Or to put it another way:- Lightwave is very much back in the race, it's had a pit stop, changed its tires to slicks and is about to do a whole lot of catching up. We should be excited, but we still have much to do! But all the signs are very good indeed!

(and yes, I am indeed totally aware that the image below is totally nerdy, fanboyish and terribly uncalled for, but I just couldn't help myself at this time of night :P Please forgive me )

131308

Cheers for the blog post Rob! :)

Kuzey
12-01-2015, 04:58 PM
Well they're in modeler so no, they're not using new mesh system.

So anything added to modeler now will have to be remade later?

Kuzey
12-01-2015, 05:01 PM
Good Question Cress. So how about it Rob?

- - - Updated - - -



Kuzey, the unified mesh engine has nothing to do with modeler.


Well it's 2am here & I found that post hard to read to be honest.

Lightwave history...blah blah unified mesh system...blah blah LW future..if you know what I mean :)


EDIT:-

OK I re read it & noticed Layout was mentioned a million times... :P

kopperdrake
12-01-2015, 05:06 PM
Excellent - made me laugh :D

Ztreem
12-01-2015, 05:18 PM
Great post Rob! This is what I have been waiting for a long time, finally we will see some work done on the foundation. A modern and solid foundation is exactly what we need. Looking forward to LW2017. :D

Satera
12-01-2015, 05:41 PM
I think it all sounds fantastic. With the considerable effort thats going into the unified system, it sounds like a lot of work on the back end. My question would be, what does that leave the users to expect as in advancements for 2016 (or if it comes later?) I'd be a bit worried that this is the long awaited update lightwave needs, but not the update which increases productivity. Almost like you have to go take a step back to take two steps forward, i understand why it has to be done, but worried feature-wise will 2016 feel a little light?

Either way, i am very excited to what I have heard and seen,and like the guys above very curious to see more! :)

cresshead
12-01-2015, 05:44 PM
so the image on the left

is that hidden line viewport like in 3dsmax/xsi etc?
see below

https://scontent-lhr3-1.xx.fbcdn.net/hphotos-xpa1/t31.0-8/12314165_10153833928687871_5030419713001915403_o.j pg



https://blog.lightwave3d.com/wp-content/uploads/2015/11/Heros_Mesh-1024x549.png

robpowers3d
12-01-2015, 05:55 PM
One thing I take from this (and all of Robs posts) is that we should keep expectations for the next release realistic. Though it IS going to be a significant release to be sure (and I mean, if the new renderer is close to Arnold, but with unlimited nodes... for the upgrade price, that alone is actually pretty huge imho), but the real magic is going to be from here on in.

Or to put it another way:- Lightwave is very much back in the race, it's had a pit stop, changed its tires to slicks and is about to do a whole lot of catching up. We should be excited, but we still have much to do! But all the signs are very good indeed!

(and yes, I am indeed totally aware that the image below is totally nerdy, fanboyish and terribly uncalled for, but I just couldn't help myself at this time of night :P Please forgive me )

131308

Cheers for the blog post Rob! :)

Well Said! :thumbsup:

wesleycorgi
12-01-2015, 05:56 PM
I'm wondering how good the SDK will be so that maybe third parties will develop new tools a little faster.

Yes, that would be cool if the 3rd Powers were able to leverage the new engine. I love LWBrush, but it creaks on higher density meshes. If it could deform polygons like Chronosculpt does, then woohoo!

hdace
12-01-2015, 05:56 PM
Did you miss the part about having a deformation stack? That means we can now choose how deformations are ordered in Layout. And that's pretty huge.

Saying, "nothing much new," does leave room for ONE thing. We already knew about the UGE being based on ChronoSculpt. Was there anything else?

However, you're right that I didn't get it. Now that you've pointed it out I realize that this is what I've been waiting for. Thanks so much for letting me see it! If I'm not mistaken, this could mean we're finally going to be able to produce proper corrective morphs. That is the one huge thing I've had my fingers crossed for for a couple of years now.

robpowers3d
12-01-2015, 05:59 PM
So anything added to modeler now will have to be remade later?

Actually our approach on any new modeling tool development currently is to develop the tools to be as compatible with both the modeler and Layout systems as much as possible to avoid what you mention. We have thought about this quite a bit.

robpowers3d
12-01-2015, 06:03 PM
I think it all sounds fantastic. With the considerable effort thats going into the unified system, it sounds like a lot of work on the back end. My question would be, what does that leave the users to expect as in advancements for 2016 (or if it comes later?) I'd be a bit worried that this is the long awaited update lightwave needs, but not the update which increases productivity. Almost like you have to go take a step back to take two steps forward, i understand why it has to be done, but worried feature-wise will 2016 feel a little light?

Either way, i am very excited to what I have heard and seen,and like the guys above very curious to see more! :)

Certainly it is not two steps back. Modernizing and moving forward on all fronts is our goal but architectural work is not feature work. That is true... but modernizing the architecture will vastly improve workflows and feature development for the future...and have added benefits right now like the mesh deformation stack workflow change mentioned.

Kuzey
12-01-2015, 06:04 PM
Actually our approach on any new modeling tool development currently is to develop the tools to be as compatible with both the modeler and Layout systems as much as possible to avoid what you mention. We have thought about this quite a bit.

Super cool :)

Kuzey
12-01-2015, 06:09 PM
Certainly it is not two steps back. Modernizing and moving forward on all fronts is our goal but architectural work is not feature work. That is true... but modernizing the architecture will vastly improve workflows and feature development for the future...and have added benefits right now like the mesh deformation stack workflow change mentioned.

Rob,

About this deformation stack workflow, is this only for Layout modeler tools at the moment..or can you edit your model in modeler & have the edits appear in the stack in layout?

Satera
12-01-2015, 06:15 PM
Certainly it is not two steps back. Modernizing and moving forward on all fronts is our goal but architectural work is not feature work. That is true... but modernizing the architecture will vastly improve workflows and feature development for the future...and have added benefits right now like the mesh deformation stack workflow change mentioned.

Thank you for the prompt reply. I'd like to clarify that I am a massive fan of this product so I appreciate the ability to talk with the developers, and be heard. This stack sounds awesome though and productive. Will this be something on the lines of Max's modifier stack?

tyrot
12-01-2015, 06:31 PM
nobody mentions him so i ll do it ... thanks david ikeda.. they started to do what you always wanted to do . i wished he - chronomaster - was also on board.

cresshead
12-01-2015, 06:55 PM
Thank you for the prompt reply. I'd like to clarify that I am a massive fan of this product so I appreciate the ability to talk with the developers, and be heard. This stack sounds awesome though and productive. Will this be something on the lines of Max's modifier stack?

probably more like the stack in Blender for now i think.

jasonwestmas
12-01-2015, 07:32 PM
Fantastic news, thank you Rob. I really do appreciate a lot about Lightwave, just needed some large replacements like this one. :)

lightscape
12-01-2015, 07:51 PM
Nice post Rob.

People who are looking to buy Lightwave will need to see proof that right now there's
"full awareness of vertices, polygons, and edges. Layout now having the ability to actually create geometry"

Not the future...now. It doesn't need to have a sexy UI or workflow. It just needd to show that ability or else it doesn't really draw people in.

hrgiger
12-01-2015, 08:06 PM
Not the future...now. It doesn't need to have a sexy UI or workflow. It just needd to show that ability or else it doesn't really draw people in.

Of course. But its not available for purchase yet so right now its just development information.

jwiede
12-01-2015, 09:02 PM
and have added benefits right now like the mesh deformation stack workflow change mentioned.

Rob, if you're going to mention this so much, can you give us some notion of what kind of UI or workflow is involved for reordering, so we can better understand actual user benefits/limitations? Will it in LW2016 offer the level of control Chris S. mentioned? If not, what level of control will be available in LW2016 in terms of deformation order of execution? Any rough idea by what factor deformation speed has been improved?

Also, can you please clarify whether the ability to create geometry or other UGE benefits will be exposed via Layout SDK / Python / LScript in LW2016?

Thanks!

djwaterman
12-01-2015, 09:06 PM
I think LW is going to be glanced at more and more by other users over the coming years, the solid base they are implementing will sort of future proof it I hope. I wonder if some of the new tools, like the lattice will be available in Layout for the next release, it's not creating geometry, only pushing it about, so it might be the case. The message I'm getting from the blog is that this next version will primarily be about the UGE and renderer. Once this is delivered they will be free to spend all their development time on shipping modeler functionality into Layout, that will be the phase 2, but presently I think we might see some deformation tools in the phase 1 release. What keeps me interested is it is obvious they are holding back on really showing us anything dazzling, everything has been very low key, not a single flashy render to show off the realism of the new renderer for example, just small self contained test samples. Imagine what we'll be seeing when they decide to launch this thing. Lightwave definitely has a future.

jwiede
12-01-2015, 09:24 PM
Of course. But its not available for purchase yet so right now its just development information.

Well, that all depends on how far off LW2016 is from release. If LW2016's coming in 2H16, sure. If LW2016 will show up by end of 1Q16, the content is much too fluffy.

jwiede
12-01-2015, 09:25 PM
<forum coma duplicate post removed>

m.d.
12-01-2015, 09:46 PM
great news Rob....structural changes much more important then flashy tools.....well they lead to better flashy tools at least :)

robpowers3d
12-01-2015, 09:49 PM
nobody mentions him so i ll do it ... thanks david ikeda.. they started to do what you always wanted to do . i wished he - chronomaster - was also on board.

I'm happy to mention David. The initial mesh system work that he did was very good and David is a talented individual. We wish him well. :)

robpowers3d
12-01-2015, 09:53 PM
Rob, if you're going to mention this so much, can you give us some notion of what kind of UI or workflow is involved for reordering, so we can better understand actual user benefits/limitations? Will it in LW2016 offer the level of control Chris S. mentioned? If not, what level of control will be available in LW2016 in terms of deformation order of execution? Any rough idea by what factor deformation speed has been improved?

Also, can you please clarify whether the ability to create geometry or other UGE benefits will be exposed via Layout SDK / Python / LScript in LW2016?

Thanks!

Yes we plan to show more detail and more video demos in future posts. I know that everyone wants all of the information instantly but it is a process that will happen as we progress through the blog posts.

robpowers3d
12-01-2015, 09:56 PM
Of course. But its not available for purchase yet so right now its just development information.

Well, it is slightly different from "development information" in that generally Lino and I are not mentioning architectural details that are not actually already implemented and in testing. So in that sense they are very real...but yes the product is not released yet so I understand your perspective.

sami
12-01-2015, 11:26 PM
Cool! The deformation stack seems like this means full UNDO and historical, nondestructive edits?

DrStrik9
12-01-2015, 11:46 PM
Sorry, I'm stupid tired. Several posts have mentioned "UGE." -- I did page searches in all the blog posts, but it didn't come up. What is UGE?

Dexter2999
12-01-2015, 11:51 PM
Sorry, I'm stupid tired. Several posts have mentioned "UGE." -- I did page searches in all the blog posts, but it didn't come up. What is UGE?

I believe it is what they are calling the "Unified Geometry Engine"

Ztreem
12-02-2015, 12:02 AM
Cool! The deformation stack seems like this means full UNDO and historical, nondestructive edits?

No, not really. I think its more like the old twist and bend tools in layout. Depending on in which order you place them the effect will be different. Like, first bend then twist or first twist then bend...

DrStrik9
12-02-2015, 12:10 AM
I believe it is what they are calling the "Unified Geometry Engine"

Ahh, OK thanks. LOL "Unified Mesh System" = UGE. Tricky. :D

DrStrik9
12-02-2015, 12:16 AM
No, not really. I think its more like the old twist and bend tools in layout. Depending on in which order you place them the effect will be different. Like, first bend then twist or first twist then bend...

I hope you're wrong about that. Can something that requires a specific procedure be accurately called a "stack"? If "stack" applies that way, then Modeler has used "stacks" for decades. The word Stack in a forward-thinking context has greater connotations than a mere "list", such as possibly being non-destructive, etc.

At least I hope so ...

Kaptive
12-02-2015, 01:04 AM
One thing occurs... once the devs have largely completed the foundational changes (which I assume is close to being done beyond testing and tweaking), then they'll get to spend their time on updating, new tools, pulling modeler tools into layout etc. Getting over the hump of the foundational changes will (I imagine) be such a huge relief for them all, and I expect they'll be chewing at the bit to move onto new and fun things! (After a good rest of course :) )

omichon
12-02-2015, 01:30 AM
Maybe I misunderstood the first two paragraphs of Rob's post but please, LW3DG, learn from the past. If you don't want to live the same crisis again, be aware that these new foundations, as good as they could be, won't last the next 20 years as previously. Since I am not sure to be retired at that time, that worries me a bit :D

lightscape
12-02-2015, 01:43 AM
Of course. But its not available for purchase yet so right now its just development information.

It doesn't need to be avail for purchase now. It just needs to show that it is already doable as stated. Not 3-5 years from now.
I hope they look at how Softimage was handled poorly. Goind dark. Not showing actual stuff going on.
In the end people lost interest and moved on and xsi couldnt recover.

Peoplr have even more choicees now. If NT wants people to invest time and money now they have to show actual working prototype not concepts.

