PDA

View Full Version : Should Animator include modeling tools?


Vizfizz
02-12-2006, 08:24 AM
Thoughts please.

Vizfizz
02-12-2006, 08:44 AM
I will admit, I'd ultimately prefer a total and complete integrated poly and sub-d modeling system, but the timeframe involved when combined with other animation tools, rendering upgrades and OS ports would push a integrated modeler into distant future I'm afraid. I'm fine with Silo or some other application handling the sophisticated stuff for now.

AzOne
02-12-2006, 09:35 AM
I am of two minds. In fantasyland, I would prefer that EI Modeler be revived (or that I could afford SolidThinking Design :hmm: )

Having integrated modeling tools in EI would be great but I don't think EITG has the resources to accomplish this currently (unless Apple or someone with deep pockets acquires them).

So, I think you summed it up best, Brian. Real world considerations and all...

Hiddenman.fr
02-12-2006, 10:45 AM
I use Ubershape combined with plugs all the time, but it is too limited. EIAS should integrate tools to make basic stuffs as lines, curves, .eps import and text tool (a real one with mesh resolution controls).

fahl5
02-12-2006, 02:12 PM
Having integrated modeling tools in EI would be great but I don't think EITG has the resources to accomplish this currently (unless Apple or someone with deep pockets acquires them).

It is not the job of our wishes to be realistic, but the job of the solution that should match those whishes as far as possible.

So I don't know if, when and how the EITG will try integrate modelling tools in the animationsystem, but I know if no one is asking for, they won't even think about.

But I personally think its absolutly worth to think about.
Steffen

Reuben5150
02-12-2006, 02:45 PM
I'd like to see "building blocks" or foundations for future modeling capabilities, something that can be built upon over time.

Maybe we are seeing part of this already, we now have muliple undo and superfast OpenGL, and probably some other relevant stuff i can't think of right now :shrug:

One of the great things about Silo, i think, is its simple straight-forward approach, but its still probably more complex than it needs to be.

I think a basic set of polygon modeling tools within EI would be a comlete revelation - really !


Reuben

Giacomo_M
02-12-2006, 03:36 PM
As I've said in another thread, I'd love to be able to easily subpatch and un-subpatch an imported cage within Animator, and I'd love to see some kind of integrated UV mapping and sub-pixel displacement (if that's the correct name for what ZBrush does) within Animator. If Animator could import NURBS and other parametric surfaces with the controls intact, that'd be great too.

Beyond that, I'm largely indifferent. There are already several great modeling apps out there and it seems a waste of time for EIAS to duplicate their efforts. Better to focus on being able to communicate with the existing modelers.

GM

halfworld
02-12-2006, 04:05 PM
I agree, I really don't support EI spending time and cash on a full modelling toolset, but I would love some some time savers - from the other thread on this issue... The ability to extrude, sweep and bevel Illustrator files would do me.

Ian

alisdair
02-12-2006, 04:18 PM
Here are some "modelling" tools I'd like to see in animator:

- a spline creation tool ( polyline, bezier and nurbs, with animatible control points ) for use in motion paths, NL's Pathfinder, and for:

- the full suite of array and copy tools that are in EIM. Some aspects of these tools could be animatible too ( number of copies over time, the scale of copies over time etc... )

-Alisdair

percy06
02-12-2006, 06:05 PM
i voted no i think that they should be working on the port to xcode and then work on the animation features. the thing that i think in order that should work on is this.

1)Porting the app
2)non liner editing
3)better preview in the matiral editor
4)better shader's (GUI controled more to make it more interactive)
5)mulipule re-do's
6)update camara (64 bit) render and 16bit render output
7)UV mapping
8)texture and light baking
9)dynamics
10)layers would be nice
11)nurbs and sub-d's modeler sort of like lightwaves
imporve performace again ( :D )

as you can see this little list i put the nurbs/sub-d's way way down the line. and i did not realy think that hard on the list there probably lots more stuff to add before number 11

