PDA

View Full Version : Should Layout / Modeller be integrated?



Matt
02-09-2003, 10:17 AM
Just wanted to start a thread to check out the Poll feature, this seemed like a good topic for it!!!

:)

Elmar Moelzer
02-09-2003, 11:19 AM
You are missing a compromise, like seperate, but with better comunication, or a masterapp that connects them so they can do everything a unified app can do but keeping the feeling of seperate apps.
There was a better discussion on this in the old forums, that actually asked people for what they imagine it to look like.
What I found weird was that there were not that many posts in it, as some ideas in it even appealed to me (and I am a seperatist).
My advice would be to get this thrad going again and then do a poll asking people which idea they like best. How about that?
From the options you gave above I would rather choose to keep it seperate though...
CU
Elmar

Castius
02-09-2003, 04:26 PM
The biggest thing that is needed is a setup mode that is independant of keyframes, and has the power and tools of modeler. This is not and easy and smooth task with the current hub setup. I love the two apps apart with a hub. But the current direction of adding tool to connect them with a plugin here and a skelgon there is really putting to many bumps in the workflow and hurting lightwave. they need a real setup mode in layout if they choose to keep them apart. With reall nice muscle bones and curves/curvetools that are usable.

Skonk
02-09-2003, 06:45 PM
Keep them seperate but improve the hub, u dont want to have all that extra system resources used up when u dont need it cos ur just animating OR modeling, if u put them into one but do it in a way that keeps un-used tools out of the memory then switching to them would require a loading time so its no diff then using 2 apps. The only problem is that the comms between them at the mo isn't always very good (tho iv never had any problems) so thats all id want changed.
There really is no reason why using 2 apps hooked together with a third app(the hub) cant work just as well as 1 intergrated app, but its just not perfect at the mo. There are no limitations that 2 apps have compared to one single app if its done right.

Only MHO :)

James..

Prosprctor
02-09-2003, 09:27 PM
OK..this may come as a surprise to some because I wanted a Federal law passed to keep them seperate..but I have come around to the otherside..
Merge them !!!!
Just as long as there are NO goofy icon pictures of some strange hiyroglyphics
And the GUI asks if we want layout opened
modeler opened or both
And the first release of this new animal is free so we can see if we want it or change it back before the official release.

faulknermano
02-09-2003, 10:04 PM
i voted "whatever".:p

DaveW
02-09-2003, 11:48 PM
I definately want them merged. I like it seperate so I'll live if it stays that way, but I think that merging them is the best way for LW to move forward and improve workflow.

Having them merged may use up a little more memory when you want to just animate or just model, but for all those times you switch between the two you're using up a lot more RAM because you need to have all the models and textures loaded twice. Wether merged or split, you run into situations where memory is being wasted but I think overall it will be more efficient to have them merged. And as for the hub, yeah it definately could be improved (synchronization with Layout should be realtime!) but that alone won't make up for all the shortcomings of having them seperate without lots of duplicate code in Layout and Modeler, so that would be more wasted memory.

I also like prospector's idea about loading the modules seperately; that would save memory and you could still have your modeler and layout shortcuts with a switch added that launches only modeler or only layouts module, or if you wanted you could have a single shortcut that launches them both together.

Alec Trevelyan
02-10-2003, 12:58 AM
DIVIDE AND CONQUER!!!!

Jockomo
02-10-2003, 01:32 AM
My vote is No, but I too believe that there needs to be more consistant communication between the two. I like the dual monitor setup, but I want them to act like they are the same application.

Just to reiterate my other thread... You should not have to save your objects just to jump from layout to modeler. Don't force me to save over something until I am ready to do it.

DigiLusionist
02-10-2003, 02:30 AM
Steve, what!!!??? YOU???

I never thought I'd live to see the day when you'd even consider integration in LW. What the heck made you change your mind?

By the way, my vote is: Integrate the hell out of the code. The interface can stay basically the same. Just make both apps run off of the say code.

How's about we use the term Integrated Code Environment, instead of Integrated Interface Environment? The integrated interface environment concept really freaks some people out.

Also, when the app code is integrated it can be programmed so those who want an integrated interface environment can customize the interface to look and behave as one app with toggled modes, while those who want LW to remain looking circa 1864 can set it up to look the same as it ever was.Integrate the hell out of the code.

Red_Oddity
02-10-2003, 05:57 AM
Merge 'm,

i can live with it should they stay seperated, but i do mis a LOT of features that would be possible if the 2 where merged.

to name a few:

-Image modeling (especially necessary for good compositing or tracking meshes over footage)
-Weight painting/tweaking when character has been setup
-Clips/texture and other references stay the way they should be than (even when you save them in the Modeler..unlike it does now)
-Animated meshes (and don't come with the Endomorph excuses again people), like extrusions, points, booleans, etc.

These are but a few, but the list goes on trust me...i find myself swithing to *cough other software cough* too many a days now for the more expensive clients as they wish things that can only be done in a fashionable time in another package but LW (i don't want to loose LW or use it exclusively for modeling in a year or so...), and yes, you can start calling me names now and tell me it's the person behind the program, not the program...bring it on.

It think it is a necessary step forward into the future.

Skonk
02-10-2003, 06:19 AM
There is nothing that cant be done with 2 programs that can be done with 1, there are no limitations to how 2 programs can interact with eachother if its implemented right. They can even share the data stored in memory if u wanted, an object loaded into memory could be displayed in both layout and modeler, both sharing the same resources. Its just that how it stands now there is a lot of room for improvement.

James..

Prosprctor
02-10-2003, 08:53 AM
1864?
nah too far back...
That was the era of Lightripple 1D.
:D
Just don't put those silly icons and I'm OK

DaveW
02-10-2003, 03:13 PM
Originally posted by Jockomo

Just to reiterate my other thread... You should not have to save your objects just to jump from layout to modeler. Don't force me to save over something until I am ready to do it.

You don't have to save your objects to jump from Layout to Modeler, I switch back and forth all the time without saving the objects so I can preview changes. You just need to save the object before you close Modeler or Layout loses any changes you made.

SpankDaddy: Yeah you can make the two apps communicate better so that they act like one program, but there would have to be a lot of duplicate code in both apps. You can also make a single program act more like two seperate programs but still have the benefits of a combined program and more efficient code.

Skonk
02-10-2003, 04:26 PM
I suppose i just dont want lightwave to end up looking like all the rest, its been 2 apps since the start so why change it? :)

James..

DaveW
02-10-2003, 09:29 PM
I don't want it to be a clone of the others either. So why change it? To improve the workflow and extend the capabilites of the two apps as a whole. Since character animation is my chief interest the main reason I want them merged is to make rigging characters less of a pain. The plugins to help resolve rigging issues are getting better but they're band-aids to solve a problem that we shouldn't have in the first place. If the apps stay seperate, then either Layout will need to implement a bunch of Modelers tools (selection, move, rotate, drag, vmap editing, ect) or Modeler would need to add a bunch of Layouts tools (bones, expressions, motion modifiers, ik). Those are just my personal wants; other people have other reasons for wanting them merged like animated modeling or point level animation.

Jockomo
02-10-2003, 11:36 PM
Originally posted by DaveW
You just need to save the object before you close Modeler or Layout loses any changes you made.


Why? That's all I want to know. Can anyone give me a valid reason for this? Alot of people seem to want to defend this point, and I have yet to see a single reason for it to do this.

I can understand losing changes you do in modeler to the object if you don't save. But why should it affect changes you made to the surface in layout?

DaveW
02-11-2003, 12:30 AM
Because the surfaces are part of the object, not the scene.

Anyways, my point was that you can freely switch between Layout and Modeler without saving changes to the object, which you said was not possible. I can understand why you would want your surface settings to stay but the reason they don't is because if you make a bunch of changes to the object (not just surface changes, but any changes) Layout needs to get the updated version version from Modeler so that the object remains consistent between the two programs; if you close Modeler or the object you're working on without saving your changes then Layout needs to get the latest version of the object from somewhere, so it gets it from the last saved version on the disk. Object changes need to be saved before exiting Modeler, scene changes need to be saved before exiting Layout. Seems pretty logical to me.

Jockomo
02-11-2003, 01:56 AM
OK, well it's starting to feel like a dead horse, so this is the last post I'll make about it.


Since you still cannot run Viper in modeler alone, or render to see all of the options that can be changed in surface settings, it still makes sense (to me anyway) that surfacing is not an issue that should be affected by closing modeler. Regardless of the fact that the surfaces are saved with the object, surfacing is really not a modeler function. (yes I know you can access the surfacing panel in modeler, but you can't effectively work with it)

Therefore closing modeler without saving your object should not affect your textures in layout any more than it should affect your keyframes in layout.

Well the bright side of this is, after lots of testing and debate, I finally found a way to deal with this problem. If I want to close modeler, not save my changes in modeler, and not loose my texture in layout (you know, like if you accidently hit the modeler launch button, when you mean to hit the zoom button) there is a solution. Ctrl+Alt+Del and shut down modeler via the task manager.

Jimzip
02-11-2003, 04:40 AM
Hey, why don't we be happy with what we've got, and not complain about wanting it like other crappy slow applications.
You're screwing up an incredible piece (or pieces as it may be) of software by changing it.

Jimzip.:)

pablof79
02-11-2003, 05:18 AM
If it ain't broke, don't fix it! :mad:

Qslugs
02-11-2003, 11:09 AM
Without integration you're missing out on the ability to do all sorts of high level vertex based animation. I feel that all this resistence against integration, is due to the fact that since Lw has been one way for so long, nobody really knows what they are missing out on.

