3D-Coat 3.3 updates thread

[Andrew Shpagin]

3.3.06 re-uploaded for owners of GTX 470/480. The longstanding issue is fixed. This is the only change in the build so it has not got new number.

[AbnRanger]

Strangely enough, I do understand Andrew is a busy man, my beef is more with your condescending tone... Grow up!

Okay, I made the mistake of suggesting something (that's normal practice): this will definitively be the last time!

I don't think it's condescending at all to state the fact that WE ALL HAVE TO WAIT...most of the time. No sweat. If you do happen to miss a (rare) re-upload cause you didn't check this update thread, it's no big deal. The next build is usually on it's way in a few days or so. There is then, no real problem here, unless you have encountered a game-stopping issue and you're in need of a quick resolution.

In such a case, Andrew will do what he can...so you're covered either way.

[splodge]

i'm just here to shrug my shoulders.

carry on...

[digman]

I agree, sometimes a quick fix is needed in a new beta release. All that would be required as mentioned is the date of the new upload. No confusion that way...

The problem in all text communication is that what you mean can be taken not in the context of what you were trying to say...

In my own statement, What I assumed would be taken as the date being added to the first page of the beta thread by the new uploaded beta, I think was taken as just adding it to the way Andrew lets us know now which is deeper in the thread. So I do agree that adding some information at the start of the thread could be helpful...

The best thing is Andrew,just keeps making this program stronger, more powerful, fixes bugs and we don't get charged a arm and leg for some new features that another company might charge us for...

[Tinker]

Жду с нетерпением новый merge. Это основа моего техпроцесса, увеличение его скорости - мечта.

I look forward to the new merge. This is the base of my workflow, increasing its speed is a dream.

[AbnRanger]

Жду с нетерпением новый merge. Это основа моего техпроцесса, увеличение его скорости - мечта.

I look forward to the new merge. This is the base of my workflow, increasing its speed is a dream.

Indeed...I hope we never have to wait several minutes-hours again. If this works and if Andrew can overhaul the the Sculpt Room (by replacing it with the engine/tools in Surface Mode?), by the start of the Siggraph convention...it would spell a major transformation for 3D Coat, and big news just in time for the show.

[Javis]

Indeed...I hope we never have to wait several minutes-hours again. If this works and if Andrew can overhaul the the Sculpt Room (by replacing it with the engine/tools in Surface Mode?), by the start of the Siggraph convention...it would spell a major transformation for 3D Coat, and big news just in time for the show.

I certainly hope so, the sooner the better. Because ATM the sculpt room is leaning toward useless save for a few things. It's slow and buggy to boot. I've been pushing Andrew to work on it and get those surface sculpting tools from the voxel room into the sculpt room for some time. I hope to see those changes in the near future.

[AbnRanger]

Who Dat?...Who Dat say a man can't get no uber-high details in Voxels?

I just posted some video tutorials covering these recent 3.3.06 additions (Cache-To-Disk and Multi-Resolution) in the following thread, for anyone interested.

http://www.3d-coat.com/forum/index.php?showtopic=5994&pid=45049&st=0entry45049

You can also find them here:

http://www.youtube.com/user/dnashj33#p/a/u/1/LynoTC18OnY

http://www.vimeo.com/12991087

[Tony Nemo]

Thanks for the demo and you proved your point!

[StOrM3]

People, should really learn to be more understanding... I think 3D Coat is one of the best quality programs out there, and the rate at which it gets updated is phenomenal... Anyone here, name another UV Mapping tool <cough> <cough> hehe that gets updated this much, and has this many features??!!?? BTW: The only issue, I had with the slow painting / voxel brushing, duh durrr.. on me, it might help if I would re-modify the Soft-Brush Speed...

So, what I'm saying is give him some breathing room, and he will do his best to make this the program that everybody, and their artist teams have to have.. < Like it isn't already..>

The only thing on my wish list is more export formats, and that can wait, as worse comes to worse, you can always export to OBJ, import it to MS3D then use that to re-export to DTS format.. Or Import it to Frag3D to Make MDL7 format model exports for 3DGS...

I'm here to test and help out all I can, with this program, it has helped me so much, I feel like I owe something back.. So please ease up on Andrew..