Ztreem
12-02-2015, 01:53 AM
I hope you're wrong about that. Can something that requires a specific procedure be accurately called a "stack"? If "stack" applies that way, then Modeler has used "stacks" for decades. The word Stack in a forward-thinking context has greater connotations than a mere "list", such as possibly being non-destructive, etc.

At least I hope so ...

In modeler you can not move the bend or twist Tools in different positions for different effects as its applied as you drop the tool. In layout they are applied as modifiers and can be rearanged for different effects, more like a stack. This was more like an example, that it does not have anything to do with the undo system. Hopefully the new deformation stack is more powerful and incoperates the possibility to stack subdivisions, bone deformation, displacements etc.

Dexter2999
12-02-2015, 01:57 AM
I hope they look at how Softimage was handled poorly. Goind dark. Not showing actual stuff going on.
In the end people lost interest and moved on and xsi couldnt recover.


I don't think that had much of anything to do with XSI's demise. Truth is, Autodesk bought it off of AVID for a few million because it was dirt cheap. They were probably intrigued with getting the ICE code and FACEROBOT stuff ported over to MAYA in one way or another. They already owned two 3D packages how much do you think they really cared about a third? Especially when you compare the numbers of the userbase MAYA and 3DMax had much larger userbases/profit margins as far as ROI. (Although per seat XSI probably made more profit compared to 3DMAX, 3DMAX outstripped XSI in volume of sales.)

And to omichon, what piece of software do you use that you think will be using the same base code in 20 years? Notepad?

hrgiger
12-02-2015, 02:40 AM
Well, it is slightly different from "development information" in that generally Lino and I are not mentioning architectural details that are not actually already implemented and in testing. So in that sense they are very real...but yes the product is not released yet so I understand your perspective.

Yeah Rob I was merely implying that I've heard people say that they can't buy software that can't even show the details that are mentioned. I was just saying that its not a case like CORE where you are asking people to buy software they havent' seen yet.

hrgiger
12-02-2015, 02:56 AM
It doesn't need to be avail for purchase now. It just needs to show that it is already doable as stated. Not 3-5 years from now.
I hope they look at how Softimage was handled poorly. Goind dark. Not showing actual stuff going on.
In the end people lost interest and moved on and xsi couldnt recover.



This is nothing like what XSI went through. XSI went through a total software rewrite where LightWave's restructuring is being done in phases. While they have been working on this new version, they have released LW 11 and LW 2015 not to mention two free feature upgrades. Not exactly 'going dark'. Its only been a year since the last version of LW came out.

And XSI was quite popular and had a much larger userbase then either Modo or Lightwave and there are a lot of people who will continue to use it until it just won't work anymore because of its flexibility and extensibility. It could have been continued but Autodesk sucks

Kinetic Shapes
12-02-2015, 02:58 AM
I am very pleased the developers are giving us info via the blog every week or two. This is so much better than before. Whilst I have not taken up any of the upgrade offers as I play wait and see, the new direction impresses with the modernisation of LW shown so far compared to where it was before. I guess animatable parametric extrusions of geometry and text won't be in 2016, but good to know that it is now a possible feature in later revisions with geometry now possible in Layout.

lightscape
12-02-2015, 03:03 AM
I don't think that had much of anything to do with XSI's demise. Truth is, Autodesk bought it off of AVID for a few million because it was dirt cheap. They were probably intrigued with getting the ICE code and FACEROBOT stuff ported over to MAYA in one way or another. They already owned two 3D packages how much do you think they really cared about a third? Especially when you compare the numbers of the userbase MAYA and 3DMax had much larger userbases/profit margins as far as ROI. (Although per seat XSI probably made more profit compared to 3DMAX, 3DMAX outstripped XSI in volume of sales.)



XSI's deathmarch happened way before AD bought them. It was during the rewrite that many users shifted to what's already functional. They went dark expecting people to stick around. :D Very bad business decision.

Right now there are even more choices for users. Blender, Modo weren't around back then. Why should users wait for a company to deliver more than concepts in a blog? They need to see its actually there and not just concepts, ideologies.

NT doesn't need t convince me. I see myself using lightwave atleast 5 more years. But they need to convince non lw users if they want them to buy into lightwave. So far the discussion in other forums is lukewarm.

raw-m
12-02-2015, 03:09 AM
Loving all this news coming out! Not used to Deformer Stacks but sounds like I could easily get used to them. Can someone point me in the direction of how these work in other apps, for now?

dsol
12-02-2015, 04:24 AM
Thanks for the update Rob, really can't wait to use this new version (and already bought Substance Live in the black friday sales so I'm ready for the sweet PBR goodness in the next release!)
A big question I do have is: Are these architectural changes being made with the idea of reworking the undo system so all tools share the same stack? I appreciate that might not be something we see in the next release, but is it a planned feature?

JohnMarchant
12-02-2015, 04:38 AM
Well i am very excited to see what we get in 2016. I must say far more xcited than i was from 11 to 2015, not that 2015 was not good. PBR, yes please and UGE, oh pretty please with cream on top.

Ztreem
12-02-2015, 05:17 AM
Loving all this news coming out! Not used to Deformer Stacks but sounds like I could easily get used to them. Can someone point me in the direction of how these work in other apps, for now?

this maybe? https://www.youtube.com/watch?v=L7L5tN8Ir5w#t=322

raw-m
12-02-2015, 05:37 AM
this maybe? https://www.youtube.com/watch?v=L7L5tN8Ir5w#t=322

Nice, finger crossed! Would be great if this deformer stack included sub-divisions.

ianr
12-02-2015, 06:20 AM
Updated SDK bible Rob,
will this be lauched first for our third party devs,
or on same day of launch release day of LW2016?

Earlier would help them.

ernesttx
12-02-2015, 06:54 AM
It would be nice to jump into the excitement and discuss...but, there's nothing to discuss. We already knew UGE was coming and it comes with architectural changes. OK. What I'm waiting for is a demonstration of it and details. Rob's post lacked specific details, it was an overview. I've been using Chronosculpt for the last 3 years, I know what it can do. I've requested updates to it, but nothing new came to it. I was even told it was not "abandoned". Ok, so this stuff is going to be in Layout. Great. Show something please. All this talk is only impressing yourself, Rob.

dsol
12-02-2015, 06:55 AM
Updated SDK bible Rob,
will this be lauched first for our third party devs,
or on same day of launch release day of LW2016?

Earlier would help them.

I'm sure the main plugin developers are already playing with the beta - under NDA of course!

lino.grandi
12-02-2015, 07:18 AM
Nice, finger crossed! Would be great if this deformer stack included sub-divisions.

Yes, you can decide when subdivision should happen for a Subpatch object. There will be a post showing how the modifier stack works as soon, probably within the next week.

raw-m
12-02-2015, 07:37 AM
Thanks Lino. How about when an object isn't sub-D'd? Will it be possible to increase/decrease the amount of polys that make up a standard box, for instance?

Wickedpup
12-02-2015, 07:38 AM
In modeler you can not move the bend or twist Tools in different positions for different effects as its applied as you drop the tool. In layout they are applied as modifiers and can be rearanged for different effects, more like a stack. This was more like an example, that it does not have anything to do with the undo system. Hopefully the new deformation stack is more powerful and incoperates the possibility to stack subdivisions, bone deformation, displacements etc.
AFAIK for you to be able to rearrange modifiers like that they have to be non-destructive....

mummyman
12-02-2015, 07:39 AM
Yes, you can decide when subdivision should happen for a Subpatch object. There will be a post showing how the modifier stack works as soon, probably within the next week.


Awesome! Looking forward to it.

lino.grandi
12-02-2015, 08:15 AM
Thanks Lino. How about when an object isn't sub-D'd? Will it be possible to increase/decrease the amount of polys that make up a standard box, for instance?

Talking about the current status of the modifier stack, no, you can only subdivide a subpatch object and use the subdivision modifier just once.

Ztreem
12-02-2015, 08:26 AM
AFAIK for you to be able to rearrange modifiers like that they have to be non-destructive....

Yes and they already is like that in layout, but not all modifiers can be rearranged like bend, twist and taper.

MAUROCOR
12-02-2015, 08:30 AM
Lino, will it be possible to make corrective morphs inside Layout in LW 2016?

Ztreem
12-02-2015, 08:32 AM
Talking about the current status of the modifier stack, no, you can only subdivide a subpatch object and use the subdivision modifier just once.

Would be nice if we could add displacement to a plane, then subdivide it, then use the normals from the first displacement to do a second displacement and then add another subdivision. Is this something that may be possible in the future?

GraphXs
12-02-2015, 08:48 AM
Yeah, I can't wait to see what the stack will allow... I know this might not make it for the next release, but would be awesome to get corrective morphs and weight painting in layout. Would we have to use this stack for that? WOOT!

Also what of the undo system? Has that gotten any love?

hdace
12-02-2015, 09:54 AM
Lino, will it be possible to make corrective morphs inside Layout in LW 2016?

Will it?

Chris S. (Fez)
12-02-2015, 10:03 AM
Talking about the current status of the modifier stack, no, you can only subdivide a subpatch object and use the subdivision modifier just once.

I hope your video will make the advantages of the modifier stack clear. Specifically, what advantages does the modifier stack in 2016 have over the "before bones" "last" etc. method used in 2015?

50one
12-03-2015, 03:25 AM
Looking forward to that video:)

prometheus
12-03-2015, 03:34 AM
Would be nice if we could add displacement to a plane, then subdivide it, then use the normals from the first displacement to do a second displacement and then add another subdivision. Is this something that may be possible in the future?

Would that imply different subdivision levels on different areas with displacement, sort of like aps? but with a stack history?

As I havenīt posted in this thread before, just want to say it is very exciting, I understand that a lot of things may not work in terms of modelling tools for a starter, but way to go..think it is a necessary process to develop better..faster geo handling, faster volumetrics, new faster and more accurate render and also overcome the old limits..it will be the foundation to compete with the others..no doubt about it...how well that competition will unfold is probably a matter of getting most of the functions from modeler working with the new geo system, as well as getting old deformers materials translated to work with the new system or being rewritten properly..there will be bugs, there will be things not working, but it seems to me the lightwave philosphy and main Feel is retained..that I like.

ianr
12-03-2015, 05:45 AM
Will it?
Corrective morphs in layout???? (Maurocor's Question)

Chronosculpt Overview LW3dG Advert:

Easily save any frame from the ChronoSculpt timeline
as an OBJ for Maya Blend Shapes
or as an Endomorph directly to LightWave object files
to create facial morph targets and deformation effects


LINO Hello Hello OVER? Bump1000%

VermilionCat
12-03-2015, 06:41 AM
What is required in order to achieve corrective morphs? I'm not familiar with this concept so I'd like to know. It's not joint morphs, right? It's not morphing based on the bone angles so what is it?

jasonwestmas
12-03-2015, 06:55 AM
What is required in order to achieve corrective morphs? I'm not familiar with this concept so I'd like to know. It's not joint morphs, right? It's not morphing based on the bone angles so what is it?

https://www.youtube.com/watch?v=vwU6aAi3mtc

You can do something like this in lightwave with the displacment node editor (using the correct displacement order) except there is a problem with the spacial math (matrices?) when the mesh itself rotates (via joint deformers) in space more than 45 degrees, you will start to get odd deformations. I suspect it's a limitation of the current geometry deformation stack.

VermilionCat
12-03-2015, 08:15 AM
https://www.youtube.com/watch?v=vwU6aAi3mtc

You can do something like this in lightwave with the displacment node editor (using the correct displacement order) except there is a problem with the spacial math (matrices?) when the mesh itself rotates (via joint deformers) in space more than 45 degrees, you will start to get odd deformations. I suspect it's a limitation of the current geometry deformation stack.

Thanks again, Jason.
It sounds complex (way over my head actually) but what video shows looks very similar to the joint morph?

jasonwestmas
12-03-2015, 08:37 AM
Thanks again, Jason.
It sounds complex (way over my head actually) but what video shows looks very similar to the joint morph?

I don't think there are any videos currently. I think most users just got fed up trying. It's kinda tedious mainly because of the modeler layout separation.

RebelHill
12-03-2015, 08:55 AM
Its not complex... its simple. you fix one wacky deformation by applying an inversely wacky one... the two cancel out, and presto. You can do this in LW, but you have to weight in a specific way, and the separation between modeler and layout makes the workflow, both for weighting and morph making exceedingly problematic. Until you can create morphs and weights in layout itself, this isnt gonna change.

VermilionCat
12-03-2015, 09:14 AM
Thank you, both of you.


Until you can create morphs and weights in layout itself, this isnt gonna change.

So this is the requirement. It doesn't happen until they migrate modeler function into layout then.
Something came into my mind though, usually people make series of corrective morph in other packages?
I mean, say you create corrective morph at 120 degrees bend, wacky one, it doesn't look convincing at 60 degrees bend, right?

jasonwestmas
12-03-2015, 09:34 AM
Thank you, both of you.



So this is the requirement. It doesn't happen until they migrate modeler function into layout then.
Something came into my mind though, usually people make series of corrective morph in other packages?
I mean, say you create corrective morph at 120 degrees bend, wacky one, it doesn't look convincing at 60 degrees bend, right?

