I'm thinking of buying the new Daz Millennium SubDragon which is on sale right now, but I'm wondering about this subdivision stuff.

Okay, I get the basics, the idea that you can dial up the polys for increased detail if the software supports it.  I guess my question is, how is this figure going to look in Poser?  Apparently it has less polys but if you use it in D/S 2.o you can dial those polys up.  Would Poser's Smoothing option give similiar or acceptable results?

In Daz going to start producing figs that look like video game characters when used in Poser, but that render fine in their own software?

Any thoughts would be great, thanks!

Poser's smoothing should help.. this feature might end up being nothing more then just Daz's version of poser smoothing.

I haven't installed D|S, but subdivision smoothing as it applies to carrara
involves both smoothing during render, to the desired level, and subdivision
of the quads into progressively smaller quads on conversion.  hence it may
be rather different than the poser 7 application (no variable levels, rendering only).

Info on DAZ|Studio Subdivision:

http://www.daz3d.com/i.x/software/studio/-/new?&_m=d

From the sound of it - and I can't say for certain because I have not experimented with D|S 2.0 - it is proper Catmull-Clark subdivision.  Poser uses REYES polygon smoothing, which behaves QUITE differently when dealing with models that have a low polygon count.  For extremely high poly models, the two approaches show pretty similar results in my experience.  Without info on how the dragon is modeled it's hard to say how the model will behave in Poser, but - hell, it's only $1.50.  Hard to go wrong there.

 In general, subdivisions are very cool, especially if your render engine supports render-time subD smoothing. Poser's smoothing appears to be more for smoothing normals (which gives the appearance of smooth surfaces but you may notice faceting if you study an alpha channel of said render) rather than smoothing faces.

Posers smoothing makes a bit of a mess with it. Specially the wing membrane. The lack of a few polys in the right places. Every poser user knows what happens with those kind of models when smoothing is turned on.

I have been doing a lot of test renders with it right now. I have found that turning Off smoothing and checking backfaces cleans the mesh up but looks like a low rez model rendered in P4. Not sure how many remember what P4 renders looked like before smoothing, but pretty much the same as DS until this latest version.

If you have something else to render in like Vue, carrara, Modo or any software that can sub divide it looks fine.

I can post some renders to show what I'm talking about.

This first one is with smoothing and no backfacing checked.

You can see the better render of the mesh. I have not done the Modo one yet but I'm sure it will look evern better because it can do better sub dividing.

CLICK ON IMAGE FOR LARGER IMAGE.

Thanks for the images. I was still debating if I should get it or not, but I'll pass since it looks terrible in Poser.

I have contacted Daz for a refund. I was happy when i saw this dragon, purchased it, transferred it in my runtime, played with it and wasn't happy with it. The model itself is nice but this is definately no dragon for Poser. I hope for Daz that they don't start to ignore the Poserusers or they will loose any clients imo.

Ah, thanks **Tashar59. ** I'm so torn with this thing because he looked so great in the promo renders, but my main appy is Poser and I do have the Millenium Dragon.  My problem is I didn't just get the LE version, I went all out and got the full version.  This will be the first thing I ever return to Daz

Having built subdivision into D/S they can now create models that will look like crap in Poser and great in Daz.  This gives them a huge market advantage.  Imagine if M4 is a sub-d character mesh.

Maybe it is time for me to suck it up and learn D/S... or Carrara.  I fear Poser's days might be numbered....

Not to start a potential argument...but, really, WHY is it a bad thing for a company to sell models DESIGNED to be used in THEIR software?  They've been selling Dynamic clothing (Poser only, anyone) for a couple years now.  I don't see it as a bad thing that they add D|S optimized products to their repertoire.

It's not at all unreasonable, and in fact it makes that item much more interesting if one wants to export to a foreign app for rendering; Poser's approach to polygon smoothing requires funky modeling habits when working with low-polygon items, as this dragon appears to be, and while this particular version may not work too well with Poser, there are other versions of the dragon work fine with it - while this new one should work substantially better in Cararra, Cinema, or any big dog renderer that you might export to (lower polygon count).

