MajinPunisher
03-30-2004, 12:38 AM
wow! this thread is so ridiculously helpful! I'm gonna start posting my current character stuff here soon.
View Full Version : Theoretical SUB-D
|
|
CarlCampbell
04-05-2004, 12:19 PM
Just read about this a page backa and it has been an issue thats had me wondering for some time now... why does game geometry precise specifically only tris?
urgaffel
04-05-2004, 02:52 PM
Basically because everything you do will be stripped down to tris when it comes to the engine and the gfxcard. So, by working with tris you can optimize your model/scenery better. Although these days the gfxcards can handle so many polys that it's becoming a non-issue. Only (well not ONLY, but almost only) rts' games still need fairly lowpoly characters these days. FPS, fighting games, sports games... Even mmorpgs have fairly large polycounts.
Anyway, it was more of an issue back in the days, it's becoming less and less of an issue now :)
Anyway, it was more of an issue back in the days, it's becoming less and less of an issue now :)
Neox
04-05-2004, 02:57 PM
Originally posted by urgaffel
. Only (well not ONLY, but almost only) rts' games still need fairly lowpoly characters these days
these days are over, well what do you call "fairly lowpoly" 1500-3000 Polygons for a charakter (as in our game) is pretty much high i guess or do you mean 7000+ polies? It all depends on how good your LOD is ;)
. Only (well not ONLY, but almost only) rts' games still need fairly lowpoly characters these days
these days are over, well what do you call "fairly lowpoly" 1500-3000 Polygons for a charakter (as in our game) is pretty much high i guess or do you mean 7000+ polies? It all depends on how good your LOD is ;)
urgaffel
04-05-2004, 03:01 PM
I was thinking of the LOTR rts games and such, they average a couple of hundred polys for each unit (I think... Definitely less than 2000 ;))
Neox
04-05-2004, 03:05 PM
ok you are right, but games like the lotr game or Rome:Total War are designed for massive fighting scenes, i guess if you want a scene with a few thousand units in a shooter you'll need pretty lowpoly units too :)
eek
04-05-2004, 03:38 PM
Its also in truth, how the game engine reads the polys. Whether it can draw true quads. Generally most can only draw tris. Also there are engines out there now that can read patch data, i.e patch models. The great thing with this is that its got inbuilt LOD.
Also you have things like critech's, polybump mapping tools which are amazing. Which in affect bake detail on, through a texture.
The key thing is that its gunna go into the engine as tris anyhow so you might as well model with them.
eek
Also you have things like critech's, polybump mapping tools which are amazing. Which in affect bake detail on, through a texture.
The key thing is that its gunna go into the engine as tris anyhow so you might as well model with them.
eek
Dave Black
04-14-2004, 06:02 PM
I've released a time lapse video of a 15 minute study using the TubeCap primative.
You can check it out here:
http://www.cgtalk.com/showthread.php?s=&threadid=137804
-Dave
You can check it out here:
http://www.cgtalk.com/showthread.php?s=&threadid=137804
-Dave
masi
04-16-2004, 10:36 AM
Originally posted by CarlCampbell
Just read about this a page backa and it has been an issue thats had me wondering for some time now... why does game geometry precise specifically only tris?
well, we're talking about realtime engines - and thats allready the answer to your question, realtime. what does realtime mean? that the engine that is running is able to render the geometry in realtime - so, what's the most simpel type of geometry? verticles and triangles - they're simple, and thats the reason why they're rendered fast - thats why game engine coders use them... best possible quallaty with best possible pc. specs
when the development of hardware in 10-15 years reachs a realtime realworld render possibilitie, than with every more progress they'll add features and possibilities of modifiers, complex and layered materials and so on...
i think in about 20 years you'll have the possibilities to take max files as realtime render engine models
Just read about this a page backa and it has been an issue thats had me wondering for some time now... why does game geometry precise specifically only tris?
well, we're talking about realtime engines - and thats allready the answer to your question, realtime. what does realtime mean? that the engine that is running is able to render the geometry in realtime - so, what's the most simpel type of geometry? verticles and triangles - they're simple, and thats the reason why they're rendered fast - thats why game engine coders use them... best possible quallaty with best possible pc. specs
when the development of hardware in 10-15 years reachs a realtime realworld render possibilitie, than with every more progress they'll add features and possibilities of modifiers, complex and layered materials and so on...
i think in about 20 years you'll have the possibilities to take max files as realtime render engine models
Zeddicus
04-16-2004, 01:16 PM
This is a really great thread (took two afternoons to read through, lol), and has helped me increase my knowledge substantially regarding subD modeling practices. Many thanks to all who are involved!
:beer:
Now for a question. Can't remember what page exactly, but at one point I think setting creases were mentioned as being bad. Just curious if this applies only to organic modeling, or all types of subD modeling. For example, what if I have an object that probably won't be animated, and even if it was it is never going to need to deform? My current WIP model is semi-organic in nature, and instead of using chamfers and what not to create hard edges, I made use of the crease setting. The results look good in areas that blend between inorganic/organic, and it also meant that the complexity of my model didn't have to get out of hand (ok, so I'm slightly lazy too hehe). I'm curious as to what kind of comments this will receive and whether what I am doing is considered "wrong" at all, and why. Thanks in advance! :thumbsup:
:beer:
Now for a question. Can't remember what page exactly, but at one point I think setting creases were mentioned as being bad. Just curious if this applies only to organic modeling, or all types of subD modeling. For example, what if I have an object that probably won't be animated, and even if it was it is never going to need to deform? My current WIP model is semi-organic in nature, and instead of using chamfers and what not to create hard edges, I made use of the crease setting. The results look good in areas that blend between inorganic/organic, and it also meant that the complexity of my model didn't have to get out of hand (ok, so I'm slightly lazy too hehe). I'm curious as to what kind of comments this will receive and whether what I am doing is considered "wrong" at all, and why. Thanks in advance! :thumbsup:
urgaffel
04-16-2004, 01:19 PM
The biggest problem with the crease setting is that you will lose it if you remove the modifier or collapse the stack. Also, I'm not sure just how reliable it is if you keep editing the mesh beneath the meshsmooth modifier...
On the other hand, if it works for you, use it.
On the other hand, if it works for you, use it.
Zeddicus
04-16-2004, 01:19 PM
There is a crease setting for ePoly's as well. ;)
urgaffel
04-16-2004, 01:22 PM
Ah, I never use the subd settings in the epoly rollout. I guess they work fine then :)
Zeddicus
04-20-2004, 10:20 PM
In general I think I prefer the mesh smooth modifier as well. FWIW, I discovered you can set crease amounts in your epoly, and then when you put a mesh smooth modifier on the model it will obey your crease settings. So no worries about losing your creases that way if you delete the modifier or collapse the stack, which is nice. Still wondering though, are creases really all that bad, or can they be our friend?
ivo D
04-21-2004, 07:01 AM
farcry.. for instance have made enourmas progress, there was a article about the bump thing, a texture that made more detail in an object, was pretty weird, dont know how it worked exactly anymore, caus it deformed some smoothed modifier ontop of it i guess.
i wont take to long anymore till we can make highpoly for realtime gaming.
zbrush can make a major roll in the future, looking at the millions of polys allready working realtime in there viewports etc.
i wont take to long anymore till we can make highpoly for realtime gaming.
zbrush can make a major roll in the future, looking at the millions of polys allready working realtime in there viewports etc.
Zeddicus
04-21-2004, 07:02 PM
Speaking of Zbrush (and subdividing in it), I just discovered a really great reason to model using proper technique... chamfered edges instead of creases I mean. OBJ and DXF (and 3DS I noticed too) formats don't support creases. One option would have been to apply mesh smooth by smoothing groups, but Zbrush doesn't support smoothing groups. So, no easy way out if you want hard edges outside of 3DS Max. Guess I answered my own question in the end, lol. ;)
Tom Pawlik
07-05-2004, 12:23 AM
i wont take to long anymore till we can make highpoly for realtime gaming.
We are allready there.
Look here :
http://www.unrealtechnology.com/html/technology/ue30.shtml
The Geforce6 with VS3.0/PS3.0 makes rendering AND very complex shaders realtime.
It's not about the Polys, it's all normalmaps. If you do not want to transform something, you can go and project it's whole geometry on a cube. You will receive an object with just 6 sides and 4 virtices, but you won't be able to see any difference between this one and the original object with it's millions of polys.
cu
Tom
We are allready there.
Look here :
http://www.unrealtechnology.com/html/technology/ue30.shtml
The Geforce6 with VS3.0/PS3.0 makes rendering AND very complex shaders realtime.
It's not about the Polys, it's all normalmaps. If you do not want to transform something, you can go and project it's whole geometry on a cube. You will receive an object with just 6 sides and 4 virtices, but you won't be able to see any difference between this one and the original object with it's millions of polys.
cu
Tom
ivo D
07-05-2004, 11:13 AM
Tom pawlik, i mean real high poly.. this looks high, but it is not.. used normal maps bumps etc etc.
i mean probably in a few years. a lot will be really high poly, so you can make all the nice deformations.
But on the other hand i have my doubt about the extend the game industrie will go to with modelling.
will there be a continues increase of polys in everything, or will it just stop after some point.
a lot of details can be mapped, but there will come a point that it wont matter anymore. looking at games 5 years back.. and now... hmm, than i think that we will be at that stage 7 years or so for sure. or maybe 5...
i mean probably in a few years. a lot will be really high poly, so you can make all the nice deformations.
But on the other hand i have my doubt about the extend the game industrie will go to with modelling.
will there be a continues increase of polys in everything, or will it just stop after some point.
a lot of details can be mapped, but there will come a point that it wont matter anymore. looking at games 5 years back.. and now... hmm, than i think that we will be at that stage 7 years or so for sure. or maybe 5...
Cyberdigitus
07-10-2004, 10:48 AM
Actually the question is, will we still be using polygons.
When a model is so detailed that the polgons take up only a pixel on the screen, it's no use to represent them with triangles eh, you can skip that and just use the points to represent the model. A lot of research is going on in this area. points are added or removed as needed, like lod on meshes. It's a bit like voxels but for surfaces.
When a model is so detailed that the polgons take up only a pixel on the screen, it's no use to represent them with triangles eh, you can skip that and just use the points to represent the model. A lot of research is going on in this area. points are added or removed as needed, like lod on meshes. It's a bit like voxels but for surfaces.
CarlCampbell
07-10-2004, 10:08 PM
Yes, I'm agree! Things will DEFINATELY have much to do with LOD and other sorts of dynamic poly-reducing.