DaveW
02-11-2003, 12:48 PM
Originally posted by pablof79
If it ain't broke, don't fix it! :mad:

I agree. But since it *is* broke, it needs to be fixed!

Matt
02-12-2003, 02:07 PM
1864?
nah too far back...
That was the era of Lightripple 1D.


LOL! :D

Panikos
02-12-2003, 02:20 PM
I think that some simple modeling tools in Layout are essential, as well as some Scene-Loading tools in Modeler.

Merging them together, requires almost rewriting of everything,
change of LWO, LWS formats, incompatibility with existing plugins

and ...
the most important : You raise the risk of bugs exponentially,
and their fixation a total chaos.

Not to mention that I prefer a more powerful rendering engine, without Modeling set occupying RAM with no reason.

PHilly[Dee]
02-12-2003, 03:01 PM
Originally posted by Matt
Just wanted to start a thread to check out the Poll feature, this seemed like a good topic for it!!!

:)

-nice thread topic... :p

hrgiger
02-12-2003, 03:15 PM
I personally like having modeler and layout seperate but I wouldn't care either way if they integrated the two. What I wouldn't want is for an integrated app that uses more system resources just to run nor would I want an interface with more drop down lists and menus then we already have.

DaveW
02-12-2003, 03:49 PM
Originally posted by Panikos
I think that some simple modeling tools in Layout are essential, as well as some Scene-Loading tools in Modeler.

Merging them together, requires almost rewriting of everything,
change of LWO, LWS formats, incompatibility with existing plugins

and ...
the most important : You raise the risk of bugs exponentially,
and their fixation a total chaos.

Not to mention that I prefer a more powerful rendering engine, without Modeling set occupying RAM with no reason.

The file formats wouldn't have to change at all unless NT added edge support (which they should). As far as rewriting goes, I'm no programmer so I can't say much about that, but quite a few seem to think a total rewrite is needed anyway. Incompatibility with existing plugins would suck for a while, but at some point you just need to get rid of all the legacy crap so that you can move forward. I suspect they could even program it so that some of the old plugins could still work, but as I said, I'm no programmer so I don't really know how much work that would be. And yeah, potential for lots of bugs like the 6.0 release, but they got ironed out pretty quickly and I think that in the long run it would be worth it. Of course putting parts of Layout's code into Modeler would likely add a lot of bugs as well. Layout could get away with using new code for the modeling tools, but that would have new bugs, and I don't know if it would save any time vs. grabbing some of Modeler's code and sticking it in there. It would take longer to merge them and there would be lots of workflow issues to figure out, but I think it needs to be done at some point and the sooner the better.

hrgiger
02-12-2003, 04:17 PM
Why does it have to be done at all?

Wouldn't it be much simpler if just certain tools worked in both Modeler and layout?

DigiLusionist
02-12-2003, 05:21 PM
I would like to hear from 3D programmers about this issue. Any out there? I never see any of you guys commenting about the practicality or viability of the suggestions made on the forum. I'd be very interested in hearing what you have to say.

There seems to be two general philosophies about integration with community members: do nothing and live with the application's limitations (cause, garsh darnit, we like it old skool!), or do something about the limitations even if it means (gasp) having to learn some new stuff about the resultant application.

There will always be something more to have to learn in order to stay current and competitive as a 3D artist. Simple fact. That means being OPEN to learning. LW simply needs to be updated in order to make it even more practical for production-level work. If by doing so, we have a few more things to learn (or unlearn), so I say, so be it.

I think it's natural for advanced users to want to push the envelope and to be able to do more things with an app. Employing workarounds is acceptable when an artist WANTS to be creative, but when an artist HAS to use workarounds to do what other apps are already integrating into their basic functionality, then I perceive that as a limitation that needs to be addressed.

Elmar Moelzer
02-12-2003, 07:26 PM
Hello, well we have done some prgramming here. A lot of LW- plugins and a few plugins for Digital Fusion too.
Still those questions are not easy to answer.
One thing is for sure: A total rewrite would take long. I mean long, like two years min. , more likely to take longer.
If it would be done, I would go for a complete C++ based core and drop the current C- based core, if I was NT. The SDK could then possibly be better integrated and not just attached to the core as it is now. I think this would make a lot of things much easier.
Now planing such a long time ahead is a bit risky, because you have to invest a lot and you will never know for sure whether the outcome will meet the demands of the times to come. Having a ready core is good (until it becomes insufficient)as one can add new features quickly.
I just want to remind you of the debacle that nearly happened to XSi. Here the complete rewrite took even longer, partially because of all those owner- problems, but also because they changed it a few times to meet new demands that arose during the dev- time.
Well, a merge would most certainly require a rewrite (though it is hard to say without knowing more facts about the code), but I could imagine that one could possibly get very far with the current code- base and an extended HUB and some doubled features (again I cant say for sure without knowing the code).
So dont imagine everything to be to easy. With complex software like LW you cant do a rewrite within a few months, no matter how big your dev- team is. You cant just say like 1 programmer will need x months and 10 will need x/10 months...
It does not work that way.
Regarding the plugins: If I was NT and I was to do a complete rewrite, I would most certainly drop the SDK as well and make a new completely C++- based. This would of course mean that all old plugins would require to be redone to some extend. Though AFAIK most of the plugin- developers code in C++ and use a C- wrapper anyway... Still they would need to make some changes and the original *.p- files will not work. So IMHO a complete rewrite would mean a complete restart from 0. For NT and for the users as well.
Maybe something worth thinking about...
CU
Elmar

Karmacop
02-12-2003, 08:57 PM
I'd hate to say anything, but just because Modeler and Layout are merged does not mean that Lightwave suddenly gets animated point editing and extrusions.

DigiLusionist
02-12-2003, 10:57 PM
Yes, Elmar. Those are some pretty serious time and resource commitments to invest into making LW better. But I think LW's worth that kind of investment. If possible, a concurrent development team could work on doing so as the current version is supported. However, this may not be financially feasible for NT to do, I understand. And the plug-in developers issue is an important consideration, too.

I suppose the bottom line is: will there be longevity with LW for NT if the current app code does not accomodate continued functionality development that is on par with the functionality development of other apps?

I use LightWave, Maya, and Max. Because I do, my perception of LW's workflow and power are influenced by this exposure to other apps. As a result, I see ways LW can and should be improved. The fact that it would be a burden to NT to make these changes doesn't make these shortcoming disappear.

I read other viewpoints by other LW users and I can see why they feel the way they do. My take on things is that I don't usually worry about what bad things MIGHT happen. My opinion is that anything can be done if the decision is made to do it. And, of course, that's just my opinion. It ain't the Gospel truth, to be sure...

jin choung
02-13-2003, 01:18 AM
wow,

can anybody just make a poll?

cooooool....

jin

faulknermano
02-13-2003, 01:43 AM
Originally posted by DigiLusionist
Yes, Elmar. Those are some pretty serious time and resource commitments to invest into making LW better. But I think LW's worth that kind of investment. If possible, a concurrent development team could work on doing so as the current version is supported. However, this may not be financially feasible for NT to do, I understand. And the plug-in developers issue is an important consideration, too.

you also have to take into consideration that if the code was re-written from the ground-up once more, we're not only talking about an increase in the number of potential bugs, but the overall reputation of lightwave as a 'workhorse.' i am speculating there is a big risk in rewriting everything.


I suppose the bottom line is: will there be longevity with LW for NT if the current app code does not accomodate continued functionality development that is on par with the functionality development of other apps?

i think the slew of great plugins say yes to this question. i think what people want is the tools to be more 'native.' at least some of them want that.



I use LightWave, Maya, and Max. Because I do, my perception of LW's workflow and power are influenced by this exposure to other apps. As a result, I see ways LW can and should be improved. The fact that it would be a burden to NT to make these changes doesn't make these shortcoming disappear.


i think of all this as an evolution. everybody keeps on asking for the 'big bang' or something. i think it's better for lw to evolute. i think, though i do not know for a fact, that the current code can be extensible enough for more improvements, in terms of SDK and native tools.

anyway, that's the end of my boring harangue. :D

Lightwolf
02-13-2003, 11:36 AM
Originally posted by Elmar Moelzer
Hello, well we have done some prgramming here. A lot of LW- plugins and a few plugins for Digital Fusion too.

:) So have I, so I guess we made the same experiences...


Originally posted by Elmar Moelzer
Still those questions are not easy to answer.
One thing is for sure: A total rewrite would take long. I mean long, like two years min. , more likely to take longer.
If it would be done, I would go for a complete C++ based core and drop the current C- based core, if I was NT. The SDK could then possibly be better integrated and not just attached to the core as it is now. I think this would make a lot of things much easier.


I would go for a completely new core and C++ based SDK as well, even though it opens up some can of worms, in the long term it will do the future development good.
From what I read in various newsgroups though, the dev team (well, some of the guys at lux, as well as Arnie Cachelin), don't really like the idea of C++ and will probably stay with C in the future. IMHO a huge mistake.


Originally posted by Elmar Moelzer
I just want to remind you of the debacle that nearly happened to XSi. Here the complete rewrite took even longer, partially because of all those owner- problems, but also because they changed it a few times to meet new demands that arose during the dev- time.


True, look at PmG, same thing, just smaller :) But look at XSI and the power behind it. True, the ui isn't that amazing either, but it is extremely flexible from a users point of view, and I think that the choice of programming language has an influence on the app..