Now back to re-studying my Unit editor for Total Annihilation, to help me remember some of my file format magick for DTS Conversion / Exporters, also see if I can talk nak into reworking snesKEY with me one more time for win 7... Yea I'm that StOrM3... So, I guess since I made the only reversed 3DO Format specs and then turned that into a DXF->3DO Conversion utility, then later on the fullblown editor with KinBoat, maybe, I could get off my lazy duff, and pull another 48 Hour straight Hack Fest, like for 3DO, well maybe I'm too old for that stuff now, but now I actually have the format specs, and maybe some help from my fellow game coders...

Ciao' 4 Now, sorry about the rants and raves, I just hate it when people hammer on Good Helpful People / Coders, expecting the world, and oh yea, they want it yesterday...

BTW: Andrew I am willing to help take some of the load off of you, if you need the help, I can do anything from coding help to just fielding bug reports / hardcore feature testing / whatever you guys need... Hit me via email if your interested...

StOrM3

Pain Killa Entertainment© 1999

Pistol Pete Penguin, MegaDroid, Seeds of Time, Brain Clay, CentiBall Hex Mini, Kombat 2010, TA Unit Editor, Among Others...

***********************************************************

[ Of All The Things I've Lost... ]

[ I Miss My Mind The Most ]

[ by: Ozzy Osbourne / Cereal Killer - Hackers Movie Pt. 1 ]

***********************************************************

[Tinker]

AbnRanger, by the way, your interface "modo style" scheme really good looking

[AbnRanger]

Thanks for the demo and you proved your point!

Not sure if that's a pat on the back or a slap to the back of the head ...

http://www.youtube.com/watch?v=4QQXIwV3TKM&feature=related

It is mostly for those relatively new to 3DC. I tried to keep it brief...didn't necessarily reach that goal, but maybe close enough.

[Javis]

AbnRanger, by the way, your interface "modo style" scheme really good looking

I did that them last month, Andrew has included it as one of the default themes when you install 3DC, you can select it in the drop down list on the start up window... Or in the "\OptionsPresets" folder. It's called "Yellow.Options.xml".

[Tinker]

Great job My Yellow.Options.xml is different, maybe left from previous version.

By the way, if change drawing method of viewport grid from color to sort of blend, multiply for example, gradient backgrounds will look much better.

I discovered this when wanted to copy colors from Silo. Although dont know what about performance.

[AbnRanger]

I did that them last month, Andrew has included it as one of the default themes when you install 3DC, you can select it in the drop down list on the start up window... Or in the "\OptionsPresets" folder. It's called "Yellow.Options.xml".

"I see you beat me to it"...said the blind man...or something.

[Tinker]

Realized thanks

[AbnRanger]

I wonder if NVidia (who owns the company that developed Mental Ray....Mental Images), could help Andrew add MetaSL Shader support natively in 3D Coat?:

http://www.lysator.liu.se/~zap/ad/MetaSL-in-max-flsh/MetaSL-in-max-flsh.html

All of 3ds Max's legacy materials were converted to MetaSL in this latest version (2011), and I'm sure they are pushing to get similar compatibility in Softimage and Maya....because the whole thrust of this is to lead the effort to have a unified, portable shader system...so shaders you use in Max will transport seamlessly to it's other applications. That could be great for Lightwave, too...if they offered support for MetaSL, in CORE, because that opens large doors for them too, as assets created in LW could easily make it into mixed pipelines.

So, imagine the implications of being the first 3D Sculpting & Painting application to be able to apply and transport shaders that render straightaway in all of Autodesk's products....via Mental Ray, and also give you a hi-fidelity viewport representation.

[Tinker]

I wish they improved global illumination algorithms. So far Vray is my choice

[philnolan3d]

I'm still perfectly happy with LW's built in renderer.

[David Walters]

If Andrew can overhaul the the Sculpt Room (by replacing it with the engine/tools in Surface Mode?), by the start of the Siggraph convention...it would spell a major transformation for 3D Coat, and big news just in time for the show.

I'm still hoping the Voxel room gets renamed to 'Sculpt', the old sculpt room gets removed entirely, and the new sculpt room gains the ability to work seamlessly with polygon and voxel meshes. Drastic I know, but think about it for a moment.

Can anyone explain to me why you wouldn't do this? What use is two sculpt rooms when there's only one Andrew? surely he would surely find it easier to deal with changes to just one consistent room. I mean that's what's happening anyway, right?