Maybe even pixels will dissappear someday too? :)
Maybe even pixels will dissappear someday too? :)
PiledotNET
07-14-2004, 10:34 PM
When a model is so detailed that the polgons take up only a pixel on the screen, it's no use to represent them with triangles
That's very very interesting.
Do you have any aditional information about this?
That's very very interesting.
Do you have any aditional information about this?
theotheo
07-15-2004, 01:00 PM
It sure would be nice if 3d-graphics cards started supporting voxels
2 cents.
2 cents.
ivo D
07-15-2004, 01:20 PM
look at zbrush, that allready is skipping a lot, you just deform in stead of model by poly.
you deform hundreds of polys with one pull or push of a part.
it wont take long till we just wont need the displacement maps to get that detail in another 3d ab.. maybe it will be all like that.
you deform hundreds of polys with one pull or push of a part.
it wont take long till we just wont need the displacement maps to get that detail in another 3d ab.. maybe it will be all like that.
Cyberdigitus
07-20-2004, 11:55 AM
Piledotnet: there are many (technical) papers to be found, search google for 'point cloud rendering' or 'point based rendering'.
here's a site showing some video's for example:
http://www.irit.fr/~Gael.Guennebaud/research.php
and another one, this one shows that the technique is interesting for interactive modifications:
http://www-sop.inria.fr/reves/personnel/Marc.Stamminger/pbr/
moreover, volumetric effects such as clouds are inherently possible with the technique. Just like the buzz now is on unified lighting (think Doom 3), soon enough the focus will be on a unified geometry representation, like points. Carmack once said that's an important aspect of voxel orcloud representations.
It even runs on pocket pc's:
http://www-sop.inria.fr/reves/publications/data/2003/DD03/?LANG=gb
Imagine what would be possible if GPU manufacturers concentrated more on this then rasterisation.
there's even a symposium on it:
http://graphics.ethz.ch/pbg/
some of their papers:
http://myweb.hinet.net/home7/hks/Papers2004/pbg2004Papers.htm
this site shows some complex scenes:
http://www.gris.uni-tuebingen.de/areas/pbr/index2.html
they mention the similarity with true volumetric voxels. Maybe point based rendering will be the death of the poly, as an inbetween step to voxels. On the other hand, all those techniques can of course coexist.
So the general idea is:
"Point based representations have been proposed as an alternative to triangles for 3D surfaces, with a number of advantages. No ’mesh’, or connectivity information has to be stored explicitly. This allows a simple and compact representation, ideal for fast rendering and editing. When combined with advanced rendering techniques such as splatting [ZPvBG01, Pau03], point based surfaces outperform triangle meshes in terms of rendering quality and data storage flexibility."
here's a site showing some video's for example:
http://www.irit.fr/~Gael.Guennebaud/research.php
and another one, this one shows that the technique is interesting for interactive modifications:
http://www-sop.inria.fr/reves/personnel/Marc.Stamminger/pbr/
moreover, volumetric effects such as clouds are inherently possible with the technique. Just like the buzz now is on unified lighting (think Doom 3), soon enough the focus will be on a unified geometry representation, like points. Carmack once said that's an important aspect of voxel orcloud representations.
It even runs on pocket pc's:
http://www-sop.inria.fr/reves/publications/data/2003/DD03/?LANG=gb
Imagine what would be possible if GPU manufacturers concentrated more on this then rasterisation.
there's even a symposium on it:
http://graphics.ethz.ch/pbg/
some of their papers:
http://myweb.hinet.net/home7/hks/Papers2004/pbg2004Papers.htm
this site shows some complex scenes:
http://www.gris.uni-tuebingen.de/areas/pbr/index2.html
they mention the similarity with true volumetric voxels. Maybe point based rendering will be the death of the poly, as an inbetween step to voxels. On the other hand, all those techniques can of course coexist.
So the general idea is:
"Point based representations have been proposed as an alternative to triangles for 3D surfaces, with a number of advantages. No ’mesh’, or connectivity information has to be stored explicitly. This allows a simple and compact representation, ideal for fast rendering and editing. When combined with advanced rendering techniques such as splatting [ZPvBG01, Pau03], point based surfaces outperform triangle meshes in terms of rendering quality and data storage flexibility."
ivo D
07-20-2004, 12:39 PM
true,, .. it already has been used i nthe movie industrie many times. but when you want to have really highres and good detail, than you have a lot of points, and it always has to be manually adjusted.. or like in the matrix/spiderman, remodelled.. and the scanned model is just used as reference.
i think zbrush is in some way point based, isnt it.. ? it doesnt really store all the poly, and tri data.
i think zbrush is in some way point based, isnt it.. ? it doesnt really store all the poly, and tri data.
Cyberdigitus
07-20-2004, 01:16 PM
Indeed, i started thinking Zbrush actually uses some of those techniques when looking up those papers. here's a paper that is more related to an editing and texturing point could model application: http://www.cg.inf.ethz.ch/Downloads/Publications/Papers/2002/p_Zwi02.pdf
it's actually a paper about an opensource application made to research those topics: http://graphics.ethz.ch/pointshop3d/
Heh, i suppose we have to appologise to hijack this tread, this isn't really about sub-d anymore is it. Oh well, it certainly is teoretical. Maybe this discussion can be contuinued in a thread 'The dead of the poly?'
edit: Actually, subd, zbrush and point clouds are pretty related if you look at the CySlice (http://www.cyberware.com/products/cyslice.html) tool from cyberware; used to create subd surfaces from dense polymeshes, created from scanned point cloud data. there's also a sample model to download with normal maps: http://www.headus.com/au/samples/killeroo/index.html
it's actually a paper about an opensource application made to research those topics: http://graphics.ethz.ch/pointshop3d/
Heh, i suppose we have to appologise to hijack this tread, this isn't really about sub-d anymore is it. Oh well, it certainly is teoretical. Maybe this discussion can be contuinued in a thread 'The dead of the poly?'
edit: Actually, subd, zbrush and point clouds are pretty related if you look at the CySlice (http://www.cyberware.com/products/cyslice.html) tool from cyberware; used to create subd surfaces from dense polymeshes, created from scanned point cloud data. there's also a sample model to download with normal maps: http://www.headus.com/au/samples/killeroo/index.html
ivo D
07-20-2004, 01:31 PM
yea, i think its the same... they also mention brushes in the document.. didnt read it carfully , just looked it over and saw it somewhere.. would make sence that zbrush is the same.. wouldnt know how the hell they could get away with so much polys etc
but oke this paper is from 2002 presentation.. so it can be from 1998 or even earlyer that they developed it.. so what are they making now...............thats the question..
pitty this forum is allmost dead... no one has sub d questions i guess
but oke this paper is from 2002 presentation.. so it can be from 1998 or even earlyer that they developed it.. so what are they making now...............thats the question..
pitty this forum is allmost dead... no one has sub d questions i guess
Rens
08-13-2004, 04:31 PM
Hey guys, I was reading this thread and came across celticdog's method of creating holes (page 28 (http://www.cgtalk.com/showthread.php?t=40373&page=28&pp=15)). I tried it, played with it and out came this image:
http://www.rensheeren.com/images/cgtalk/subdball-s.jpg
Pretty cool, yesno? I like the way how it keeps the reflections quite neat (except the bit just right of the middle).
Stats:
Meshsmooth: 3x
Total polygons: 95K
All quads: yea baby!
See the large image (1600x1200) here (http://www.rensheeren.com/images/cgtalk/subdball-l.jpg).
Aaaand I made an utterly cool desktop image (1600x1200) for all sub-d freaks out there ;)
Get it here (http://www.rensheeren.com/images/cgtalk/subdball-t.jpg).
- Rens
http://www.rensheeren.com/images/cgtalk/subdball-s.jpg
Pretty cool, yesno? I like the way how it keeps the reflections quite neat (except the bit just right of the middle).
Stats:
Meshsmooth: 3x
Total polygons: 95K
All quads: yea baby!
See the large image (1600x1200) here (http://www.rensheeren.com/images/cgtalk/subdball-l.jpg).
Aaaand I made an utterly cool desktop image (1600x1200) for all sub-d freaks out there ;)
Get it here (http://www.rensheeren.com/images/cgtalk/subdball-t.jpg).
- Rens
SimonReeves
08-13-2004, 06:48 PM
ah nice, I shall have to try celticdog's method too :)
tzktime
09-05-2004, 01:35 PM
hey! that one really cool looking sphere! :thumbsup:
CarlCampbell
09-05-2004, 05:13 PM
Neat image, Rens! I know this thread is about modelling but would you mind sharing how you lit the scene and what did you use for material settings?
AlexMateo
09-29-2004, 05:10 PM
cool impresive and good look :D
SwanKnight
10-14-2004, 11:54 AM
Pardon me for my inexperience and perhaps the inconvenience, but what are Sub-d's???
Can anybody direct me to a tutorial link?
I want to learn very much.
Can anybody direct me to a tutorial link?
I want to learn very much.
Rens
10-24-2004, 10:28 PM
Thanks guys. :)
Neat image, Rens! I know this thread is about modelling but would you mind sharing how you lit the scene and what did you use for material settings? Sorry about the late reply.
It's pretty basic really. The lighting is a only skylight with a reflection card above it so the sphere has something to reflect. The material is a raytrace material with a falloff map in the reflection slot (fresnel type, IOR 1,6).
You can read more about reflection cards here (http://www.cgtalk.com/showthread.php?t=159426).
Hope this helps!
- Rens
Neat image, Rens! I know this thread is about modelling but would you mind sharing how you lit the scene and what did you use for material settings? Sorry about the late reply.
It's pretty basic really. The lighting is a only skylight with a reflection card above it so the sphere has something to reflect. The material is a raytrace material with a falloff map in the reflection slot (fresnel type, IOR 1,6).
You can read more about reflection cards here (http://www.cgtalk.com/showthread.php?t=159426).
Hope this helps!
- Rens
Ashwin101
11-23-2004, 07:29 PM
Hello everyone, i am new to this forum and would love as much advice and help as possible...
Cutting to the chase, i am an animation student in london, but i am having problems with the later stages of rigging in 3DS max, using normal bones, not biped. I have created solvers and skined the model to it, but now i can't adjust the envelopes and paint weights?? I have been up nights trying to make the model deform correctly to the bones, but no luck...
Can anyone help as this is a small part of a much bigger animation which is due in soon.
Thank you.
Regards
Doves
Cutting to the chase, i am an animation student in london, but i am having problems with the later stages of rigging in 3DS max, using normal bones, not biped. I have created solvers and skined the model to it, but now i can't adjust the envelopes and paint weights?? I have been up nights trying to make the model deform correctly to the bones, but no luck...