Originally posted by Elmar Moelzer
Well, a merge would most certainly require a rewrite (though it is hard to say without knowing more facts about the code), but I could imagine that one could possibly get very far with the current code- base and an extended HUB and some doubled features (again I cant say for sure without knowing the code).


Well, when lux came out, I thought that there were 2 dev teams, one for legacy lw, one for "mecca". But since the current situation isn't clear, a true next gen (i.e. rewritten from the ground up") lightwave is pure speculation.

But, on the C vs. C++ subject I agree with you. My second DF plugin I wrote, I ported over to LW (an image filter). The DF plugin took me 2 hours to code, and I had to suss out the algorithm as well. The LW counterpart took me 4 hours, basically just for user interface and load/save stuff.
Maybe a decent C++ wrapper for the LW API would help us (imagine, automatic UI creation and loading and saving of variables, object id tracking etc....) Oh well, whish I had the time :)

Cheers,

Mike

DaveW
02-13-2003, 12:11 PM
Originally posted by hrgiger
Why does it have to be done at all?

Wouldn't it be much simpler if just certain tools worked in both Modeler and layout?

It might be simpler. But they would need to add a bunch of Modelers tools and change the way a few things in Layout work. They would need selection modes, action centers, drag, weight map and endomorph tools. For people who want animated modeling you'd need to add pretty much all of Modelers tools.

And what about adding multiple undos? Wouldn't that call for rewriting a significant amount of Layout's code? And I have heard from someone at NewTek (I think it might have been Arnie) that adding edge support would require a new model format and a rewrite of Modeler.

So if it's going to take so much rewriting to get the features we need, why not do it right and do a complete rewrite/merge rather than rely on more duct tape?

Panikos
02-13-2003, 01:51 PM
Its easy for all of us to make assumptions, guesses.

Only developers have a complete idea of what "total rewrite" means.

Remember how solid was LW5.6 and how fragile was "total rewrite LW6.0".

It took a long time until 6.0 evolved into the current state, 2 years or more ...

Considering the trends in 3d-industry, the new technology,
the competition, financial flexibility of Newtek, the rewriting from scratch is a very serious decision.

There is a main core of LW which is very powerful.
Imho, polishing this, improving some weaknesses and adding some new features is the wisest policy.
I am too naive to imagine how long and complex is the source-code of LW. Only typing the code requires very long time.

Look at pmg, they started from scratch.
They started with a very powerful animation system and luckily they built on mistakes of old-technology.
But still pmg-studio is in progress ...

spr0g
02-13-2003, 05:05 PM
this is one of those things that will have the community split in oppinion...

one half likes the idea... the other doesn't... and those that dont fall into these categories... ur... just... well.. awkward! hehe

I gotta say tho... it CAN have its advantages... and it CAN have its disadvantages... I think its fine the way it is... newtek have decided on this for a good reason whatever that reason may be... and most of the users seem to like it like that.

my 2 cents

Chris S. (Fez)
02-13-2003, 05:14 PM
Though I am all for integration, this does not necessarily mean that Newtek must nix the Hub. The hub is a form of integration, even if it is severely lacking. I HATE the HUB in its current incarnation.

But...seeing as Newtek just released the Aura/Lightwave connection, it seems reasonable to speculate that the Hub is here to stay. Fine. If Newtek feels the HUB is the way to go, fine. But I have been fighting the HUB almost daily since 6.0 was released (and no, not on the same computer!) and I am getting tired. If Lightwave 8 features a HUB even half as buggy and slow and limited and cumbersome as the current incarnation, I will be...um...realllly dissapointed. :)

Rory_L
02-13-2003, 11:42 PM
I`m not going to vote:eek:

Panikos was close to answering the Why? I`m an artist, as are most of us here. I can clearly see the things I like about Lightwave and also have an idea that it could be improved in some areas. However I am not in any respect a programmer, nor, obviously, am I a Newtek developer. The only people in a position to say which road is the best to take for our much loved application are those LW developers.

By now we`ve all said enough about what`s wrong with LW and how we`d change it. Now we just have to sit back and be patient. The next version of LW will be better than this one: have faith and count yourselves lucky you`ve already got such a good 3D app. as 7.5!

R

DigiLusionist
02-14-2003, 12:27 AM
Actually, Rory, users' involvement and input into the development of a product is indeed important. Otherwise, there wouldn't be customer support or customer feedback departments in companies.

Automobile manufacturers, for instance, rely on polls and surveys to decide what features to add to their models, and even what models to continue to produce.

Newtek wouldn't have Feature Request forums if they didn't know how vital user input is to their products.

You are right: LW 7.5 is the best incarnation we've had so far.

Elmar Moelzer
02-14-2003, 03:10 PM
Hello Lightwolf!
Hmm we have actually been thinking about realeasing our Wrapper and the Templates (say: "one line- plugins") to the public one day. I will let you know when.
Gruesse nach Stuttgart!
CU
Elmar

Rory_L
02-14-2003, 08:25 PM
Actually, Rory, users' involvement and input into the development of a product is indeed important

DigiLusionist, YES! :) I agree, but I think my second point was hinting, (not clearly enough, obviously) that we`re just going round in circles now. We HAVE let Newtek know our feelings and ideas for improvements. This was important and it`s done! Now we should sit back and let the developers implement those improvements in the way they see best.

Pointing out the goal to them was good; stipulating the route is bad! OK?

I`m not voting partly because there isn`t an appropriate choice to check for me. I don`t want to check NO or YES and I certainly ain`t going to check NOT BOTHERED! I care strongly about Lightwave; just, for lack of inside knowledge, can`t see which way to go.

If Matt were to edit the choices to include

Whichever, just make it great!

I`d vote:)

R

DigiLusionist
02-14-2003, 09:30 PM
I understand now, Rory. Sorry for jumping the gun.

PHilly[Dee]
02-14-2003, 09:32 PM
It's ALL GOOD....

wiremuse
02-15-2003, 02:07 AM
I do see advantages to them being seperate. However there are a lot of bad things about it as well. The lack of communication between the apps. The lack of proper boning and animation tools available inside modeler. However I think we can keep those who want it merged and those who want it seperate happy.

To get to the point I would like it so they are built around each other in a more unified fashion say using a shared libraries so each app can have access to the others functions or be able to pull up windows from the other etc. and only having to load what parts it needs into memory or run when the user wants it. While still being able to keep the UI Apart or even together if another front end was built around it and users would be able to break it down into a selective and customizable interface.. A more unified workflow like this would increase productivity to a great degree. I think one good example of something some-what similar is how the Softimage|XSI Interface works. Absolutely grand methodology.

wiremuse
02-15-2003, 02:10 AM
Originally posted by CIM
The smart thing to do is to integrate Lightwave; not waste time making some kind of super-duper Hub. However, Newtek/Luxology would have to avoid the slowness/lag that plagues Max and Maya. XSI is the only program that is successfully integrated.

Newtek/Luxology must re-write Lightwave to live up to todays standards, so why not get rid of the Hub and go all-out! That way Lightwave (or whatever it'll be called) can be unified and will work as a whole.

I agree 99%

I supose a complete rewrite would depend on how modular the code is. However it does seem quite certain that some areas need a full overhaul and complete rewrite and others where I think a fair amount of code reuse is safe.

Jimzip
02-15-2003, 02:16 AM
Dear Matt,
Thank you for this topic.

Dear others,
read it, and please NEVER POST IT AGAIN!

Jimzip:D

Afalk
02-15-2003, 12:53 PM
I've worked with Max, Maya, and Lightwave, and am a
dedicated Lightwave guy.

The seperation of Modeler and Layout is fantastic for
organization, and the HUB makes everything work pretty
well.

I would love to see tighter integration between
Modeler and Layout, and a smarter HUB would be
great!

The only peeve I've ever had w/the seperate apps
is the Interaction Tools -->> i.e., what you need to
do to move around in 3D space in modeler vs. layout,
and the shortcuts that power those moves EX:
I'd love to see the addition of the lock view button
in Modeler (ala auto-focus from layout) and the
autofit ("a") command in Layout.

I think these things, as well as the much lamented
lack of vertex mainpulation in Layout are all possible
with tighter integration of code beneath the hood.

Just my .02

Tony

TerryFord
02-15-2003, 03:40 PM
I voted for integration. The hub seems like a kludge to me; clumsy and slow.

I don't think we'll see an integrated LightWave though, I believe that it's two aps for historical and legal reasons (two different copyright holders). That's just supposition on my part though, I'd like to be wrong.

As for edges support requiring a modified model format, what's so terrible about that? Although I have a huge amount of respect for the LW development team, changing the .lwo format seems to be anathema to them for some reason, considering how long it took to add UV/vertex mapping support.


Regards,
Terry

Gregg "T.Rex"
02-15-2003, 04:52 PM
I'm currious...... What whould be easier and better to do?

Take all the animation and rendering capabilities of Layout and throwing them into Modeler?
Or, making Modeler work INSIDE Layout, among the other tabs Layout have?

I vote for the second option, ONLY if Modeler's speed and easy-of-use is NOT sacrificed (much:D), for the sake of integration.

Of course, leave them as they are now, is not a step forward, me thinks...

After all. Maya and XSI do have "different" tabs for Model and Animation. The difference is that they dont switch to onther app window and they have all the animation and rendering arsenal also when Modeling things.....

