Saw this on maxunderground...
The new extension includes: ProOptimizer modifier for optimizing high-poly models (a reworked and enhanced version of Mootool's Polygon Cruncher), 12 new Particle Flow operators (based on Particle Flow Tools: Box#1, featuring tighter integration with Max, improved memory and performance) and a ProSound toolset that incorporates multi-track audio capabilities with Maxscript support (Boomer Labs' most useful Sound Trax plugin). What's great news is that Autodesk has hired Oleg Bayborodin to once again develop software for the base Max application (or in this case to better integrate software he had already developed for Max), which could hint at ongoing efforts in that direction. Taking into account ICE-related developments on the Softimage front this would not be a bad idea.
Replies
I have a few monster scripts I wrote for soundtrax when doing facial animation. It's nice to see you won't need to buy the plug-in any more.
Oleg Bayborodin is on staff huh... wow that could be awesome! I really like particle flow. Maybe they're serious about kicking it up a notch.
I'm downloading them now. I'm VERY excited that they decided to include boomer labs plug-ins. 3dsmax2009 just got interesting.
i've googled to no avail, so if anyone knows any better....!
*edit* xnormal helps a bit, just because it doesnt crash with super-detailed OBJ's but it'd be nice to get stuff into maya in a slightly more optimised state
Mesh labs doesn't have batching available afaik, but I think its planned for future releases.
If Polycruncher is integrated into max, how are you supposed to get the model in there in the first place for reduction?
About the MeshLab vs Polycruncher.
I got exactly the opposite feeling. MeshLab output seemed be better than the PolyCruncher one. MeshLab is able to manage very large meshes and that for a tool aimed to reduce the size of large meshes is a rather basic feature.
Simplifying big meshes in chunks affect the final quality in a significant way.
The only missing feature of MeshLab simplification is the symmetry preserving option.
As I pointed out above yes breaking down meshes to batch process in polycruncher shouldn't be necessary however the functionality to view face decimiation and control it via a slider certainly arent available in meshlabs. You can have a far greater degree of accuracy in that regards, which can be crucial when dealing with delicate areas such as a face.
There is no undo in meshlabs and you cant preview the decimation before committing. All combined with the lack of batch processing make for a bit of a disappointment. Hopefully they will make some improvements
Yes, I was speaking about the quality of the results, and imho polycruncher simplified meshes are not as good as meshlab ones (tendency to open/enlarge holes http://www.mootools.com/plugins/forum/viewtopic.php?f=12&t=54, flipping faces on planar regions http://www.mootools.com/plugins/forum/viewtopic.php?f=12&t=27 ) and so on...). But this is just a personal opinion I do not want to start a flame here .
On the other hand I agree, the absence of an undo feature is a striking annoying omission in MeshLab.
But use what you want... like I said I have used both and I have simply tried to explain what I think of them.
Meshlabs to get down in polys. Then Polycruncher to finely tune.
I personally find Meshlabs good enough to get my meshes into Max. Cool thing is it allow u to manually select areas you want to optimize.
My comp starts to freaks out on 1mil. I don't like to work on models over 500k as it slows my work flow.
What's this modest computer spec?
Mine:-
Geforce 7600
DualCore E6400
2 Gigs Ram
7800 GTX
Pentium 4 3.2ghz
on the machine here
at home quadcore
4gig ram 8800 GT
as said very mediocore, no need for polycrunch for that stuff, i understand using it for lods when it has to go really quick, but nor for that stuff
Any news on perhaps and update to the archaic realtime renderer/viewport? That's really my only gripe with Max.
same goes to geezus and anoon, when you only want to bake it, why do you have to actually see it in the viewport? Thats just nonesense, just grab a lower res of your mesh, set up the cage and import the highres without showing it in screen to bake it
almost 30 mio and still decent framerates, simple as that...
pre-08-09 well you're taking your chances...
...I guess you just import everything with the veiwport set to bounding boxes. Then just parent the low res reference objects, to the high ones? Then hide the high res objects?
Any aligning would have to be done in external software (zbrush)?
...ha ha. I found what I was looking for the diplay as bounding box button. Under object properties.
I believe Epics main reason for optimising was the file sizes. Also they applied colours to the high res, for easy texturing.
baking that without lighting is fast as hell, having it optimized is just a error source i'd avoid