another light test ( took quite a bit of time to get there )
this is part of the announcement :
======
* * * mcjTeleblender2 update coming up soon * * *
Daz Studio spotlights will be (optionally ) exported as Blender Spotlights, including correct orientation and spread angle
Daz Studio DirectLights will be (optionally) exported as Blender Sun Lights including correct orientation
...point lights ... ... Blender point lights
Uber AreaLights .... as Blender Area Lights
also the Opacity strength of the surface/materials will be added to the material node trees as a transparency shader +mix shader
previously all transparencies were obtained using maps ( gray images )
this was because i didn’t want to modify Blender’s build-in obj importer, to avoid unforeseen problems down the road
the opacity strength is transfered from Daz Studio the same way texture tiling is : calls to a new function in blenderBot
the default materials for Aiko4 use opacity strengths to make the eye surfaces transparent
up to now mcjTeleblender only accepts opacity maps ( an all-black opacity map in this case )
but the upcoming version will take the 0% opacity strength into account
create any type of light and parent to it a primitive-plane
the plane may be rectangular
mcjTeleBlender and mcjBlendBot will re-create this in Blender as an area light with the color and intensity of the light
but with the size, position and orientation of the are-alight plane
usually you will want to make this plane invisible before the export
this way it wont be exported as a plane, just as an area-light
fig 4 had mostly spotlights ( exported from Daz Studio! ) and an area light
note that since always you could do arealights by applying a random image to the ambient channel of any surface in the daz scene ( it's in the manual )
Thanks for the effort you put on this. It's very useful and a great time saver.
On the next update, do you have plans to solve the duplicated materials problem? I started making my own script to solve that problem, but if it will be on the update, I will stop making it. My approach is to read the obj and mtl files and found all the materials with the same features. Then, the script will edit that files grouping identical materials, replacing it with a new material named with the image file name (or the first material if no image). I read your code, but finally decided to make an independent script that run using node.js.
Thanks for the effort you put on this. It's very useful and a great time saver.
On the next update, do you have plans to solve the duplicated materials problem? I started making my own script to solve that problem, but if it will be on the update, I will stop making it. My approach is to read the obj and mtl files and found all the materials with the same features. Then, the script will edit that files grouping identical materials, replacing it with a new material named with the image file name (or the first material if no image). I read your code, but finally decided to make an independent script that run using node.js.
the upcoming update will have transparency strength support ( instead of requiring image maps ) and point/spot/area/distant-sun lights
for materials, ... me too someday i'd like a script to speed things up i suspect there are existing scripts doing just that
lets write one right now often my main issue is a figure's skin materials so lets make this grouping work only on selected objects, not on everything i select Aiko3 oops, i want the blender console, so i close blender and re-launch the scene using the batch file created by mcjTeleblender i open a new text in the text editor i'm not very experienced in python programming
i type and run this
import bpy
objects = bpy.context.selected_objects
for obj in objects:
print( obj )
it works ! i see
now i'll cycle through the materials i = [0 ...last] and cross check materials j = [1..last] and each time i find a material with the same diffuse color and the same texture image name ( this should be enough ) i'll swap mat[j] with mat
import bpy
def swapDuplicates( mats, mat, i, n ):
for j in range( i, n ):
print( " %d" % j )
print( mats[j] );
def mergemats( o ):
materials = o.material_slots
i = 1
n = len( materials )
for m in materials:
mat = m.material
if mat:
print( ( i - 1 ) )
swapDuplicates( materials, mat, i, n )
i = i + 1
objects = bpy.context.selected_objects
for obj in objects:
mergemats( obj )
i'll complete this tomorrow
note that the first step will be to build the list of materials for all selected objects instead of processing each object's list of materials separately
Thanks for the effort you put on this. It's very useful and a great time saver.