Uhmm.... I wonder what happened to that nice place called... Mecca?:)

faulknermano
02-15-2003, 09:42 PM
Originally posted by Gregg "T.Rex"
I'm currious...... What whould be easier and better to do?

Take all the animation and rendering capabilities of Layout and throwing them into Modeler?
Or, making Modeler work INSIDE Layout, among the other tabs Layout have?

I vote for the second option, ONLY if Modeler's speed and easy-of-use is NOT sacrificed (much:D), for the sake of integration.



dunno what's easier. but i'd like the feel of modeler better than layout. i also prefer, without doubt, the navigation tools of modeler, though i prefer to be able to swtich navigation 'modes'.

Rory_L
02-16-2003, 10:17 PM
Vertex manipulation IS already here in Layout: it`s called Weight Maps!

R

faulknermano
02-17-2003, 12:12 AM
Originally posted by Rory_L
Vertex manipulation IS already here in Layout: it`s called Weight Maps!

R

best flame bait i've seen thus far in this forum. :D

i think what is meant is to be able to manipulate vertices as easily as we do in modeler.

faulknermano
02-17-2003, 12:20 AM
you know what, even if we did have an interface to select point in layout, the movement / navigation behavior in layout is so ungainly that it wont really matter. have you guys tried what's it like move bon-vertices in layout in perspective view? ugh.

Rory_L
02-17-2003, 12:45 AM
best flame bait i've seen thus far in this forum.

Hehe! You`re so right, but I can`t resist the inner demon!:D

...and I really should try harder! I hate people who defend the present LW to the death, by pointing out complex ways of getting functionality, instead of agreeing a change would be better...and there I went and did it myself! Bad Rory!:)

R

Lightwolf
02-17-2003, 02:41 AM
Originally posted by Elmar Moelzer
Hello Lightwolf!
Hmm we have actually been thinking about realeasing our Wrapper and the Templates (say: "one line- plugins") to the public one day. I will let you know when.

Hallo Elmar,

That sound very cool.

I've been thinking about something in-between, currently even done in c, to at least get some of the typing out of the way.

I wish at least LScript would make writing plugins easier, but it doesn't really.

Cheers and Gruessa nach Graz,

Michael (Light-) Wolf

faulknermano
02-17-2003, 02:44 AM
Originally posted by Rory_L
Hehe! You`re so right, but I can`t resist the inner demon!:D

...and I really should try harder! I hate people who defend the present LW to the death, by pointing out complex ways of getting functionality, instead of agreeing a change would be better...and there I went and did it myself! Bad Rory!:)

R

i'm something like that, in a way.. (defending lw). but less of 'defending' as opposed to just offering an alternative workflow. sometimes i do not know whether a person is simply asking for a feature request or is actively looking for a solution that's why he / she is bringing it up in the feature requests forum.

labuzz
02-17-2003, 05:23 AM
. but i'd like the feel of modeler better than layout. i also prefer, without doubt, the navigation tools of modeler, though i prefer to be able to swtich navigation 'modes'.

I agree 100%

JMarc
02-18-2003, 01:57 PM
Originally posted by DaveW
You don't have to save your objects to jump from Layout to Modeler, I switch back and forth all the time without saving the objects so I can preview changes.

This isn't completely true. Try this:

Open Layout and Modeler. In Modeler create a box. Without saving it, go to the dropdown arrow in the top right corner and try to "Send Object to Layout". You can't. That command is ghosted out until you save the object. I think this is what the previous post-er was referring to.

DaveW's workflow only functions if you have previously saved the object. This means if we just want to mess around with some temporary objects in Layout we still have to save them to disk first. No big problem really except it creates an extra useless step or two, since I have to go and erase those temp objects to prevent them from accumulating.

Jean Marc Rodrigue

Matt
02-19-2003, 12:30 PM
Originally posted by Rory_L
If Matt were to edit the choices to include

Whichever, just make it great!

I`d vote:)

R

I can change my message, but not the poll catagories, sorry!

The only reason I did this thread was because it's such a hot debate that it would be easier for NewTek to see a simple poll of how people were feeling.

You can take the "Whatever" option to mean that!

:)

Antimatter
02-22-2003, 03:01 AM
I personaly preferes them to be seperate so i can just focus on modeling then i switch over and animate it without other execess stuff geting in way but i can live if they merge it.

but what i do not want is maya, its horrious i tried it but all of the icons drove me nutty, also the menus were like HUGE, i mean it had like 4 different type of menu each type had like 20 menu in it each menu had like minum 50 option, plus suboption and it just turned out to be pure hell for me to use.

that's the main reason why i like lightwave is the interfance is nice and simple, also very effective to.. nothing of this 40000 menu all over the freaking place plus 5000 icons which i got no clue wtf they do. i want it to stay textual like it is right now.

when i want to use bevel tool i just look for bevel and i click on it and use it instead of hover and wait for 5 min for popup help to popup over each icon

heh i know there's a shortcut and i do use it but that was just a example.

if they stay seperate great, if they get merged together great but i just mainly want the interfence to stay the way it is, its very effective imho, i tried max, tried maya, their interfence sucked imho. i hate menus, and i hate icons. and i hate pictures i like the textual interfence of lightwave.

anyway enough ranting i'm shutting up.

ToonShady
02-22-2003, 02:02 PM
Hi, I like to just throw out this idea. Get it out of my head. It's sorta the integrated separatist approach, building on top of some of the ideas that had be exposed.

The idea is the modular approach. But before I describe it any further, I like to say, XSI is not modular in my book. It uses the procedural methodology, and the tools are divided in tabs (model, animate, texture, etc) like the tabs in LW layout or modeler. Even in the areas of rigging the character mesh to the skeleton, it's still no cheese cake. If you get into a scenerio where you want to add details to your mesh while it is bind to the skeleton, you would have to mute the envelope operator and shape animation (morph) operator and other stuff inorder to get back into that editable state. Even then, you will to manually take care of the weigthing for the added details when reactivate those operators. In my experience, this history/node base procedural workflow is good for alot of things. Specifically, I would really like LW's displacement plugins/modifiers to respect this workflow and bones would be a type of displacement modifier so as to be procedurally evaulated along with other displacement plugins. But for what I described of changing geometry while it's bind to the skeleton is something that the modular approach of LW would do better.

So having that out of the way, I'll go on with the modular idea that I wanted to describe. Currently, LW layout and modeler are programs, not modules. Both programs don't really have access to the same .lwo data. Modeler makes something, save it, the hub makes a copy of the file somewhere in harddisk and pass it to layout.
But take a look at something like the graph editor and scene editor for comparison. They are two editors that have access to similar data, like the motion key frame data for example, and they are just displaying and interacting to the user in different manners.
Now perhaps we can look at the current version of VertexPaint, 3.5.4? That's the kind of modularity that I am refering to, except it should be non-modal. Just imagine for a moment now that you open up a program call lightwave3d. You then open Modeler and Layout just like you would open a GE or Scene Editor. In essence, Modeler and Layout are just editors that operates on the same .lwo data. Modeler will have acess to the .lwo data in the way that it is now: sub level manipulatiion (points, polies) and layers (object slots). Layout will have access to the .lwo data but more towards displacement and motion, as well as building other scene related data like nulls, camera, Lights. If LW were to work like this, I would imagine that VertexPaint would be as another non-modal editor (module) that would take care of rigging. Currently, it already is to some extend. VP feels like something that bridges Layout and Modeler for character setup, with vetex and poly selection (modeler), weight paint(modeler), joint doformation (layout), positioning of lights(layout). VP as a separate module could evolve to a setup mode environment for LW.

with this design approach, I am sure there will be some issues that will need to be addressed. For now I can think of keyboard shortcuts access to commands have to respect mouse over behavior (as oppose to mouse clicking) to a large degree. For example, if Modeler and Layout modules are both open and visible, you have to be able to execute a modeler keyboard shortcut when the mouse is over at the modeler's window, and vise versa for Layout.

And hopefully, we'll get a sexy looking interface to go with like the ones that Matt did. :)

hrgiger
02-23-2003, 12:49 AM
Antimatter- what a great avatar.

faulknermano
02-23-2003, 01:30 AM
Originally posted by ToonShady


The idea is the modular approach.

i agree to all the larry said.

to add, i wrote something before in defense of the separatists, or something like that.

the idea was that modeler and layout were not performing their supposed duties. the argument is not so much if lw should be integrated into one - meaning modelling, animating, etc, were in one interface or module, rather than the current being able to accomplish what they are supposed to.

for instance, modeler doesnt not do real rigging, which, imo, it should. there, it fails, again imo, to be a complete modeler, especially in relation to layout (dont compare wings3d with lw modeler because modeler OUGHT to compliment layout).

to me it is an identity crisis, not _exclusively_ an integration issue. i propose to look at the two apps' INTENDED roles and go from there.

Qslugs
02-23-2003, 01:03 PM
You know guys, in Maya you can just shut off what you aren't using and it dosen't even load (modeling, animation, rendering, dynamics). Just something to think about. Just because the program is integrated, dosent mean you have to have it all loaded all the time.

Antimatter
02-23-2003, 05:50 PM
Hrgiger: heh thx some one stamped that icon on someothing i said it was appearly something very obivious and it sort of stuck :)

i like toonshady idea the best its the best idea ive seen so far,

the people who wants a seperate program now can have something similar to this, sort of a sub program. yet people who wants it merged will have no problem trasfering data and stuff over to the other window cos its techicaly the same program just seperate window.

now about the shortcut thing i don't think that is a problem, good example is like active and deactived windows, if you use a shortcut its for the active window so that shouldn't be a problem. imho.