Can anyone help as this is a small part of a much bigger animation which is due in soon.
Thank you.
Regards
Doves
boyjian
12-01-2004, 12:15 PM
perfect! great work ! I like it ,
ivo D
12-01-2004, 12:22 PM
Hello everyone, i am new to this forum and would love as much advice and help as possible...
Cutting to the chase, i am an animation student in london, but i am having problems with the later stages of rigging in 3DS max, using normal bones, not biped. I have created solvers and skined the model to it, but now i can't adjust the envelopes and paint weights?? I have been up nights trying to make the model deform correctly to the bones, but no luck...
Can anyone help as this is a small part of a much bigger animation which is due in soon.
Thank you.
Regards
Doves
this is the sub d thread.. post your question in a new thread , i nthe max forum.
but.. do vertex weighting, dont use envelopes, there is no straight answer to your questrion, so post an image of your model in a new thread etc.
Cutting to the chase, i am an animation student in london, but i am having problems with the later stages of rigging in 3DS max, using normal bones, not biped. I have created solvers and skined the model to it, but now i can't adjust the envelopes and paint weights?? I have been up nights trying to make the model deform correctly to the bones, but no luck...
Can anyone help as this is a small part of a much bigger animation which is due in soon.
Thank you.
Regards
Doves
this is the sub d thread.. post your question in a new thread , i nthe max forum.
but.. do vertex weighting, dont use envelopes, there is no straight answer to your questrion, so post an image of your model in a new thread etc.
sundialsvc4
12-01-2004, 01:21 PM
Pardon me for my inexperience and perhaps the inconvenience, but what are Sub-d's???
Can anybody direct me to a tutorial link?
I want to learn very much.
Sub-D's are subdivision surfaces. Do an advanced-search on Google for the exact-phrase and you'll be inundated with tutorial materials.
The most basic idea is that with these you are able to model good, organic shapes (consisting of curves and so forth) without dealing directly with true mathematical curves ... a process that has been described as trying to build a model with spaghetti and Jell-O. :scream:
Subdivision works by starting with a wireframe model and creating a smooth, curvy-looking model "inside of it," based on the frame. The mathematics of it all definitely looks like PhD stuff. But it works well, giving you a nice mixture of curvy appearance, and control.
Look at that sphere on page-52. (Wow! Is that a length-record for one thread, or what?...) Notice how all of the edges are slightly, nicely, convincingly beveled. Nothing takes a precise 90-degree turn; it's very slightly rounded. All physical molded objects have these, partly so that you won't cut your hand and partly to enable the material to be easily removed from the mold. But ordinary polygon-geometry not only doesn't have that; it's very hard to do. Subdivision techniques can do it precisely.
Another very, very nice feature of subdivisions is that, since the final point positions are calculated, based on a much-smaller set of reference data (the polygons), they can produce arbitrary visual precision without piling-on more geometry. Reposition your camera and ugly flat surfaces won't start showing up. You get a visual effect analogous to having thousands more polygons than you actually built. The concept is loosely similar to that of a NURBS-surface (where points influence a curve by "attracting" it). You get the benefits of a mathematically generated surface, but you have much, much more control.
Can anybody direct me to a tutorial link?
I want to learn very much.
Sub-D's are subdivision surfaces. Do an advanced-search on Google for the exact-phrase and you'll be inundated with tutorial materials.
The most basic idea is that with these you are able to model good, organic shapes (consisting of curves and so forth) without dealing directly with true mathematical curves ... a process that has been described as trying to build a model with spaghetti and Jell-O. :scream:
Subdivision works by starting with a wireframe model and creating a smooth, curvy-looking model "inside of it," based on the frame. The mathematics of it all definitely looks like PhD stuff. But it works well, giving you a nice mixture of curvy appearance, and control.
Look at that sphere on page-52. (Wow! Is that a length-record for one thread, or what?...) Notice how all of the edges are slightly, nicely, convincingly beveled. Nothing takes a precise 90-degree turn; it's very slightly rounded. All physical molded objects have these, partly so that you won't cut your hand and partly to enable the material to be easily removed from the mold. But ordinary polygon-geometry not only doesn't have that; it's very hard to do. Subdivision techniques can do it precisely.
Another very, very nice feature of subdivisions is that, since the final point positions are calculated, based on a much-smaller set of reference data (the polygons), they can produce arbitrary visual precision without piling-on more geometry. Reposition your camera and ugly flat surfaces won't start showing up. You get a visual effect analogous to having thousands more polygons than you actually built. The concept is loosely similar to that of a NURBS-surface (where points influence a curve by "attracting" it). You get the benefits of a mathematically generated surface, but you have much, much more control.
ivo D
12-01-2004, 01:28 PM
basicly, we look for solutions for problems that we encounter, its a sort of puzzling, we all work with polygons(4 sided faces (square) and with that we want to build something.
to do this correctly there are some rulez and solutions for typical problems you could encounter. those solutions are posted and showns here, or discussing a point to get one.
look through this thread and youll found numerous things we came upon, and found solutions for and discussed.
if you have a problem, in terms of how to get a shape build with polygons, you can post it here. (not total structures, human body etc) but specific points.
just read the thread youll get the picture
to do this correctly there are some rulez and solutions for typical problems you could encounter. those solutions are posted and showns here, or discussing a point to get one.
look through this thread and youll found numerous things we came upon, and found solutions for and discussed.
if you have a problem, in terms of how to get a shape build with polygons, you can post it here. (not total structures, human body etc) but specific points.
just read the thread youll get the picture
CoreyJAvitar
12-03-2004, 07:32 PM
:bounce:
I love this thread guys! Thanks so much!
I love this thread guys! Thanks so much!
一张报纸
12-09-2004, 06:44 AM
good!!ok
fei chang hao
fei chang hao
vasquez888
01-18-2005, 09:44 PM
Hello
Just want to say that this thread is extremely informative and I have learnt quite a bit from it.
Not being rude or rushing 3DZ, but is the pdf still going to be released or has it been released elsewhere? I ask this cos I joined up late and there hasn't been a post in this thread for quite some time.
One thing that I don't quite understand is what 3DZ said back on page 44 (yeh I read through every single page!).
"Poles are really important, else we might as well be using nurbs. It's all about locate refinement, and if there is no way to break the loop, the darn thing will stretch across your entire model."
What I don't quite understand is from what I've read, poles are quite bad in that it'll result in shading problems or the like when smoothed. So why are they important?
Also, the impression I got is that edge loops are quite important for the topology when it comes to deforming the model, so how come it's a bad idea for it to stretch across the entire model?
If anyone could explain these two points and in which cases you would want to terminate loops then I'd be most grateful! =D Images to help aid the explaination would also be a bonus! =P
Just want to say that this thread is extremely informative and I have learnt quite a bit from it.
Not being rude or rushing 3DZ, but is the pdf still going to be released or has it been released elsewhere? I ask this cos I joined up late and there hasn't been a post in this thread for quite some time.
One thing that I don't quite understand is what 3DZ said back on page 44 (yeh I read through every single page!).
"Poles are really important, else we might as well be using nurbs. It's all about locate refinement, and if there is no way to break the loop, the darn thing will stretch across your entire model."
What I don't quite understand is from what I've read, poles are quite bad in that it'll result in shading problems or the like when smoothed. So why are they important?
Also, the impression I got is that edge loops are quite important for the topology when it comes to deforming the model, so how come it's a bad idea for it to stretch across the entire model?
If anyone could explain these two points and in which cases you would want to terminate loops then I'd be most grateful! =D Images to help aid the explaination would also be a bonus! =P
ivo D
01-18-2005, 11:27 PM
Hello
JWhat I don't quite understand is from what I've read, poles are quite bad in that it'll result in shading problems or the like when smoothed. So why are they important?
Also, the impression I got is that edge loops are quite important for the topology when it comes to deforming the model, so how come it's a bad idea for it to stretch across the entire model?
If anyone could explain these two points and in which cases you would want to terminate loops then I'd be most grateful! =D Images to help aid the explaination would also be a bonus! =P
the pdf was.. or is in progress. last time ive heard of him it was something of. allmost done.. but i think it will still take a long time before it comes out.
from what ive seen, hes very busy..
but ow well.. just paste the best parts you want to know in a txt file, print it. and look at it every once in a while .
about poles: yes poles are bad and good in another way.
when modelling and getting a 5 sides poly or some other weird non square poly, with some other, some weird effects can occure.. in simple words, when not every little puzzle shape = a(poly/square) , and so not the same sort of shapes in a row, than you will see some disturbans when smoothed. its like throwing a cloth over the floot, than its flat.. but when putting it over a grass field, it wont be ONE NICE FLUID LINE. bad example. but thats the bad part, in not havinga continiues flow of the same shapes, that will than give a bad effect when smoothed.
but when you want to create wrickles ,than you can make poles, to create edges in your surface. there are ways to create them right, but dont have the time to explain that at the moment.
edge loops: edge loops always point out regions, like in a head, you always have edge loops around the eyes, nose, mouth etc. they dont stretch over the whole head, because the eye is a separe part of the head, its also allmost impossible to do that.
edge loops are use to isolate things.. to model a character you have to follow the flow of the skin/muscles, to create a natural look. this flow will result in edge loops automaticly. because you want to islote eyes, they are round, so you will get a loop around them.
also muscles mostly get a loop of polys around them, to connect them with others, and islote them. also the abs,have loops etc. just look at some wire frames and try to see all the loops, and look why they are there, what for effect do they have etc.
JWhat I don't quite understand is from what I've read, poles are quite bad in that it'll result in shading problems or the like when smoothed. So why are they important?
Also, the impression I got is that edge loops are quite important for the topology when it comes to deforming the model, so how come it's a bad idea for it to stretch across the entire model?