Isn't surface mode really similar to polygon sculpting anyway?

Bonus benefit of this merging would be: Painted objects alongside voxels -- great for background environments!

[splodge]

I have mixed feelings about a new sculplt room. But when I step back and look at the big picture I see voxels as being in a awkward stage right now.

So there's two possibilities:

Option 1. Improve the decimator so that it can produce cleaner meshes and respect edge flow. No need for retopo! Hurray!

Option 2. Add a proper mesh modeling mode.

My reasoning is this - At the moment the idea of creating a mesh primitive and then turning it into voxels and then retopoing it back into a mesh again is kinda silly and a waste of time. The whole retopo process is questionable. I'm still finding it easier (and quicker) using a poly modeler to create a mesh model with the correct topology as opposed to creating a voxel model and then retopoing it. And some of the more detailed voxel models that can be created in 3D Coat would take several years to retopo into a mesh model. So retopo can't be the answer..

So it has to be all or nothing for me. Either we can create a model entirely in voxels and hit a magic button and have it transformed into a nice clean mesh, or we just push voxels to the side and go back to regular poly modeling. Obviously we would hate to see Andrew's efforts on voxels go to waste and so option 1 (decimation) has to be the best option..

ps - Some may be wondering why I'm in favor of decimation over quadrangulation. The quadrangulate algorithm can't handle sharp edged stuff. It's only useful for blobs.

[Amber]

I have mixed feelings about a new sculplt room. But when I step back and look at the big picture I see voxels as being in a awkward stage right now.

So there's two possibilities:

Option 1. Improve the decimator so that it can produce cleaner meshes and respect edge flow. No need for retopo! Hurray!

Option 2. Add a proper mesh modeling mode.

My reasoning is this - At the moment the idea of creating a mesh primitive and then turning it into voxels and then retopoing it back into a mesh again is kinda silly and a waste of time. The whole retopo process is questionable. I'm still finding it easier (and quicker) using a poly modeler to create a mesh model with the correct topology as opposed to creating a voxel model and then retopoing it. And some of the more detailed voxel models that can be created in 3D Coat would take several years to retopo into a mesh model. So retopo can't be the answer..

So it has to be all or nothing for me. Either we can create a model entirely in voxels and hit a magic button and have it transformed into a nice clean mesh, or we just push voxels to the side and go back to regular poly modeling. Obviously we would hate to see Andrew's efforts on voxels go to waste and so option 1 (decimation) has to be the best option..

ps - Some may be wondering why I'm in favor of decimation over quadrangulation. The quadrangulate algorithm can't handle sharp edged stuff. It's only useful for blobs.

QFA

I have no idea why so much time and effort is invested into voxels.I would rather see more improvements in paint/retopo/uv area.

I have not used voxels even once but I do use Zbrush all the time, baking normal dis maps in Zbrush is no brainer try to bake some from voxels.Voxels is a niche tool, but seems people who use it have some vocal majority on the forums.If 3DC would improve paint mode and have at least Mudbox poly scult features it would be much better than whole voxel sculpt room.

[Javis]

I have mixed feelings about a new sculplt room. But when I step back and look at the big picture I see voxels as being in a awkward stage right now.

So there's two possibilities:

Option 1. Improve the decimator so that it can produce cleaner meshes and respect edge flow. No need for retopo! Hurray!

Impossible. We'll always need manual retopo tools. You can't use auto retopo for anything that is going to be deformed during animation. If you never do any animation than I'm sure auto retopo should suit your needs, but anyone that does... It's a no go.

Option 2. Add a proper mesh modeling mode.

I know this is on Andrews long term agenda. But that's a lot of work, and frankly, what is there right now should be fixed so it works properly before moving on to another larger project such as that.

My reasoning is this - At the moment the idea of creating a mesh primitive and then turning it into voxels and then retopoing it back into a mesh again is kinda silly and a waste of time. The whole retopo process is questionable. I'm still finding it easier (and quicker) using a poly modeler to create a mesh model with the correct topology as opposed to creating a voxel model and then retopoing it. And some of the more detailed voxel models that can be created in 3D Coat would take several years to retopo into a mesh model. So retopo can't be the answer..