ToonShady
02-23-2003, 07:57 PM
You mean something like this?



Believe me, I try working like this setup, and often loose track of which window is active. And consider the single undo issue in layout, it's too easy to execute something that you can't go back. And sorry about all those typos in my first post.:)

faulknermano
02-23-2003, 08:24 PM
Originally posted by ToonShady
You mean something like this?

Believe me, I try working like this setup, and often loose track of which window is active. And consider the single undo issue in layout, it's too easy to execute something that you can't go back. And sorry about all those typos in my first post.:)

larry, search the web for TweakUI. it will enable this thing called mouse-over activation where you simply move your mouse over the window and it will be active. it takes some time getting used to and sometimes exhibits funky behaviors (that you'd get used to after a while). this is great for things like move your mouse over grapheditor and it becomes active. the downside is that you MUST be paying attention to where your mouse is because as soon as graph editor is active, the shortcuts for GE come into play.

ToonShady
02-23-2003, 09:02 PM
Thanks Lernie, this sounds fantastic. I run into that GE and Layout problem all the time when I animate. This becomes a real issue whenever there're alot of panels. The mouse over behavior is how it works in XSI.

Antimatter
02-24-2003, 05:26 PM
how on earth can you fellows run two copy of lightwave, i have enough headache with one copy open :)

Qslugs
02-24-2003, 08:45 PM
Thats not two Lw's, it's one Modeler, one Layout with all the support windows opened. If LW were integrated you wouldn't need that. You could have one monitor for floating windows, one for 1-4 viewports, or 1 big app stretched across 2 windows. I am sure I am going to get flamed for stating that btw.

faulknermano
02-24-2003, 10:47 PM
Originally posted by Qslugs
Thats not two Lw's, it's one Modeler, one Layout with all the support windows opened. If LW were integrated you wouldn't need that. You could have one monitor for floating windows, one for 1-4 viewports, or 1 big app stretched across 2 windows. I am sure I am going to get flamed for stating that btw.

considering the integrationists and separtists are so close in the votes, you dont have to worry so much. :D

Jimzip
02-24-2003, 11:04 PM
Dear lord toonshady! What resolution is your monitor on?
P.S, I'm going to use that setup now...:rolleyes:

Jimzip

DigiLusionist
02-24-2003, 11:18 PM
If you add up all of the votes for integrating LW (like May, XSI, or otherwise), the votes are 86 for integration and 65 against, with 13 uncommitted.

Qslugs
02-25-2003, 07:27 AM
faulknermano, where do you feel the Abolitionists stand?

Antimatter
02-25-2003, 07:36 AM
heh i'm one of the uncomitted :) if they remain separate great, if they get merged together great, i can adapt.

i just mainly want the layout to stay the same, maybe addition of more tabs and more button but i want NO icons and NO menu. Maya is a nightmare with all of its menu got like 20 menu for each mode, and has like 4 mode and blah blah. its nasty. i like lightwave flow.

anyway what i ment was right now theyre seperate apps i just refers to both modeler and the layout as lightwave :)

but anyway my puter isn't the hottest computer around esp for rendering, decent enough 1ghz and blah blah but it has a VIA chipset so that really limits the bandwidth so....

but anyway if NT does goes with the merged mode i would like to see a option to unload certain things i don't need or want at the time. Ie i am primary modeling so it would be nice to unload like the ah layout, the lighting, and texturing and pretty much everything not related to modeling. then later when i start to texturize i can load up texturizer stuff.

would really help keep my cpu from choking up whenever i start working on anything more than 20k on the screen or on a layer :) yeah ive modeled things approaching 80, 90k polygon but they were split into like 8 or 9 layers each... no problem with each piece but when i had them all together it lagged horriably.

hrgiger
02-25-2003, 08:20 AM
Yeah, I don't know. It seems like there are much more needed features then a rewrite or integration of modeler and Layout. I would seriously doubt that NT is even thinking of integrating the two, or at least I would be surprised. That's what they wrote the hub for. I think the hub just needs updated like everything else to talk more between the two programs.
Like Lernie said, I think that modeler and Layout should do what they're supposed to. Certain things should work in both programs. Bones should work in modeler so we can do rigging. Trying to rig in Layout is difficult at best. In fact, we should eliminate anything that causes us to jump back and forth between modeler and layout on a trial and error basis, i.e. getting good deformations in layout by the bones you setup in modeler. That doesn't make sense.
Question is, if you had lights in modeler to see your model like it's intended to be in an environment in layout, if modeler had bones so you could see how your character deforms and do corrections right there, if you could use all of modelers tools in layout so they could be animated, etc... would you still feel the need for integration? If everything worked in one program that worked in the other and there was never a need to jump back and forth between modeler and layout, what reason would you have for integrating?

faulknermano
02-25-2003, 08:21 AM
Originally posted by Qslugs
faulknermano, where do you feel the Abolitionists stand?

on shaky ground. :D

faulknermano
02-25-2003, 08:24 AM
Originally posted by faulknermano
on shaky ground. :D

but really... what would i know? ;)

DigiLusionist
02-25-2003, 08:51 AM
If LW worked in the manner in which you just listed, there wouldn't be an issue. And in order for LW to function optimally, integration of the code seems necessary. As I have written before, if keeping the interfaces separate will allay fears or squash learning curve discomfort, then keep them separate. I'm talking about unifying the code. Exactly what is wrong with having the code integrated?

hrgiger
02-25-2003, 11:09 AM
Originally posted by DigiLusionist
If LW worked in the manner in which you just listed, there wouldn't be an issue.

Exactly. If these things worked, there wouldn't be an issue, or in other words, no reason to have people think that the code needs integrated. Modeler and Layout don't need to be merged to get bones to work in Layout or lights or anything else. I don't actually care if they're integrated or not. I mean, I guess I prefer having them seperarte but like a lot of others have said, it ultimately doesn't matter to them. I just don't think that taking months to re-write the whole program at the expense of probably a lot of other more needed features, is kind of a waste. The program was re-written for 6. Seems a little early to be starting over again. I think that the idea around the hub just needs developed more instead of scrapping it altogether for something else. Features need updating and I think it's no different with the hub. That's all I'm saying.
I've never had issue with the two seperate programs except when I have to jump back and forth between the two to make adjustments. That gets old. Especially when I rig a character, and then realize, one of my bones in a chain is maybe not pivoting from the right point. You can correct it in Layout but sometimes, it's usally quicker to just adjust it in modeler and re-convert. That back and forth gets old. Even more so with point weighting. Ugh.

DigiLusionist
02-25-2003, 11:27 AM
You want the hub to be improved so Modeler and Layout work better together. Got that. Yet you don't think it's important to integrate the code to enable the hub to do so.

Exactly how else is the functionality in LW going to be streamlined unless the code is streamlined? And simply because the hub was implemented in the first place to translate data from one set of code to another doesn't mean that is the best implementation for LW. There wouldn't have been a need for the hub in the L6 partial rewrite if the code was unified to begin with.

The logic seems painfully simple to me.

hrgiger
02-25-2003, 12:31 PM
I guess. I don't really care. It's probably not going to happen anyway, so there really isn't a dispute.

Qslugs
02-25-2003, 06:51 PM
I don't believe the program got a full rewrite for 6. It never got that far from what I can remember.

Qslugs
02-25-2003, 06:51 PM
However, it was said at the time that it was a rewrite......

faulknermano
02-26-2003, 02:00 AM
Originally posted by DigiLusionist
You want the hub to be improved so Modeler and Layout work better together. Got that. Yet you don't think it's important to integrate the code to enable the hub to do so.

i dont know exactly what hrgiger was trying to get across, but my idea / opinion was not "modeler and layout to work better together". RATHER, modeler and layout accomplishing a certain functionality in full. for example: modeler > model, rig layout > animate, render.

imho, i think that's the best setup, because modelling is so close to rigging. what i'm thinking is the less back and forth movement, the better.

you argue: why just not integrate and get it over with? okay, fine: integrate. but really, how _feasible_ is that. let's talk about reality: let's HELP newtek on some ideas that can work with their framework.

i voted "i can adapt" because mainly i know i could AND i wasnt sure which had more benefit IN REALITY, not in theory. i'm afraid of bugs when you need less in the middle of a production. in theory, integration sounds promising. but in reality, how do we really know? we just guess and dream (which is fine, but dont stake everything on that).

now i was thinking of ideas on how a segregated lightwave environment could look like. i see modeler with a full suite of deformation tools. you can drag your IK goal and you'll see your model deform. skelegons will really be bones. we can use the super intuitive tools of modeler WITHIN modeler to set up a real rig. layout will accept such objects and rigs and will enable animation, effects and lighting. the HUB will still play a part which means it updates geometry and rigging information to layout. you can pop a bone at any point in the work stream and it will reflect. the problem with the hub right now is not just speed, but the kind of information that it has been burdened with. the less you use the hub the more successful is the design, i think.

do you see the 'modular' advantages?

DigiLusionist
02-26-2003, 02:30 AM
Yes, faulknermano, I definitely see the modular advantages. And I agree workflow can be improved with the sorts of changes you and hr are proposing.

You two are taking issue with my contention that the underlying code needs to be unified. From your last post, you seem to be taking the position that it isn't feasible for NT to integrate the app in that way. I disagree. I'm taking the position that it is not only feasible, but it is a market necessity for NT to do so.

