"FlyerX" <fly### [at] yahoocom> schreef in bericht 
news:46741223@news.povray.org...
> Thomas de Groot wrote:
>>
>> 1) Bump mapping:  When applying an image map as bump map to, e.g. a 
>> cloth, one may often want to scale down (or up) the bump map in order to 
>> obtain the desired effect. In Poser, this is done by changing the U_Scale 
>> and V_Scale of the bump image map in the Material Room. However, this 
>> scaling is not taken on by Poseray and, to obtain the desired effect, one 
>> needs to add manually a scale parameter to the map in the exported 
>> xxx_POV_mat.inc file.
>
> I may implement scaling into the basic materials in the next version but 
> now you can go into the POV-Ray tab and copy the material into the text 
> window and edit the bump map UV scaling there as a transformation. The 
> copy command is in the POV-Ray tab as insert->materials->copy from basic 
> tab.
Oh! That is nice indeed! I realize that there are many Poseray features that 
I have never used yet! Excellent solution.
In a general case, I think that one should always consider the necessities 
to edit the model to one's needs.
>
>>
>>
>>
>> 2) Bump_size: By default, Poser defines a white background parameter in 
>> the Material Room for all the used image maps. This has unexpected 
>> consequences when using a bump map, even if the used image has 
>> traditionally black as background and white as foreground. Poseray 
>> translates the bump sizes into negative values which may disconcert at 
>> first, but renders correctly in POV-Ray. However, the values are very 
>> high (e.g. -20, or -50) for some reason (even if the bump strength in 
>> Poser is set to 1.0) and they may have to be changed to more conventional 
>> values manually.
>
> I will look into this. Since I do not have Poser 7 I assumed bump_size 
> behaved the same as in Poser 6.  You can scale the imported bump size in 
> the Poser scene import options. The default value is what Poser 6 uses.
> I noticed some inconsistencies between the bump_size implementation using 
> different models and whether the bump map was an image or a proprietary 
> BUM file.
Yes, this seems to start with Poser 7. It took me a while to realize what 
happened. Even so, I am not entirely sure what the role of the background 
color is really, only that it matters in the way Poser renders it itself. 
Also I became convinced only yesterday of the need for negative bump_size 
values to obtain the correct result in POV-Ray.
>>
>>
>>
>> 3) Procedural textures: Procedural textures in Poser are only shown in 
>> Poseray (and exported to POV-Ray) as pigments and finishes. This is 
>> understandable as the procedural parameters used by Poser may not be 
>> easily translated into POV-Ray code. When using procedural textures in 
>> Poser, best is to find an equivalent POV-Ray texture and use that 
>> instead.
>
> I agree. Would be nice if Poser could bake and export its procedural 
> textures. This and strand hair are features that makes Poser less portable 
> to other applications.
Yes, that would be nice. Strand hair in particular would be important, in 
the same way as dynamic clothes is.
>
>>
>>
>>
>> 4) G2 hair: Poser 7 introduced G2 figures and also G2 conforming hair. 
>> For many reasons G2 hair is more realistic than the conforming hair in 
>> Poser 6. However, when exported to POV-Ray through Poseray, the result is 
>> rather disappointing. I discovered now why this is so. G2 hair uses a 
>> bump map for displacement mapping. Why for displacement is debatable as 
>> the hair mesh is fine but not particularly so. Better is to use the map 
>> as a bump map. The results are much better. However, there is a problem 
>> with G2 conforming hair and Poseray. First, the hair geometry (in the 
>> example I used at least) has its morphs centered on the neck body part of 
>> the hair figure. In Poseray the neck of the human figure AND the neck of 
>> the hair figure are translated into one single mesh geometry group. 
>> Second, and as a consequence probably, the displacement map for the hair 
>> group is not loaded. While the material maps for the hair are properly 
>> loaded, the bump map is not. When adding the proper bump map in Poseray 
>> and exporting to POV-Ray, the results are excellent.
>>
>
> I will look into this since I have one of those G2 models with hair. Maybe 
> I could route the displacement call into the bump map automatically. I 
> will add this to the manual in the export to Poser tutorial section.
After I sent this post, I found out that the hair object I used for the 
comment above is a bit atypical compared to other G2 hair figures (or props 
as the case may be). I have not tested all the G2 hair however, and it may 
be that for the G2 Jessie for instance, or all the *long* hair, the morph 
building is identical. So, the solution might be a bit more complicated...
>
>
> Thank you very much for your feedback.
Well, I thank you very much indeed for a great product that is really 
invaluable for crisscrossing amongst applications! That is also one of the 
reasons I take the time to stay alert to possible improvements and to note 
down what I believe might be worthwhile to the community.