If anyone could explain these two points and in which cases you would want to terminate loops then I'd be most grateful! =D Images to help aid the explaination would also be a bonus! =P
the pdf was.. or is in progress. last time ive heard of him it was something of. allmost done.. but i think it will still take a long time before it comes out.
from what ive seen, hes very busy..
but ow well.. just paste the best parts you want to know in a txt file, print it. and look at it every once in a while .
about poles: yes poles are bad and good in another way.
when modelling and getting a 5 sides poly or some other weird non square poly, with some other, some weird effects can occure.. in simple words, when not every little puzzle shape = a(poly/square) , and so not the same sort of shapes in a row, than you will see some disturbans when smoothed. its like throwing a cloth over the floot, than its flat.. but when putting it over a grass field, it wont be ONE NICE FLUID LINE. bad example. but thats the bad part, in not havinga continiues flow of the same shapes, that will than give a bad effect when smoothed.
but when you want to create wrickles ,than you can make poles, to create edges in your surface. there are ways to create them right, but dont have the time to explain that at the moment.
edge loops: edge loops always point out regions, like in a head, you always have edge loops around the eyes, nose, mouth etc. they dont stretch over the whole head, because the eye is a separe part of the head, its also allmost impossible to do that.
edge loops are use to isolate things.. to model a character you have to follow the flow of the skin/muscles, to create a natural look. this flow will result in edge loops automaticly. because you want to islote eyes, they are round, so you will get a loop around them.
also muscles mostly get a loop of polys around them, to connect them with others, and islote them. also the abs,have loops etc. just look at some wire frames and try to see all the loops, and look why they are there, what for effect do they have etc.
vasquez888
01-18-2005, 11:45 PM
Wow thanks Ivo D for the quick and informative reply!
Really cleared up the things I didn't quite understand. =)
Really cleared up the things I didn't quite understand. =)
vasquez888
01-19-2005, 11:48 PM
Hello I was trying out Chin's method of cutting holes in sub-d on page 25. I downloaded his max file and had a play with it. When I added Conform in step 5 to his model, it worked fine.
I then decided to play around with it again, but this time I thought I'd cut the holes into the model instead. This time, when I added the conform modifier, I found that it messed up the edges of the model.
http://www.core16.com/images/conform_error.jpg
Does anyone have any ideas on why this is happening and how to prevent it?
Thanks
I then decided to play around with it again, but this time I thought I'd cut the holes into the model instead. This time, when I added the conform modifier, I found that it messed up the edges of the model.
http://www.core16.com/images/conform_error.jpg
Does anyone have any ideas on why this is happening and how to prevent it?
Thanks
vasquez888
01-20-2005, 12:23 AM
Ok, I figured out why this is happening. It happens if I put the holes on one side of the model and then I mirror that side and weld both sides together.
If I cut the holes on the other side seperately, then it doesn't happen.
How am I supposed to keep both sides symmetrical if I can't do it without the Conform modifier screwing up the edges?
If I cut the holes on the other side seperately, then it doesn't happen.
How am I supposed to keep both sides symmetrical if I can't do it without the Conform modifier screwing up the edges?
vasquez888
01-20-2005, 01:04 AM
Ok, I finally figured out when I get the problem, although I'm not sure why it happens. This is Max 5 by the way. I guess I should have made that clearer from the start.
Anyway, the problem with the messed up edges with Conform occurs when I do a mirror of one side and then I weld both sides together. However, it only happens if I choose the mirrored copy, and attach the original to it. If I choose the original and attach the mirrored copy to it, then I don't get the problem.
Now my question is whether this is a bug with Max 5 or is there a reason why the behaviour is different if I attach in a different order?
Anyway, the problem with the messed up edges with Conform occurs when I do a mirror of one side and then I weld both sides together. However, it only happens if I choose the mirrored copy, and attach the original to it. If I choose the original and attach the mirrored copy to it, then I don't get the problem.
Now my question is whether this is a bug with Max 5 or is there a reason why the behaviour is different if I attach in a different order?
Texlon
01-22-2005, 07:41 PM
Hey guys I wanna participate in this wonderfull thread regarding theoretical Sub-D techniques with a question I got. How in the world is it possible to create a triangle shape on a heavily curved surface with Sub-D? for a better understanding: basically an arrow/triangle type of thing on a sphere or a similar curved object, while that triangle should follow the original shape.
A second question is: How would you approach the shapes shown in the attached image? I really need those shapes for my work. I hope you can help.
btw. If those problems have already been discussed and I looked over it in the thread, please post a diret link to the page (the thread got pretty big).
Thanks,
Leon
A second question is: How would you approach the shapes shown in the attached image? I really need those shapes for my work. I hope you can help.
btw. If those problems have already been discussed and I looked over it in the thread, please post a diret link to the page (the thread got pretty big).
Thanks,
Leon
Texlon
01-22-2005, 08:42 PM
Well,
because of the lack of participation (at this time) and the urgency I tried to solve it myself, and this is what I came up with. Not the best solution (you can still see some little bumps on the surface) but I am kinda happy with the result :)
Now I need help with the second question in the post above this one... couldn't solve it as for now.
bye,
Leon
Edit: Forgot to attach the images ...
because of the lack of participation (at this time) and the urgency I tried to solve it myself, and this is what I came up with. Not the best solution (you can still see some little bumps on the surface) but I am kinda happy with the result :)
Now I need help with the second question in the post above this one... couldn't solve it as for now.
bye,
Leon
Edit: Forgot to attach the images ...
aliasali
01-30-2005, 07:23 PM
Heeey! im so interested to read this thread.but incredible! 54 pages????
how can i start readinem! plz! i want to learn them too!:D
kidding.thank u dave black to start up such a great and usefull thread.gonna read em all ! (yeah?)!:D
how can i start readinem! plz! i want to learn them too!:D
kidding.thank u dave black to start up such a great and usefull thread.gonna read em all ! (yeah?)!:D
jcjimenez
03-08-2005, 01:58 PM
definitely this is one of the most great thread I've ever seen :)
Dervish
04-13-2005, 09:20 AM
great thread :thumbsup: learned more from it then reading endless tutorials, and only at page 10 yet , someone ( not me, im the lazy guy ) really should compile all these really helpfull stuff into a couple pages without bumps ( like this comment ), im still dazed of all that helpfull input.
considering to completly rework my current WIP now. cause a lotta mistake i couldve avoided knowing what i know now ( http://www.cgtalk.com/showthread.php?t=229482 ).
considering to completly rework my current WIP now. cause a lotta mistake i couldve avoided knowing what i know now ( http://www.cgtalk.com/showthread.php?t=229482 ).
Lincoln Hawk
07-20-2005, 12:42 PM
My contribution, ipod u2 only sub-d, no aditional pieces.
http://i6.photobucket.com/albums/y220/Lincoln_Hawk/ipod_u2_web.jpg
http://i6.photobucket.com/albums/y220/Lincoln_Hawk/ipod_u2_wire_web.jpg
http://i6.photobucket.com/albums/y220/Lincoln_Hawk/ipod_u2_web.jpg
http://i6.photobucket.com/albums/y220/Lincoln_Hawk/ipod_u2_wire_web.jpg
Ian Jones
07-27-2005, 11:52 AM
Lincoln Hawk: There's some smoothing issues around this corner. I can't really help as I'm a newbie at sub-d but im just pointing it out. See attached image.
ToddD
07-27-2005, 02:27 PM
Lincoln(and Ian) Here is how I would model the corner you are refering to, hope this helps:
http://img.photobucket.com/albums/v99/tbonz/podcorner-01.jpg
Todd
http://img.photobucket.com/albums/v99/tbonz/podcorner-01.jpg
Todd
LouisCho
07-28-2005, 11:52 PM
Hi all
I am making a piston and i want to make a hole for the connecting rod. I am trying to make the hole with the method of celticdog at the page 28 of the theorical sub-d thread http://forums.cgsociety.org/showthread.php?p=483183&mode=linear#post483183
Yeah with a brand new sphere it's rock but it's make a strange oval in a cylinder. I don't understand how place verticles to avoid this longitudinal deformation and to have a great round hole.
So I make my verticles
I collapse the vertex
I chamfer
And I drop my &%$ CPU in the stairs :twisted:
Surely that i miss something in the celticdog method but what? I tried to move verticles by hand but it's make a kind of heavy mess when i apply meshsmooth.
Front wires, before chamfer
http://img179.imageshack.us/img179/9995/wiresfront2cr.jpg
Wires after chamfer
http://img179.imageshack.us/img179/2223/wireschamfer1lj.jpg
render, no hole
http://img179.imageshack.us/img179/5715/pistonrender7nd.jpg
Is there someone who can explain me that?
I am making a piston and i want to make a hole for the connecting rod. I am trying to make the hole with the method of celticdog at the page 28 of the theorical sub-d thread http://forums.cgsociety.org/showthread.php?p=483183&mode=linear#post483183
Yeah with a brand new sphere it's rock but it's make a strange oval in a cylinder. I don't understand how place verticles to avoid this longitudinal deformation and to have a great round hole.
So I make my verticles
I collapse the vertex
I chamfer
And I drop my &%$ CPU in the stairs :twisted:
Surely that i miss something in the celticdog method but what? I tried to move verticles by hand but it's make a kind of heavy mess when i apply meshsmooth.
Front wires, before chamfer
http://img179.imageshack.us/img179/9995/wiresfront2cr.jpg
Wires after chamfer
http://img179.imageshack.us/img179/2223/wireschamfer1lj.jpg
render, no hole
http://img179.imageshack.us/img179/5715/pistonrender7nd.jpg
Is there someone who can explain me that?
zukezuko
07-29-2005, 12:31 AM
i am no expert at this
but as far as i can see you have no horizontal lines on your
cylinder. Those horizontal edges coming out of the hole should continue further.
no wonder you get weird results. :)
but as far as i can see you have no horizontal lines on your
cylinder. Those horizontal edges coming out of the hole should continue further.
no wonder you get weird results. :)
Zeko
07-29-2005, 02:37 AM
Louis.cho did u extruded the polys inwards and deleted them? And then chamfer the edge loop?
LouisCho
07-29-2005, 01:31 PM
Louis.cho did u extruded the polys inwards and deleted them? And then chamfer the edge loop?
after the chamfer i want to bevel ,01 extrude inward and bevel again. Like the trick to the page 28.
I can do it now because the oval shape, i want a perefct cercle.
I will try to do like zukezuko said and give some news(after the job!)
after the chamfer i want to bevel ,01 extrude inward and bevel again. Like the trick to the page 28.
I can do it now because the oval shape, i want a perefct cercle.
I will try to do like zukezuko said and give some news(after the job!)
JonyK
09-13-2005, 10:59 AM
Excellent thread (no duh), but I saw most people posting about circles, thats fine, but what about some kind of oval or triangle shape?
http://www.jonkirkpatrick.com/oval_triang.jpg
wire:
http://www.jonkirkpatrick.com/oval_triang_wire.jpg
http://www.jonkirkpatrick.com/oval_triang.jpg
wire:
http://www.jonkirkpatrick.com/oval_triang_wire.jpg
ToddD
09-13-2005, 02:35 PM
JonyK, that topology is fine on a flat surface, but if you had curves in 1 or 2 directions that wouldn't work well.
JonyK
09-13-2005, 02:56 PM
JonyK, that topology is fine on a flat surface, but if you had curves in 1 or 2 directions that wouldn't work well.