On the next update, do you have plans to solve the duplicated materials problem? I started making my own script to solve that problem, but if it will be on the update, I will stop making it. My approach is to read the obj and mtl files and found all the materials with the same features. Then, the script will edit that files grouping identical materials, replacing it with a new material named with the image file name (or the first material if no image). I read your code, but finally decided to make an independent script that run using node.js.
the following script seems to do the job
first it collects all the materials for the selected objects
then
for each material in that list, ( which we will call the reference mat ) for each object, for each material slot if that slot has the same diffuse texture and the same diffuse color as the reference mat it makes this material slot share the reference material
note that this is based on the premise that the diffuse_color texture map is prefixed with "kd" which is the case for objects imported from .obj files
one nice feature is that it should work for non-node-based materials too ( blender internal etc )
import bpy
materials = []
#---------- getMap ----------
def getMap( mat, key, keydot ):
texSlots = mat.texture_slots
for t in texSlots:
if t:
if ( t.name == key ) or ( t.name.startswith(keydot) ):
return( t.texture )
return 0
#---------- getImage ----------
def getMaterialUID( mat ):
KdMap = getMap( mat, 'Kd', 'Kd.' )
im = 0;
if KdMap:
im = KdMap.image.filepath
return( [im, mat.diffuse_color] )
#---------- getMap ----------
def swapDuplicates( refmat, slots, refUID ):
for slot in slots:
myUID = getMaterialUID( slot.material )
if ( myUID[0] == refUID[0] ):
if ( myUID[1] == refUID[1] ):
slot.material = refmat
#---------- mergeSelectedObjectsMats ----------
def collectMats( o ):
for slot in o.material_slots:
materials.append( slot.material )
#---------- mergeSelectedObjectsMats ----------
def mergeSelectedObjectsMats( o ):
slots = []
for slot in o.material_slots:
slots.append( slot )
for mat in materials:
myUID = getMaterialUID( mat )
swapDuplicates( mat, slots, myUID )
objects = bpy.context.selected_objects
for obj in objects:
collectMats( obj )
for obj in objects:
mergeSelectedObjectsMats( obj )
when i select Aiko3, most of the skin materials were made to share the 'SkinHands' material
Normal Maps -- Interesting... How are you getting the Normal Map Information? I don't believe that Normal and Specular Map information is exported to the .mtl file.
Normal Maps -- Interesting... How are you getting the Normal Map Information? I don't believe that Normal and Specular Map information is exported to the .mtl file.
for each scene exported, mcjTeleblender
writes
the obj file
the mtl file
a .py file
a .bat file
the py file will now contain calls to new mcjBlendBot.py functions
Normal Maps -- Interesting... How are you getting the Normal Map Information? I don't believe that Normal and Specular Map information is exported to the .mtl file.
for each scene exported, mcjTeleblender
writes
the obj file
the mtl file
a .py file
a .bat file
the py file will now contain calls to new mcjBlendBot.py functions
it;s also in the .py file that cameras and lights are transfered
and other Blender/Cycles settings
Interesting... It sounds like you are pulling the information from the DS scene. If that's the case, why not pull ALL of the material information from the scene and forget about the ,mtl file? That way you can make a better translation to the Cycles nodes, since the information you have is more comprehensive.
Interesting... It sounds like you are pulling the information from the DS scene. If that's the case, why not pull ALL of the material information from the scene and forget about the ,mtl file? That way you can make a better translation to the Cycles nodes, since the information you have is more comprehensive.
because this way the Blender python importers ( mcjBlendBot ) can still be used with Poser, Carrara etc etc
or any program that exports obj/mtl
-------
image : the cube uses a normal map that didn't need inversion
you know the switching-the-R-and-G-color-channels trick,
i found it on the net and i dont think it makes sense
i think what it does is rotate the normals instead of inverting up-and-down
i did try to use a negative strength setting
oh my ! am i stoopid or what, i didn't think ...... Daz Studio can do render involving normal maps , cant it? :)
all i have to do is make sure the renders in Blender look similar to the way Daz Studio handles given specific test normal maps !