Personally I cannot justify spending the money on a new dragon.  Even with PC discount it is not worth it if it won't render right in Poser and since I cannot get D|S 2 to install correctly. I have way to many dragons in my runtime that I don't even use.

Who needs a new dragon anyway? Did a quick count in my runtimes and I've counted over 10 dragons. Honestly, all they release at DAZ lately is stuff we all have tons of already. Dragons.... dinosaurs, dresses, temples, man, I've got loads of those, how much does DAZ think i need of those? is any store releasing anything new, innovative and interesting these days? All I see is the same stuff over and over again.

*It's not at all unreasonable, and in fact it makes that item much more interesting if one wants to export to a foreign app for rendering; Poser's approach to polygon smoothing requires funky modeling habits when working with low-polygon items, as this dragon appears to be, and while this particular version may not work too well with Poser, there are other versions of the dragon work fine with it - while this new one should work substantially better in Cararra, Cinema, or any big dog renderer that you might export to (lower polygon count).

*That's the whole point. You may have noticed that DAZ wants to tap into a new market, the game design market. To do so, they need to move away from the old high poly stuff and come down to the level the game designer is at. SubD is a move into that direction and I'm sure we will see more of those kind of models released. They're not geared at the poser community at all, they're setting the stage to move to a new market. Once DAZ tapped into that market succesfully, they don't need poser users anymore to sustain their business. At that time D/S users and game designers will provide enough income for DAZ to finally break with Poser.
 
The only bad side of the coin for poser users is that once DAZ accomplishes this and completely drops Poser support (which is just a matter of time), Poser will go into oblivion and the final curtain for poser will fall. Poser's main right of exsistance, how crude it may be, is DAZ figures being compatible with Poser. Once DAZ pulls the plug, it's lights out for Poser. The G2 figures (or any other non-DAZ figures) do not have a large enough customer base to support the survival of Poser.

Personally, I don't really care. I don't use G2 figures, I dislike Appolo, have nothing with V4/A4 and could care less about the coming release of M4 or any other DAZ figure. I'm still using the good old whole P4 - G1 range figures when needed. I'm still a fervent M1/V1-M3/V3 user. I've got David, Hiro, Stephanie, Aiko3, the Freak, Behomoth and more of these figures in my runtimes. I've got gigabyte after gigabyte of clothes, hair, shoes, textures and so on for them to keep me happy for years to come.

Which ever move anyone in the poser community makes is fine with me, I'm not following it, i'm not buying more stuff for figures.

The only thing which still interests me is new landscaping items for Poser, some interior and exterior items and a vehicle once in a while. That's where my money goes, but only if it's something new something innovative, like Cyllan's Terraformers - Altitudes, that's innovative stuff, something new. If only people would release more of this kind of interesting, innovative stuff, instead of the same old, I've seen it all before.

So, please DAZ go ahead, pull the plug and perhaps we will see a great new move because of the lurking thread and poser will once again gain my attention.

So, please DAZ go ahead, pull the plug and perhaps we will see a great new move because of the lurking thread and poser will once again gain my attention.

So ... I guess there is no possibility that Poser might actually support proper Catmull-Clark subdivision in the future?  And that DAZ's release of a subdivided model isn't an advancement that might propel similar innovations in Poser? That's the good thing about competition ... innovations on one side push similar innovations on the other. 

That's where my money goes, but only if it's something new something innovative, like Cyllan's Terraformers - Altitudes, that's innovative stuff, something new.

So what you're saying is it's only innovative if it uses the same old technology? What about the innovations that drive improvements in products as a whole and move them forward?

To put it in perspecive ... there once was a day when there was NO such thing as a layered graphics image editor. Everything was done on one flat image. Then (if memory serves me correctly) a graphics app named Altimira Composer introduced layers and all that nifty stuff.  Other graphics apps couldn't use it at first. But then we saw similar features in Photoshop, Painter, Paint Shop Pro, and so on.

Sometimes you have to do something different to improve things in the long run.