For more realistic results riggers can add in several morph targets for a single appendage that blend in and out depending on the rotation channels automated via "set driven key". It's not like you can prepare for every single situation in a rig using "set driven key" but that's not the point. The point is to make your corrections on a per pose basis when your rig needs it in extreme circumstances and only if the problem is noticeable in a camera shot.

ernesttx
12-03-2015, 09:41 AM
Barry Zundel has a good write up on his blog - http://barryzundel.blogspot.com/ (scroll down a bit to read it)
Here is the video he shows - https://vimeo.com/5828261

If only LW2016 can do this simple type of corrective morph with the new UGE, I would be so happy.

RebelHill
12-03-2015, 09:47 AM
say you create corrective morph at 120 degrees bend, wacky one, it doesn't look convincing at 60 degrees bend, right?

It usually depends on the situation, but oftetimes it can look fine, yes. This is due to the fact that both the morf displacement and the bone displacement are actually linear transforms, thus a correct cancellation at one point in the transform will work at another, since the relation between the two is linear. If the result is a lil off or ugly in the middle somewhere (as is most often the case in more extreme base>max ranges) then you can simply dial the two together in a non linear fashion so the morph correction ramps more slowly than the bone rotation.

hrgiger
12-03-2015, 10:22 AM
I started on Hash Animation Master back in 99 and it always had smartskin which was its corrective morph solution. It would operate in a separate window where you would bend a bone say to 120 degrees then you could reshape the joint to look correct at that extreme. Then you would scrub back to say 60 degrees and then make corrections there. Your changes would interpolate(linear) through each degree in between and usually a couple of corrections was enough to get a nice result.

ernesttx
12-03-2015, 10:36 AM
yes I still use Animation Master in some respects for character animations. For stuff like this, it's now easy to export out .mdd file and import to LW. If only I could do this in LW.

hdace
12-03-2015, 11:29 AM
My corrective morphs are always triggered by Joint Morph Plus. And within that interface one can apply subtlety different morphs for different levels of rotation. I actually do try to create morphs that cover any eventuality. Creating morphs for one shot is too inefficient in a big project.

The real problem is that when an elbow (for example) rotates (and keep in mind the shoulder may be rotating too) there are two areas of the arm, above and below the elbow, that are now at multiple different angles from the original morph angle. Since morphs can't be easily rotated weight adjustments don't help. You end up with a morph that is supposed to compensate for a bad deformation that actually makes the deformation even worse. Solving this problem isn't just a case of being able to create the morphs in Layout (which we should already be able to do anyway). There needs to be a way to apply the morph AFTER the rotations occur, and that should be possible with 2016 given what Lino has said about the new deformation stack. But it would be nice to get confirmation.

PS - I would like to make it clear that I learned almost all of this stuff from RH & Pooby.

hdace
12-03-2015, 11:42 AM
A morph is just an offset. A vector for each point.
If the limb has been bent by the bone, and the deformation cached out, then you offset the point, it will look a certain way,
But if you save that as a morph and apply that offset before the bend (as you are forced to in LW), it will be totally wrong.
In order to make a morph after the bone has been bent, but that will produce the same result when applied before the bone bend, their has to be a reverse calculation of the bone weighting, to mix the influence of their matrices and invert that influence, (also taking into consideration the rest pose) and multiplying the vector offset (morph) by the result, to make the morph.

I deal with this kind of thing day in day out in ICE, in Softimage. You are welcome to try it out, but I guarantee it wont work as you hope. The bones and the morphs have a relationship to each other, to produce the final result and this can't be overlooked.

It IS possible to apply morph after bones ( in ICE for example), but to make it work you still have to deal with rotating the morph's vectors by the bones and youd have to code it as LW has no way of dealing with this out of the box.
watch these videos I made, if you are interested to see what I mean https://vimeo.com/67402407 https://vimeo.com/70124932

I actually argued with pooby about this at first, but later realized he was right.

jasonwestmas
12-03-2015, 11:42 AM
There needs to be a way to apply the morph before the rotations occur, and that should be possible with 2016 given what Lino has said about the new deformation stack. But it would be nice to get confirmation.

PS - I would like to make it clear that I learned almost all of this stuff from RH & Pooby.

Technically you can create morphs before bone rotations (if that is what you are referring to) using the Edit Nodes environment w/ Morph Map Node inside the deform tab. Change the "Node displacment Order" to "Before Bones".


I doubt there are enough nodes to make this work natively however. Perhaps the Bone Info node from DP Kit.

hdace
12-03-2015, 11:48 AM
Later in the same thread Lino said this:


Creating precise corrective morphs in LightWave at the moment is not possible. For now.

http://forums.newtek.com/showthread.php?137894-Creating-a-corrective-morph-using-Chronosculpt

What I want to know, and I've asked this twice previously in this thread and its "father" thread that is now a sticky, is this now going to change? Apparently Lino & Rob are not able or do not want to answer.

- - - Updated - - -


Technically you can create morphs before bone rotations (if that is what you are referring to) using the Edit Nodes environment w/ Morph Map Node inside the deform tab. Change the "Node displacment Order" to "Before Bones".

I always use Joint Morph Plus for this. I'll try what you suggest but not sure if bone rotation can easily be thrown into the mix.

I misspoke and edit/corrected it. One would want AFTER bones. The problem is applying a morph after rotation renders it useless because the points are shooting off in the wrong direction. So in local space the points will move correctly if they're applied AFTER the rotation. Right now this happens in world space.

RebelHill
12-03-2015, 12:01 PM
I misspoke and edit/corrected it. One would want AFTER bones. The problem is applying a morph after rotation renders it useless because the points are shooting off in the wrong direction. So in local space the points will move correctly if they're applied AFTER the rotation. Right now this happens in world space.

Nope... you want BEFORE. Here's why...

The morphs exist as a transformation on the mesh in its base state, relative to that base state (aka, in object space). If you apply the morph after the bone displacement, then you have a space mismatch. You can correct for that by transforming the morph by the bone deformation, but since verts can be weighted to multiple bones, and verts wont therefore necessarily have moved exactly as if they were stuck solidly to the bone (very common when smooth/spread weighting across joint boundaries) then you'd have to calculate all that too (not efficient). Thus, you need a morph in object space, which is applied first, and this "wonky" version of your mesh is then displaced by the bones, with the 2 together giving the pleasing result.

There is another option, which involves storing morphs in tangent space to the mesh itself, but LW does no have tangent space morphs (nor do most, messiah had them cant think of many others).

The problem you get in LW however, most commonly, is not that you cant set morphs before bones (you can now and will be able to with a deformation stack obvs), the problem is that the most common method of weighting in LW is to use the bone falloff tools, which help to soften deformations at joints. But because of this, if a morph moves a vert before the bone displacement, then the distance to the bone also changes, which makes the effect of the bone deformation different... which means the morph no longer properly compensates for that deform.

The only solution to this in LW is to use explicit weights, normalised with WMO, negating LWs falloff effects. This makes getting good weights and base deformations much more laborious (amplified by there being no weight painting in layout), and thus, makes the whole process that much harder.

jasonwestmas
12-03-2015, 12:16 PM
As long as the process is a direct path, laborious (within reason) is fine for me. Granted nobody wants to work without weight painting when using WMO.

RebelHill
12-03-2015, 12:19 PM
Personally... I dont think its "reasonably" laborious. Hence, I find the better method to be to forego it and work with LWs falloff/weight spreading, which can create smoother, better bends in the first instance which require much less in the form of any correction.

hdace
12-03-2015, 12:45 PM
Nope... you want BEFORE. Here's why...

The morphs exist as a transformation on the mesh in its base state, relative to that base state (aka, in object space). If you apply the morph after the bone displacement, then you have a space mismatch. You can correct for that by transforming the morph by the bone deformation, but since verts can be weighted to multiple bones, and verts wont therefore necessarily have moved exactly as if they were stuck solidly to the bone (very common when smooth/spread weighting across joint boundaries) then you'd have to calculate all that too (not efficient). Thus, you need a morph in object space, which is applied first, and this "wonky" version of your mesh is then displaced by the bones, with the 2 together giving the pleasing result.

There is another option, which involves storing morphs in tangent space to the mesh itself, but LW does no have tangent space morphs (nor do most, messiah had them cant think of many others).

The problem you get in LW however, most commonly, is not that you cant set morphs before bones (you can now and will be able to with a deformation stack obvs), the problem is that the most common method of weighting in LW is to use the bone falloff tools, which help to soften deformations at joints. But because of this, if a morph moves a vert before the bone displacement, then the distance to the bone also changes, which makes the effect of the bone deformation different... which means the morph no longer properly compensates for that deform.

The only solution to this in LW is to use explicit weights, normalised with WMO, negating LWs falloff effects. This makes getting good weights and base deformations much more laborious (amplified by there being no weight painting in layout), and thus, makes the whole process that much harder.

I'm afraid that you've inadvertently shown that you want AFTER.

It is illogical to apply a compensating morph BEFORE bone deformation when the compensation is for a deformation that happens AFTER bone rotation. If the compensation is going to have any meaning it should be applied when it actually fixes the problem. If you apply the fix, then repeat the problem, the fix is no longer valid.

In a big production with lots of long scenes with characters moving all over the place "laborious" is a big problem. One wants a series of "standard" correction morphs that just work automatically. One needs to use LW's advantages, not focus on its drawbacks. The automatic bone weighting system, and JointMorph Plus, are both huge advantages. All we need now is a way to apply morphs after bones in local space.

You said this would create a space mismatch. We have space mismatches all the time and Layout corrects them at the flick of a switch. Why can't we just have a space mismatch correction switch? Maybe that's what the engineers are working on (specific to this issue).

(btw it's really scary disagreeing with your guru)

Tranimatronic
12-03-2015, 12:47 PM
honestly though for volume preservation, you are better off using the technique outlined here:
http://forums.newtek.com/showthread.php?127947-Eyelid-Bulge-Solution-%28Almost%29/page3
there is an awesome python script by LightFreeze that will cover the outside of the elbow tip. The inside spread will be more complicated, but may be better using animated displacements?
If I get time I will try to knock together an example

hdace
12-03-2015, 12:48 PM
RH: I've tried adjusting the bone options a thousand different ways but usually find it just makes bad deforms even worse. The default settings in your rigs are almost always the best.

RebelHill
12-03-2015, 12:52 PM
It is illogical to apply a compensating morph BEFORE bone deformation when the compensation is for a deformation that happens AFTER bone rotation... All we need now is a way to apply morphs after bones in local space.)

I'm afraid you're wrong... before is exactly how it should be for local space morphs. I suggest that you misunderstand what local space actually is... it's object space (before any deforms) NOT the "space" AFTER bone deforms. That would be mesh tangent space.

hdace
12-03-2015, 01:03 PM
But one creates the compensating morph in world space. Doesn't that mean it should be applied in world space (or mesh tangent space, which I obviously don't understand!)?

I thought local & object space were the same thing. In the end it doesn't really matter if I'm wrong (that is a frequent occurrence!), I would still like to hear Lino definitively state that precise corrective morphs are now possible.

RebelHill
12-03-2015, 01:11 PM
No... you create morphs in object space (call it local space if you will... object-local space). You certainly dont want to apply it in world space, as then it would be invariant under transforms to the object (so a morf that moved verts behind the mesh would move them in front if the mesh were rotated 180). What local space is not, is local to the sense of the "bit" of mesh that has been deformed... thats surface (or tangent) space.

https://en.wikipedia.org/wiki/Tangent_space

And I can tell you definitively... making precise corrective morphs will be no more or less doable than now (they can be done now, but you must use explicit weighting, and the modeler/layout split makes it a very inefficient task), unless that is there are tools for modeling and weighting in layout (which dont make it any more possible, but only make it easier), and those are already noted as not going to be in 2016.


https://www.youtube.com/watch?v=q_Z4yoTidXY

wyattharris
12-03-2015, 01:32 PM
That vid was much clearer than the Wikipedia rabbit hole you just shoved me in. :D

Tangent space -> Manifold -> Klein bottle -> Real projective plane.
Learned a lot along the way though.

RebelHill
12-03-2015, 01:37 PM
That vid was much clearer than the Wikipedia rabbit hole you just shoved me in.

GET IN THE F-ING HOLE!!!... PUT THE LOTION IN THE F-ING BASKET!!!

sorry *twitch*

ernesttx
12-03-2015, 01:45 PM
This is what I'm hoping for - https://vimeo.com/27100314 - Skip to :40 sec in to see it. Chronosculpt can do the sculpting part but no bone. LW can do the bone, but no sculpting. LW2016 should be able to do both in Layout.

hdace
12-03-2015, 01:45 PM
RH: I see what you're saying. There is one problem though. Your demo skips the nature of a corrective morph which is applied on top of a previous deformation.

If there was some kind of method to calculate the rotation of the affected vertices then I believe it would be better to apply it afterwards. However, I hear what you're saying. That isn't going to happen.

RebelHill
12-03-2015, 01:52 PM
If there was some kind of method to calculate the rotation of the affected vertices then I believe it would be better to apply it afterwards. However, I hear what you're saying. That isn't going to happen.