-----------
figure 4 : a bump map, recently created using my free mcjPlanar Daz script - converted into a normal map using the freeware xNormal - rendered in Daz Studio / 3 Delight
---
figure 5 - There ! i'm now confident i don't need the 2-node contraption in Blender to invert the normal maps
maybe someday there will be a "live" version of mcjTeleBlender
which would possibly be a plugin
you move a camera in DS, it moves in Blender
you re-pose a figure, the plugin uploads the vertex deltas to Blender
and there may be functions to transfer objects moved/modified/created from Blender to Studio
....animated renderings .... maybe
mcjTeleblender3 will also transfer the Daz Studio materials "glossiness" setting to blender
by setting Blender-BSDF-Glossy Roughness = 1 - ( Daz DefaultMaterial Glossiness )
also, in the Daz Studio material if you apply an image map to the glossiness channel
this map will appear in the Blender Cycles materials and will be connected to the Roughness input of the BSDF-Glossy shader
this should help when making aged/worn painted/varnished materials
also i fixed the mcjTeleBlender checkbox which lets you decide if the BSDF-Glossy nodes are mixed with the BSDF-Diffuse nodes using an Add-Shader or a Mix-Shader
With regards to the glossy diffuse mix -- I'm experimenting with setting glossy node roughtness=0 and using the DAZ Glossiness setting to control the mix of the glossy and diffuse nodes. If there is a specular map, then I use it to control the mix between the glossy and difuse nodes. This works well for skin, since skin specular maps tend to be dark.
With regards to the glossy diffuse mix -- I'm experimenting with setting glossy node roughtness=0 and using the DAZ Glossiness setting to control the mix of the glossy and diffuse nodes. If there is a specular map, then I use it to control the mix between the glossy and difuse nodes. This works well for skin, since skin specular maps tend to be dark.
for now i added the image map that can feed the roughness input of the GlossyBSDF channel
and as before, if there can be an image map that feeds the color input of the GlossyBSDF channel
but in Daz Studio there's up to 3 image maps
( trying to think this through )
so i guess i could say if( ( there's an image map on the daz Specular Strength ) and ( the Use MixShader Option is On ) )
then add a texture-image node it to the Blender Material and connect it to the MixShader ratio input
i may also route map-less Strength values to the MixShader ratio input
shown below, a render using an image on the BSDF Glossy-Roughness input
when glossy roughness is 1, it still looks glossy, me too i was expecting it to look like chalk at full roughness
fig 3 - it would make more sense, the color input gets a color and the mix strength input received a strength
the opacity control is designed to satisfy the needs of Daz Studio's opacity maps for hair, eyelashes, etc
where the opacity map is tones of gray
getting colored transparencies can be done
for example by setting the opacity strength to 50%
and supplying the colors through the diffuse color map
using colored transparency maps does not give results similar to the behavior in Daz Studio
note that when a channel does not have an image texture in Daz Studio,
the corresponding material in Blender wont have one either
so there wont be useless computations during cycles rendering
the diffuse / specular-glossy / ambient / Opacity
node tree will probably be like this
i think we'll get colored-transparencies similar to those in Daz Studio !
so, easy colored lights and stained glass
in teleblender 2 when an image map was present at the ambient color input
mcjBlendBot would switch the diffuse node for an emissive node
but in teleblender 3, the ambient tecture image, if there is one, will be used
and else the ambient color ( and strength?) will be used
in figure 2 you can see the transparency is indeed yellow
the specularity is pink, the glossiness has a very low but mom mull roughness, so we get a focused light uhhhh thing
if the roughness was 0 ( glossiness 100% ),the specular contribution of the specular node would be a mirror
ambient is blueish
diffuse color is orangeish
all the inputs with white colors can be connected to their respective image map
============
Fig 4 : same settings, full opacity
Fig 3 - zeroed the ambient component of the shader
fug 5 - a black diffuse channel was given a greater importance in the diffuse/specular mixer
glossy roughness was increased
I've got an odd error where the opacity map that I have in DS is coming out differently in Blender. I realized though that there is a newer mcjteleblender than I have, so I will update, and try again.