Vizfizz
02-12-2006, 06:31 PM
Your list is impressive and all items are required, no doubt. Even more reason why any attempts to create geometry should be provided by a separate application or a plugin. A plugin by a 3rd party vendor fills the void and allows those who want it, the ability to buy it. If I need advanced modeling, I'll jump into Maya, Silo or FormZ.

percy06
02-12-2006, 07:33 PM
so viz your a pro user and all yes my list is impressive :bowdown: what would you think EI needs most of all other than a big cash injection.

i stand by non liner editing to be no1 whats your thoughts

yhloon
02-12-2006, 07:49 PM
I won't agree a full complex modelling toolset

but i would like to be able to select vertex/edge/face and tweak & scale it, surface soft selection similar to silo 2, and also a preset library, so i can save my modified model, and use it as morph target...

Vertex painting and Sub-D direct from EIAS.

Loon

Vizfizz
02-12-2006, 09:03 PM
so viz your a pro user and all yes my list is impressive :bowdown: what would you think EI needs most of all other than a big cash injection.

i stand by non liner editing to be no1 whats your thoughts


Wow... big question. It can be approached from so many angles and a lot depends on EITG's intended marketing strategy. (which I don't know) I don't think one tool will do it, but rather a set of tools, methodologies and company marketing agressiveness are in order.

However, I'll give it go. (Remember these are my opinions...not EITG's)

EIAS has certain "infrastructure" limitations due to its age. None of these limitations are uncermountable, but changing EIAS into something its not could potentially cascade and effect a number of systems that are dependant on its current infrastructure. (Camera, plugins, etc) Picture EIAS like the x86 architecture. It can grow and offer new things, but it has to remain sensitive to the old x86 core. Luckily we're seeing advancements to the Shader and Plugin APIs and this is a great thing. FBX also goes a really long way to bridging gaps in the character animation field. Right now, EIAS is better than its ever been. So where can they improve?

1. The program needs to be more open to both the user, the programmer, and other applications. The number one need pro users require in the film industry is an open architecture program. Notice I didn't say open source, there is a difference. The reason why Maya is such a success in the film industry is advanced users can get into the guts of the program through MEL and through Maya's plugin API. EIAS now offers Xpressionist, which is a first step, but Xpressionist more about controlling animation channels than permitting a method to construct new tools which tap into nearly EVERY aspect of the program.

2. EIAS is a polygon based program. In the past, this was somewhat of a detriment in the film community. EI countered by being able to handle huge numbers of polys. However, polys were generally avoided because of their finite resolution. Studios would utilize NURBS patch modeling instead and then either tessalate to a required resolution or just simply render NURBS on the fly. Today, machines are stronger, and polys for film animation are making a big comeback. For the typical user, polys are easier to use. SubDs are popular because they offer a nice hybrid between the two technologies. If EI wants to be seen as a complete package, some form of modeling within the program is required. This is mainly because animation users want the ability to modify and manipulate geometry down to the vertex level. Bouncing back and forth between a separate modeler and animation package is cumbersome for the character and organic animator, but not necessarily a problem for the illustrator or hard surface animator. Organics require the abilty to control the surface of a mesh to the nth degree. That currently isn't available in EIAS. So, EI must continue to market itself to a niche group of people that are either doing hard surface animation, matte or set design, motion graphics, lighting and rendering tasks, or CAD. If this is EI's desire, then they are on the right path. But you must also look to what the market wants.

3. The CG market, at least in the entertainment industry, wants character animation. CG is the preferred method of storytelling for animated movies now. The rise of Pixar is proof of this. However, there is difference between visual effects and organic animation. EIAS is definitely better for VFX than character work...but to the aspiring filmmaker, which is more appealing? Character work in my opinion. Everyone wants to create the next Toy Story, and right now, EIAS isn't the right tool for the job. Could it be? Yes. is it getting there? Yes. Is it too late? Difficult to say.

4. Next, EITG needs to focus its direction towards its desired market. If it is to remain an advanced rendering solution for other packages through FBX then make that known. For years, users of EIAS have wanted Camera to be available as a separate application to compete with Maxwell, Turtle, Renderman, etc.. but EITG refused to open it up. At first I thought this was a mistake, but I think they took a page from Apple and did the right thing. If they can make data exchange between EIAS and other packages as seemless as possible, they can start to offer Camera to other programs without giving up its secrets.