Its not that it is or isnt going to happen... there would simply be no point. To actually have a morph it has to be stored as a set of vectors for each vert, and so you need a set of co-ords (a space) in which to make that transformation. If you created the morf in object space, the deformed the mesh, then calculated the result of that deformation on each point, and added that onto the morf stored vector THEN, you would apply this modified morph after... getting exactly the same result as if you had applied it before the deformation, only having used up more compute time to do it.

Oh, and btw, tangent space morphing isnt some kind of panacea here (even though it seems like it'd be the fix all) because ofc when the bone (or whatever) deforms the mesh, the surface shape changes, which alters the orientation of the tangent plane, which in turn means the morph wouldnt make the same sense, and all you've done is create an "after bones" specific type of morphing with its own set of problems.

If you leave behind LWs handy bone falloffs, go for precise, explicit weights, then you can have precise corrective shaping.

hdace
12-03-2015, 02:06 PM
Then what is pooby talking about regarding ICE in XSI? I'm not convinced the result would be the same. In your demo the rotated points don't spread out evenly. Transformation can't be the only thing they're using in ICE or he wouldn't have mentioned it. They must be using the rotation data to spread out rotated points so you do get a better result.

"their has to be a reverse calculation of the bone weighting, to mix the influence of their matrices and invert that influence, (also taking into consideration the rest pose) and multiplying the vector offset (morph) by the result, to make the morph."

RebelHill
12-03-2015, 02:15 PM
Its because in ICE he's creating an additional displacement which is yet another thing thrown into the mix alongside the bones and the morph. To just blanket get the exact transforms applied to each vert by the bone, add that to (or subtract rather in the final thing) the object space morf, and apply it all after bones is exactly the same as applying the morf before bones and doing no such calculation. And yes, it is only transformation, because all displacement is transformation.

hdace
12-03-2015, 02:23 PM
Now I've got a headache.

wesleycorgi
12-03-2015, 02:36 PM
My new mantra for successful relationships, "morph first, bone after." :D

Tranimatronic
12-03-2015, 02:46 PM
You are better off forgetting corrective morphs and doing corrective displacements.
Save your problem pose, bring it into zbrush, apply any original displacement maps, sculpt the correction, save the new displacement.
In the displacement node graph using a weight map to localize the area of effect, subtract the new corrective displacement from the default displacement (giving us the delta). Apply this as additive with the percentage tied to the bones rotation. Everything working in the correct order/space.

hdace
12-03-2015, 03:04 PM
Hmmm... That sounds good.

RebelHill
12-03-2015, 03:43 PM
You can do that, as its a surface based effect which does happen (or should happen) after the other deformations. However, you are limited to simply moving points back n forth along their own normal, you cant smudge points around in space, so its much less flexible than a morf. Unless ofc you want to use a vector displacement map, in which case you're back to issues of the co-ordinate space.

pinkmouse
12-03-2015, 04:34 PM
Now I've got a headache.

Indeed. One of the reasons CA leaves me completely cold.

erikals
12-03-2015, 05:16 PM
this is the best way (imo) but wish i could do it (properly) in Layout


https://www.youtube.com/watch?v=313H-xNh8y8

there is also the other "less correct" Layout / EditFX way >
https://www.youtube.com/watch?v=rM5t5rBGBpE

RebelHill
12-03-2015, 05:25 PM
technically not a "corrective" morf. Try using that method to untangle the overlapping geo inside the elbow itself (the geo that actually spans the joint)... whole different story.

erikals
12-03-2015, 05:30 PM
true, certainly not going to work, but works alright for this though

ernesttx
12-03-2015, 05:53 PM
dear god all these methods are counter intuitive and counter productive to a character animator's workflow. please LW team make this a priority to get right. get a character animator's opinion/design specs/workflow in the mix for ideas. make this a seamless operation since corrective morphs are vital and used everywhere. that's why i've been pushing for chronosculpt updates. you've got the tech there, just make the needed tweaks and you've got the best toolset for this stuff.

RebelHill
12-03-2015, 06:15 PM
make this a seamless operation

easier said than done unfortunately. There are further issues to contend with than those mentioned here. For the minute, the only real thing that LW itself needs is morph and weight creation in layout... that'd be enough. Im sure that's coming in some future release, but it seems its def not going to be in 2016.

jwiede
12-03-2015, 07:14 PM
Guys, maybe take the theoretical discussion of whether and why corrective morphs are an issue to another thread? You're quite far from discussing Rob's UGE blog post at this point.

hdace
12-03-2015, 08:30 PM
I thought it connected to Lino's hints about the deformation stack, but not according to RH. Sorry if I caused this. But it is important.,

jeric_synergy
12-03-2015, 09:21 PM
Now I've got a headache.
Oh good, it's not just me.

lino.grandi
12-04-2015, 12:51 AM
Guys, maybe take the theoretical discussion of whether and why corrective morphs are an issue to another thread? You're quite far from discussing Rob's UGE blog post at this point.

I agree. Just wait for the next blog post. I'll show an example where corrective morphs are used in the new modifier stack.

Ztreem
12-04-2015, 03:33 AM
Would that imply different subdivision levels on different areas with displacement, sort of like aps? but with a stack history?


No, more thinking of stacking several displacements and subdivisions after each other. like this.

RebelHill
12-04-2015, 04:43 AM
Guys, maybe take the theoretical discussion of whether and why corrective morphs are an issue to another thread? You're quite far from discussing Rob's UGE blog post at this point.

You do realise that the means of creation for this kind of corrective workflow, and they way to implement it on a setup is something that is addressed by the addition of the UGE, and the mesh editing capabilities it will allow for in layout (albeit in a later reelase), right?

Surrealist.
12-04-2015, 05:22 AM
Clearly the point in my opinion. :hey:

ernesttx
12-04-2015, 06:13 AM
Lino, thank you. Will definately be keeping an eye out for that video. But, can it at least have SOME information about what it is you are going to show; instead of silent video. Throw us character animators a 'bone'. <-- intentional, oh yeah

jasonwestmas
12-04-2015, 07:07 AM
I agree. Just wait for the next blog post. I'll show an example where corrective morphs are used in the new modifier stack.

"Quote Originally Posted by jwiede View Post
Guys, maybe take the theoretical discussion of whether and why corrective morphs are an issue to another thread? You're quite far from discussing Rob's UGE blog post at this point."

Thank you although our discussion is UGE based. Not sure what else there is to talk about, any suggestions?

prometheus
12-04-2015, 07:35 AM
No, more thinking of stacking several displacements and subdivisions after each other. like this.


Ahh..think I get it, first I thought..why not just use additive layers, that is of course limited to be in use by the one and same subdivision level, while in your suggested case it would work to pick the mid-low res level and change main shape procedural without being to costly in deformation process..and still maintaining the higher detailed procedural on the highest subdiv level, and also much easier to follow what level is controlled by respective procedural.

hdace
12-04-2015, 07:58 AM
I agree. Just wait for the next blog post. I'll show an example where corrective morphs are used in the new modifier stack.

That's what I'm talkin' about!

ernesttx: Personally I really enjoy the music with the videos. But, yeah, this one might need narration. It's all kinda complex, huh?

lino.grandi
12-04-2015, 07:58 AM
Lino, thank you. Will definately be keeping an eye out for that video. But, can it at least have SOME information about what it is you are going to show; instead of silent video. Throw us character animators a 'bone'. <-- intentional, oh yeah

This one will not be silent. ;)

jasonwestmas
12-04-2015, 08:13 AM
Whoah a video on lightwave deforms from Newtek LW3DG! Stop the press, I'm looking forward to this one.

ernesttx
12-04-2015, 08:39 AM
This one will not be silent. <---- oh yeah. Lino, looking forward to this one more than the new Star Wars !!!

50one
12-04-2015, 08:40 AM
I thought there will be something to watch when I'll check back today...C'mon Lino this is too good to just keep it for yourself only!;)

hrgiger
12-04-2015, 09:42 AM
Lino said on Skype it might be another week. He's a busy man.

50one
12-04-2015, 09:46 AM
Lino said on Skype it might be another week. He's a busy man.

As long as it's in the works, quite excited to see the modifier stack in Layout. Good times.

ernesttx
12-04-2015, 10:00 AM
If Lino needs help, I'm sssoooooo available to tackle this one. Just send me beta version hehe

Luc_Feri
12-05-2015, 06:17 AM
Wasn't impressed at all with the last blog. Still looks like a hack rather than a revelation. Also really, radial arrays and a torus tool?? Tools that have been in other packages for light years, lattice deformers er yep seen it before. Hope for more.

MAUROCOR
12-05-2015, 07:11 AM
Wasn't impressed at all with the last blog. Still looks like a hack rather than a revelation. Also really, radial arrays and a torus tool?? Tools that have been in other packages for light years, lattice deformers er yep seen it before. Hope for more.

All that kind of frustrated comments about tadial array, torus tool, lattice deformer, etc have been answered before. It will be necessary to research that in previous posts.

erikals
12-05-2015, 07:37 AM
Still looks like a hack rather than a revelation
nope, it's a Revelation, make no mistake about that.


Also really, radial arrays and a torus tool??
Tools that have been in other packages for light years, lattice deformers er yep seen it before.
Hope for more.
yep, they will show more later on, hope to see more Modeler attention.
currently they do not have a dedicated Modeler Developer at NT, maybe they could get help from some plugin developers.
PicTrix - Tama - Artur - TrueArt - Hurley - Worley - Viktor - Ino - Takahiko ...etc...

jasonwestmas
12-05-2015, 07:46 AM
well actually the real revelation will be when all this modeler stuff actually works inside of layout and without tacky non-interactive manipulation; interrupting, impractical workflows and can be stacked and animated.

erikals
12-05-2015, 10:00 AM
wondering, will modeling operations in Layout use Multiple Cores?
or will it will be One Core only (~15% of cpu) like in the current Modeler

hmm....

hdace
12-05-2015, 10:10 AM
Moore's law is meaningless if you can't multi-thread.

probiner
12-06-2015, 04:45 AM
wondering, will modeling operations in Layout use Multiple Cores?
or will it will be One Core only (~15% of cpu) like in the current Modeler

hmm....

while multi-threading is a great speed up for things like deformers, I believe many modeling operations are inertly single-threaded and therefore, I would guess the methods you use to structure mesh data, memory allocation, etc, is what gives you more bang for the buck.
,
Having an environment where you don't have to invest on time consuming optimizations and, in a pinch you can throw a lot of things and it without losing responsiveness is a great ground for creativity.


Looking forward the techdemos. These blog posts are a nice tease, but we're totally left to our imagination.

Luc_Feri
12-06-2015, 06:23 AM
When I see Lino tumbling around geometry into the millions of polys (i.e. a big scene without having to hide stuff) like Chronosculpt, then I will believe in the revalation of a new layout and geometry engine.

The news in october came in with a bang, promising a revolution of sorts, that would stand out in the 3D world. At the moment apart from PBR and render engine, which looked good mind, every software from substance designer to Poser, DAZ and iClone is getting PBR, this is not that much of a deal really is it.

I haven't seen anything yet which is truly game changing and making LW stand out. Chronosculpt did stand out, I want this geometry engine in modeler/layout, I've tumbled 80 million polys around inside of it. Then LW will be supreme again.

And just for the record, I'm not bashing, I'm wanting LW3DG to prove me wrong and then I'll say yes you guys rock. I've given enough positive vibes a few months ago towards LW3DG, they need to hear the crux of it also. I've invested money and time in this software, bought Octane to use with LW and also Houdini Indie to help complement it. I have turned my back on MODO for now (Foundry cash cow), didn't even flinch with the black friday stuff. I want this company to be the best again, just being realistic. :D

hrgiger
12-06-2015, 07:20 AM
Well they're not asking you to buy it yet Luc so just try being patient.

And tumbling large amounts of polygons doesn't mean much. Editing and deformation is what is important.

PBR may not sound revolutionary but then as has been stated, this is the beginning of what they're saying will be a revolution, replacing the render engine, volumetrics, and a new geometry engine capable of producing new geometry are all foundational changes that had to be made which will be the basis for future growth. People want revolutionary features they're going to accept that a foundation has to be laid for those features. Modeling was tried in Layout before for version 9. And it failed because they didn't want to lay the proper groundwork first.

Luc_Feri
12-06-2015, 07:43 AM
Well they're not asking you to buy it yet Luc so just try being patient.

And tumbling large amounts of polygons doesn't mean much. Editing and deformation is what is important.

Huh?? I know you love LW and are defensive towards it but I'm just trying to be realisitic if they want to get back into big time again.

You see this is the problem with forums. I'm not asking to buy it now or real soon, just show more real substance after the grandoise claim LW2016 was going to be a game changer. The momentum it had will start to wane is all I'm saying unless they back up it up with real substance as the last video IMO was weak.

As for my comment about tumbling polygons doesn't mean much like I am some dumb *** who didn't actually know editing and deformation of polygons is important. Of course I know that. :D

Tumbling geometry is important, you telling me you don't rotate around the scene your working on?? The denser the scene the more I have to hide right now for scene management.