I've got an odd error where the opacity map that I have in DS is coming out differently in Blender. I realized though that there is a newer mcjteleblender than I have, so I will update, and try again.
In any case, this is what I was seeing
maybe it's something that happens at the edges of the UV map
for example, it UV coordinates exceeds 1.0, in some cases it loops back to 0 in others it may do the strangest things
i'm almost sure that mcjTeleblender has an issue with the Tiling offsets, which i will study and fix in teleblender 3 ( this weekend!? hopes are high ) also i want to see if negative tilling works like daz studio
note that you could try to set vertical tiling to 0.9 or 1.1 and play with vertical offset maybe it will solve a cross-UV-border issue
Hey MCJ,first off would like to thank you for opening up my options by creating Teleblender,I now do not see a future for me without Blender or Cycles!My issue with Teleblender as of lately has been giving me hours of stress!I am doing complex animations in layers to achieve a highly detailed scene that I compile in AE and am staying within the parameters of my cuda cores by layering my objects,is there a limit on how much information Teleblender can handle transferring?When I try to use a scene file per your directions with genesis2 deleted and just lights it wipes my genesis2 skin files to white when rendering within blender and in batch file.For a few days I was able to get around this by using cycles nodes instead of mat.lib for G2 but as of the last 48hours if I use any type of a scene file for lighting it will not allow me to have genesis skin too.I have downsized my G2 maps and they were working perfectly,and now this!I seriously have tried every combination but NOTHING is working,I have been fighting progress and would greatly appreciate any advice you have!Also does DS cam to Blender work with 4.6?I have had no luck getting Blender to swallow the script in Blender 2.70!All apologies for making first contact whining but if the results I have achieved thus far were not so amazing I would have ditched Teleblender and have been done with it,but it is now an irreplaceable part of my work flow.........
if the problem is with Blender 2.71 maybe you could try installing an older version of Blender like 2.69 elsewhere on your disk, and tell mcjTeleblender to use that one ( and you have to install the mcjBlendBot scripts in the proper sub-folders )
oh and i dont use DS4.6 very often because it always always crashes after less than an hour - so it's possible the problem is with DS 4.6 , since i tested it with DS 4.5
so i'll test this today
you seem to say the problem is mat-lib + genesis 2 + Blender 2.71 + DS 4.6 + scene-files + in animation batch files that's a lot of variables ... but with luck i'll find the problem
here's something that may help
say you exported a 30 image animation
open the batch file that mcjTeleblender created for you in wordpad
Comments
another light test ( took quite a bit of time to get there )
this is part of the announcement :
======
* * * mcjTeleblender2 update coming up soon * * *
Daz Studio spotlights will be (optionally ) exported as Blender Spotlights, including correct orientation and spread angle
Daz Studio DirectLights will be (optionally) exported as Blender Sun Lights including correct orientation
...point lights ... ... Blender point lights
Uber AreaLights .... as Blender Area Lights
also the Opacity strength of the surface/materials will be added to the material node trees as a transparency shader +mix shader
previously all transparencies were obtained using maps ( gray images )
this was because i didn’t want to modify Blender’s build-in obj importer, to avoid unforeseen problems down the road
the opacity strength is transfered from Daz Studio the same way texture tiling is : calls to a new function in blenderBot
Aiko4
the default materials for Aiko4 use opacity strengths to make the eye surfaces transparent
up to now mcjTeleblender only accepts opacity maps ( an all-black opacity map in this case )
but the upcoming version will take the 0% opacity strength into account
In the upcoming version
---
in the case of area lights, the system will be
create any type of light and parent to it a primitive-plane
the plane may be rectangular
mcjTeleBlender and mcjBlendBot will re-create this in Blender as an area light with the color and intensity of the light
but with the size, position and orientation of the are-alight plane
usually you will want to make this plane invisible before the export
this way it wont be exported as a plane, just as an area-light
fig 4 had mostly spotlights ( exported from Daz Studio! ) and an area light
fig 5 hardware render
yes it's teacups on their heads ( https://sites.google.com/site/mcasualsdazscripts4/mcjteacupandsaucers )
note that since always you could do arealights by applying a random image to the ambient channel of any surface in the daz scene ( it's in the manual )
Hi Casual
Thanks for the effort you put on this. It's very useful and a great time saver.