5. Tools. Ugh. So many I want. I agree with you that breaking away from the linear approach to animation is important, but so is including nodal methodology. EIAS' infrastructure needs methods to route animation channels more effectively through a nodal approach in addition to the xpressionist route of writing expressions and also through constraints. Expressions and constraints must be evaluated on a frame by frame basis and they slow things down and users are scared of them (expressions at least). Redirecting animation channels like a telephone switchboard would go a long way. Then integrate modeling, followed by more character tools. We know that Camera can make things look pretty.

halfworld
02-13-2006, 07:34 AM
I don't know if Jens browses this forum but if he does maybe he can chime in on this/ correct me...

I think the goal for XP is that it will eventually allow users to program + compile their own plugins... I'm sure I read it somewhere.... If true, it really would open the flood gates as it were, imagine simply attaching your camera to a custom plugin and instantly getting handheld shakes (obviously a scene scale global would be needed)... But things like that, little utility plug-ins would make EI an even more attractive app in my opinion.

And no, I don't think it is too late for EI in character animation, people do switch their 3D programs, I've done it. And lets remember that 3D isn't about one 'do all' app, it is increasingly about being able to tap different animation apps for their strengths, FBX really helps us with that (FBX export is coming!).

Edit: A lot of the back and forth to modeller tedium could be resolved with better asset management: An update model button for instance.

:)
Ian

Martin Kay
02-13-2006, 03:49 PM
EI needs to acquire more users and keep the ones its got... I don't know whether it can aqquire new users from other high end CA apps very easily, but can it also compete with other apps that hobbyists or 'middle of the road' dtp graphic designers might buy. These latter type people would be looking for an all in one application that doesn't cost too much, (although c4d isn't really an all in one program, it used to be fairly cheap). EI is really a specialist rendering application. Its extremely unlikely that modelling tools will ever be included in Animator and I'd have thought It'd be a waste of their limited resouces to go up this path. Much better to bundle something like Silo and produce GOOD documentation on how to use it- for those totally new users.

Also, for new users, Animator should have its interface redesigned. Forget all the 'techno correct' way of changing parameters, (like the camera aperture scale) and design a consistent to use method of things being 'off or 'on' on a scale of 1 to 100, small or big, 1 to 100. They need to get away from the ethos of, 'sometimes when we say 'right' we mean 'right', BUT sometimes we actually mean 'left'. Sometimes we mean neither... As I remember this sort of stuff went through the TripleD shaders as well.
One reason c4d caught on was not because it was good quality, but because it was so easy to use, (mostly). I don't say EI is hard to use, but there are confusing elements.

Assuming EI can keep its current users, its big problem is getting new users. It's really a marketing problem, getting the product into the head of new potential users. Unfortunately EI and also c4d, don't seem to have the 'shop window' that Maya, XSI and Max have. All things being equal, (and I know that things, as far as animation go, probably aren't equal in all apps), the visual quality of work you get out of all these apps is mostly down to the operator. This aside, the perception one gets, from looking at user galleries for both EI and c4D is that the work is largely amatuerish and mediocre. This obviously tends to reflect badly on the product. EI need to get the very best work into adverts in magazines like 3D World and to have some kind of profile on sites like CG Talk. Paul S was a great ambassador
for EI, and for what EI could produce, in the right hands, at a time when EI really needed it, unfortunately and unbelievably he was hounded of the postforum, no doubt busy enough with commisions...

Martin K

Reuben5150
02-13-2006, 04:05 PM
Paul S was a great ambassador for EI, and for what EI could produce, in the right hands, at a time when EI really needed it, unfortunately and unbelievably he was hounded of the postforum, no doubt busy enough with commisions...

Martin K


Hounded off postforum ?, think it was more a case of being distracted by Max and Brazil !


Reuben

plsyvjeucxfw
02-13-2006, 08:22 PM
But EI's Global Illumination got him BACK!