Finally, it ain't no revolution if we are years behind everyone else is it? As I said loads of software has some PBR implementation as standard, tons these days. This ain't no revolution yet. If the development is slow then it won't matter at all, they'll always be catching up. If they get a Chronosculpt type of geometry integration with massive polygon scene capabilities , then superb, that would be a revolution. This was hinted at by Rob, removing the past limitations, so far I have seen ZERO evidence to back that up. This is what I thought the first blog was telling me, what all this unified geometry engine was about, not some gradual hack away from the current Modeler/Layout workflow. An engine that could handle massive amounts of data, this is what I thought the big excitement was claiming to be coming to us with LW2016 and soonish, next year maybe, not 5/10 years from now.

I'm sure I'm not the only one who thought this massive change was coming with LW2016 with a chronosculpt style geometry engine capable of removing all the previous limitations we had, to be a game changer, now it seems they want us to wait years before it lands. By then who knows what Autodesk, Foundry, C4D, Houdini will have done, might be too late again. They need to get this going right now to be the game changer or they will flounder and miss the boat. I want them to do it, more than anything as Chronosculpt has some amazing tech inside of it. If I have been misled and have this all wrong then, ha, goes back to what I was saying before about marketing practices and the way it impacts our opinions.

Luc_Feri
12-06-2015, 08:42 AM
'Another relevant thing to point out is that the unified mesh system which we have implemented in LightWave Layout is an improved version of the modern mesh system which we developed for ChronoSculpt. Some of you may have witnessed the standing room only crowds that we had at SIGGRAPH when we demoed ChronoSculpt. Folks were amazed that we were manipulating such complex geometry so quickly and excited by the animated geometry deformation and sculpting tools. Because the unified mesh system now implemented in LightWave Layout was developed from the ChronoSculpt mesh system Layout will get the same advantages for memory usage and deformation speed as well as the benefits of reducing the overhead which was required for the previous two systems. That overhead impacted memory usage, performance, as well being much more complex to interface with when developing. Now instead we are able to work with one single proper modern mesh system which greatly simplifies the development process.'

This is from the new blog. All I am saying is if this is coming and I believe it will, I hope to see more impressive demos and videos than the last one. They are saying an 'improved' geometry engine upon Chronosculpt, well it sounds amazing as it already handles geometry really dam well. An improved engine should be mind blowing, but I honestly felt that LW2016 would see this mind blowing engine already in place in some form. I was hoping if not for a unified Modeler/Layout, a Layout which could handle millions of polygons really easily. I'm sorry if I got my hopes up too quickly. :D

hrgiger
12-06-2015, 09:02 AM
Jfc Luc I'm not in the least bit defensive(although annoyed with your overly simple analysis of my response), I use other apps besides LW, including Modo for where LW modeler is lacking. This is just being realistic. Removing limitations that Rob and Lino are speaking of is changing the architecture of LW to allow for things that weren't possible before. Creating geometry in Layout, allowing for faster deformations, integrating volumetrics into the lighting and rendering. That doesn't mean however you're going to see a full suite of modeling tools in Layout for 2016 or even that chronosculpt functionality is now inside of Layout. Rob stated it, this is a phased plan of modernizing LW, changing the foundation is a huge and necessary first step. If you don't see putting a entirely new geometry engine, render engine or volumetrics as significant development or see the bigger picture of what work LW needs to address peoples needs, then nothing anyone can say will change that. I don't care and doubt anyone else does either.

Yes tumbling polygons is important but it's not a huge problem in LW now either. Certainly not on the scale of where editing and deformations is currently.

Im always amused when people compare LW to Auto desk apps or Houdini and think it's a winner take all situation. You just have to ask yourself if the software does what you need it to and for many, it's just one app in a many app pipeline.

Luc_Feri
12-06-2015, 09:24 AM
Oh come on mate. You might forget I read this forum and the luxology ones. I've seen you defend LW time after time. I think you are honest enough to correct someone if they are misinformed and stick up a bit for LW if people are bashing it for no reason. There is nothing wrong with that at all, I feel passionate about LW just as I did MODO for a while. You don't have to be defensive about being defensive. :D

I thought you were doing the same a little with me and my comments, somewhat trying to belittle me. I speak my mind, I said this to Lino. I think he felt I was a bit harsh but I do care about the future of this product and I do wish them all the best at LW3DG. I sometimes feel they don't like the negatives but I like to try and be constructive with my arguments.

I agree use the software to get the results that you want to get and for the work that you need to get done is all that matters. I just feel that LW3DG could blow out the competitors if they got the same type of engine as Chronosculpt in LW and it's performance. It's a business after all, and as a CEO of Newtek right now I would personally be pulling out the stops to get this tech out developed before the boat has sailed. I feel a massive amount of interest and capital could be made with LW should they manage to pull it off reasonably quickly and not to dally on this.

jeric_synergy
12-06-2015, 10:13 AM
I believe many modeling operations are inertly single-threaded and therefore,
Probiner, i know that might have just been a typo, but in case it wasn't and to further perfect your very fluent English: "inherently".

In some other thread I'd like to see a simplified explanation of what processes can be multi-thread, and why some cannot.

hrgiger
12-06-2015, 10:25 AM
I dont see it as defending LW as much as im just being realistic about expectations for the LW 2016 release. Ive been saying it for years, im tired of seeing new features piled on top of a disjointed architecture where everything seems like its own island and not addressing legacy issues that LW has suffered for years. My biggest complaints over the last several years (and there have been many) have to do with performance, integration issues, fiberfx, lack of modeling fixes and improvements... am I excited about a new torus tool or a live array tool? Not really(i have wanted a natiive lattice tool for 10 or more years though). But to me, the best part about what has been shown thus far is that theyre actually talking about whats been wrong with Lightwave and what theyre doing about it. And thats talk ive wanted to hear from them in some time.

Autodesk has little to fear from LightWave or Modo, and theres nothing wrong with that. No matter how good LW or Modo is, it would take a lot more then simply features to change the market.

hrgiger
12-06-2015, 10:32 AM
Probiner, i know that might have just been a typo, but in case it wasn't and to further perfect your very fluent English: "inherently".

In some other thread I'd like to see a simplified explanation of what processes can be multi-thread, and why some cannot.

Eric, it could just be because something like a bevel or other transform tool cant be split up into multiple threads. In a lot of cases you might be just looking at a simple vector transform so 1 math operation= 1 thread.

RebelHill
12-06-2015, 10:36 AM
I'd like to see a simplified explanation of what processes can be multi-thread, and why some cannot.

There really is no "simplified" explanation, the reasons are kinda complex (and some of the tricks that can be applied to achieve parallelism even moreso). In short though, think of it like this. You can have one person chopping the onion while another is peeling the potato... You cannot have one peeling the potato at the same time as another is chopping the potato. Some tasks can be parallelised, some must be queued.

MAUROCOR
12-06-2015, 10:45 AM
I think LW3DG is doing an amazing work with LW. They are changing the software from inside, as stated by Rob Powers. And they have solid plans and an ambicious target, thanks God. It took a lot of time, yes, but finally we will be able to see (and experiment) what they are talking about. A little bit of patience is required. So stay calm, please. Donīt fight, take a deep breath and no panic.:D

KurtF
12-06-2015, 11:04 AM
The architecture may be in place, but creating a full toolset will likely require several versions worth of upgrades. Everybody should keep their expectations realistic. Just saying.

Luc_Feri
12-06-2015, 11:41 AM
I dont see it as defending LW as much as im just being realistic about expectations for the LW 2016 release. Ive been saying it for years, im tired of seeing new features piled on top of a disjointed architecture where everything seems like its own island and not addressing legacy issues that LW has suffered for years. My biggest complaints over the last several years (and there have been many) have to do with performance, integration issues, fiberfx, lack of modeling fixes and improvements... am I excited about a new torus tool or a live array tool? Not really(i have wanted a natiive lattice tool for 10 or more years though). But to me, the best part about what has been shown thus far is that theyre actually talking about whats been wrong with Lightwave and what theyre doing about it. And thats talk ive wanted to hear from them in some time.

Autodesk has little to fear from LightWave or Modo, and theres nothing wrong with that. No matter how good LW or Modo is, it would take a lot more then simply features to change the market.

For sure. I agree with most of that. I'm just being impatient to actually see this evidence rather than it be pure talk. Yes they are totally addressing and saying the right things but that's was my point, I haven't seen any evidence yet of this from the videos so far. They were just snippets of bits and bobs. I didn't think the fiberfx was that breathtaking on performance, the scene didn't have that many fibers on show. So far for me more talk than real evidence of this new geometry engine wow factor. Yes other features are taking shape with the render engine, and that is a good thing and a few others but very small snippets.

When you say Autodesk has little to fear, that may be so, until that is people can see the sheer performance of a Layout I envisage with a massive amount of geometry in it for a full scene. That is a feature any artist would want, as most of the products on the market seem to choke on massive polygon counts.

I know it will take a while to get the full modelling suite integrated with this new engine but Chronosculpt is a few years old now, so how long will it take to integrate. I would be happy with just a layout for now that could handle the massive amount of polygons as it mainly chokes when you populate on full scenes with lots of props.

hdace
12-06-2015, 12:05 PM
There really is no "simplified" explanation, the reasons are kinda complex (and some of the tricks that can be applied to achieve parallelism even moreso). In short though, think of it like this. You can have one person chopping the onion while another is peeling the potato... You cannot have one peeling the potato at the same time as another is chopping the potato. Some tasks can be parallelised, some must be queued.

Great analogy!

jeric_synergy
12-06-2015, 12:27 PM
I'm amazed it can happen at ALL, and counter-intuitively it seems like complex processes like rendering seem more amenable than what would seem to be simpler, eg modeling, tasks. But that's probably because I don't understand what's going on at any level. (At a guess, rendering is so complex there's probably all sorts of semi-separate pipelines that can be addressed by specific threads.)

I'm really more interested by inclination in the UI: no matter what's under the hood, if the steering wheel is replaced with a tiller, it's not pretty.

Academically, it'd be interesting to see what issues, if any, the new capabilities raise in the UI. Historically I'd say LW has issues dealing with multiple (layout) selections, and making that both consistent and convenient.

Surrealist.
12-06-2015, 12:52 PM
When I see Lino tumbling around geometry into the millions of polys (i.e. a big scene without having to hide stuff) like Chronosculpt, then I will believe in the revalation of a new layout and geometry engine.

The news in october came in with a bang, promising a revolution of sorts, that would stand out in the 3D world. At the moment apart from PBR and render engine, which looked good mind, every software from substance designer to Poser, DAZ and iClone is getting PBR, this is not that much of a deal really is it.

I haven't seen anything yet which is truly game changing and making LW stand out. Chronosculpt did stand out, I want this geometry engine in modeler/layout, I've tumbled 80 million polys around inside of it. Then LW will be supreme again.

And just for the record, I'm not bashing, I'm wanting LW3DG to prove me wrong and then I'll say yes you guys rock. I've given enough positive vibes a few months ago towards LW3DG, they need to hear the crux of it also. I've invested money and time in this software, bought Octane to use with LW and also Houdini Indie to help complement it. I have turned my back on MODO for now (Foundry cash cow), didn't even flinch with the black friday stuff. I want this company to be the best again, just being realistic. :D

If I were you then, I go back to the lobby and get a couple of supersized cokes and the largest bucket of buttery popcorn they have. :D

This is going to be a long show.

Seriously, I take all of this news within the overall context of a projected time frame. Turning the "Way Back Machine" to the year 2009, which is where this all started, we are merely 6 years into a 10-15 year plan to revamp LightWave.

And no one in these six years has uttered a single realistic projection at NewTek or LW 3D Group, that reflects the real task at hand until now. Finally Rob is saying something about LightWave for the next 25 years. It is real nice to hear someone thinking in those terms. He is not saying this particular program is going to take that long. But he at least it realistically projecting out into the future beyond just the next few years.

And in reality if you consider it really will take at least 10 years to do this, you can see that they are right on time now to release the first real glimpse of the ground work being laid in place. And then it could be reasonable to expect that over the next 3 years we will gradually move toward a more complete toolset using the new UGE.

So I see all of the actions they have done under Rob's direction as reflecting this fact. And reflecting the fact that along the way they have to somehow balance a parallel development to improve LightWave that is. For example addressing some issues and fixes with Modeler. A camera in modeler for example. They all add up to it isn't going to happen over night.

So I just think, yeah, wait and see. I have confidence. But the main point is I have to use tools that work now. And I can never wait for anyone. But I don't think there is anyone who will not welcome new healthy competition with something innovative.

Please somebody shake up the monopoly that is the main apps available. Please.

probiner
12-06-2015, 07:03 PM
Probiner, i know that might have just been a typo, but in case it wasn't and to further perfect your very fluent English: "inherently".

In some other thread I'd like to see a simplified explanation of what processes can be multi-thread, and why some cannot.

Yes, I meant inherently or in portuguese, inerente.

I would say RH explanation of the problem was quite fit. So you probably better optimizing how the guy that peels potatoes, gets access to new potatoes, what he uses to peel them and how he disposes the peels and stores the processed potatoes than how many people can peel a single potato to make it faster.