On the next update, do you have plans to solve the duplicated materials problem? I started making my own script to solve that problem, but if it will be on the update, I will stop making it. My approach is to read the obj and mtl files and found all the materials with the same features. Then, the script will edit that files grouping identical materials, replacing it with a new material named with the image file name (or the first material if no image). I read your code, but finally decided to make an independent script that run using node.js.
now i'll cycle through the materials
i = [0 ...last]
and cross check materials
j = [1..last]
and each time i find a material with the same diffuse color and the same texture image name ( this should be enough )
i'll swap mat[j] with mat
i'll complete this tomorrow
note that the first step will be to build the list of materials for all selected objects
instead of processing each object's list of materials separately
the following script seems to do the job
first it collects all the materials for the selected objects
then
for each material in that list, ( which we will call the reference mat )
for each object,
for each material slot
if that slot has the same diffuse texture and the same diffuse color as the reference mat
it makes this material slot share the reference material
note that this is based on the premise that the diffuse_color texture map is prefixed with "kd"
which is the case for objects imported from .obj files
one nice feature is that it should work for non-node-based materials too ( blender internal etc )
when i select Aiko3, most of the skin materials were made to share the 'SkinHands' material
-- new -- new -- new -- new -- new -- new -- new -- new -- new -- new -- new -- new --
-- free experimental beta no-guarantees Blender Script to combine identical materials into one ---
https://sites.google.com/site/mcasualsdazscripts4/mcjmergeselectedobjectsmats
the upcoming version of mcjTeleBlender
will support Normal Maps
below you see a simple material with a beige diffuse color, a bit of gloss, and a sand-dune normal map
the separate/combine nodes are used to invert the displacement produced by the normal map
i guess i should make that inversion easier to disable hmm
---
in figure 3
image map on the Normal, Diffuse and Opacity plus glossy and opacity strength controls ( maybe not the final design )
in figure 4
a very map-intensive material
diffuse, specular, opacity, bump, normal maps
Normal Maps -- Interesting... How are you getting the Normal Map Information? I don't believe that Normal and Specular Map information is exported to the .mtl file.
for each scene exported, mcjTeleblender
writes
the obj file
the mtl file
a .py file
a .bat file
the py file will now contain calls to new mcjBlendBot.py functions
mcjBlendBot.setOpacityStrength('Plane', 'Default', 1 )
mcjBlendBot.setNormalMap('Plane', 'Default', 'i:/sandnormal2.jpg' )
mcjBlendBot.setBumpMapStrength('Plane', 'Default', 1 )
mcjBlendBot.setTiling('Plane', 'Default', 4, 0, 4, 0 )
it;s also in the .py file that cameras and lights are transfered
and other Blender/Cycles settings
for each scene exported, mcjTeleblender
writes
the obj file
the mtl file
a .py file
a .bat file
the py file will now contain calls to new mcjBlendBot.py functions
mcjBlendBot.setOpacityStrength('Plane', 'Default', 1 )
mcjBlendBot.setNormalMap('Plane', 'Default', 'i:/sandnormal2.jpg' )
mcjBlendBot.setBumpMapStrength('Plane', 'Default', 1 )
mcjBlendBot.setTiling('Plane', 'Default', 4, 0, 4, 0 )
it;s also in the .py file that cameras and lights are transfered
and other Blender/Cycles settings
Interesting... It sounds like you are pulling the information from the DS scene. If that's the case, why not pull ALL of the material information from the scene and forget about the ,mtl file? That way you can make a better translation to the Cycles nodes, since the information you have is more comprehensive.
because this way the Blender python importers ( mcjBlendBot ) can still be used with Poser, Carrara etc etc
or any program that exports obj/mtl
-------
image : the cube uses a normal map that didn't need inversion
Trying to make sense of normal maps
you know the switching-the-R-and-G-color-channels trick,
i found it on the net and i dont think it makes sense
i think what it does is rotate the normals instead of inverting up-and-down
i did try to use a negative strength setting
oh my ! am i stoopid or what, i didn't think ...... Daz Studio can do render involving normal maps , cant it? :)
all i have to do is make sure the renders in Blender look similar to the way Daz Studio handles given specific test normal maps !