That's why I had previously asked 3D programmers to chime in. And I mean programmers experienced in creating a complete app like LW. They know a hell of a lot better than I about the practicality of what I and other users are requesting.

As a professional user and LW instructor, I know a lot about why the workflow issues are a problem for me. It's to those issues and experiences that I base my opinions.

As to whether or not LW will ever be integrated codewise, I will never know whether NT will ever do it until they actually do it. Nor does any Non-NT community member here actually know they will not. Yet some here vociferously proclaim there is no point in conjecture, as "it will never happen, anyway."

The fact that a new team of programmers is being pulled together, however, opens up the possibility of eventual integration. A can-do approach wins out over a can't-won't course of action any day.

hrgiger
02-26-2003, 06:29 AM
I guess what I mean Lernie is that sometimes the hub doesn't always work for me. Sometimes I have to save a model before it's updated in Layout, or sometimes even close and reopen Layout before it will update. I just want that 'cross talk' between modeler and Layout improved.

And I totally agree with you. I wish everything could be done in modeler like rigging. It would be great to see the effects of bones and IK so that we could fix the deformations in modeler. Layout is for animation. Everything should be ready to go once it's loaded into layout. Maybe everything with the exception of texturing which seems easier in Layout right now anyway. If each modeler and layout worked fully at what is was doing, and I agree skelegons in modeler isn't really even half way there when it comes to rigging, there would really not be a need to integrate the two.

Disilusionist, I don't see why it is a market necessity to merge the two apps. Just because Max is, or XSI is maybe? Why don't we charge as much as them too? Would that be good? Lightwave has done pretty well for itself for a long time being the two seperate programs, I just don't see that it's a 'necessity.'

Qslugs
02-26-2003, 07:29 AM
faulknermano, I highly doubt that Newtek is taking this conversation into account for Lw 8.x, after all, they never asked us what we though. This was just some guys idea. I also wouldn't be surprised if they were already well into integration, having at least the enviorment fulliy integrated. Solely working on tools now. Thus, all the job posts for specific things, like booleans for instance. This is just speculation of course.

faulknermano
02-26-2003, 07:49 AM
Originally posted by hrgiger

Disilusionist, I don't see why ....

LOL! digilusionist... i think this is steve trying to hit you below the belt. :D


I'm taking the position that it is not only feasible, but it is a market necessity for NT to do so.

supposing it was (and i say it isnt only out of pure speculation: what would i know about app-building and marketing), this doesnt automatically make it a marketing improvement. i can imagine agreeing that it is feasible.. but how about bugs? coming across bugs because of young code is not an exclusive experience one could tell you. like steve wondered, i opine: i do not think integrating is a marketing necessity; i think you take your stand too far to where you have less and less basis to draw from. is it because of the 'potential' of what could be?


As a professional user and LW instructor, I know a lot about why the workflow issues are a problem for me. It's to those issues and experiences that I base my opinions.

as do i. though i am not an lw instructor.


Yet some here vociferously proclaim there is no point in conjecture, as "it will never happen, anyway."

when i talked about 'newtek's framework' my presupposition was for segregation, since that's how it has been for many years now. i would think that a person would be in a stronger position to suggest things based on that kind of presupposition than one we have no basis for (other than dissenting voices of integrationists).

my concern with rewrites is the amount of bugs it generates as opposed to working with existing code to squash bugs. if you think lw6.0 had a lot of bugs, then you're in for a treat. i'm not squeamish about changes. an integrated lightwave will have great intended functionality. but it's only intended. how much stuff will prove worthless in production? i dont know. indeed maybe they'll strike gold and will have a very stable release. but statistically, even with other progs, this isnt the case (cases in point: XSI and LW6). is newtek as large as avid? can they bear the brunt of a shaky release. ask chuck, i wouldnt know. :) but this is to address your marketing issue, not the potential functionality of the future of lw.

it goes without saying that integration is risky; promising, but full of unknowns. and just because it is risky, it doesnt mean one should take it. what if workflows can be addressed efficiently without having to take the risk? (though i admit that's a big "what-if")

hrgiger
02-26-2003, 10:03 AM
Oh crap! Digilusionist,(not disilusionist) that definately was a typo! Sorry... Thanks for pointing that out Lernie.

DigiLusionist
02-26-2003, 10:16 AM
Faulk, your points are well taken. Your concerns are valid, and I hope I'm not minimizing your concerns about the potential downsides to the integration process. My view is that they are POTENTIAL downsides, not automatic eventualities. Bugs could and probably would happen, as they did with L6. However, bugs will happen with any continued development of LW regardless of code integration or not. Again, does one act based on desire and purpose or not act because of potential fears?

I would think that if NT did work on unifying the code, it would not recklessly release a unified code version until it had a solid measure of stability. At least, I would be willing to give NT the benefit of the doubt for not doing so.

As to why I posit that it is a market necessity for LW to integrate the underlying code: NT does not operate in an industrial vaccuum. 3D app companies are vigorously competing for business and users. Therefore, a 3D app company must develop tools that have competitive features sets in relation to competing packages. I think that is not an unreasonable statement to make. Any profesional users who utilize more than one 3D app understand the inherent limitations of the current implementation of LW. Simply because some users like LW's "quirkiness" doesn't negate its need to be improved. LW is a good package, and I certainly think NT does considers the viability of its products in the marketplace.

How LW is improved from this point on seems to be the bone of contention in this thread. NT can either continue with the current two-code development approach, or it can plan out and implement a unified code development approach. My view is that the functionality of LW will increase manyfold if the code were unified.

I know I will continue to be roasted for thinking the wisest approach is to unify the code. But that's my view. As it's my view, others can accept it as my view and let it go at that, or they can argue counterproductively that their views are the only valid views.

I appreciate your posts, as they don't get personal or are ungrounded. They are well thought out, and give me reason to consider them. So, thank you for that.

Ultimately, this is just a discussion on one of many threads on one of many forums. I enjoy engaging in discussion, but I certainly don't think we're going to save the world by talking about this silly stuff. And I certainly don't think NT will be damaged by dissension of views by users. Contrary to what has been brought up before by others, I believe NT is strong enough of a company to handle such difference of opinion.

Russel
02-26-2003, 07:55 PM
How about having one application but "separate sections" - For example:
Instead of having Modeler and Layout as 2 separate entities, combine them and add several buttons to the top (above the tabs?):
"Model"
"Texture"
"Rigging"
"Animation"
As you click on any of these your workspace would alter showing the appropriate tools to use for that purpose - sort of like like clicking on the tabs - only more defining/specific in scope.
If you were animating a model(s) you would be in the animation pane and only those commands and tabs would be seen that deal with animation, and if you wanted to make some particular texture change to a single model, just select the part you want to modify, click the "texture" button and commence with the change, then when finished, click back onto the animation button and continue with animating... and so forth.
It should be seamless and would help in being intuitive.

Zach
02-27-2003, 02:53 AM
keep um separated, mm k.

Just allow access to point manipulation in layout (including weightmap changes while model is being deformed, ability to animate and group points by using a "point select" tool) and be able to take a snapshot of a scene to import into modeler for easy quick modeling reference to scene scale.

JMarc
02-27-2003, 07:47 AM
Russel is right on the money! That is the most efficient setup a LightWave user could ever want.

I have yet to see a single good reason for keeping Layout & Modeler separate. People keep suggesting things they want added to LW in order to make the separate apps better but why create all that redundant coding when what they are really asking for is intergration?

cavalos
02-27-2003, 08:05 AM
52.81% of pelope here wants integration so what are you waiting for!

Best
Christian

hrgiger
02-27-2003, 08:51 AM
Actually, all the poll tells us that 94 people wish that Lightwave was made to be like one of the other 3D applications. 59 of us wish you would leave it the hell alone. I don't see any votes to integrate LW.

DigiLusionist
02-27-2003, 09:10 AM
There is a number after this choice.

Russel
02-27-2003, 10:51 AM
Jean Marc, you've been added to my Christmas list! ;)

hrgiger
02-27-2003, 11:31 AM
Yeah, I added that one in there Digi.

DaveW
02-27-2003, 11:31 AM
Originally posted by hrgiger
Actually, all the poll tells us that 94 people wish that Lightwave was made to be like one of the other 3D applications. 59 of us wish you would leave it the hell alone. I don't see any votes to integrate LW.

Those 94 votes are for integration. The question is should LW be integrated, and the options are YES, like Max, YES, like XSI, ect.

Actually after reading some of your posts, it sounds like you want integration too. Rigging and lighting in Modeler? That would be a ton of Layout code added to Modeler. At that point it would be more of a hassle to maintain two seperate apps with almost exactly the same functionality than it would be to have one app that does it all.

hrgiger
02-27-2003, 12:02 PM
Originally posted by DaveW
Those 94 votes are for integration. The question is should LW be integrated, and the options are YES, like Max, YES, like XSI, ect.

Actually Dave, I could care less if modeler and layout were together or seperate. If modeler worked like it should and the same with layout, then there wouldn't be a need to integrate the two. I mean, I wouldn't mind a program that has a mode change, from modeling to texturing to animating, etc.. I think the way it stands though, and the fact that I don't think modeler/layout is going to happen anytime soon anyway, we should be able to do complete rigging in modeler.
People have their preferences about how they would like to work, but it's not modeler/layout being seperate that's holding the program back from being great or anything. People get in their heads that if these programs were merged that everything would be hunky dorey and it would be Euphoria. Fact is, and I"m not a programmer so correct me if I'm wrong, but a lot of Lightwaves plug-ins and add-ons would also have to be rewritten to work with the new architecture. Everything. I'm not talking about things like extender.p and unwrap.p, I"m talking about things like Darktree, Sasquatch, Particle Storm, Messiah, WCS, Real flow, bodypaint, etc.... How long is that going to take?