The less the user has to pay for what he is trying to do the the faster and more mistakes he can make to trace the right path.


The architecture may be in place, but creating a full toolset will likely require several versions worth of upgrades. Everybody should keep their expectations realistic. Just saying.
We don't know what's in place or not, really. But my guess is that the foundation takes much more time than doing the tools, specially if every tool uses the SDK instead of being it's own separate world. But, tools take a lot of design decisions. There's a lot of time invested in their philosophy, interaction and even their feeling, all which take time to mature.
Now, if the possibilities are at least in place that will also mean third parties doing great things you won't have to wait for LW3DG to do. And this is why this whole thing is overall exciting, because for a while now, some things coming out felt a bit like a third parties' approach or copy, without addressing what only LW3DG can touch. I would guess that even with the known skepticism, we're all hoping for a good crop.

Cheers

DrStrik9
12-06-2015, 08:38 PM
The less the user has to pay for what he is trying to do the the faster and more mistakes he can make to trace the right path.

Yes. Art is an ongoing chain of experiments.

wyattharris
12-07-2015, 10:25 AM
There really is no "simplified" explanation, the reasons are kinda complex (and some of the tricks that can be applied to achieve parallelism even moreso). In short though, think of it like this. You can have one person chopping the onion while another is peeling the potato... You cannot have one peeling the potato at the same time as another is chopping the potato. Some tasks can be parallelised, some must be queued.
Also keep in mind that the slowest part of the process is the user. To keep the analogy, you have a factory of 1000s (billions realistically) of workers ready to chop, peel, dice and puree a variety of food. Unfortunately they are all waiting for the foreman to decide what they should do.

Usually billions of cycles are wasted for every second you are holding that mouse button. BUT certain tools may benefit from multithreading. (sculpting on millions of points? IDK)

It was a cool feeling the first time I hit F9 on a multi-proc system and saw all cores shoot up to 100% utilization.

Chuck
12-07-2015, 10:37 AM
So anything added to modeler now will have to be remade later?

I seem to recall a comment from the LightWave team on the matter already, to the effect that Modeler has not been changed over to the Unified Mesh System, but that the tools being added to Modeler are being written in a manner to be easily adaptable for the Unified Mesh System, so that they are neither "hacks" nor is the work wasted, with them having to do the tools again from scratch on the new system. I'm sure someone can pipe up to let us know if this old fogey's memory is working properly or no. :)

Edit: and I see Rob commented on this just a few posts further down the line. Memory indeed working properly (so far)! :)

prometheus
12-07-2015, 10:54 AM
Nah..letīs dump it all and rewrite a new program from scratch..will only take a few years, LOL.

1988 those were the days it started right? (lightwave) with videoscape that is if that is valid to count in as predecessor? 27 years...thatīs a lot more years than some younger folks have been living just to put things in perspective :)

ernesttx
12-09-2015, 11:20 AM
Waiting for video on this. Thread seems to have died out, so time to revive it. :)

50one
12-09-2015, 01:01 PM
Waiting for video on this. Thread seems to have died out, so time to revive it. :)

Ditto. Bumpity Bump hehe.

What's the progress Lino?

OnlineRender
12-09-2015, 04:19 PM
Ditto. Bumpity Bump hehe.

What's the progress Lino?

When it's been more than 1 day after a blog post http://i.giphy.com/26tPsCKenvnFFUGyI.gif

MAUROCOR
12-09-2015, 07:18 PM
When it's been more than 1 day after a blog post http://i.giphy.com/26tPsCKenvnFFUGyI.gif

That is great!

50one
12-10-2015, 01:20 AM
Why is that I always get a slap on the wrist whenever I politely ask if there's an update coming?:D
Just curious about that modifier stack:)

Every4thPixel
12-10-2015, 07:45 AM
hahahaha!! Lol :D

lightscape
12-10-2015, 07:59 PM
Modo 902 just came out with its easy to use camera matcher and projection tools.

I hope UGE can enable us to finally model in camera.


https://vimeo.com/148503217

djwaterman
12-11-2015, 12:11 AM
Modo 902 just came out with its easy to use camera matcher and projection tools.

I hope UGE can enable us to finally model in camera.

You don't have to hope because it will, once the modeling tools are brought into Layout, which is going to happen.

In the meantime they have put a camera view into modeler so you can do it in Modeler.

Ztreem
12-11-2015, 04:21 AM
Modo 902 just came out with its easy to use camera matcher and projection tools.

I hope UGE can enable us to finally model in camera.


https://vimeo.com/148503217

In the next version we can do that. We have the match perspective tool in layout for matching the camera with the image then you will have the sync camera with modeler so you can model in camera view. Then front Project and bake. done!

pinkmouse
12-11-2015, 04:23 AM
What about lens distortion?

Luc_Feri
12-11-2015, 07:08 AM
Jeez have I been torn between MODO/Lightwave.

I dismissed the recent 20% MODO upgrade and now this.

MODO I used to love but stability got on my nerves and Lightwave did seem simpler/easier, older but more robust.

MODO under the foundry started to be a little showy, more pricey and less convincing and still just as buggy. LW was amazing value, I bought 11, got a free upgrade to 11.5 and some great point releases in 11.6 on towards 11.6.3 with no forum noise, no blog and little showy stuff from LW3DG just great new things delivered and bug fixes.

Now we have a flashy blog, conflicting marketing deals that seem disjointed from the previous months ones and very little substance beeing shown. I'm worried about this side company from Newtek, truth be known.

mummyman
12-11-2015, 07:14 AM
Stick with whatever you can get your jobs done with! If it's Lightwave...so be it. If it's MODO... then cool. If both... even better!

erikals
12-11-2015, 07:23 AM
Now we have a flashy blog, conflicting marketing deals that seem disjointed from the previous months ones and very little substance beeing shown. I'm worried about this side company from Newtek, truth be known.

i'm on the opposite side, i've never had more faith in LWG than now
( even though i'm verbal about missing modeler fixes / updates )

Luc_Feri
12-11-2015, 07:26 AM
Stick with whatever you can get your jobs done with! If it's Lightwave...so be it. If it's MODO... then cool. If both... even better!

I know mate your right totally. I just can't keep upgrading both hoping one will outshine the other eventually. The truth is even just recently I was exporting to Marvelous Designer, making cloth, animating, exporting back but LW was screwing up the original .mdd or something changed. The only fix was to import the seperate objects into MODO, add the .mdd influences and then export from MODO in alembic!!!

I guess we all need more than one main application sometimes as importing/exporting formats is still not 100% robust. :D

erikals
12-11-2015, 07:30 AM
first suggestion, always post the file here and make forum users help you

either they find the solution, or...
- you contact LightWave and ask for a fix
- you contact Marvelous Designer and ask for a fix

as for updates, nothing wrong with skipping 2 updates in the LW world...
i skipped 2015, i might skip 2016, depending on "X" circumstances

Wickedpup
12-11-2015, 07:32 AM
What about lens distortion?



https://vimeo.com/148503214

Luc_Feri
12-11-2015, 07:32 AM
i'm on the opposite side, i've never had more faith in LWG than now
( even though i'm verbal about missing modeler fixes / updates )

Fair enough. I'm only speaking as I feel. As I said before this is not bashing, I prefer LW to MODO for a lot of things and I just hope cracks aren't being covered up and this revolution really is not smoke and mirrors and is what I hope you, me and others want it to be.

In a way I wish The Foundry had mentioned some great new additions were coming in 902 but they didn't and they just dropped them. This was almost like LW3DG use to be, no real news and then boom a tremendous new feature or update.

Luc_Feri
12-11-2015, 07:36 AM
first suggestion, always post the file here and make forum users help you

either they find the solution, or...
- you contact LightWave and ask for a fix
- you contact Marvelous Designer and ask for a fix

as for updates, nothing wrong with skipping 2 updates in the LW world...
i skipped 2015, i might skip 2016, depending on "X" circumstances

No I don't complain or ask for bug fixes that have not been completely ratified and wait until user workflow generated procedure errors have been totally ruled out. Too many 3D folk want bug fixes for things that are actual user errors or just misunderstandings on how a new tool is supposed to be used.

I mainly mentioned that MD issue as way to justify the need for more than one main application at times. It's the reason I bought LW in the first place, MODO could not handle certain things and was not robust enough with the I/O pipeline.

Wickedpup
12-11-2015, 07:39 AM
In a way I wish The Foundry had mentioned some great new additions were coming in 902 but they didn't and they juts dropped them.
They gave several sneak peeks about this in the Direct Connect forum as far back as in August. You can not blame them for your lack of attention. :D

Ztreem
12-11-2015, 07:46 AM
I know mate your right totally. I just can't keep upgrading both hoping one will outshine the other eventually. The truth is even just recently I was exporting to Marvelous Designer, making cloth, animating, exporting back but LW was screwing up the original .mdd or something changed. The only fix was to import the seperate objects into MODO, add the .mdd influences and then export from MODO in alembic!!!

I guess we all need more than one main application sometimes as importing/exporting formats is still not 100% robust. :D

Why not export in alembic from LW?

erikals
12-11-2015, 07:48 AM
if not wrong, i think he referred to exporting from MD to LW...

Luc_Feri
12-11-2015, 07:48 AM
They gave several sneak peeks about this in the Direct Connect forum as far back as in August. You can not blame them for your lack of attention. :D

Haha. I don't log on anymore, my loss totally. I do read the forums lurk style a little bit, I couldn't remember reading about these upcoming changes. Oh well. :D

You need to log in for the direct connect part of forum, ah that's why I hadn't seen it. Still I have been on the fence with 901 as it was buggy/crashy.

Luc_Feri
12-11-2015, 07:57 AM
if not wrong, i think he referred to exporting from MD to LW...

Oh the issue from MD is this, not wanting to derail thread.

I have originally created a figure walk cycle and baked out the .mdd for it inside Layout. Exported the figure .obj to MD and loaded the cache animation. Created some cloth and recorded an animation in MD. Exported the new cloth item from MD with its own .mdd cache file. Reloaded original figure.obj into layout and then loaded my cloth .obj and the corresponding .mdd cache files and the cloth intersects the figure during the walk cycle and it's slightly out of sync.

I've tried and tested this a few times same result. What I found was if I loaded these .obj's and cache files into MODO as .mdd influences it didn't have a glitch at all and worked properly, great. I then saved out of MODO as alembic, loaded this into LW and it worked fine!!! :D

I've tried .fbx from MD and it did not work at all on several tests.

Crazy I know. This is why I said, having two main applications can sometimes give you the workarounds, but I wish I didn't have to.

erikals
12-11-2015, 08:03 AM
then loaded my cloth .obj and it's corresponding .mdd cache file and the cloth intersects the figure during the walk cycle and it's slightly out of sync.
sounds like the mdd simulation is 1 frame off for some reason

try frame offset with value "1" or "-1"
https://www.lightwave3d.com/static/lw/gfx/features/11-5/interchange/image_03.jpg

Luc_Feri
12-11-2015, 08:15 AM
Thanks erikals, I'll have to test later as I have deleted these files as it was for testing purposes only. Thanks again though for a suggestion, maybe I was a little tired the other night I was testing. :D

This is my final nemesis cloth. Ernest Chan created a wonderful script to help me with polygonised hair guides to curves and that went superb with my testing. I can do everything in the pipeline now because MD is a better/faster substitute than bullet cloth and looks great. Octane Render for the animations and now some power can be unleashed finally for my portfolio work.

erikals
12-11-2015, 08:27 AM
glad to help, the LightWave forum rocks, people always help. don't forget... :]

very true, using LightWave with Octane can make stuff shine http://erikalstad.com/backup/misc.php_files/king.gif

Luc_Feri
12-11-2015, 08:42 AM
Cheers for help, it gives me something to try out and sounds possible a 1 frame mismatch, if not for this issue it's a little reminder maybe for features inside of LW. I also like to post issues or WIP's for others to read because maybe they have the same struggles sometimes. :D

Octane combo with LW is great your right.

ernesttx
12-11-2015, 09:27 AM
Will there be a blog update today? Not counting Rob's rather lackluster post, it's been 3 weeks since any substantial posting.

Photogram
12-11-2015, 10:34 AM
Modo 902 just came out with its easy to use camera matcher and projection tools.

I hope UGE can enable us to finally model in camera.


https://vimeo.com/148503217


This is exactly what we need here for our everyday work in 3D!!!

A sculpt tool similar to 3rd power will be welcome! ;)
An axis and projection constrain option for the brush can be a good thing to help adding volume to an organic shape..
Simply by constraining the brush normal to always point directly toward the center of the camera view. ;)

LW3DG Please integrate a texture painter, this will help to fill the texture outside the camera projection.

Cheers!