-----------
figure 4 : a bump map, recently created using my free mcjPlanar Daz script - converted into a normal map using the freeware xNormal - rendered in Daz Studio / 3 Delight
---
figure 5 - There ! i'm now confident i don't need the 2-node contraption in Blender to invert the normal maps
upcoming version
( which will be called mcjTeleblender V 3 )
will use tabs
main tab is pretty much the only things i change when i render scenes
------------
figure 2 - i realized i have to plug the normal map to the glossy node of the materials and reduce the strength parameter
.............. maybe tomorrow !!!!!!!!! mcjTeleBlender 3 .... then again maybe not, you never know :)
maybe someday there will be a "live" version of mcjTeleBlender
which would possibly be a plugin
you move a camera in DS, it moves in Blender
you re-pose a figure, the plugin uploads the vertex deltas to Blender
and there may be functions to transfer objects moved/modified/created from Blender to Studio
....animated renderings .... maybe
free Ozite normal map
mcjTeleblender 3's default glossy parameters wont be as plastic as that
Figure 4 - i will leave the separate/combine nodes in the normalBranch
Fig. 5 - ah HA! someone knewed how to summon the powerful super double secret back door
while we're at it, i'll improve handling of specularity/glossiness
in case you missed it, Nakamuram already has a version of mcjBlendBot that handles glossiness and opacity
http://www.sharecg.com/v/72122/browse/8/Script/Mods-for-mcjTeleblender-DAZ-Studio-to-Blender
mcjTeleblender3 will also transfer the Daz Studio materials "glossiness" setting to blender
by setting Blender-BSDF-Glossy Roughness = 1 - ( Daz DefaultMaterial Glossiness )
also, in the Daz Studio material if you apply an image map to the glossiness channel
this map will appear in the Blender Cycles materials and will be connected to the Roughness input of the BSDF-Glossy shader
this should help when making aged/worn painted/varnished materials
also i fixed the mcjTeleBlender checkbox which lets you decide if the BSDF-Glossy nodes are mixed with the BSDF-Diffuse nodes using an Add-Shader or a Mix-Shader
Thanks for incorporating my mod, Casual!!
With regards to the glossy diffuse mix -- I'm experimenting with setting glossy node roughtness=0 and using the DAZ Glossiness setting to control the mix of the glossy and diffuse nodes. If there is a specular map, then I use it to control the mix between the glossy and difuse nodes. This works well for skin, since skin specular maps tend to be dark.
Note this fairly recent article: http://www.blendercanvas.com/a-study-of-glossy-and-anisotropic-blender-cycles-shaders -- Like the author, I used to think that a glossy node with roughness=1 was the same as a diffuse node. I cannot remember if this was the case with earlier versions of Blender/Cycles.