After all Paul's no fool. He uses whatever tool is required to get the job done.

Vizfizz
02-13-2006, 08:40 PM
Martin,

Yes, I do believe Paul S. ran off to work with Max and Brazil...though on several occassions he's popped into the Post Forum to tell us that he's had to jump back into EIAS to get certain things done.

I agree with you in that EITG should probably avoid providing a complete modeling solution for Animator. However, as you've probably already seen, I am in favor of EITG providing some limited tools and vertex level editing. Its just really necessary for character work. I'm also working with Konkeptoine to possibly have them provide users with a modest modeling subset in the form of plugins. If we want EIAS to compete for the ilustrator and mid level artist, some kind of internal modeling capabilities will be necessary. Bundling Silo as a dedicated sub d / high end modeler is wise and should continue.

I think EIAS' interface is good, albeit old. It doesn't have the aqua factor, and I'd only support aqua if it will improve the user's experience and interaction. If all we're doing is a asthetic face lift, I think I'd rather put energy into something else. Though ultimately, I agree with you. The interace should be updated, ultimately.

Obtaining new users:

I can agree with you that this is a marketing issue. However, we all know that marketing costs money. Its a catch 22 situation. You have spend money to make money. I believe EITG should continue to conserve its energy and work on mandatory OS ports, new tools, and paying salaries. As much as we want to see flashy ads in CG trades, they're gonna have to wait until EITG is a larger company.

We as users, however, have a bit of responsibility ourselves. This is why I've started this forum. I don't think we should do EITG's job, but we are ALL ambassadors here. Trying to place a single person as EIAS' primary evangelist is not wise. The moment that person moves to another package or has dropped off the scene (like Paul) its seen as a major set back or defection. I was upset at Paul for leaving EI and that's just stupid. He has specific needs for what he does and if he can't get them with EIAS then he should move to something that can do what he wants. This kind of reaction is what I see a lot of EI users falling prey to. The thing is, as a community we're stronger than any one person. Our community must act as ambassador. This show of support energizes people from the outside. They see what we have and want to be apart of it. Humans are social creatures and want to be defined somehow. If EIAS users are whiney little bitches, who wants that?

So, I've said it once, I'll say it again. Everyone needs to do their part. No matter how small.

harryb
02-13-2006, 09:29 PM
For me, I'd like to have vertex and edge editing capability for sds imports. It would be sooooo useful for creating morph targets and hand-animating cloth.

And an updated version of transporter. I know FBX is here but it would be massively helpful to import relevant data like IGES and native Maya, Silo, Rhinoceros, Modo and Soft Image files. Rhino does a good job of importing and exporting to a wide variety of formats -- maybe something like that?

Just my 2 cents.

percy06
02-14-2006, 01:20 AM
Martin,

This is why I've started this forum. I don't think we should do EITG's job, but we are ALL ambassadors here.

man this is so right i have always tried to do this in any thing i have done be it skateboarding using truespace or going out with a fat chick. i always trying to promote the good in any activity that i do. i wish people of today had the D-day spirt like my nan.

Vizfizz
02-14-2006, 07:29 AM
For me, I'd like to have vertex and edge editing capability for sds imports. It would be sooooo useful for creating morph targets and hand-animating cloth.

And an updated version of transporter. I know FBX is here but it would be massively helpful to import relevant data like IGES and native Maya, Silo, Rhinoceros, Modo and Soft Image files. Rhino does a good job of importing and exporting to a wide variety of formats -- maybe something like that?

Just my 2 cents.

Unfortunately all of those programs you listed are all NURBS/SubD centric programs. EIAS only speaks polygons right now. Spline entities are not understood. In order for Transporter to handle those model formats it needs to understand NURBS and tesselation. FBX was written specifically for EIAS to do this. It will import .fbx data from these packages and any NURBS entities will be converted into polygons for EIAS to understand. Tesselation is the same process that EIM and SILO have to use when exporting files to EIAS as .fact files. Polys only.