Thomas

Post a reply to this message

"Janet" <par### [at] attnet> schreef in bericht 
news:web.4674ba66d112495dc4e58e10@news.povray.org...
> "Thomas de Groot" <t.d### [at] internlDOTnet> wrote:
>> 4) G2 hair: Poser 7 introduced G2 figures and also G2 conforming hair. 
>> For
>> many reasons G2 hair is more realistic than the conforming hair in Poser 
>> 6.
>> However, when exported to POV-Ray through Poseray, the result is rather
>> disappointing. I discovered now why this is so. G2 hair uses a bump map 
>> for
>> displacement mapping. Why for displacement is debatable as the hair mesh 
>> is
>> fine but not particularly so. Better is to use the map as a bump map. The
>> results are much better. However, there is a problem with G2 conforming 
>> hair
>> and Poseray. First, the hair geometry (in the example I used at least) 
>> has
>> its morphs centered on the neck body part of the hair figure. In Poseray 
>> the
>> neck of the human figure AND the neck of the hair figure are translated 
>> into
>> one single mesh geometry group. Second, and as a consequence probably, 
>> the
>> displacement map for the hair group is not loaded. While the material 
>> maps
>> for the hair are properly loaded, the bump map is not. When adding the
>> proper bump map in Poseray and exporting to POV-Ray, the results are
>> excellent.
>
> It's often advantagious to import items one at a time. An obj for hair, an
> obj for the figure, the poncho, etc. For example, if you are working on 
> the
> poncho material in POV-Ray, you can comment out the other objects and it
> parses much faster.
> I use a method like this:
> //Assembled objects ...
> #include "BodyVase3_POV_geom.inc" object{BodyVase3_}
> #include "V4Head_POV_geom.inc" object{V4Head_}
> #include "Hat_POV_geom.inc" object{Hat_}
> #include "Flowers_POV_geom.inc" object{Flowers_}
>
You are absolutely right, Janet. I have been thinking along this line, but 
not used it really just because everything went smooth and fast enough not 
to bother yet. However, I am often at the limits where Poser begins to slow 
down noticiably because of too many things to manage at the same time. This 
is particularly the case with several dynamic simulations applied to the 
same figure. So, breaking up the workflow where possible is an excellent 
option and easy to reassemble in POV-Ray later on.

Thomas

Post a reply to this message

High!

Thomas de Groot schrieb:

> The attached image resumes the comments above and shows how the figure and 
> the hair should be.

Wow... the first real PoV longhair in about seven years since I read 
this group! Awesome... now grow him a Vercingetorix-style Gaul moustache 
and send him in full armour to Gergovia (not Alesia, that would be too 
cruel...)!

A more practical question: his hair looks as it was partially tied up in 
the back - is it so?

But at picture tells me that with PoseRay, a true self portrait would be 
  at least theoretically possible...

See you on www.khyberspace.de !

Yadgar

Now playing: The World Spins, I'm Part Of It (China Crisis)

Post a reply to this message

news:467522be@news.povray.org...
> High!
>
> Thomas de Groot schrieb:
>
>> The attached image resumes the comments above and shows how the figure 
>> and the hair should be.
>
> Wow... the first real PoV longhair in about seven years since I read this 
> group! Awesome... now grow him a Vercingetorix-style Gaul moustache and 
> send him in full armour to Gergovia (not Alesia, that would be too 
> cruel...)!

There are a couple of long hair figures/props (for males!) available in 
Poser. This is one of the latest coming with James G2.
Yes a moustache is indicated indeed  :-)
However, a truly Celtic hairdress would involve limed hair, combed backwards 
and up into rigid spikes.

>
> A more practical question: his hair looks as it was partially tied up in 
> the back - is it so?

That is true. That is why the morphs are attached to the neck for this 
particular hair figure.

>
> But at picture tells me that with PoseRay, a true self portrait would be 
> at least theoretically possible...

Not only theoretically!! Not in Poseray, but in the Poser Face Room, you can 
scan in a portrait of yourself and apply that to the figure. You will obtain 
a (fairly) correct avatar of yourself. However, from comments I have read in 
the Poser forums, this is not as straightforward as it seems...
>

Thomas

Post a reply to this message

Thibaut Jonckheere <tua### [at] MAPSONyahoofr> wrote:
> I am sure than there are many functionnalities I am not aware of, but I
> know that:
>
> - It can correct models that have problems, by re-calculating the
> normals (very useful in many situations, as bugged model are quite common).
It can? Thanks for pointing it out, I've been trying to import some
Makehuman .obj items in Pov-ray, but they have some serious normal
problems. Guess I'll have to read Poseray's.

Post a reply to this message

"Thomas de Groot" <t.d### [at] internlDOTnet> wrote:
> "Janet" <par### [at] attnet> schreef in bericht
> > It's often advantagious to import items one at a time. An obj for hair, an
> > obj for the figure, the poncho, etc. For example, if you are working on
> > the
> > poncho material in POV-Ray, you can comment out the other objects and it
> > parses much faster.
> > I use a method like this:
> > //Assembled objects ...
> > #include "BodyVase3_POV_geom.inc" object{BodyVase3_}
> > #include "V4Head_POV_geom.inc" object{V4Head_}
> > #include "Hat_POV_geom.inc" object{Hat_}
> > #include "Flowers_POV_geom.inc" object{Flowers_}
> >
> You are absolutely right, Janet. I have been thinking along this line, but
> not used it really just because everything went smooth and fast enough not
> to bother yet. However, I am often at the limits where Poser begins to slow
> down noticiably because of too many things to manage at the same time. This
> is particularly the case with several dynamic simulations applied to the
> same figure. So, breaking up the workflow where possible is an excellent
> option and easy to reassemble in POV-Ray later on.
>
> Thomas

Well, you bring up an interesting point because I had never thought of using
more than one Poser file for a single image. But, its good to keep in mind
if I ever get that much going in a Poser scene. I also heard that lowering
the max undos in Poser 7 can help. I think the default is 100, which is a
lot!!
And thanks for starting a thread on PoseRay. I think it's a great program
and I can learn a few things here, like scaling the bump mapping.