(and yes, memory did serve me correctly ...)

http://en.wikipedia.org/wiki/Altamira_Software

So what you're saying is it's only innovative if it uses the same old technology? What about the innovations that drive improvements in products as a whole and move them forward? <<<<

Thought I'd better add to this ... I am NOT implying that anyone who tries something new within the constraints of Poser's capabilities is not innovative.  That's something we don't have control over as content creators. We HAVE to create something that works in the programs we create content for ... and the innovations we try HAVE to work within what the software allows.

SOFTWARE innovations are a different matter indeed ... others have already posted in this thread that Poser currently handles smoothing differently than (say) DS or Carrara or Cinema 4D, or any other major 3D app. This is a feature that has ALREADY BEEN THERE for many other apps, but that was brought in as an option for our community by DAZ.  I'm glad to see it, too.

Quote - So ... I guess there is no possibility that Poser might actually support proper Catmull-Clark subdivision in the future? 

Since it won't benefit the enormous base of existing poser content, I very much doubt it...

I hope that's not the case. It would be a very cool feature to add.

Well, it benefits very low poly content. For our "normal" stuff, it won't look much different at all.

The biggest advantage would be using very low poly content in multitudes, and smoothing them to look acceptable with the system.

And if you really want a huge orc horde bearing down in your renders, vue would be the best solution for the best quality.

Quote - > Quote - So ... I guess there is no possibility that Poser might actually support proper Catmull-Clark subdivision in the future? 

Since it won't benefit the enormous base of existing poser content, I very much doubt it...

I do not see why they would not. Adding a new method of smoothing does not affect existing content. DS draws on the same base of existing content, and they did add this method, so what's different for Poser?
Advantages are evident, especially when it comes to morphs that would only need to define the morphs for the original points.

Let's say you want to create an image that has a whole horde of dragons flying around in the distant sky. Why load 10 instances of a 90000 polygon dragon (if you can), when you can load 10 14000 polygon dragons? That's the equivalent of one and a half of the high polygon ones.

And ... for those Poser users who have already purchased him ... turn on smoothing for everything BUT the wing parts (you can disable them in the Properties window). Here's the result.

Daz claim in the product Description that the sub-d Dragon works in Poser 5+.  I was curious to see if Poser smoothing would approximate the Daz subd option. From those renders it appears not, so Daz should be advertising that these models are Studio only.

Why say it's Studio only? Poser loads it, and as you can see above, you CAN use smoothing on the body ... the wings look better with it turned off, though.

You may not get the same results as you do in DAZ Studio or Carrara, but you get results.

thats the crux : *You may not get the same results as you do in DAZ Studio or Carrara, but you get results.

*a simple line on the product of :

'Results may vary outside of the Daz Studio Application' would have avoided alot of this bitterness thats been seen.*

I showed these images to show what it looks like NOW in Poser. Not what might be in the future. I have been accused of setting it up to look bad on those first 2 images at Daz on purpose.  I used a light color light so it would show the results better. Guess I am supose to use dark lighting. LOL.

I also made sure to post an image rendered in something else, Vue in this case, to show that the dragon can look good used in other software that can do subD. They only used that at Daz to fuel the set up conspiracy of the first 2 images.

I don't see anything wrong with Daz going in this direction and I agree that maybe it will help Poser get the same thing. Good thing for all of us. This extreme high rez mesh is getting a bit out of hand I think. I like using my RR mil figures. Best of both world. Low enough poly to work play nice in most systems but look great with Poser smoothing. I remember Daz saying they were going to have a low rez V4 but now after Dan Farrs replies at the MM thread, I think it will be the same as the Dragon.

As I pointed out I was answering the first question but made sure that it was understood that it looked fine in other sofotware. If your like me that poses and creates the scene in poser and then exports to another app, it may be something you can use.

No consiracy here. or is there.

I just remembered I do have a Wings3D tutorial on this forum that shows how to subD a figure. It was for Teri Yuki 1 before they made the highrez version. You could always go that rout and Then use the Dragon in Poser.

^^ as you say, I think all the excitement here is a bit misdirected.  This is just one version of a model that DAZ has had for sale for a long time, it's not like they're discontinuing all poser-type content.