for now i added the image map that can feed the roughness input of the GlossyBSDF channel
and as before, if there can be an image map that feeds the color input of the GlossyBSDF channel
but in Daz Studio there's up to 3 image maps
( trying to think this through )
so i guess i could say if( ( there's an image map on the daz Specular Strength ) and ( the Use MixShader Option is On ) )
then add a texture-image node it to the Blender Material and connect it to the MixShader ratio input
i may also route map-less Strength values to the MixShader ratio input
shown below, a render using an image on the BSDF Glossy-Roughness input
when glossy roughness is 1, it still looks glossy, me too i was expecting it to look like chalk at full roughness
fig 3 - it would make more sense, the color input gets a color and the mix strength input received a strength
a full shader will probably be something like this
in the diagrams here http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Shaders
they show that glossy with roughness scatters rays, but not with a 180 degrees spread, instead, it's a 90 degrees spread
the opacity control is designed to satisfy the needs of Daz Studio's opacity maps for hair, eyelashes, etc
where the opacity map is tones of gray
getting colored transparencies can be done
for example by setting the opacity strength to 50%
and supplying the colors through the diffuse color map
using colored transparency maps does not give results similar to the behavior in Daz Studio
note that when a channel does not have an image texture in Daz Studio,
the corresponding material in Blender wont have one either
so there wont be useless computations during cycles rendering
adjusting glossiness multipliers
i'll strive to get close to the same specularity as in teleblender 2
the skin shader i use is very simple because i usually try to keep Amy non-realistic though not cartoony
maybe i should read what clamping is about :)
the diffuse / specular-glossy / ambient / Opacity
node tree will probably be like this
i think we'll get colored-transparencies similar to those in Daz Studio !
so, easy colored lights and stained glass
in teleblender 2 when an image map was present at the ambient color input
mcjBlendBot would switch the diffuse node for an emissive node
but in teleblender 3, the ambient tecture image, if there is one, will be used
and else the ambient color ( and strength?) will be used
in figure 2 you can see the transparency is indeed yellow
the specularity is pink, the glossiness has a very low but mom mull roughness, so we get a focused light uhhhh thing
if the roughness was 0 ( glossiness 100% ),the specular contribution of the specular node would be a mirror
ambient is blueish
diffuse color is orangeish
all the inputs with white colors can be connected to their respective image map
============
Fig 4 : same settings, full opacity
Fig 3 - zeroed the ambient component of the shader
fug 5 - a black diffuse channel was given a greater importance in the diffuse/specular mixer
glossy roughness was increased
I've got an odd error where the opacity map that I have in DS is coming out differently in Blender. I realized though that there is a newer mcjteleblender than I have, so I will update, and try again.
In any case, this is what I was seeing
maybe it's something that happens at the edges of the UV map
for example, it UV coordinates exceeds 1.0, in some cases it loops back to 0 in others it may do the strangest things
i'm almost sure that mcjTeleblender has an issue with the Tiling offsets, which i will study and fix in teleblender 3 ( this weekend!? hopes are high ) also i want to see if negative tilling works like daz studio
note that you could try to set vertical tiling to 0.9 or 1.1 and play with vertical offset maybe it will solve a cross-UV-border issue
i bet mcjTeleBlender 3 will be posted Sunday
but i bet uhhh just one chocolate chips cookie
Figure 2 : Oh my Bob i think i got the Opacity noodles all noodled the way Daz Studio noodles them !
if the problem is with Blender 2.71 maybe you could try installing an older version of Blender like 2.69 elsewhere on your disk, and tell mcjTeleblender to use that one ( and you have to install the mcjBlendBot scripts in the proper sub-folders )
DS cam to Blender script was very recently ( June 26th ) updated to work with Blender 2.71
https://sites.google.com/site/mcasualsdazscripts/mcjexportcamtoblender-ds1-2-3-4
oh and i dont use DS4.6 very often because it always always crashes after less than an hour -
so it's possible the problem is with DS 4.6 , since i tested it with DS 4.5
so i'll test this today
you seem to say the problem is mat-lib + genesis 2 + Blender 2.71 + DS 4.6 + scene-files + in animation batch files
that's a lot of variables ... but with luck i'll find the problem
here's something that may help
say you exported a 30 image animation
open the batch file that mcjTeleblender created for you in wordpad
it should look something like this
keep only the first line, and remove the "-b" which makes Blender work in invisible mode
it should now look like this
launch the modified batch file
you should see Blender and a second window, smaller with a black background
that's the Blender console
it may contain messages that warn you of problems encountered during the import process
you could also copy this and show it to me
---------------------
ok now i go in that scary place called DS 4.6 and will try to find what's up