It's actually on a sphere, a pretty high def sphere so it is curving at an angel, but probably nothing to radical.
edit: I did take a few more looks at it and it seems it does create a deformation around the object, thanks for your input.
It's actually on a sphere, a pretty high def sphere so it is curving at an angel, but probably nothing to radical.
edit: I did take a few more looks at it and it seems it does create a deformation around the object, thanks for your input.
Lincoln Hawk
09-15-2005, 11:58 AM
Thanks ToddD, sorry for a long time. Help me a lot.
Velocamonkey
09-19-2005, 03:01 PM
Cool this seems like the place to be, I'm trying to model some detail on a curved surface. So I've got some circular indents in it which I can model fine, see the first couple of pics. I can cut and end my edges efficiently and get a nice smooth circle. When it smooths it also keeps the curve of the surounding area and so the surface smoothly blends into this circular indent area. All is good using similar techniques to previous responses.
PIC
http://www.michael-hull.co.uk/PROBLEM_01.jpg
My problem comes when I need to add more detail, have something that doesn't cut into the surrounding geometry quite so cleanly, like a square shaped indent as before, not a circle. So I need 4 edges for the square, but then more to get the corners and to stop it turning into a circle when smoothed, so that's 8 edges I guess.
This means I'll need more cuts in the corners to maintain it's sharp edges when it is smoothed, but I can't seem to cut these in an efficient way without messing up the surrounding topology when smoothed. Please see my different approaches in the pics. I also tried as in the first picture, but finishing the edge loops at these corners, creating tris which is bad, and also it's not great when smoothed as pinching occurs.
PIC
http://www.michael-hull.co.uk/PROBLEM_02.jpg
I would greatly appreciate it if you could shed even the smallest light on my problem, maybe I'm just missing something really obvious, but this is the only thing in modelling that keeps tripping me up!
I've tried using the polysculpt in Maya and smoothing down or my bad edges, but this doesn't help. I guess the question is how do you make cuts into an all ready regular curved surface without messing it up when it's smoothed?
Thanks for reading this, I appreciate the time.
PIC
http://www.michael-hull.co.uk/PROBLEM_01.jpg
My problem comes when I need to add more detail, have something that doesn't cut into the surrounding geometry quite so cleanly, like a square shaped indent as before, not a circle. So I need 4 edges for the square, but then more to get the corners and to stop it turning into a circle when smoothed, so that's 8 edges I guess.
This means I'll need more cuts in the corners to maintain it's sharp edges when it is smoothed, but I can't seem to cut these in an efficient way without messing up the surrounding topology when smoothed. Please see my different approaches in the pics. I also tried as in the first picture, but finishing the edge loops at these corners, creating tris which is bad, and also it's not great when smoothed as pinching occurs.
PIC
http://www.michael-hull.co.uk/PROBLEM_02.jpg
I would greatly appreciate it if you could shed even the smallest light on my problem, maybe I'm just missing something really obvious, but this is the only thing in modelling that keeps tripping me up!
I've tried using the polysculpt in Maya and smoothing down or my bad edges, but this doesn't help. I guess the question is how do you make cuts into an all ready regular curved surface without messing it up when it's smoothed?
Thanks for reading this, I appreciate the time.
mustan9
09-19-2005, 03:39 PM
My problem comes when I need to add more detail, have something that doesn't cut into the surrounding geometry quite so cleanly, like a square shaped indent as before, not a circle.
One thing I hate about modelling is when you have to stop and ask yourself "How do I keep this quad?".
I can see to ways of doing this. One is to pinch the edges into the four corners, or to create extra parellel edges. As for the shading bumps you must keep your edges more spaced apart, and then manually try to keep the curved surface shape you want.
NOTE: I kept things apart so you can see the edges. So no tight corners.
http://img53.imageshack.us/img53/7718/wireframe4lt.jpg (http://imageshack.us)
http://img53.imageshack.us/img53/7831/shaded2pg.jpg (http://imageshack.us)
The above is the best way I can think to insert a sqaure shape into a quad polygon. If you want to avoid shading problems then I think you have to create more edges then you were orginally doing.
http://img53.imageshack.us/img53/9681/wireframe25am.jpg (http://imageshack.us)
http://img53.imageshack.us/img53/7969/shaded21pe.jpg (http://imageshack.us)
One thing I hate about modelling is when you have to stop and ask yourself "How do I keep this quad?".
I can see to ways of doing this. One is to pinch the edges into the four corners, or to create extra parellel edges. As for the shading bumps you must keep your edges more spaced apart, and then manually try to keep the curved surface shape you want.
NOTE: I kept things apart so you can see the edges. So no tight corners.
http://img53.imageshack.us/img53/7718/wireframe4lt.jpg (http://imageshack.us)
http://img53.imageshack.us/img53/7831/shaded2pg.jpg (http://imageshack.us)
The above is the best way I can think to insert a sqaure shape into a quad polygon. If you want to avoid shading problems then I think you have to create more edges then you were orginally doing.
http://img53.imageshack.us/img53/9681/wireframe25am.jpg (http://imageshack.us)
http://img53.imageshack.us/img53/7969/shaded21pe.jpg (http://imageshack.us)
Velocamonkey
09-19-2005, 04:05 PM
Yeah definitely that works in terms of adding all the detail and terminating it in the corners, it keeps it all quads BUT those corners are pointy when it's smoothed which I really don't want or solve my problem. Also adding in those extra loops at the end have made your whole sphere have a square feel in your last picture where they are now. These are the problems I can't defeat, please someone find the solution!!
m@Xist
09-25-2005, 04:06 PM
I tried before to get this thingy on the sphere. The prob is that when you extrude the face and then apply the smooth modifier, it gets round because the faces around that face ar not sub-ded.
I started with a new sphere and converted it to a poly. I then selected a face and applied the MSmooth 3 times, which gave me smooth results so far.
Then beveled and extruded. (You can refine the edges for better form of the thingy like yours)
Then made nurms sub-d with 2 iterations.
It's not perfect at all, sub-d seems to have probs with the tris caused by MSmooth. I probably could try to use more inserts. Any suggestions for improvment?
I don' check this one by Jonky! Could you explain the details plz.
http://brainworx.rules.li/brainscan/cgtalk/sphereRectangle.jpg
I started with a new sphere and converted it to a poly. I then selected a face and applied the MSmooth 3 times, which gave me smooth results so far.
Then beveled and extruded. (You can refine the edges for better form of the thingy like yours)
Then made nurms sub-d with 2 iterations.
It's not perfect at all, sub-d seems to have probs with the tris caused by MSmooth. I probably could try to use more inserts. Any suggestions for improvment?
I don' check this one by Jonky! Could you explain the details plz.
http://brainworx.rules.li/brainscan/cgtalk/sphereRectangle.jpg
Muhneer
10-18-2005, 02:32 PM
I had a problem which looked a bit like that, i fixed it with changing the corner
its not perfect. Only the right top corner is done :
http://www.muhneer.nl/3D/subdivision_square.jpg
http://www.muhneer.nl/3D/subdivision_square_2.jpg
its not perfect. Only the right top corner is done :
http://www.muhneer.nl/3D/subdivision_square.jpg
http://www.muhneer.nl/3D/subdivision_square_2.jpg
Muhneer
10-18-2005, 02:36 PM
At the moment i have a problem with Sub-D modelling i was building a round
cylinder with square parts coming out of it, i am almost done but i cant get it
finished because it smooths not good.
If ya like here is the url to the post : http://forums.cgsociety.org/showthread.php?t=286716
I hope this thread gets its speed back, like in the beginning, its a wonderfull thread.
cylinder with square parts coming out of it, i am almost done but i cant get it
finished because it smooths not good.
If ya like here is the url to the post : http://forums.cgsociety.org/showthread.php?t=286716
I hope this thread gets its speed back, like in the beginning, its a wonderfull thread.
mustan9
10-18-2005, 02:59 PM
What software are you guys using?
I was playing around with SoftImage XSI 5, and they have a hard edge feature. Solves this problem real quick! Wish they had that in Max.
I was playing around with SoftImage XSI 5, and they have a hard edge feature. Solves this problem real quick! Wish they had that in Max.
Dzignguy
10-18-2005, 03:38 PM
I was playing around with SoftImage XSI 5, and they have a hard edge feature. Solves this problem real quick! Wish they had that in Max.
How about an example of how this "hard edge" thing works. Perhaps a demonstration of the same example, Square hole in sphere, would be appropiate. A smoothed and wireframe view would be nice! Pictures are worth a million words.
Thanks,
john
How about an example of how this "hard edge" thing works. Perhaps a demonstration of the same example, Square hole in sphere, would be appropiate. A smoothed and wireframe view would be nice! Pictures are worth a million words.
Thanks,
john
Dave Black
10-18-2005, 03:51 PM
MAx does have a "hard edge" feature. It's called edge tension. Or, you can use the smooth by smoothing groups option in MS. However, these methods, as well as XSI's hard-edge feature do not do what we are talking about in this thread. Hard edges are not a good thing in all cases. Usually, one wants a soft beveled edge, which this method give you.
-Dave
-Dave
bardur
10-18-2005, 04:06 PM
Dave... have you finished you Sub-D FAQ?
If so could i have a copy of it
Regards
Bardur
If so could i have a copy of it
Regards
Bardur
Y0da
10-25-2005, 09:15 AM
Regarding the square thing on the sphere. I have discovered a neat little trick that solves the square edges problem when you Sub-D it. Apply a 'Spherify' modifier after you Sub-D! It works wonders for holes cut into spheres as well. This is in Max of course. ;)
Tony Richardson
10-31-2005, 04:41 AM
Here's a little something I was playing with just to test a few things. It's very simple but it does show what different types of loops there can be for basic sub'd. Obviously, everything is one mesh.
If your interestred I can email the file. Sorry MAX8 only.
Tony
If your interestred I can email the file. Sorry MAX8 only.
Tony
Tony Richardson
10-31-2005, 07:33 AM
Well, I think I finally figured out how to do a respectable looking square whole in a sphere! I haven't been keeping up with the thread so if this is old news I apologize.
The first image is a very close-up view of the edges rendered.
The second image is a screen shot of the square hole with meshsmooth(2).
The third is a screen shot without meshsmooth applied.
It turned out to be a very simple process, once I figured out how to do it. That took a little time. There is no distortion at the corners and the smoothing looks good, at least to me.
If anyone is still interested in making a square hole in a sphere, I can do a step by step to show how it is done. This process should also work for a cylinder.
Tony
The first image is a very close-up view of the edges rendered.