That really depends on the sculpt you have created, or the mesh you have modeled. Some things would be just flat out stupid to sculpt with voxels or polys, when it would just be better to actually model it. You can't use voxels for everything... You must weigh the time you'll be spending on something and the ease of the tools for the job at hand. Using all voxels for something, is not a good idea.

So it has to be all or nothing for me. Either we can create a model entirely in voxels and hit a magic button and have it transformed into a nice clean mesh, or we just push voxels to the side and go back to regular poly modeling. Obviously we would hate to see Andrew's efforts on voxels go to waste and so option 1 (decimation) has to be the best option..

That's a very limited view of things, which I do not agree with. We can have good voxel sculpting tools, poly sculpting tools AND a decent poly reduction tool. These are not mutually exclusive. Why should they be?

ps - Some may be wondering why I'm in favor of decimation over quadrangulation. The quadrangulate algorithm can't handle sharp edged stuff. It's only useful for blobs.

I agree with this point.

[splodge]

Impossible. We'll always need manual retopo tools.

Some people once thought that it was impossible for a computer to beat a human at chess.

I accept that it's still going to take some human guidance when dealing with models that are intended for animation. But there is no reason why a computer couldn't detect tubular forms and place clean edge loops around them.

200 years later..

"Retopo my model, Hal"

"I'm sorry Dave, I'm afraid I can't do that"

[Javis]

Indeed. I'm sure at some point we'll reach that plateau, but alas that is some years off.

[AbnRanger]

I'm still perfectly happy with LW's built in renderer.

Well, if you watch the video (I think he loses a lot of folks the moment he starts talking "Coding"), he talks about other renders being able to use the shaders as well, with their own "Backend" as he put it. In the later third of the video, you see the major benefit is being able to see a close to rendered result...in your viewport. I mostly use FinalRender R3, but I'm all for a unified, portable shader system...that is again, render independent. I just think NVidia would take great interest in helping Andrew in such an effort, as that would result in two of their technologies being on display (along with CUDA).

[moska]

*My reasoning is this - At the moment the idea of creating a mesh primitive and then turning it into voxels and then retopoing it back into a mesh again is kinda silly and a waste of time. The whole retopo process is questionable. I'm still finding it easier (and quicker) using a poly modeler to create a mesh model with the correct topology as opposed to creating a voxel model and then retopoing it. And some of the more detailed voxel models that can be created in 3D Coat would take several years to retopo into a mesh model. So retopo can't be the answer..*

correct me if im wrong:

i think people using mudbox will have to retopo again since the base mesh doesnt correspond with the sculptedmesh(polys move during sculpt) ?

[Taros]

Indeed. I'm sure at some point we'll reach that plateau, but alas that is some years off.

No, not years... it's a question of time. Andrew already knows the following document, but he has no time to rework the current quadrangulation routine, until now.

-> The future of 3DCoats quadrangulation routine (No guarantee... it's my assumption )

[AbnRanger]

Splodge, I really surprised to hear that level of cynicism over Voxels and the Retopo tools, coming from you. I get frustrated too, sometimes...but I have to say, in practice, it makes a LOT of sense to utilize Voxels for sculpting. The ability to trim and cut away sections of an object, with no concern for topology is phenomenal. And wasn't it you that informed me some time ago that you could import your base mesh into the Retopo Room and essentially use that as your Retopologized mesh? Now that you can temporarily disable Auto snap, any major deformation changes done in the Voxel Room, you can re-adjust your mesh to fit in the Retopo room.

There are some things I'd like to see done in the Retopo Room, and I agree...some more modeling functions would help a great deal (like the ability to create a "Shell" so you only have to Retopo one side of an object, Soft Selection, and accurate transform gizmo's....that are centered on the object...not out in space somewhere), but on the whole, they work quite well...when there are no bugs to contend with.

I'm really excited about the work Andrew is doing with Multi-Threaded merging, but the one thing that I think will eventually set 3DC apart is when or if Andrew replaces the engine/tools in the Sculpt Room with those in Surface mode. The ability to CHOOSE from 3 different sculpting methods is absolutely unprecedented, and will be hard to overlook...as long as the performance is there and it's stable.

[splodge]

correct me if im wrong:

i think people using mudbox will have to retopo again since the base mesh doesnt correspond with the sculptedmesh(polys move during sculpt) ?

Mudbox is a detailing program. The idea is that you're supposed to create your base mesh in Max or Maya and then take it into Mudbox for detailing.