FelixCat
02-14-2006, 02:36 PM
hi
A bit OT, i know, but i think a real improvement in CA tools 4 EiAS could be very important right now. Since Pixar, clients seems very interested in animation. For instance, one of my clients is asking me about some number of CA shorts for cine and TV, and im not sure if EiAS is the right tool for the job. I will make my best for fit in but i think i will miss something that can put some kind of handlers on the face for more subtle control, besides morph animation... maybe something like poors man SOFTIMAGE|FACE ROBOT. :drool:
I will have rigid clothes an stiff hair (life is hard :cry: ) but the expresions on the faces are crucial.
Maybe too much work a feature like this, i now, but dreaming is free.

FelixCat

harryb
02-14-2006, 06:54 PM
Unfortunately all of those programs you listed are all NURBS/SubD centric programs. EIAS only speaks polygons right now. Spline entities are not understood. In order for Transporter to handle those model formats it needs to understand NURBS and tesselation. FBX was written specifically for EIAS to do this. It will import .fbx data from these packages and any NURBS entities will be converted into polygons for EIAS to understand. Tesselation is the same process that EIM and SILO have to use when exporting files to EIAS as .fact files. Polys only.

Hi Brian,

I know EIAS is polygon only but EIM used to be able to import IGES files (NURBS data) very well toward the end of its life. And it used to export FACT files very well, of course. Maybe EITG could take the SAT, SAB and IGES import features that were in Modeler and repurpose them in Transporter. And the SDS looks like it's being handled well by the clever folks at Konkeptoine. Their "Encage" system would probably make Ed Catmull blush with pride. Maybe EITG could put all this in Transporter, taking the place of some outdated formats. I mean, come on . . . is there anyone out there who still uses Sculpt and Swivel 3D? ;)

FBX is great but it would be nice to be able to get data from other time-tested, universal formats too.

Vizfizz
02-14-2006, 06:59 PM
Not a bad idea...but Transporter is Blair's baby. Its best to pose that question to him.

harryb
02-14-2006, 07:35 PM
Thanks Brian,

I'll do that but I can't find his contact info on the EI forums or the main site. Would you happen to have his contact info?

Vizfizz
02-14-2006, 07:42 PM
I think its just simply info@northernlights3d.com.

I don't want to post his personal email address here.

manuel
02-14-2006, 10:11 PM
Maybe EITG could take the SAT, SAB and IGES import features that were in Modeler and repurpose them in Transporter.
I wouldn't be surprised if they were build on Spatial Technology libraries, so EITG probably can't use that.

harryb
02-14-2006, 10:30 PM
I wouldn't be surprised if they were build on Spatial Technology libraries, so EITG probably can't use that.

Hi Manuel,

I just got an email back from Blair at Northern Lights. It turns out the Spatial libraries ARE extremely expensive and that IS why they aren't used. That's probably the main reason why EITG let the modeler go -- the Spatial ACIS stuff. There are other libraries, even open-source ones, that might be used instead. Rhino and other NURBS packages use the AGlib NURBS library to great effect. Maybe there's some way EI could use that?

Oh well, we'll see. . . .

manuel
02-15-2006, 09:22 AM
Rhinoceros has had to remove all AGLib technology from Rhino because Alias, who owns the library, didn't like the idea of a low-cost Nurbs modeller competing with their super expensive high-end stuff. So I don't think you'd want to license any technology from those guys.

harryb
02-15-2006, 06:08 PM
Really?! Wow, I wonder what made them license it in the first place. Oh well. . .

juanxer
02-27-2006, 02:35 PM
If that's so, one would asume there is a market for an EIModeler sucessor. If there was an affordable modeling engine around, it could be interesting for EITG to market such a modeler as a low cost conceptual design tool (and EIAS as a rendering and animation "upgrade"?), implementing the features such customers demand without making them getting in our way too much.

I am finally learning EIM's ways and getting to like it a lot. I long for more versatile surfacing tools and being able to tweak curves and surfaces as in Maya's malleable NURBS. The ideal would be something quite like Maya, modeling-wise: NURBS, poly and SubD toolsets and being able to convert from one to another.