The second image is a screen shot of the square hole with meshsmooth(2).
The third is a screen shot without meshsmooth applied.
It turned out to be a very simple process, once I figured out how to do it. That took a little time. There is no distortion at the corners and the smoothing looks good, at least to me.
If anyone is still interested in making a square hole in a sphere, I can do a step by step to show how it is done. This process should also work for a cylinder.
Tony
Dzignguy
10-31-2005, 12:25 PM
Tony Richardson wrote:
If anyone is still interested in making a square hole in a sphere, I can do a step by step to show how it is done. This process should also work for a cylinder.
I'm deffinately interested! I just tried your implementation, at least as far as I can see, and still get a flatened looking edge around the square. It certainly does not look like your final render. Perhaps what you are getting is in the order or method of your implementation...... I'm all eyes.....
thanks,
John
If anyone is still interested in making a square hole in a sphere, I can do a step by step to show how it is done. This process should also work for a cylinder.
I'm deffinately interested! I just tried your implementation, at least as far as I can see, and still get a flatened looking edge around the square. It certainly does not look like your final render. Perhaps what you are getting is in the order or method of your implementation...... I'm all eyes.....
thanks,
John
Tony Richardson
10-31-2005, 04:55 PM
Ok, here is my method. It only takes a minute or two once you get the steps down correctly. This was done on a 24" dia. sphere with 32 segments.
1# Cut the sphere square so that there are four triangles.
2# Select each cut and use the connect command set to one to form a square. Adjust the size of the square as needed with scale. I also used edge constraint(under edit geometry) for all this to make sure it stays flat.
3# Select the four faces within the square you created and inset by .05 inches. Do this again to from three squares within each other as shown.Now, cut from the corners of the inner most square out to the edge of the sphere square as shown in image 2.
5# Select the inner four triangle faces and then hit grow one time to select the faces surrounding them as shown in image 2.
6# Extrude the faces inward -.05 inches and close the dialog box.
Extrude again the depth you want the square minus .05 inches and close the dialog box.I did 1/2" for this.
Extrude again -.05 inches and close the dialog box.
(NOTE) Be sure to use Local Normal when extruding.
7# IF you want to, you can collapse(weld) the 2 vertices for each outer edges and move them to the center of each outer edge of the square.It's not neccassary, but it looks cool.
Thats it! You should not have to adjust anything to get a smooth square hole.
Tony
1# Cut the sphere square so that there are four triangles.
2# Select each cut and use the connect command set to one to form a square. Adjust the size of the square as needed with scale. I also used edge constraint(under edit geometry) for all this to make sure it stays flat.
3# Select the four faces within the square you created and inset by .05 inches. Do this again to from three squares within each other as shown.Now, cut from the corners of the inner most square out to the edge of the sphere square as shown in image 2.
5# Select the inner four triangle faces and then hit grow one time to select the faces surrounding them as shown in image 2.
6# Extrude the faces inward -.05 inches and close the dialog box.
Extrude again the depth you want the square minus .05 inches and close the dialog box.I did 1/2" for this.
Extrude again -.05 inches and close the dialog box.
(NOTE) Be sure to use Local Normal when extruding.
7# IF you want to, you can collapse(weld) the 2 vertices for each outer edges and move them to the center of each outer edge of the square.It's not neccassary, but it looks cool.
Thats it! You should not have to adjust anything to get a smooth square hole.
Tony
Tony Richardson
10-31-2005, 05:45 PM
This also works using 4 squares on a sphere for large hole.
Steve Green
11-01-2005, 08:52 PM
This is a great thread - Has anyone seen these kind of methods scripted yet?
If you're only working on a single quad to start with, I wonder if it is possible to do all the steps for creating a square or circular hole in a quad in a nicely scripted modifier.
- Steve
If you're only working on a single quad to start with, I wonder if it is possible to do all the steps for creating a square or circular hole in a quad in a nicely scripted modifier.
- Steve
urgaffel
11-02-2005, 04:55 PM
Tony, great job. That's a really good method, thanks :)
Mutagen Media
11-04-2005, 02:17 AM
Not sure if this was covered, since I gave up on reading this whole thread about half way through, but I can't help but postulating:
It seems to me that the deal with Ngons reduces to this: Ngons don't stay Ngons (at least in Max, from my experience). Max converts all geometry to tris eventually. The point with dealing with Ngons is this: do you want to explicitly control how the Ngon is divided, or do you trust Max to use its best judgement. The answer is obvious: Max is known to get this wrong, so you should always divide an Ngon into Tris and/or Quads. So factor out the whole Ngon part of the Sub-D argument.
Is there some flaw to this reasoning?
- B
It seems to me that the deal with Ngons reduces to this: Ngons don't stay Ngons (at least in Max, from my experience). Max converts all geometry to tris eventually. The point with dealing with Ngons is this: do you want to explicitly control how the Ngon is divided, or do you trust Max to use its best judgement. The answer is obvious: Max is known to get this wrong, so you should always divide an Ngon into Tris and/or Quads. So factor out the whole Ngon part of the Sub-D argument.
Is there some flaw to this reasoning?
- B
urgaffel
11-04-2005, 02:34 AM
Max converts everything to tris once you render, sure but the turbosmooth and meshsmooth and hsds modifiers doesn't just turn everything into tris... I think you got the wrong impression somewhere along the line. The thing about ngons is that they can cause weird shading and they can act up when deformed. As it is with everything else, you have to know when and where to use them. No one is saying that you can't use them, just use them wisely.
eek
11-04-2005, 02:36 AM
no, but max's own subdivider is based on a Pixar (sabre) algorithm which, adaptively divides the face. And therefore very good. Its always best to use quads i find, and tris very rarely. Tris and ngons cause rendering artifact.
The hardest part of all is having the right amount of poles to each vertex (4), man that is hard!
eek
The hardest part of all is having the right amount of poles to each vertex (4), man that is hard!
eek
SyaPed
11-09-2005, 11:56 PM
Hi there, I tried searching through this awesome monstrous thread but couldn't find out if the .pdf compilation of all the great info had been completed?
Thanks!
Thanks!
urgaffel
11-10-2005, 04:54 PM
Dave Black was working on it but I think he's been too busy to finish it.
voyer
11-30-2005, 02:19 AM
cool...
thanks for this.
thanks for this.
Marzocchi
12-04-2005, 02:37 PM
Now convert it to an editable mesh.
Now convert it to an editable poly.
Noobish question:
Why do you first convert it to a editable mesh and then poly?
IS there any benifits doing this instead of mkaing it into a editable poly directly?
Now convert it to an editable poly.
Noobish question:
Why do you first convert it to a editable mesh and then poly?
IS there any benifits doing this instead of mkaing it into a editable poly directly?
Dave Black
12-05-2005, 04:58 PM
What was the context in which I said that? Probably when dealing with cylinders. There was a bug in max 5 that duplicated the bottom faces of the cylinder when converted directly to EP. In later versions, this is not necessary.
How that helps.
-Dave
How that helps.
-Dave
johanflood
12-28-2005, 10:15 AM
Ok, here is my method. It only takes a minute or two once you get the steps down correctly. This was done on a 24" dia. sphere with 32 segments.
1# Cut the sphere square so that there are four triangles.
2# Select each cut and use the connect command set to one to form a square. Adjust the size of the square as needed with scale. I also used edge constraint(under edit geometry) for all this to make sure it stays flat.
3# Select the four faces within the square you created and inset by .05 inches. Do this again to from three squares within each other as shown.Now, cut from the corners of the inner most square out to the edge of the sphere square as shown in image 2.
5# Select the inner four triangle faces and then hit grow one time to select the faces surrounding them as shown in image 2.
6# Extrude the faces inward -.05 inches and close the dialog box.
Extrude again the depth you want the square minus .05 inches and close the dialog box.I did 1/2" for this.
Extrude again -.05 inches and close the dialog box.
(NOTE) Be sure to use Local Normal when extruding.
7# IF you want to, you can collapse(weld) the 2 vertices for each outer edges and move them to the center of each outer edge of the square.It's not neccassary, but it looks cool.
Thats it! You should not have to adjust anything to get a smooth square hole.
Tony
thanks for your post and I had to test it with only quads.
from the bottom of the hole I deleted edges so I got 9 quads remaining. Got rid of the x-shape. have you tried that ?
1# Cut the sphere square so that there are four triangles.
2# Select each cut and use the connect command set to one to form a square. Adjust the size of the square as needed with scale. I also used edge constraint(under edit geometry) for all this to make sure it stays flat.
3# Select the four faces within the square you created and inset by .05 inches. Do this again to from three squares within each other as shown.Now, cut from the corners of the inner most square out to the edge of the sphere square as shown in image 2.
5# Select the inner four triangle faces and then hit grow one time to select the faces surrounding them as shown in image 2.
6# Extrude the faces inward -.05 inches and close the dialog box.
Extrude again the depth you want the square minus .05 inches and close the dialog box.I did 1/2" for this.
Extrude again -.05 inches and close the dialog box.
(NOTE) Be sure to use Local Normal when extruding.
7# IF you want to, you can collapse(weld) the 2 vertices for each outer edges and move them to the center of each outer edge of the square.It's not neccassary, but it looks cool.
Thats it! You should not have to adjust anything to get a smooth square hole.
Tony
thanks for your post and I had to test it with only quads.
from the bottom of the hole I deleted edges so I got 9 quads remaining. Got rid of the x-shape. have you tried that ?
custinyakis
01-13-2006, 02:35 AM
Hello everyone ..! I've been newly working on modeling and I have a simple problem again but it's crucial for me to understand this to get the principles and basics of polygonal Modeling. If you look at the images below you'll get the idea . . .
I applied solid chamfer to an edge group and it gave me 4 pentagons.
My question is : What cuts do i need to quadrify it to a stage where it doesn't give me the slightest pinching when meshsmoothed?
I really need to figure this out. Thanks everyone . . . Take care !
http://www.tr3d.com/dosyalar/TryElbW7n.jpg
http://www.tr3d.com/dosyalar/72KS7c3Qc.jpg
http://www.tr3d.com/dosyalar/HNiYXvh9a.jpg
I applied solid chamfer to an edge group and it gave me 4 pentagons.
My question is : What cuts do i need to quadrify it to a stage where it doesn't give me the slightest pinching when meshsmoothed?
