Grafikimon
-
Posts
23 -
Joined
-
Last visited
Content Type
Profiles
Forums
Events
Posts posted by Grafikimon
-
-
From what I remember from teh book the character should have an emotionless face. Something completely neutral. The eyes right now are so wide that they will make it hard to show suprised since that will be your default expression for them.
I would close them to a neutral position
the rest looks good. Can't wait to see how you do the rigging of the face.
-
Here is my current progress.
No not talbot.
-
yeah it si for fun mosty and to explore some of the more odd things I learned when I started tehn unlearned later on. Dangling splines and other things that most people I asked told me to avoid.
The images have been around for years up on the resource website. link is on the main hash page. Might be called tutorials. In any case the site is gone and teh images are no longer available. I ad downloaded them a while ago and just decided to build one of them
-
removed
-
removed
-
Been looking at these reference images for years and finally decided to see how effective they actually are.
Trying to develope a low spline style to work in.
The refrence image is the one on the right
-
don't forget it is twice the price and crashes on consumer nvidia cards and runs slow on pro cards.
the morphs are still linear so it is still not as revolutionary as they may think they are!
-
Normal mapping was used on the trolls in LOTR. They also used it for alot of the Massive army shots.
Normal mapping from what I understand and from my own experiments require you to build your low rez model then export it to Brush
detail on. You can go up to millions of polys doing intricate detail that is alot hard without the Z depth info. then you export this model. There are free utilities you use to map the high rez model onto the low rez model. the resulting normal map is then applied to the low rez model. There are several reasons to make a low rez model look high rez
a: Ease of rigging and making poses
b: faster to animate
c: Faster render times
well that's my take on the whole normal map usage.
I would say implement it since it couldn't hurt.
-
It always came as a dual pc/mac cd. Just remember that that Randy has been pounding away on the OS X version (which, BTW has just been released as a beta so he is close to having it finished. This is version 10.5 he is working on)
So you can download the OSX version and use the same serial and CD to run both. Check the news forum as I think it is keyed to the older CD not the current one.
So give Steve at hash a call and see what the deal is with an OSX version for the the ones you are getting coming out.
-
Jack,
If you are bringing a organic model into AM from lightwave you can smooth it so it comes in as if it was sub-patched. What you do is set the modeler Patch divisions to 1. this smooths the model without dividing it. You then Freeze it. when you bring this model into AM the splines will look exactly like your sub-patched model.
a good technique for any sub-patched model that has to come into AM.
Thanks for the plugin. I'll try it and see how much better it works.
-
Then import the mesh into AM with the .lwo plugin.
.lwo plugin??? is this a plugin for AM? where can i download it.
-
I posted a long winded explanation of how to import organic models into AM so I'll just go over a beif explanation of how to do it here.
The best results i'e had are with dxf formatted models. 3ds models which are tripled usually come in terrible unless they are used as a prop and not edited.
Now the dxf importer does not bring in textures so it is best to do them in Animation Master (the projection mapping ability of AM is better anyway so this is a minor issue)
First thing is to soften your model in the original app. roughly 30% softer than the original. (This will take some trial and error to figure out for the other software.) The reason to soften is that Subdivisional surfaces form cages, that is, they form a box of CPs around the smooth version which never come in contact with the CPs, while splines the surface passes through the cps so to get the CPs to the right position requires softening your SDS cage. You also want to make sure you have as many quads as possible and remove as many poles as you can (Pole: a CP with more than 4 edges passing through it) Tris should also be hunted down and moved to hidden locations.
Export to DXF, there is a modified crossroads app on the resources pages that can export quad dxf models. Most software can do this natively. One free app that exports mdl files makes millions of short segments so it is better to export an obj then import that into crossroads to make a dxf.
Now once you have your dxf file go to the project tree. rigth-click (ctrl-click, mac) on the object item and select import dxf.
This will create a new mdl file from your imported model.
you may have to select any cps and reapply any smoothing to them (they occasionally come in peaked) and fix any spline flows that went wrong.
Here is a model that I imported
as you can see it just has a few creases to fix along the arm and on the waistline. about an hour of clean-up. Not too bad.
-
The best base codec for quicktime is sorenson 3
use the normal fps unless you need it to be smaller in which case use half the frame rate. (So 24fps becomes 12fps. 30 becomes 15fps.) try not to go below 10 fps as it becomes noticeable.
cinepak is only good for 320x240, which it was designed for BUT it will contain any size movie. It will just not be as small. Cinepak is less processor intensive as sorenson so it is an option for 3-400Mhz computers (pretty rare in these 1-2Ghz days)
Divx requires people to go find divx codecs for their computer which is a pain. Divx despite all the claims is not much smaller than sorenson 3, well as much as I could get it without loosing quality.
When doing AVIs for pcs you can do mpeg4 and mpeg1. mpeg 1 is an old codec so it needs some tweaking to get good results in movies with fast action. slow action and movement it is great for. use tmpgenc to convert any AVI to a mpeg1 since it can accurately scan a movie for fast parts and tweak a better image quality out of those parts.
mpeg4 is comparible to sorenson 3, but not really. it can get as small but there is some quality loss there.
sorenson, mpeg4 are good for any size. mpeg1 and cinepak are good for 320x240 sizes.
quicktime has a habit of loosing the sound when converting an mpeg1 to some other format so keep this in mind if you change your mind and want to convert an mpeg1 file, easier to use the original or rerender.
-
QT Pro is like $30 so it is pretty cheap.
THat said you can probably just map a movie to a face or as a backgound and just rerender it out of AM.
I did it once ages ago but it should still be doable.
-
Jeff's DVD is great. nothing in depth on rigging. just a note to use an extra bone in the upper arm. It is mostly about learning to animate. The 1st section is for people completely new to 3d. the 2nd goes into channels and other animation techniques/theory and finally a sample of how he goes about animating.
He has free models up on his site to use
-
Thanks for the help.
I'll have to try this.
Jay
-
Hi,
I am half the way through Stop Staring a book on facial animation and wanted to know if there is a way to imitate the controls he uses in the book.
Basicly he uses expressions tied to a null so that as he moves it various sliders (poses) will be manipulated. So instead of having 6 sliders you move this control and it activates varous sliders dependign on position or rotation.
Simon Mills also did a demo reel using the same controls.
Also the author mentions Animation Master so I hope there is a way to do the same here.
Jay
HASH 3D CITY
in Work In Progress / Sweatbox
Posted
I am going to rough out a city layout in illustrator just to get the ball rolling. No one has made any decisions yet so lets throw some ideas out there and see what we come up with.
I am going to use meters since it is a good rule of thumb and will accommodate windows and doors within one grid unit.
I am also going down to barnes and nobles to browse through that Batman Begins book because it had some great maps.
Google Earth is also a great piece of software since we can look at existing city layout for inspiration. It has 3d models of buildings in NY and some cities in California so it will help see the general layout.
You cant extract the buildings.