Subdivision in Poser would be a GREAT idea. One area where Poser has a strong lead on DAZ|Studio is animation and dynamic cloth and hair, and those calculations are orders of magnitude faster when used on low poly models. Calculate in low poly, render in high poly, best of both worlds!
I hope SM reasons the same way.

I can get Poser to load a subd-mesh made in Lightwave. It will look like a big pile of polygons . Daz should have the courage to say that this model will NOT produce decent results in Poser , or include instructions in the readme of what needs to be done . A subd cage may not be a usable model unless subdivided because it is NOT a mesh in its own right, its a CONTROL mesh for the subdivision surface .

3/20/08

Seeing that some folks think I buy figures and applications just to complain about the bugs and flaws, for $1.05 this new subDragon seems to have a phenominal price to whine ratio. Maybe I should get it while the complaints are still flying.

It looks like the more expensive version has more morphs and textures. Is that about right?

Actually this is a fascinating topic, and I would love to see some more discussion on it. It sounds like it would have great potential for future versions of Poser. One irritating characteristic of Poser is that as one adds more figures, props and complexity to a scene, the graphics interface slows down. This subdivision technology sounds as if it could address that problem, as well as other things.

Thanks Tarsha59 and Deceey for the renders. With sellective use of smoothing, I like the looks of that dragon just fine.

LMK

Seeing that some folks think I buy figures and applications just to complain about the bugs and flaws, for $1.05 this new subDragon seems to have a phenominal price to whine ratio. <<<<

Indeed!   LOL

Actually, I hope that DAZ will release more subD figures and objects. I like to fill my scenes, and as it is right now I'm constrained by the 4 GB limit of Poser. Most of my Poser scenes have to be stripped down until they render, even after reducing texture map resolution and replacing texture maps by pure procedurals wherever possible.
I'd especially like subD bushes and trees.

3/20/08

svdl:

Are there any subD figures (trees and bushes) that are cheap or free that one could play with?

LMK

The extra texture in the full bundle is the same body texture as the first but with some white lines painted on it. If you look very close at my images you will notice some faint white lines. I thought that was a bit cheep and sleasy for hyping 2 textures and it's only the body map that is the second texture. You still use the same one head map.

You do get all the morphs and poses. The FBM's are pretty intersting because it changes the dragon to 5 complete different dragons.

Just a quick thanks to Deecey for your render and thoughts.  I've decided to keep the Dragon (I did pick up the more expensive verson.)  He might do fine in background renders and he will work nicely in Carrara and D/S, both of which I have and just need to sit down and master.

My only complaint with the whole one application vs another is the learning curve.  I've spent years learning the nuances of Poser.  I love the Material Room and have learned a workflow to get the best results from this piece of software.  As time marches on and things change (which is inevitable) I see that I really have to start all over again and learn these things again.

All the power to Daz if they can gain a market advantage, as someone pointed out, Poser depends heavily on Daz and if they can't keep up, well...

And even better if competition drives improvements.  This is idealogy behind free enterprise, isn't it?

Thanks all.

I would rather see instancing in Poser, rather than sub-d .

...and I spend most of my time in Poser trying to prevent smoothing...;) 'least on my stuff...;)

um it was too much to expect I guess...
sub-d is not smoothing
and a sub-d mesh will look terrible when "smoothed" with poser cause poser smoothing is not sub-d
Daz studio has sub-d to a certain extent

sub-d was invented by pixar to allow a low poly "control" mesh with a much more detailed and subdivided mesh on top avaialble at rendertime. So you animate and work with a low polygon mesh and then render a high poly mesh.

daz studio does this in a very clumsy way. But at least it has sub-d.

this sub-d dragon will look terrible no matter what you do with it in poser.
unless you have access to the high poly smoothed mesh - it will look chunky and low poly - cause thats what sub-d is all about.
Raj

..and I spend most of my time in Poser trying to prevent smoothing...;) 'least on my stuff...;)  <<<

The way it should be handled is to "dial in" the level of subdivision you want to use. (Most apps do it this way). If you don't want smoothing, dial in 0. If you do, you dial in the level of subdivision.