Post a reply to this message

"Janet" <par### [at] attnet> schreef in bericht 
news:web.46760350d112495ca7c783b0@news.povray.org...
>
> Well, you bring up an interesting point because I had never thought of 
> using
> more than one Poser file for a single image. But, its good to keep in mind
> if I ever get that much going in a Poser scene. I also heard that lowering
> the max undos in Poser 7 can help. I think the default is 100, which is a
> lot!!
> And thanks for starting a thread on PoseRay. I think it's a great program
> and I can learn a few things here, like scaling the bump mapping.
>

Up till now, the maximum number of figures I have been manipulating at the 
same time in Poser is two. I think that one can easily do scene building. 
However, personally, I prefer to that in POV-Ray. Where dynamic cloth 
simulation is concerned, the choice could be to do some of them separately. 
Only where clothes are overlapping and interacting with each other, you have 
to keep them together. Remember though, that each simulation is done 
separately nonetheless and saved in dynamic files. Loading those files takes 
time. A couple of times I thought that Poser had frozen, but it was just 
loading them, taking a couple of minutes. Once that is done, the application 
runs smoothly.

Another issue that you may come across when using several Poser figures in 
the same POV-Ray scene, is that you reach fairly rapidly the swapping limit 
of your system during render.

Thomas

Post a reply to this message

On Tue, 19 Jun 2007 14:20:28 +0200, "Thomas de Groot" <t.d### [at] internlDOTnet>
wrote:

>Another issue that you may come across when using several Poser figures in 
>the same POV-Ray scene, is that you reach fairly rapidly the swapping limit 
>of your system during render.

I've got a gig of memory and I've not noticed any swapping even with 10 poser
models referenced to give about 50. I've got OpenGL problems with the modeller
but not had a thrashing HD.
I generally use one model each for the body and each dynamic cloth.


Regards
	Stephen

Post a reply to this message

"Thomas de Groot" <t.d### [at] internlDOTnet> wrote:
> "Janet" <par### [at] attnet> schreef in bericht
> news:web.46760350d112495ca7c783b0@news.povray.org...
> >
> > Well, you bring up an interesting point because I had never thought of
> > using
> > more than one Poser file for a single image. But, its good to keep in mind
> > if I ever get that much going in a Poser scene. I also heard that lowering
> > the max undos in Poser 7 can help. I think the default is 100, which is a
> > lot!!
> > And thanks for starting a thread on PoseRay. I think it's a great program
> > and I can learn a few things here, like scaling the bump mapping.
> >
>
> Up till now, the maximum number of figures I have been manipulating at the
> same time in Poser is two. I think that one can easily do scene building.
> However, personally, I prefer to that in POV-Ray. Where dynamic cloth
> simulation is concerned, the choice could be to do some of them separately.
> Only where clothes are overlapping and interacting with each other, you have
> to keep them together. Remember though, that each simulation is done
> separately nonetheless and saved in dynamic files. Loading those files takes
> time. A couple of times I thought that Poser had frozen, but it was just
> loading them, taking a couple of minutes. Once that is done, the application
> runs smoothly.

I'll remember that! Sometimes I wonder what Poser is doing when I open a
scene. lol
If you ever do use Poser for your scene building, and you like a certain
camera angle, PoseRay can duplicate it for you, as long as you load the
Poser scene, as opposed to just the 3D model (obj). One thing though, it
loads the camera from Frame 1. However, you can always
edit->memorize->camera in the frame you like, and then use
edit->restore->camera in Frame 1.

> Another issue that you may come across when using several Poser figures in
> the same POV-Ray scene, is that you reach fairly rapidly the swapping limit
> of your system during render.
>
> Thomas

With 4 Gigs of RAM, it'll take a lot of figures. :) I bought a new computer
recently, and, yeah, maybe I went a little nuts with the RAM. But I like
it!

Janet

Post a reply to this message

"Thomas de Groot" <t.d### [at] internlDOTnet> schreef in bericht 
news:4677ca0d$1@news.povray.org...
>
> Another issue that you may come across when using several Poser figures in 
> the same POV-Ray scene, is that you reach fairly rapidly the swapping 
> limit of your system during render.
>

To answer to both Stephen and Janet:
I have 1 GB of memory and normally that is more than enough. I think that 
the scene were I got a swapping mention (corrected automatically by my 
system) also contained several ivy models. Probably the combination of all 
those heavy-duty elements asked a lot.

Thomas

Post a reply to this message