So I guess you could say, I lean towards keeping modeler and layout seperate. Not to mention, that when I'm done modeling and ready to move onto layout, I like leaving everything in the modeling interface behind and not having to wade through any of that stuff while I'm trying to animate. It's just simpler.

Matt
02-27-2003, 02:13 PM
Russel said : How about having one application but "separate sections" For example:

Instead of having Modeler and Layout as 2 separate entities, combine them and add several buttons to the top (above the tabs?)
"Model"
"Texture"
"Rigging"
"Animation"


Russel . . . that's kinda how SoftImage XSI works, this is why I put the option in the Poll.

The reason I started this poll was because this is a hotly debated subject with many points of view, if NT were serious about looking at integration, I thought the poll feature would be a simple way for them to gauge what their customers want.

So far it's looking like it's staying put!

QSlugs Said: Just because the program is integrated, dosent mean you have to have it all loaded all the time

IF it were to be integrated, this is the ONLY way to do it IMO, different "modes"!

Russel
02-27-2003, 02:53 PM
Oh, I never used XSI to know (just 3DSmax).
:D

I think that this poll is great... it just may be hard for some with no other 3d application experience to vote specifically... thus the other 3 choices are appropriate - as well as people adding their suggestions that don't seem to fit otherwise.
:)

JMarc
02-27-2003, 02:59 PM
Originally posted by Russel
Jean Marc, you've been added to my Christmas list! ;)

Heh-heh!

If we get what we want in the next version of LightWave it will be better than Christmas! :D

If we don't get it... I'll take a Widescreen TV, please. ;)

Gregg "T.Rex"
02-27-2003, 04:51 PM
Judging from what i see at this poll, users want Lightwave to be left in its current state, but also take all the good things other apps have and implement them to make the ultimate 3D software.
IMHO, this is, if not imposible, extremely difficult and time consuming. I realy hope NT have the resources for this GIANT task and manage to make it posible...
Regards,
Gregg "T.Rex"

DigiLusionist
02-27-2003, 06:40 PM
According to the numbers in the poll, there are currently 98 for integration of some sort vs. 71 against, and 14 non-committal. For statistical purposes, the 14 noncommitted weigh little in the determination, as they would go along with integration if it happened.

Therefore, according to this poll, the majority of users who participated in the vote DO want LW integrated.

Unless my addition skills are completely lacking...

Antimatter
02-27-2003, 09:58 PM
that sounds about right :)

i'm one of those uncommitted one. i like them seperated but if NT merge them great.

there are really only two or three thing i am worried about.

the interfance staying mostly the same, aka text buttonover, nothing of that fancy icon crap, and nothing of those 200000 menu with 2000 submenu and 20000 option per menu. aka maya and 3dstudio max for the icon thing.


i have never tried out softimage xsi but from the description here its sounding pretty cool.

anyway if they merge the interfence, they could have like four or 5 or something like that super tab, then within each tab several option. like for under the modeler supertab you would have all of the modeler tab that we got in modeler right now, and blah blah.

now imho that would actualy be very sweet cos i wouldn't need to load/unload each apps i have to cos only 256 meg of memory on this puter, i'm sort of a hobbist and this is like the first puter ive ever really rendered or worked with lightwave on so... anyway the new one is geting 1gb+ of ram but different story.

faulknermano
02-28-2003, 10:31 AM
Originally posted by DigiLusionist
According to the numbers in the poll, there are currently 98 for integration of some sort vs. 71 against, and 14 non-committal. For statistical purposes, the 14 noncommitted weigh little in the determination, as they would go along with integration if it happened.

Therefore, according to this poll, the majority of users who participated in the vote DO want LW integrated.

Unless my addition skills are completely lacking...

actually it's what you dont see. think of all the other lightwave users out there. got to be more than a two hundred. :)

DigiLusionist
02-28-2003, 12:48 PM
Uh, ya, maybe three hundred or even four hundred users...

?

And, had they participated in the poll, I'd have counted their votes...

hrgiger
02-28-2003, 03:18 PM
True. Lots of users not keying in. So actually, the results are 98 for, 71 against, and two or three hundred that are just fine apparently with Lightwave the way it is.

faulknermano
02-28-2003, 08:24 PM
Originally posted by hrgiger
True. Lots of users not keying in. So actually, the results are 98 for, 71 against, and two or three hundred that are just fine apparently with Lightwave the way it is.

or just too busy to care. :D

DigiLusionist
02-28-2003, 09:39 PM
stopping the madness...

click

jburford
04-29-2015, 08:31 AM
Hey Guys, the Poll from 2003 about Unification still works!

Did you vote? (sorry, I am being terrible now)

jasonwestmas
04-29-2015, 12:12 PM
This isn't the issue as much as the issue that animation and camera matching tools that are in modeler need to be in layout also. e.g. weight map assignments, creating endomorphs and basic modeling abilities should be in layout. That should be obvious by now.

By all means be creative and come up with better ways to do these things than what is currently available in other applications.

jwiede
05-02-2015, 07:07 PM
This isn't the issue as much as the issue that animation and camera matching tools that are in modeler need to be in layout also. e.g. weight map assignments, creating endomorphs and basic modeling abilities should be in layout. That should be obvious by now.

Sorry, but as you seem to believe that comprises the complete set of "missing functionality" needed to solve most users' unification needs, here's an counter-example:

Architectural and commerial/industrial visualization desperately needs instancing (ideally, nesting-capable) while modeling, it's the only way to keep polycounts managable while dealing with complex entities composed of hundreds to thousands (or more) of replicated assemblies, all requiring precise placement. Picture an apartment complex where individual buildings are composed of dozens to hundreds of replicated instances of doors, windows, railings, lighting units, and so forth. Picture a large industrial machine composed of hundreds to thousands of screws, nuts, bolts, gears, and other parts (perhaps shared amongst multiple lines of machines, etc.). Modeling such entities entirely as geometry is possible (though perhaps not efficiently within Modeler), but the resultant polycounts would likely make any subsequent use or modifications between inefficient and excruciatingly difficult.

Neither of these are cases where it would be even remotely reasonable let alone efficient to model the individual objects and then go into Layout and attempt to precisely assemble them together (essential in visualization) into the complete buildings or machines -- Layout's instancing placement options simply don't efficiently allow for that kind of precise large-scale assembly. If there were an instancing engine available for use and scripting (or better, SDK access) within Modeler, it would be fairly direct to enable the kind of precise placement and usage needed to enable visualization requirements in such cases, but unfortunately, instancing is exclusively a Layout-side feature.

Unification isn't just about granting canned feature access within both Modeler and Layout, it's also (perhaps even more) about resolving these situations where users, scripters and third-party plugin developers simply cannot provide important functionality (functionality that in many cases competitors have had for ages) because it would require underlying APIs and features that aren't available in the app where they need it. Duplicating a few canned features in both Modeler and Layout won't address these kinds of broad infrastructure access issues.

Little of the new functionality added from v10 onwards even offers API access from scripts or third-party plugin code, an orthogonal but related problem that stems from focusing too much on the "local aspects" of problems, as I believe is happening with the unification issue. Belief that a few quick fixes can stall or even eliminate deep infrastructure issues only serves to worsen the situation in the long-run, because while such fixes might satisfy a few customers, they also add internal limitations making the eventual deep rewriting/reconstruction needed even greater in scope (and thus less likely to occur).

jasonwestmas
05-02-2015, 10:17 PM
Sorry, but as you seem to believe that comprises the complete set of "missing functionality" needed to solve most users' unification needs, here's an counter-example:

Architectural and commerial/industrial visualization desperately needs instancing (ideally, nesting-capable) while modeling, it's the only way to keep polycounts managable while dealing with complex entities composed of hundreds to thousands (or more) of replicated assemblies, all requiring precise placement. Picture an apartment complex where individual buildings are composed of dozens to hundreds of replicated instances of doors, windows, railings, lighting units, and so forth. Picture a large industrial machine composed of hundreds to thousands of screws, nuts, bolts, gears, and other parts (perhaps shared amongst multiple lines of machines, etc.). Modeling such entities entirely as geometry is possible (though perhaps not efficiently within Modeler), but the resultant polycounts would likely make any subsequent use or modifications between inefficient and excruciatingly difficult.

Neither of these are cases where it would be even remotely reasonable let alone efficient to model the individual objects and then go into Layout and attempt to precisely assemble them together (essential in visualization) into the complete buildings or machines -- Layout's instancing placement options simply don't efficiently allow for that kind of precise large-scale assembly. If there were an instancing engine available for use and scripting (or better, SDK access) within Modeler, it would be fairly direct to enable the kind of precise placement and usage needed to enable visualization requirements in such cases, but unfortunately, instancing is exclusively a Layout-side feature.

Unification isn't just about granting canned feature access within both Modeler and Layout, it's also (perhaps even more) about resolving these situations where users, scripters and third-party plugin developers simply cannot provide important functionality (functionality that in many cases competitors have had for ages) because it would require underlying APIs and features that aren't available in the app where they need it. Duplicating a few canned features in both Modeler and Layout won't address these kinds of broad infrastructure access issues.