Each level up divides each side of a polygon in half (or looking at it another way, one four-sided polygon would become four four-sided polygons when subdivided once).

Taking a 1500 polygon model made of all four-sided polygons  ...

Subdivided one level it looks like it's 6000 polygons (1500X4)
Subdivided two levels it looks like it's 24000 polygons (1500 x 4 x 4)
Subdivided three levels it looks like it's 96000 polygons (1500 x 4 x 4 x 4)

In actuality, the cage model is still 1500 polygons. But you can make it look like more when you render it with "true" subdivision smoothing. And the nice thing about it is, you can increase or decrease the smoothing as needed for your renders. If you don't want to smooth something that is far from the camera you don't have to. Closer to the camera? Yeah, dial it up.

3/20/08

And a similarly understandable explaination if instancing would be...what?

LMK

"Instancing" would mean that the software stores the data relating to an object only once ... and then keeps track of the number of times the same object is used in the scene.

It's a good idea if you want to use a lot of the same content in a scene, but doesn't address conserving memory if you have a 100,000 polygon character that's wearing 300,000 polygons worth of high resolution clothing, fighting a different character with an entirely different set of clothing that is 300,000 polygons.

In the long run, subd would be a better solution than instancing, IMHO.  Both would be even nicer. LOL

Quote - He might do fine in background renders and he will work nicely in Carrara and D/S, both of which I have and just need to sit down and master.

If you want to use it in poser, you can always use one of the free modelers to subdivide it a few times. This will give you the exact same results as using SubDs, though you will need to pose it first cause the process will screw up any morphs and rigging it may have.

Instancing works great for a group of marching soldiers at midrange or long distance, for example. Close range wouldn't work, you'd immediately see that they're all identical.

A really smart instancing algorithm can introduce variations per instance. Even better, if the geometry can be made procedurally, you can vary some input parameters per instance. Vue uses this for ecosystems of plants and rocks - those are instance of procedural geometries, and each rock/plant in an ecosystem is slightly different.

Vue 6 Infinte has both instancing AND sub-D. Then again, it's significantly more expensive than Poser...

ghonma wrote: "If you want to use it in poser, you can always use one of the free modelers to subdivide it a few times. This will give you the exact same results as using SubDs, though you will need to pose it first cause the process will screw up any morphs and rigging it may have."

Not true. My tutorial to do just that proves that rigging is not screwed. You can also subD the figure with the morphs set to create the SubD morph. If there is an out of order vert problem, nothing UVMapper can't be used to correct/prevent it.

3/20/08

Isn't there a python script or two to subdivide polygons in Poser? Why wouldn't that work on a low poly figure?

LMK

Just because a model has been subdivided doesn;t mean it will look better. great examples some people have pointed out are VERY high poly models that don;t bend as well, or look as good as lower poly counterparts.

An excellent example that comes to mind, is the daz martian war machine. if you still have the original around, load it up and render it.

Then load up the improved rigging and optimised mesh version I did for daz, and render that.

The original model have almost 100,000 polys, and the optimised one had less then 26,000 if I recall.

Yet both look identical rendered.

Sub d is great gaming tech, for VERY low poly models to make them look better, but once you get past a certain poly count level, you don;t seem much (if any) improvement at all.

Also, if you do subdivide a model, bear in mind all the morphs are lost.

Python script for Catmull-Clark subdivision: can be done, the algorithm is fairly simple.
Adjusting existing morphs - can be done too in the same script, but takes quite a bit of programming (and it'll slow down the script by orders of magnitude...)

 The entire subD workflow rules! Save system resources for your "hero" objects, using Level Of Detail for all others. I think the best think DAZ could do right now is to create low poly versions of all their characters. 

3/20/08

In an imaginable work-flow, could one subdivide a figure or prop that is already textured, positioned in a scene, morphed, and posed for the sake of improved rendering quality?

LMK

Not my taste in dragons.

I won't be parting with the Mil Dragon anytime too soon.

The sub dragon looks fugly if you ask me.

(MHO)