There are a lot of specialized new modelers around, lately: SubD ones, mostly, ranging from costly to quite cheap. Would it be interesting to provide a cheap new "Rhino-like" conceptual design one?

juanxer
02-27-2006, 02:42 PM
About modeling tools in Animator: perhaps an ideal first tool would be a versatile Bezier curve primitive (it being able to, say, parent its endpoints or selected knots to other objects, simulate tension under gravity, etc.), and ways for plugins to read it and do things such as, say, produce tubes of several kinds, distort an object along, distribute objects along, etc.

SteveW928
02-27-2006, 10:01 PM
I think the basic problem here is that a rendering / animation package is a HUGE project, and a modeler / CAD type app is another HUGE project. Either of these is quite a bit more complex than most other kinds of apps on the market.

Having one company do BOTH requires some pretty major resources (even doing one of them does). EIAS is what EITG does, and does well. They have a long history and quality they are building on. Doing a modeler really isn't their speciality. EIM was cool in some ways (Ubers for example), but IMHO quite lacking in its tool-set and workflow. I think this is just a matter of resources and this not being their primary skill set.

I use Concepts Unlimited, and feel the same way about their rendering implementation. They are experts at modelers, but the 3D/rendering part just doesn't have the feel of a dedicated rendering package like EI.

Sure, one can get good results out of CSi's rendering app... and for industrial designers that just need a few conceptual renderings... they might really love the built-in ability. Same for EIM, some great stuff was done with it, and was nice to have it right in the EI package (though not really built in)... but it just doesn't match the dedicated modelers.

So, I voted for building a few modeling tools into EIAS, but leave the job of a real modeler to a company focused on doing it at a high level. Otherwise, you just end up with a mediocre product, and the primary app suffers a slow-down because of resources.

-Steve

percy06
02-28-2006, 09:43 PM
What are these libraries I here them mentioned a lot but what are they. I can guess there like sud-ds mathematics Catmull and clark but I don't know there names) if so why not one of you clever 3rd parties make such a lib probably be a lot of work but if you can sell if to other people then that must be an incentive or is that lvl of math far beyond mere mortals. Me I find my 9x table hard

manuel
02-28-2006, 10:44 PM
Well, first of all, all the libraries mentioned here were for NURBS modelling, not Sub-D's. And yes, I don't think that creating a NURBS- library is something you would do in an afternoon with the help of your first-grade math book. Hence they're all so expensive.

DickM
03-02-2006, 12:24 AM
Well, obviously a poly modeler and encage integrated would be best but I know that's not in the cards. For my character animation, vertex control is almost a necessity. Some kind of smart skin. Also, EI needs to tweak/fix some of the features that are already there such as weight maps. Also, I'd like to see deformers look at weight maps for there influence. Lattice def would be nice too. Personally I love EI but it's almost like it's pushing me away. :banghead:

Richard

http://www.morleyarts.com/testinglab/toodyshot2.html

Reuben5150
03-02-2006, 04:30 PM
[QUOTE=DickM]Well, obviously a poly modeler and encage integrated would be best but I know that's not in the cards.


Nothing's been said, even the beta team do not know whats planned for each "major release", after the mactel port who knows where development might go, i would like to think EI will have poly modeling tools one day, but not expecting it any time soon :sad: , 3rd party developers appear to do their best to fill this void, sure, Mr Revolver looks like being a really usefull tool, but its not enough IMO.


Reuben

Vizfizz
03-02-2006, 04:39 PM
Some internal modeling tools are definitely required.. I agree. But until that day comes.. I think we need something.

BJMonkey
03-06-2006, 04:07 PM
I'd really like some soft modelling tools and the ability to export the edited mesh.

Mostly for making blend shapes... Imagine being able to bend a bone, edit the mesh to eb how you want it to look when the bone is bent, put the bone back to normal and export the mesh. You have a perfect fixer for muscle deforms etc.

This is just about all I'd use it for... Making blend shapes and maybe a few simple deforms.

Brook,

CGTalk Moderation
03-06-2006, 04:07 PM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.