Little of the new functionality added from v10 onwards even offers API access from scripts or third-party plugin code, an orthogonal but related problem that stems from focusing too much on the "local aspects" of problems, as I believe is happening with the unification issue. Belief that a few quick fixes can stall or even eliminate deep infrastructure issues only serves to worsen the situation in the long-run, because while such fixes might satisfy a few customers, they also add internal limitations making the eventual deep rewriting/reconstruction needed even greater in scope (and thus less likely to occur).

Of course you are right but what you are describing sounds to me to be a near total rewrite. At the time I was thinking of the next best thing, assuming that this rewrite is unlikely to happen any time soon especially when we have people who don't want this division to change.

shrox
05-02-2015, 11:20 PM
It's simple.

Do the "unified app", then allow users to choose a "dual mode" where it behaves as if it's the separate apps.

erikals
05-03-2015, 07:49 AM
It's simple.

Do the "unified app", then allow users to choose a "dual mode" where it behaves as if it's the separate apps.

yup. +10

jasonwestmas
05-03-2015, 08:47 AM
It's simple.

Do the "unified app", then allow users to choose a "dual mode" where it behaves as if it's the separate apps.

Sounds like a near total rewrite again.

erikals
05-03-2015, 09:19 AM
"unified app"

so i take it it was meant more like integration, or something...

jasonwestmas
05-03-2015, 09:47 AM
so i take it it was meant more like integration, or something...

well I suppose we need to be more specific with what we want to see happen. Integration can mean simply putting a few modeler features inside of layout to enable what we see happening with these new 3rd party plugins from 3rd powers and ToolChefs. . . but still having the whole "localized tools" thing getting in the way of developing more practical (for the one man band), high performing and efficient workflows. More "advanced features" and artistic control would come about because of a more global acceptance between deformers and other effects which would mean something closer to a rewrite eventually. My guess is that lw3dg is working on such things if common sense and funding prevails, but how deeply they are is anybody's guess.

drako
05-03-2015, 03:10 PM
UNIFY,MERGE,ONE SOFTWARE...I cant see anymore lightwave like before Videoscape 3D and Aegis Modeler 3D.We are in 2015 and if some old guys cant follow the 3D era we have to
build an honor version of lightwave with 2 apps just for them.But for us freelancers and studios that we try very hard to take jobs with vfx and animation lightwave sometimes lacks of the new technology.
New artists they cant understand the philosophy of the software.Why Lightwave has two apps and why layout goes well with a lot of polys and not the modeler.Why we cant have one software
with three tabs specialized in whatever you want.Animation,Modelling,Render.Whatever...Animati on with undo that dont work in all the panels it is not good.Modeler with no history really
does not help.How you want the software to roll if we dont have new blood of artists to feed the old.First comes the UNIFICATION then comes the new Lightwave.
If we dont have a merge i think that a lot of us we continue to help Lightwave as we buy it...But for how long...I remind you the project CORE just for the records.

shrox
05-03-2015, 03:32 PM
....We are in 2015 and if some old guys cant follow the 3D era we have to build an honor version of lightwave with 2 apps just for them....

Yes. Now jog on, whippersnapper...

erikals
05-03-2015, 04:33 PM
hm, had to google that... https://www.google.no/search?q=whippersnapper.&ie=utf-8&oe=utf-8&gws_rd=cr&ei=C6JGVZbSCafNygOx5oBA#q=whippersnapper

drako
05-03-2015, 04:40 PM
Shrox if you like to insult someone for what he believes thats not a problem.Thats totally normal. I think that a I m not someone big you know or someone that dont give respect to the old lwavers but its time to move on.Any way for me who keep a company with 5 people it is difficult.
You dont think that we have to move on also?

Lewis
05-03-2015, 05:03 PM
12 years has gone since Matt started this topic/poll and this issue is still unresolved :(. Also many of those who voted for separation back then are now "converted" to unification and many have left LW due separated apps (among other reasons).

LWG3D - it's TIME to move on to next level.

erikals
05-03-2015, 05:43 PM
LWG3D - it's TIME to move on to next level.
ain't gonna happen yet, might as well wait until Siggraph to see what they have in mind, and then wait until LW2016

personally i'm skipping LW2015, maybe LW2016 also...

shrox
05-03-2015, 08:01 PM
Shrox if you like to insult someone for what he believes thats not a problem.Thats totally normal. I think that a I m not someone big you know or someone that dont give respect to the old lwavers but its time to move on.Any way for me who keep a company with 5 people it is difficult.
You dont think that we have to move on also?

I was being silly, I apologize.

drako
05-04-2015, 01:39 AM
Shrox dont apologise, we know how much you love lightwave and we all want to see our software getting better.I know that you are one of the most productive lightwaver in the community.
MODEL, ANIMATE, RENDER .
Lets move on...
LIGHTWAVE 2016 new era.
:thumbsup:

lightscape
05-04-2015, 03:43 AM
The funny thing is some people are modelling in modo, sculpting in zbrush, animating in maya, rendering with vray.

Unification is the end goal ofcourse for lightwave.
But how can these people who work with different specialized appz see past lightwaves split personality? Newtek should be developing powerful tools while working towards the end goal.

Text tools
Mograph
Fluids
Particles

jasonwestmas
05-04-2015, 09:23 AM
The funny thing is some people are modelling in modo, sculpting in zbrush, animating in maya, rendering with vray.

Unification is the end goal ofcourse for lightwave.
But how can these people who work with different specialized appz see past lightwaves split personality? Newtek should be developing powerful tools while working towards the end goal.

Text tools
Mograph
Fluids
Particles

How can they? Because it's not the division of the two apps. that is the problem, it's HOW they are divided. You wouldn't want half of your modeling tools or particle tools in one app. and the other half inside of another app. Ideally most of us animators want all of our animation tools in the same environment. But for animation it's doubly troublesome to have your animation tools divided imo.

Modeling in one app. Animating in another and Rending in a third makes more sense as long as the appropriate tools are grouped together under the same roof most of the time.

Hail
05-04-2015, 09:29 AM
The funny thing is some people are modelling in modo, sculpting in zbrush, animating in maya, rendering with vray.

Unification is the end goal ofcourse for lightwave.
But how can these people who work with different specialized appz see past lightwaves split personality? Newtek should be developing powerful tools while working towards the end goal.

Text tools
Mograph
Fluids
Particles

Try doing complex motion graphics or adjusting weight maps in layout and then maybe you'll understand why. :P

lightscape
05-04-2015, 10:03 AM
Try doing complex motion graphics or adjusting weight maps in layout and then maybe you'll understand why. :P

I do with max and maya.
I think my post is more on accepting the state of lightwave as it is. I don't really see lightwave getting unified in the future. There's just no incentive or resources to pull it off. I'm on the 5th stage for lw which is acceptance. :D
Not even unification would attract new users anymore imho. But if its the end goal sure that would be great but I don't see it happening anymore.

A mograph module or an improved weightpainting tool is not impossible to do in layout. Heck even in modo the setup mode, weightpainting aspect feels like I'm going back to modeller but on the same screen. :D

shrox
05-04-2015, 11:40 AM
Shrox dont apologise, we know how much you love lightwave and we all want to see our software getting better.I know that you are one of the most productive lightwaver in the community.
MODEL, ANIMATE, RENDER .
Lets move on...
LIGHTWAVE 2016 new era.
:thumbsup:

You called me old...but since I am Highlander, I don't really worry.

drako
05-04-2015, 02:40 PM
You called me old...but since I am Highlander, I don't really worry.

I dont think that we are old if we started Lightwave at the era of Amiga.I think that we are immortals...:thumbsup:

hrgiger
05-24-2015, 10:07 AM
I of course support integration and that would be my preferred course of direction but at the very least they need to eliminate the numerous weaknesses by having modeling and animation in separate interfaces. But I dont see any other way to do that efficiently without a unified interface.

erikals
05-24-2015, 10:10 AM
I of course support integration and that would be my preferred course of direction but at the very least they need to eliminate the numerous weaknesses by having modeling and animation in separate interfaces. But I don't see any other way to do that efficiently without a unified interface.

unfortunately, i was thinking the same thing...

something like this doesn't really solve it... :/


https://www.youtube.com/watch?v=l3m1FGKVhiE


but they have to decide which one to go for >

- Modeler tools in Layout
- Camera View in Modeler

both "solutions" have limitations...

erikals
05-24-2015, 10:30 AM
i'm sure some folks here also remember this one >


https://www.youtube.com/watch?v=6IYmspJP1Zk

erikals
05-24-2015, 01:40 PM
...or is it somehow possible, to select points / polygons in Layout, and automatically launch an external Modeler function...

...hm

...it should be doable actually, the question is if the operation could be fast enough, probably not, unfortunately

...this again brings up the LightWave Hydra engine... > https://www.lightwave3d.com/chronosculpt


...or adding Camera View in Modeler, followed by something like PointOven >


https://www.youtube.com/watch?v=K1M_RsBFomM

jwiede
05-24-2015, 06:49 PM
...or adding Camera View in Modeler, followed by something like PointOven >

"Camera View in Modeler" only benefits a single, very specific problem around view-matching, it doesn't address the majority of the "having modeling and animation in separate interfaces" set of problems. Something like "Modeling tools in Layout" could solve a much broader set of the problems, but as we've discussed, providing a decent/modern modeling environment in Layout requires much more than just modeling tools (also required are equivalents for the environmental support which allow those tools to be useful -- layers, selection and snapping systems, and so forth).