I really need to figure this out. Thanks everyone . . . Take care !
http://www.tr3d.com/dosyalar/TryElbW7n.jpg
http://www.tr3d.com/dosyalar/72KS7c3Qc.jpg
http://www.tr3d.com/dosyalar/HNiYXvh9a.jpg
Unkle Bunkle
01-16-2006, 09:33 AM
I know exactly what you're talking about, check this thread out:
http://forums.cgsociety.org/showthread.php?t=308658&page=1&pp=15
The conclusive answer seems to be to build it as cleanly as you can, then just fiddle until it hopefully doen't pinch.
http://forums.cgsociety.org/showthread.php?t=308658&page=1&pp=15
The conclusive answer seems to be to build it as cleanly as you can, then just fiddle until it hopefully doen't pinch.
Dave Black
01-17-2006, 04:09 PM
custinyakis:
You have ngons at the corners. Tie-off those ngons with quads or tris and it will eliminate or reduce the effect.
-Dave
You have ngons at the corners. Tie-off those ngons with quads or tris and it will eliminate or reduce the effect.
-Dave
custinyakis
01-19-2006, 04:07 PM
A Lovely Thread !!! . . . Thank you so much Unkle Bunkle.
Dave Black, can you show me what I can do to reduce the pinching of the edges ??? with some simple lines in photoshop on the image ? I would really appreciate that. I've been reading this thread for a very long time and I learned a lot from it. To make a square hole on a sphere for example( This is a very similar problem with this one isn't it ? )
By the way I changed the previous images' resolutions . . . they were disturbing the overall design of the page. Sorry about that : )
Some people are using Rhinoceros NURBS for complex industrial designs instead of Sub-divs. Why are they using Rhino ? Are NURBS tools in Rhinoceros better than the NURBS tools in Max ?
And what what happens if I animate, bend, twist a NURBS model ? Are they animation friendly ?
Thanks everyone who type even a single word to this thread. Take care !
Dave Black, can you show me what I can do to reduce the pinching of the edges ??? with some simple lines in photoshop on the image ? I would really appreciate that. I've been reading this thread for a very long time and I learned a lot from it. To make a square hole on a sphere for example( This is a very similar problem with this one isn't it ? )
By the way I changed the previous images' resolutions . . . they were disturbing the overall design of the page. Sorry about that : )
Some people are using Rhinoceros NURBS for complex industrial designs instead of Sub-divs. Why are they using Rhino ? Are NURBS tools in Rhinoceros better than the NURBS tools in Max ?
And what what happens if I animate, bend, twist a NURBS model ? Are they animation friendly ?
Thanks everyone who type even a single word to this thread. Take care !
Optigon-cg
01-19-2006, 08:23 PM
I would try this (see image). What appears to be a tri is actually a quad. The edge running through the center of your chamfer terminates there giving it its 4th vert.
Unkle Bunkle
01-20-2006, 09:01 AM
Methinks that cut will create pinching, that's a very narrow polygon within the rest that are mostly quite huge.
I've found that you want to keep all the faces as similarly sized as possible, this way when it smooths it won't pinch so much.
Think about it this way, when it smooths it adds lot's of divisions between edges, if you have wide spaced out edges then 2 that are really close, when it smooths this'll pinch (unless on a flat surface).
I've found that you want to keep all the faces as similarly sized as possible, this way when it smooths it won't pinch so much.
Think about it this way, when it smooths it adds lot's of divisions between edges, if you have wide spaced out edges then 2 that are really close, when it smooths this'll pinch (unless on a flat surface).
custinyakis
01-20-2006, 03:59 PM
Option-cg thank you ... I tried that but it did not prevent the pinching. Moreover by this way you only make the triangle a quad polygon but others remain pentagons. Through this conversation I come up with a new question . . .
As I'm reading the tutorials in some 3D CG Websites I see that lots of the modelers use triangles & ngons in mechanical polygonal models. See the images below . . .
http://www.tr3d.com/dosyalar/Nz4y1G89U.jpg
http://www.tr3d.com/dosyalar/X94H8w6h4.jpg
What about the ''All-Quads'' ? Can we use traingles & ngons in some cases ?
Unkle Bunkle you're right !! The same-sized polygons are definitely better.
** Actually I don't have permission to use these images. They are from the tutorials of Munkmotion and David Melchor Diaz. Hope they don't get mad ( :
As I'm reading the tutorials in some 3D CG Websites I see that lots of the modelers use triangles & ngons in mechanical polygonal models. See the images below . . .
http://www.tr3d.com/dosyalar/Nz4y1G89U.jpg
http://www.tr3d.com/dosyalar/X94H8w6h4.jpg
What about the ''All-Quads'' ? Can we use traingles & ngons in some cases ?
Unkle Bunkle you're right !! The same-sized polygons are definitely better.
** Actually I don't have permission to use these images. They are from the tutorials of Munkmotion and David Melchor Diaz. Hope they don't get mad ( :
Ian Jones
03-14-2006, 11:25 AM
custinyakis: I think that in certain situations it works just fine. The examples you give show poly's which are in a co-planar <? (parralel / uniform surface) with which any distortion in the resultant subdivision would be pretty much flat and hardly visible if at all. Reflections may be distorted though? I think...
nudelsalat
04-01-2006, 02:40 PM
Hi, i resently started to try using sub-divisions for some technical illustrations, but the results arent what i expected. isnt a box supposed to turn into a perfect sphere when subdivided? my problem is, that i need to create a cylinder shaped plug and i wanted to use a subdivided box(since modeling with a box would be much easier) to create one but it does not turn into a cylinder. =/
tedzbug
06-07-2006, 10:58 PM
Ok, here is my method. It only takes a minute or two once you get the steps down correctly. This was done on a 24" dia. sphere with 32 segments.... .....Tony
Awesome, this is a big help for me......... i was having some real tough issues figuring out how to make this sort of geometry. Thx
http://i72.photobucket.com/albums/i162/Zdimensional/new-shapes.jpg
Awesome, this is a big help for me......... i was having some real tough issues figuring out how to make this sort of geometry. Thx
http://i72.photobucket.com/albums/i162/Zdimensional/new-shapes.jpg
amos_chid
03-05-2007, 08:44 AM
This is a very great thread, i've spent all day reading this valuable resource
my only concern is, the first 20-30 pages arealmost useless since most of the images are already gone.
i guess it will be very useful if anyone use image attachment so we can still see the image
all the time (i know it will be hard on cgtalk server though).
my only concern is, the first 20-30 pages arealmost useless since most of the images are already gone.
i guess it will be very useful if anyone use image attachment so we can still see the image
all the time (i know it will be hard on cgtalk server though).
Ashford_Butler
03-18-2007, 04:31 AM
Thought this may help http://www.tweakcg.com/forum/viewtopic.php?t=251
middle of the page.
middle of the page.
youngy
06-04-2007, 04:10 PM
Following on from what Tony Richardson has been posting, how about this arrangement for extruding a square shape on a curved surface etc and keeping everything 4-sided? Such a tricky thing to achieve without distortion! Interesting thread though. Would love to see some more options.
Laa-Yosh
06-05-2007, 09:39 AM
Time to revive this thread again...
I've been doing a lot of technical modeling lately, and to make things worse it's for Mental Ray which in our current pipeline won't support n-sided polygons. Cool :)
(another fun part is if you want to use Zbrush on the result... :P)
Some notes...
- In general, it's very important to choose a right level of detail for your control cage. Everything is fine and cool as long as you're trying it on a cube, maybe even on a cylinder; but build something more complex and you'll discover that the techniques explored in this thread can fall apart very quickly. Drilling holes into an 8-sided cylinder will cause a lot of pinching no matter how clever you are; using 16 or 20 sides or even more will make it a lot easier. The same goes for more complex shapes too, like cars, armor and such.
It's pretty much the key part IMHO. Yeah, the point of subdivs is to use a low res control cage, to keep the final poly count down, and it's easier to work with less polys, too - but give it a try. See, as you start to add extra edges around your holes and extrudes, they'll create small, thin polygons near the large, wide polygons and that's what distorts your original shape. Now if you already have small, thin polygons, the difference will become smaller and thus the surface distortion will be a lot less noticeable.
- To get a nice smooth shape with properly flowing curves, it's a good way to start as simple as possible and work your way up procedurally. Build a low poly version of your model, and apply a meshsmooth to get nice, evenly distributed quad polygons. Use open edges to influence how the meshsmooth works. Use the shell modifier a lot, too.
- Use turbosmooth's isoline display and the lowres cage together to study how far edges are moved away from their original position because of the subdiv smoothing. Add extra loops when neccessary to keep them where you want them.
- Surprisingly, triangles can work pretty well in many cases.
- Constraints are your friend, use the edge constrain to preserve the surface shapes as much as possible in the final stages of the modeling. Avoid manual, 'freehand' point pushing like the plague.
- Another surprise is that messy stuff -can- work pretty well. Sometimes it's the only way.
- I imagine that polyboost must have a lot of cool tools to aid you in technical modeling, too...
Now, the thing that'd probably close this thread forever would be to somehow dump Fausto's mind into here. You know what I mean:
http://www.hardwired.hu/img/wg/2/1098/Starcraft_2_72.jpg
I'd love to see a wireframe of THAT.
I've been doing a lot of technical modeling lately, and to make things worse it's for Mental Ray which in our current pipeline won't support n-sided polygons. Cool :)
(another fun part is if you want to use Zbrush on the result... :P)
Some notes...
- In general, it's very important to choose a right level of detail for your control cage. Everything is fine and cool as long as you're trying it on a cube, maybe even on a cylinder; but build something more complex and you'll discover that the techniques explored in this thread can fall apart very quickly. Drilling holes into an 8-sided cylinder will cause a lot of pinching no matter how clever you are; using 16 or 20 sides or even more will make it a lot easier. The same goes for more complex shapes too, like cars, armor and such.
It's pretty much the key part IMHO. Yeah, the point of subdivs is to use a low res control cage, to keep the final poly count down, and it's easier to work with less polys, too - but give it a try. See, as you start to add extra edges around your holes and extrudes, they'll create small, thin polygons near the large, wide polygons and that's what distorts your original shape. Now if you already have small, thin polygons, the difference will become smaller and thus the surface distortion will be a lot less noticeable.
- To get a nice smooth shape with properly flowing curves, it's a good way to start as simple as possible and work your way up procedurally. Build a low poly version of your model, and apply a meshsmooth to get nice, evenly distributed quad polygons. Use open edges to influence how the meshsmooth works. Use the shell modifier a lot, too.
- Use turbosmooth's isoline display and the lowres cage together to study how far edges are moved away from their original position because of the subdiv smoothing. Add extra loops when neccessary to keep them where you want them.
- Surprisingly, triangles can work pretty well in many cases.