DrStrik9
12-11-2015, 02:31 PM
maybe I was a little tired the other night I was testing. :D

I think all of us can say the same thing from time to time, so no worries, you're not alone in this. :+)

raw-m
12-12-2015, 08:51 AM
I have a question on the new HVs system (although I'm not sure if this is HVs or an Emitter issue). If you animate the Particle Age along an objects verts, for instance, animate too quickly and you get horrible stepping. Will the new system prevent the "stepping" or blobby issue? I've been struggling with it recently, the only way is to animate slowly and speed it up in post - not always possible for syncing other elements.

erikals
12-12-2015, 09:01 AM
doubt it, you'd need in-between particles

i seem to recall reading some time back DPont had a plugin that created in-between particles...

raw-m
12-12-2015, 09:36 AM
Damn, it's an Emitter thing then, even though the particles are bunched up together. Hope they find time to update the particle system next :)

Thomas Helzle
12-12-2015, 10:24 AM
Hm. I'm late to the show it seems.

Not completely sure yet what to expect, the UGE information is rather superficial so far and feels more like preparing the audience for a not-that-exciting update.

Yes, the architectural overhaul is needed like the pope finally needs a wife after talking about family business for hundreds of years, but I somehow see another year go by without a usable (for me) modeller and that could easily become several until it's fully there (?).

And the new renderer doesn't use the GPU? Hm...

Now I personally have used the best unified modifier stack on the planet in Softimage XSI for years and I also still think it's one of the best subD modellers (with WORKING edge weighting even, wink, wink) which it already was in 2004 when I started using it.

I used Lightwave in one single job this year, still love the renderer, still hate modeller, still hate the split between the two (makes it harder and harder to justify using it at all).
Now from how I read the posts we will see an architectural overhaul under the hood in 2016 which will have very limited actual relevance on the surface in this iteration (correct me if I'm wrong) other than faster tumbling speeds or deformation (not my main problems in Layout)?

I bought into Core in 2009, I bought all updates since, I was happy when Rob took over since I found his approach refreshingly realistic after the lunatic stuff before, I think I am quite patient actually.
But that only goes so far - we're talking 6+ years now with still no unified application but only first steps towards it.

It's also real that all my jobs but one this year were done in Rhino 3D & Grasshopper, rendering with Thea Render which is now fully integrated (including rendering directly in the viewport) and is using GPU & CPU combined while having the most easy and efficient network rendering I've ever used with basically zero setup. It even has bucket rendering for efficient large-scale rendering on the GPU/CPU.

So I'll probably have to toss a coin to decide if I should continue to support Lightwave as a kind of personal pet project and anti-Autodesk-show-of-hands or finally drop into lurker mode and wait and see if it will get back to relevance for my range of work and only update if something comes up that actually needs it.

And yes, the volumetrics look great, but we had that in XSI ages ago as well.

Somehow I miss some real bacon... ;-)

Cheers,

Tom

MAUROCOR
12-12-2015, 11:22 AM
Hm. I'm late to the show it seems.

Not completely sure yet what to expect, the UGE information is rather superficial so far and feels more like preparing the audience for a not-that-exciting update.

Yes, the architectural overhaul is needed like the pope finally needs a wife after talking about family business for hundreds of years, but I somehow see another year go by without a usable (for me) modeller and that could easily become several until it's fully there (?).

And the new renderer doesn't use the GPU? Hm...

Now I personally have used the best unified modifier stack on the planet in Softimage XSI for years and I also still think it's one of the best subD modellers (with WORKING edge weighting even, wink, wink) which it already was in 2004 when I started using it.

I used Lightwave in one single job this year, still love the renderer, still hate modeller, still hate the split between the two (makes it harder and harder to justify using it at all).
Now from how I read the posts we will see an architectural overhaul under the hood in 2016 which will have very limited actual relevance on the surface in this iteration (correct me if I'm wrong) other than faster tumbling speeds or deformation (not my main problems in Layout)?

I bought into Core in 2009, I bought all updates since, I was happy when Rob took over since I found his approach refreshingly realistic after the lunatic stuff before, I think I am quite patient actually.
But that only goes so far - we're talking 6+ years now with still no unified application but only first steps towards it.

It's also real that all my jobs but one this year were done in Rhino 3D & Grasshopper, rendering with Thea Render which is now fully integrated (including rendering directly in the viewport) and is using GPU & CPU combined while having the most easy and efficient network rendering I've ever used with basically zero setup. It even has bucket rendering for efficient large-scale rendering on the GPU/CPU.

So I'll probably have to toss a coin to decide if I should continue to support Lightwave as a kind of personal pet project and anti-Autodesk-show-of-hands or finally drop into lurker mode and wait and see if it will get back to relevance for my range of work and only update if something comes up that actually needs it.

And yes, the volumetrics look great, but we had that in XSI ages ago as well.

Somehow I miss some real bacon... ;-)

Cheers,

Tom

If you are so happy about what you have, why do you bother?!?

Thomas Helzle
12-12-2015, 11:25 AM
If you are so happy about what you have, why do you bother?!?

Because I think giving feedback to the devs is important and better than just fading away.

Cheers,

Tom

50one
12-12-2015, 01:29 PM
Because I think giving feedback to the devs is important and better than just fading away.

Cheers,

Tom


Let me be that click and ask...

....Have they ever listened to your feedback? :)

prometheus
12-12-2015, 01:39 PM
I have a question on the new HVs system (although I'm not sure if this is HVs or an Emitter issue). If you animate the Particle Age along an objects verts, for instance, animate too quickly and you get horrible stepping. Will the new system prevent the "stepping" or blobby issue? I've been struggling with it recently, the only way is to animate slowly and speed it up in post - not always possible for syncing other elements.

you could use the dissolve channel and use local density gradients, that will fade out particles where they are less dense, or use speed gradients..but it will not add more particles in anyway.



doubt it, you'd need in-between particles

i seem to recall reading some time back DPont had a plugin that created in-between particles...
REally? would like to see that...I thought he only made the hvīs with node hack and some blending mode?

- - - Updated - - -


Because I think giving feedback to the devs is important and better than just fading away.

Cheers,

Tom

love your fractals and digital sculptures on your site Thomas.

jeric_synergy
12-12-2015, 02:07 PM
....Have they ever listened to your feedback? :)
It's a valid question: of the hundreds (thousands?) of UI tweeks I've suggested, I can only point to a couple that actually got into the app.

The light-up key-frame indicator on top of the timeline cursor is one of them.

Thomas Helzle
12-12-2015, 02:09 PM
love your fractals and digital sculptures on your site Thomas.
Thanks Prometheus :-)
Not much commercial work on my site since it's all under NDA these days...

----
As for why I write what I wrote: It's just the thoughts that came to me when reading the blog and some of the comments.
Since Softimage is dead (I stopped paying maintenance in 2011) and I don't know how much longer it will work, I'd love to have a fully working Lightwave.
Right now I'm simply unsure what way to go.

Peace, over and out,

Tom

MAUROCOR
12-12-2015, 02:43 PM
Thanks Prometheus :-)

Since Softimage is dead (I stopped paying maintenance in 2011) and I don't know how much longer it will work, I'd love to have a fully working Lightwave.
Right now I'm simply unsure what way to go.

Peace, over and out,

Tom

That is a better answer to me. ;)

lightscape
12-12-2015, 07:41 PM
You don't have to hope because it will, once the modeling tools are brought into Layout, which is going to happen.

In the meantime they have put a camera view into modeler so you can do it in Modeler.

They haven't shown UGE in layout. Nobody knows if its there at all. Ideas and concepts are all there is.
Devs roadmap are subject to change unless you see it actually there.




In the next version we can do that. We have the match perspective tool in layout for matching the camera with the image then you will have the sync camera with modeler so you can model in camera view. Then front Project and bake. done!

The video shows a unified app workflow where its so seamless as expected from any unified app. Modo is actually pretty late to have match perspective. Its been in 3dmax for ages. Even sketchup has one.

djwaterman
12-12-2015, 07:54 PM
[QUOTE=lightscape;1458911]They haven't shown UGE in layout. Nobody knows if its there at all. Ideas and concepts are all there is.
Devs roadmap are subject to change unless you see it actually there.

Wrong. It's been stated quite clearly and repeated a number of times that it is there right now in the next version.

Oedo 808
12-12-2015, 08:41 PM
I bought into Core in 2009, I bought all updates since, I was happy when Rob took over since I found his approach refreshingly realistic after the lunatic stuff before

Careful now, people still worship that ideology.


So I'll probably have to toss a coin to decide if I should continue to support Lightwave as a kind of personal pet project and anti-Autodesk-show-of-hands or finally drop into lurker mode and wait and see if it will get back to relevance for my range of work and only update if something comes up that actually needs it.

Do you think the new upgrade pricing will help with that decision? Assuming it remains as it is.

lightscape
12-12-2015, 10:13 PM
[QUOTE=lightscape;1458911]They haven't shown UGE in layout. Nobody knows if its there at all. Ideas and concepts are all there is.
Devs roadmap are subject to change unless you see it actually there.

Wrong. It's been stated quite clearly and repeated a number of times that it is there right now in the next version.

Have you seen it? Is there another blog we don't know about?
So far what's been revealed since October about UGE is sounding like PMG's marketting campaigns.
Non-lwver's are not buying it in other forums.
The longer they don't show layout with UGE(vertex, edge, poly awareness) the more people are thinking its not there.

erikals
12-13-2015, 02:23 AM
The longer they don't show layout with UGE (vertex, edge, poly awareness) the more people are thinking its not there.
The longer they wait, the better the UGE presentation will be, realistically. so, i'll wait... http://erikalstad.com/emoti/popcorn.gif

Wickedpup
12-13-2015, 04:33 AM
Nah! The amount of wrapping will not change what is on the inside of a present. :bday:

Surrealist.
12-13-2015, 04:42 AM
I think he simply means, the more time they work out the various bugs and other issues that may be in the way of presenting what is there, and the more they work on the presentation itself, the better it will be. As opposed to folding to demands and releasing some kind of lame video with all kinds of excuses as to why such and such interface is not done yet, or ah yeah there is still a bug in this but it is supposed to...

2016 is not even close to being released yet. And there are a lot of things that need to get done before they can move from "we have it working, and it is here" to it is working in a form with interface and bugs worked out.

As to projections, how often and how much, well what else is new? It often takes longer than one would expect to do something.

They are working on it. And also don't forget the Holiday season is upon us these next 2-3 weeks.

roctavian
12-13-2015, 05:08 AM
The longer they wait, the better the UGE presentation will be, realistically. so, i'll wait... http://erikalstad.com/emoti/popcorn.gif


For a perfect presentation we should come back next year.

RebelHill
12-13-2015, 05:35 AM
Have you seen it?.. The longer they don't show layout with UGE(vertex, edge, poly awareness) the more people are thinking its not there.

You wont actually SEE it yet (not in those terms) and thats been made clear, from the blog, as well as clarifications from the odd team member here. There will be no mesh editing in layout in 2016... thats a given now, but layout does now use the UGE for its meshes. So, all you can see for the moment, is a mesh, which will look no different to meshes as they've always been. What's different is the actual code that underpins the mesh's "existence" in layout.

You cant "see" the UGE itself, and the tools its changes will allow will not be in 2016.

erikals
12-13-2015, 07:02 AM
For a perfect presentation we should come back next year.

yep,... in 17 days

jasonwestmas
12-13-2015, 07:02 AM
You wont actually SEE it yet (not in those terms) and thats been made clear, from the blog, as well as clarifications from the odd team member here. There will be no mesh editing in layout in 2016... thats a given now, but layout does now use the UGE for its meshes. So, all you can see for the moment, is a mesh, which will look no different to meshes as they've always been. What's different is the actual code that underpins the mesh's "existence" in layout.

You cant "see" the UGE itself, and the tools its changes will allow will not be in 2016.

Makes me wonder if 3rd parties will be able to make new tools with this new found power right away.

RebelHill
12-13-2015, 07:38 AM
Makes me wonder if 3rd parties will be able to make new tools with this new found power right away.

Interesting wonder and currently an unknown... however, Im going to guess... no. When making plugins for modeler, you cant just pick some poly/vert and do something to it, it has to be done, you have to send that instruction, through modelers mesh edit system, which actually performs the edits you instruct. I suspect that while the new geometry engine will be in layout for 2016, the actual mesh edit components will be coming later (since obv you need the mesh system in place before the mesh edit system).

jasonwestmas
12-13-2015, 08:25 AM
Interesting wonder and currently an unknown... however, Im going to guess... no. When making plugins for modeler, you cant just pick some poly/vert and do something to it, it has to be done, you have to send that instruction, through modelers mesh edit system, which actually performs the edits you instruct. I suspect that while the new geometry engine will be in layout for 2016, the actual mesh edit components will be coming later (since obv you need the mesh system in place before the mesh edit system).

So it sounds like python calls won't have useful access mainly because there is nothing low level and tangible for python to adhere to inside of layout. I guess we'll see if this is even being considered.

hrgiger
12-13-2015, 10:24 AM
Well even if the UGE affects deformation speed and with the addition of a deformation stack, those are pretty significant improvements to layout even without any work being done yet on mesh editing tools (if that is the case).

Chris S. (Fez)
12-13-2015, 12:09 PM
I'd like to see the same heavy deforming mesh in 2015 and 2016. Then upload the mesh so we can "benchmark" viewport performance in Modo and Max etc..

lightscape
12-13-2015, 07:30 PM
You wont actually SEE it yet (not in those terms) and thats been made clear, from the blog, as well as clarifications from the odd team member here. There will be no mesh editing in layout in 2016... thats a given now, but layout does now use the UGE for its meshes. So, all you can see for the moment, is a mesh, which will look no different to meshes as they've always been. What's different is the actual code that underpins the mesh's "existence" in layout.