- Constraints are your friend, use the edge constrain to preserve the surface shapes as much as possible in the final stages of the modeling. Avoid manual, 'freehand' point pushing like the plague.
- Another surprise is that messy stuff -can- work pretty well. Sometimes it's the only way.
- I imagine that polyboost must have a lot of cool tools to aid you in technical modeling, too...
Now, the thing that'd probably close this thread forever would be to somehow dump Fausto's mind into here. You know what I mean:
http://www.hardwired.hu/img/wg/2/1098/Starcraft_2_72.jpg
I'd love to see a wireframe of THAT.
urgaffel
06-05-2007, 09:12 PM
The funny thing is that I think that the wireframe for the marine is actualy quite simple. Take the lower leg armor for example. "All" you have to do is to get the general shape done, then subdivide it and/or cut the panel lines in. Clean it up, add the little holes for the bolts and you're more or less done. The detail on the lower edge (above the "boot") can easily be done with the shell modifier and a spline for the profile (you can see it in the waist area too). What makes it look awesome is the stupid (and I mean that in the best possible way :D) amount of detail and the general design. If you disect the separate pieces of the armor, you realize that it's a lot simpler than it looks.
And Laa-Yosh is indeed right, technical/hard surface modeling can benefit from tris, pentagons (5 sided) and all kinds of "no-no" techniques :)
And Laa-Yosh is indeed right, technical/hard surface modeling can benefit from tris, pentagons (5 sided) and all kinds of "no-no" techniques :)
shroomer
06-23-2008, 05:46 PM
this thread would answer so many of my questions if only the images were still available (and the same goes for a few other threads)! does anyone know if the pdf/faq was ever made?
in trying to learn 3d modeling (and in particular, subd modeling), and i'm discovering many of the fantastic issues that subd'ers get to wrestle with. say i want to model the robot from Laa-Yosh's post (2 posts up from here) with all that kind of detail. would you guys say it's quicker with subd or with nurbs?
i haven't really used nurbs as i prefer the 'mouldability' of subd's. i have heard that nurbs are tedious to work with and that there are issues at render time - what are the main problems and do you think they are less or more of an issue than tweaking verts to get mechanical/hard edge details to look right on subd models?
also, i've bought a few modeling dvds that go into hard surface modeling, but there is very little dedicated to subd DETAILING and the bit that i did see i thought showed very poor technique. can anyone recommend a good (and thorough) dvd (grooves/ports on CURVED surfaces)?
TIA
in trying to learn 3d modeling (and in particular, subd modeling), and i'm discovering many of the fantastic issues that subd'ers get to wrestle with. say i want to model the robot from Laa-Yosh's post (2 posts up from here) with all that kind of detail. would you guys say it's quicker with subd or with nurbs?
i haven't really used nurbs as i prefer the 'mouldability' of subd's. i have heard that nurbs are tedious to work with and that there are issues at render time - what are the main problems and do you think they are less or more of an issue than tweaking verts to get mechanical/hard edge details to look right on subd models?
also, i've bought a few modeling dvds that go into hard surface modeling, but there is very little dedicated to subd DETAILING and the bit that i did see i thought showed very poor technique. can anyone recommend a good (and thorough) dvd (grooves/ports on CURVED surfaces)?
TIA
PiXeL_MoNKeY
06-24-2008, 02:49 PM
I would reccommend Chris Thomas' CG Academy DVD Advanced Modeling 2 - Technical Modeling DVD (http://www.cg-academy.net/pages/topic_modelling/dvds_modelling_advanced_02/dvds_modelling_advanced2.php). For 3ds Max tutorials, they really are the best stuff out there.
NURBS/Solids modeling has its place in 3D, and modeling. You need to use what works for you, not what works for other people. Everyones pipeline and workflow is different so don't base your decision based on what other users do.
-Eric
NURBS/Solids modeling has its place in 3D, and modeling. You need to use what works for you, not what works for other people. Everyones pipeline and workflow is different so don't base your decision based on what other users do.
-Eric
shroomer
07-01-2008, 07:56 PM
PiXeL_MoNKeY, thanks for the reply. i have seen the dvd you suggested and although it's really good, there isn't much subd modelling technique (unless i missed it!)
i'm comfortable poly and subd modelling on 'regular' surfaces (flat, perfect sphere/cylinder etc) but what about when things get more complex? what i'm trying to understand is whether nurbs are quicker than subds when the model gets complicated. (for example the 'robot' in the picture that Laa-Yosh posted:
http://www.hardwired.hu/img/wg/2/1098/Starcraft_2_72.jpg
is that even a subd model? maybe it started as subd, was then converted to high poly and then detailed?
are there any dvds around that show how to model these sumptuous curves and details?
i'm comfortable poly and subd modelling on 'regular' surfaces (flat, perfect sphere/cylinder etc) but what about when things get more complex? what i'm trying to understand is whether nurbs are quicker than subds when the model gets complicated. (for example the 'robot' in the picture that Laa-Yosh posted:
http://www.hardwired.hu/img/wg/2/1098/Starcraft_2_72.jpg
is that even a subd model? maybe it started as subd, was then converted to high poly and then detailed?
are there any dvds around that show how to model these sumptuous curves and details?
Steve Green
07-03-2008, 08:01 PM
PMJI, but yeah I'm 99% sure that is a subd model.
I've not had to do anything that complex, but there was some pretty good info in this thread about cutting into curved surfaces, but I'm not sure if the relevant images are still there.
I seem to recall some people lift parts of the model and use the shell modifier with bevel profiles to get overlapping plates, which is a nice technique.
I don't know of any specific DVDs, but I haven't really looked to be honest.
- Steve
I've not had to do anything that complex, but there was some pretty good info in this thread about cutting into curved surfaces, but I'm not sure if the relevant images are still there.
I seem to recall some people lift parts of the model and use the shell modifier with bevel profiles to get overlapping plates, which is a nice technique.
I don't know of any specific DVDs, but I haven't really looked to be honest.
- Steve
PiXeL_MoNKeY
07-08-2008, 10:26 PM
The majority of Fausto's work, AFAIK, is subd's w/ mudbox for detailing. Check out the wire from this thread (http://forums.cgsociety.org/showpost.php?p=4051182&postcount=36), or the wires you can find on his website (http://www.fausto3d.com/home.html).
-Eric
-Eric
shroomer
07-10-2008, 06:39 PM
PMJI, but yeah I'm 99% sure that is a subd model.
- Steve
PMJI??
there was some pretty good info in this thread about cutting into curved surfaces, but I'm not sure if the relevant images are still there.- Steve
most are gone unfortunately, but there are still some good'uns left. it seems though that cutting detail into curvy surfaces (especially complex, non-uniform curves) is best left to nurbs to avoid bumps and other imperfections - what do you guys think? i mean, it is possible with subd but it takes a lot of tweaking and it never really looks 'perfect' (e.g. even that Fausto model has 'bumpy' legs - but maybe this was the desired effect).
in any case, i'm gonna be trying out MoI (nurbs) as i've heard nothing but praise for it and i'd like to see how that compares to subd.
- Steve
PMJI??
there was some pretty good info in this thread about cutting into curved surfaces, but I'm not sure if the relevant images are still there.- Steve
most are gone unfortunately, but there are still some good'uns left. it seems though that cutting detail into curvy surfaces (especially complex, non-uniform curves) is best left to nurbs to avoid bumps and other imperfections - what do you guys think? i mean, it is possible with subd but it takes a lot of tweaking and it never really looks 'perfect' (e.g. even that Fausto model has 'bumpy' legs - but maybe this was the desired effect).
in any case, i'm gonna be trying out MoI (nurbs) as i've heard nothing but praise for it and i'd like to see how that compares to subd.
Steve Green
08-03-2008, 08:47 PM
PMJI (Pardon Me Jumping In)
and spammers can get to f**k.
- Steve
and spammers can get to f**k.
- Steve
urgaffel
08-04-2008, 07:53 AM
Wow, that's the first time I've seen spam here in the forum but I guess it's more common than I think...
About subds, which images do you need? Any technique in particular?
About subds, which images do you need? Any technique in particular?
EricChadwick
09-01-2008, 04:57 PM
Some links that might help, cribbed from the Tech-Artists.Org wiki
(http://tech-artists.org/wiki/Subdivision_Surface)
* Sub-Division Primer (http://www.subdivisionmodeling.com/page1.htm) from Subdivisionmodeling.com (http://www.subdivisionmodeling.com)
* Digital Sculpture Techniques (http://www.theminters.com/misc/articles/derived-surfaces/index.htm) by Bay Raitt (http://cube.phlatt.net/home/spiraloid/) and Greg Minter (http://www.izware.com)
* 3D Tutorials (http://www.per128.com/pages_tutorials/index.html) from Per Abrahamsen (http://www.per128.com)
* Hard edge - highpoly (http://boards.polycount.net/showthread.php?t=47041) from the Polycount Boards (http://boards.polycount.net)
* Modeling hi poly weapons is painful, any tips? (http://boards.polycount.net/showthread.php?t=37457) from the Polycount Boards (http://boards.polycount.net)
* Mechanical SUB-D tips and tricks (http://boards.polycount.net/showthread.php?t=42381) from the Polycount Boards (http://boards.polycount.net)
* Technical highpoly workflow tutorial and scripts (http://boards.polycount.net/showthread.php?t=38222) from the Polycount Boards (http://boards.polycount.net)
(http://tech-artists.org/wiki/Subdivision_Surface)
* Sub-Division Primer (http://www.subdivisionmodeling.com/page1.htm) from Subdivisionmodeling.com (http://www.subdivisionmodeling.com)
* Digital Sculpture Techniques (http://www.theminters.com/misc/articles/derived-surfaces/index.htm) by Bay Raitt (http://cube.phlatt.net/home/spiraloid/) and Greg Minter (http://www.izware.com)
* 3D Tutorials (http://www.per128.com/pages_tutorials/index.html) from Per Abrahamsen (http://www.per128.com)
* Hard edge - highpoly (http://boards.polycount.net/showthread.php?t=47041) from the Polycount Boards (http://boards.polycount.net)
* Modeling hi poly weapons is painful, any tips? (http://boards.polycount.net/showthread.php?t=37457) from the Polycount Boards (http://boards.polycount.net)
* Mechanical SUB-D tips and tricks (http://boards.polycount.net/showthread.php?t=42381) from the Polycount Boards (http://boards.polycount.net)
* Technical highpoly workflow tutorial and scripts (http://boards.polycount.net/showthread.php?t=38222) from the Polycount Boards (http://boards.polycount.net)
vBulletin v3.0.5, Copyright ©2000-2013, Jelsoft Enterprises Ltd.