You cant "see" the UGE itself, and the tools its changes will allow will not be in 2016.

I personally don't expect a full blown modeller inside of layout in 2016.
But people do expect to see layout having awareness of vertex, edge, poly selection even in a locked alpha state. This will just show that its there and being worked on and not all concepts up to now.
"Experimental features" that are locked. Its pretty common.

Deformers aren't really going to impress people looking for a unified app.

Lw 10 came out 2010, lw 11 was out early 2012. They have almost 5 years to work on UGE and unification by now. That's halfway through Modo's lifespan. Things are progressing very slowly compared to Modo development.

Surrealist.
12-13-2015, 08:44 PM
I wouldn't go that far. In 2006 Modo was still very much in its infancy. But I don't think you can compare them blow for blow. LightWave development is on a completely different path than starting a new app out from scratch. And they are also starting out with newer tech which has the potential to surpass a lot of offerings. It is also on a path to offer things that no other app has. It is not about just being unified. There are already plenty of apps for that. People will by and large be attracted to new tools they don't have I think.

lightscape
12-13-2015, 10:52 PM
I wouldn't go that far. In 2006 Modo was still very much in its infancy. But I don't think you can compare them blow for blow. LightWave development is on a completely different path than starting a new app out from scratch. And they are also starting out with newer tech which has the potential to surpass a lot of offerings. It is also on a path to offer things that no other app has. It is not about just being unified. There are already plenty of apps for that. People will by and large be attracted to new tools they don't have I think.


Modo 601 released in 2012 was the point where its a direct competition to lightwave.
And I think they started around 2003-2004. So that's around 8 years.
We're going 4-5 years into lightwave development.

Someone in this forum mentioned that they should've continued Core in the background and by this time it would have been like Modo 301-401 by now.

A different path or not the results is what matters. Its perception, its what people think of a product that matters.
If they think the progress is slow then they will not buy into it, or if they did its too late...RIP XSI.

Lewis
12-13-2015, 11:13 PM
.. And they are also starting out with newer tech which has the potential to surpass a lot of offerings. It is also on a path to offer things that no other app has.

Hmm, I'm curious what do you mean that LW 2016/2017 or whatever it's gonna be called is gonna offer that no other app has and will surpass others? And Please, please don't say "under the hood" :).

Surrealist.
12-13-2015, 11:48 PM
Modo 601 released in 2012 was the point where its a direct competition to lightwave.
And I think they started around 2003-2004. So that's around 8 years.
We're going 4-5 years into lightwave development.

Someone in this forum mentioned that they should've continued Core in the background and by this time it would have been like Modo 301-401 by now.

A different path or not the results is what matters. Its perception, its what people think of a product that matters.
If they think the progress is slow then they will not buy into it, or if they did its too late...RIP XSI.

Actually they started in 2001 so it took them 3 years to release the first Modo which was a Modeling application. It took them another 8 years - total of 11 if we are counting to release 601. I dove into it at that point. The animation tools were not really done yet. That took a couple of more years. So I'd say more like at least 13 years to bake Modo into a competition for other animation programs.

Even with Core, LW 3D Group is moving along quite well, 6 years into the cycle. It might have seemed slow at this point, but the work done is ground work for faster development. It is far too soon to be making comparisons as far as speed. It would be more fair to wait and make that assessment in 1-2 years after 2016 is released.

Lewis
12-14-2015, 12:23 AM
It would be more fair to wait and make that assessment in 1-2 years after 2016 is released.

Well we probably still won't have unified app, Sculpting, proper UV tools, pixarSubDs, fixed CCs, edges support........ in 2016 so what do you think it would be different in comparison then ? Also when is that release ? Q1, Q2, Q3, Q4 ?? And who says other companies won't have their 2016 release by then also ?

So only thing not "fair" is to compare "future" LW version with current version of competition (any not just modo), so we can only compare what's NOW and that is fair.

P.S. Also saying that later updates will be much faster and easier for DEVs later is just talk, there is really no proof of that nor it can be measured until time pass and proves it to be right or wrong :).

spherical
12-14-2015, 12:34 AM
And Please, please don't say "under the hood" :).

+ 1,000,000,000. Or anyone else, for that matter.

50one
12-14-2015, 01:31 AM
Wondering why is it taking so much time to produce a short, few seconds video - showcasing how the modifier stack works or the UGE - how the meshes can be created modified in Layout - just don't get it, since it's already there & working?
Maybe a few screenshots instead, will be easier and faster to produce to cut down the speculation?

Surrealist.
12-14-2015, 02:03 AM
From what I have seen, having had a chance to follow David Ikeda on Facebook, I will wager there is a large gap between "there and working" and there and presentable. Some of his videos were rather crude, but showed that the technology is there, it works. But had yet to be connected to other tools and the final polished interface. He had a very primitive interface that just allowed him to showcase the technology. And that works for him.

Judging by the harsh and unforgiving negativity around here I will further wager that the last thing they are interested in is producing some shoddy example using a temporary interface, and having to explain things away. People would jump all over that.

I imagine the wait is nothing more that working out some bugs, getting a working presentable interface for that feature and putting together a presentation.

Reading anything else into the wait is just not fair. And after all, did we not scream for some communication?

Well we are getting it. Slowly at the pace of real development.

If they break down and show some crude videos people will jump all over that and say the interface is lacking -even if it was explained 5 ways from Sunday that this is not the final interface. People picked he last video apart down to the nth detail in the interface.

This is a tough room. I'd never want to work it.

50one
12-14-2015, 02:20 AM
From what I have seen, having had a chance to follow David Ikeda on Facebook, I will wager there is a large gap between "there and working" and there and presentable. Some of his videos were rather crude, but showed that the technology is there, it works. But had yet to be connected to other tools and the final polished interface. He had a very primitive interface that just allowed him to showcase the technology. And that works for him.

Judging by the harsh and unforgiving negativity around here I will further wager that the last thing they are interested in is producing some shoddy example using a temporary interface, and having to explain things away. People would jump all over that.

I imagine the wait is nothing more that working out some bugs, getting a working presentable interface for that feature and putting together a presentation.

Reading anything else into the wait is just not fair. And after all, did we not scream for some communication?

Well we are getting it. Slowly at the pace of real development.

If they break down and show some crude videos people will jump all over that and say the interface is lacking -even if it was explained 5 ways from Sunday that this is not the final interface. People picked he last video apart down to the nth detail in the interface.

This is a tough room. I'd never want to work it.

Sure it makes sense, but given that they work on UI wouldn't it be nice to see just the viewport capture with some cool stuff happening - I know it's an Alpha tech but still since it's already there then there must be some cool scenes floating around showcasing that stuff, just would be cool to see that.

Dexter2999
12-14-2015, 02:24 AM
Sure it makes sense, but given that they work on UI wouldn't it be nice to see just the viewport capture with some cool stuff happening - I know it's an Alpha tech but still since it's already there then there must be some cool scenes floating around showcasing that stuff, just would be cool to see that.
Deleted

Surrealist.
12-14-2015, 02:38 AM
Sure it makes sense, but given that they work on UI wouldn't it be nice to see just the viewport capture with some cool stuff happening - I know it's an Alpha tech but still since it's already there then there must be some cool scenes floating around showcasing that stuff, just would be cool to see that.

I am with you I'd like to see anything. Most people would. But enough people would just hack it to pieces it would hardly be worth it for LW 3D Group in my opinion.

Let them put something together presentable. They have been listening to us about communication and they are responding in their way. I can hardly blame them not just letting loose with all kinds of half baked coolness.

This is actually the first I can recall that the general public has gotten this much of a glimpse before the fact as to what they have cooking.

Things are looking up in my opinion.

But to some people it is a glass half empty, others a glass half full. I choose the latter.

50one
12-14-2015, 02:48 AM
yup, some folks choose to smash the glass, spill the water, cut their faces with shattered glass and then complain - that's how the interenet works these days unfortunately.

I do agree, it's soo much different from what we had in the past, just being very eager to see more - which is a good thing I guess thatt means their new strategy is working - for me that is haha.

pinkmouse
12-14-2015, 03:30 AM
... And Please, please don't say "under the hood" :).

Yup. It's a bonnet. :)

ianr
12-14-2015, 06:42 AM
Okay Guys,we know we have Chronoscupt tech under the bonnet.fact

I remember in old David i's fairwell posts being referred to as Hydra

Engine,now a defunct code name as the devs have moved on.fact

But I think that what we should carry from his post was ,due to

the positioning of Hydra in the overall code of things, the porting

of tools will be able to happen at a faster genisis rate.

So what do we make of all our speculation, its a good

parlour game. Hopefully LW3dG can pull say 2015 in two

big lifts,the first being layout.I wish them goodwill & a

,little rest over the fesitival season,so they can return

in the new year with a firecracker of The next Rev.

50one
12-14-2015, 06:53 AM
I know what we've got....I just want to see the meaty stuff lol

jasonwestmas
12-14-2015, 07:15 AM
Well even if the UGE affects deformation speed and with the addition of a deformation stack, those are pretty significant improvements to layout even without any work being done yet on mesh editing tools (if that is the case).

yes, performance speed during deformation playback is a huge bonus for any animation technique. Looking forward to this the most probably. Having a lot of control is nice but if the app doesn't have good playback I tend to loose interest fast.

Photogram
12-14-2015, 07:30 AM
Someone in this forum mentioned that they should've continued Core in the background and by this time it would have been like Modo 301-401 by now.
.

I remember before 2000, the programming of XSI 1.0 took 7 years!
Today's with more modern toolkit i think it is faster!

50one
12-14-2015, 07:54 AM
I remember before 2000, the programming of XSI 1.0 took 7 years!
Today's with more modern toolkit i think it is faster!

Not really, I think the law where more demand is generated despite the more power put into users applies to programming as well( can't remember the guy name - it's not Moore) - thanks to more features and more complicated algorithms - 10 years ago Path tracing, ambient occlusion were not that embedded into industry as they are now, not to mention PBR and few other gems that appeared like 3 yrs ago.

probiner
12-14-2015, 12:25 PM
But to some people it is a glass half empty, others a glass half full. I choose the latter.
Some say there is no glass, others that is the second coming. So, like Judgment Day, we have to wait, and by then it will be undeniable. :P

I'm sure something nice will come along, we just can't grasp now, how it will mature and impact the whole application. :)

Surrealist.
12-14-2015, 01:57 PM
Well what I am referring to specifically is that in order for LightWave to grow from here they have had to do a certain amount of groundwork. It took them 4 years working on this specific project of the UGE and apparently the rendering to arrive at a place where some serious changes can take place. With the added benefit of what is some new groundbreaking tech that grew out of Hydra.

And there is really no way to avoid this passage. They have to come to this place and move on to the other side. It is more or less a half way point if you will. And to some that is half way there, others it is still half way behind.

I see it as half way there and moving into something potentially different that other offerings once it arrives near its final destination in say 3-6 years. Some will say that is not soon enough. I say it is inevitable. Why does it matter what we think. They have to move forward.

Line from one of my favorite movies. "You have to get busy living, or get busy dying".


https://www.youtube.com/watch?v=7tkzc983aE0

Paul Brunson
12-14-2015, 02:58 PM
I've seen these same type of conversations take place on other companies forums across the internet in many industries. I've come to believe its because customers are looking for entertainment. In this case, they're looking to Newtek to excite them.

The trouble is creating exciting products and features for the developers means pushing through a lot of mundane and challenging stuff. Its just not realistic to expect a single software company to provide a constant supply of entertainment and exciting material for us.

But at the same time I think its important for a company in Newtek's (LW3D Group's) position to recognize that their customers are looking to them as a source of entertainment and excitement. The more excitement they can provide on a consistent basis the better it is for them. Even programmers are in the business of entertaining people these days. :)

jasonwestmas
12-14-2015, 03:54 PM
I've seen these same type of conversations take place on other companies forums across the internet in many industries. I've come to believe its because customers are looking for entertainment. In this case, they're looking to Newtek to excite them.

The trouble is creating exciting products and features for the developers means pushing through a lot of mundane and challenging stuff. Its just not realistic to expect a single software company to provide a constant supply of entertainment and exciting material for us.

But at the same time I think its important for a company in Newtek's (LW3D Group's) position to recognize that their customers are looking to them as a source of entertainment and excitement. The more excitement they can provide on a consistent basis the better it is for them. Even programmers are in the business of entertaining people these days. :)


"Enterainment" is too generic and childish a term to describe what most users on this forum are looking for. IMO I've read enough posts to understand that these users are more interested in the technology (and not mere sensationalist entertainment perspectives) and how it will help them keep using Lightwave more often and in more ways than ever before. Seeing as it is one of their favorite packages for a multitude of reasons it is exciting to know that Lightwave will open more door for them in the future and not just slap on more band-aids like in the past. Not to mention we have been waiting a long *** time for something like this to manifest. So yes it is exciting in that respect and I can understand the impatience.