Needless to say, as none of the problems I had with PP-2014 so far was addressed, I un-installed it immediately.

So, make sure to back up your preferences and any modifications you made so far in case you have to roll-back, too.

And check your figures carefully from every side after installing it.

Well, it's not that SM ever made an SR that didn't break something, but so far it never made a figure completely unuseable for me.

Wow !   :-0

What mesh is that in the .jpg  ?

"What mesh is that in the .jpg  ?"

3rd Gen DAZ mesh. Modified, of course, but nothing extraordinary.

Both figures are custom sculpted (DAZ morphs & ZBrush), re-rigged and one is weightmapped, the other not.

Must have something to do with normals, as it happens on the raw object file, too. I tried to smooth it out and do some other tricks I have, but it was persistant.

Doesn't happen on "vanilla" David 3 for example, but of course this is way beyond my tolerance threshold for "Poser Peculiarities".

Especially as there was no other performance improvement or flaw fix whatsoever for me. (Tested render speed several times)

Well, SR1 is supposed to fix standard V4's exploding butt, but I didn't bother checking after discovering this. If I really feel the insatiable need to use V4, I either use V4WM or one of my own custom V4s anyway.

:-(

Do your modified .obj have vertex normals in them?  It could be why the vanilla .objs are fine.  I don't think Poser has used normals or smoothing from the obj for a long time and maybe that's the cause of the anomalies (like a subD tweak or something that is causing them to show up).

Thank you, Netherworks. I'll check that.

Not feeling like re-installing SR1 right now, but I need to install PP-2014 on my backup laptop anyway, and that'll be a good opportunity to test your idea.

Yeah man, hopefully that will help :)  It's been a while since the discussion has come up but if I'm not mistaken, Poser hasn't used normals information since Poser 3.  It also has the benefit of making the .obj significantly more trim to leave them out.  I do recall issues from normals in back in the good old days.

Anyways, it's a complete guess on my part.  I haven't have a bad case to fiddle with.

I hate to think it's something we (testers) missed so I checked Aiko 3, Victoria 3 and David.  I also checked your WM David offering (thanks for sharing that by the way).  Of course, the WM david is on the cr2 level but that rules it the WM causing it (at least that iteration of it).  The other gen 3's I mentioned all had no issues.

I know we looked at V4 because that was a big part of this SR.  I have had Roxie in there, Cookie, Chip and My Michelle prior to the SR-1 release... along with Jessica and hmm... Alyson 2.

There are quite a few little tweaks under the hood including refresh issues and overlapping fixes for the morphing tool, rendering tweaks, and a nasty python bug where python scripts would lose the current actor when moving the cameras.

Thanks again, Netherworks.

I sent you a PM in case you want to have a closer look at the problem.

I really would like to use SR1 even though PP-2014 so far has been remarkeable bug-free right from the start, but not being able to use my custom figures is of course a deal-breaker.

Does Poser email us about SR's updates I didn't get a email.It's not good not knowing.
As far as I know SR1 didn't come with the PP14 download.don't know how to check.
Guess I alt to wait for the SR1.1.

JoePublic : All the customizations you make to others meshes ,
Why don't ya make ya own meshes ?

"JoePublic : All the customizations you make to others meshes ,
Why don't ya make ya own meshes ?"

I once made a head in Wings perhaps 12yrs ago and it actually looked mostly like the tutorial, but making a complete figure to the same technical level as the DAZ figures would take me way too long.

Simply don't have the energy left and I don't feel the need to make "my very own figure" as an "artistic statement".

I make a lot of things, but I rather build things not available as 3D objects yet.  :-)

Quote - "What mesh is that in the .jpg  ?"

3rd Gen DAZ mesh. Modified, of course, but nothing extraordinary.

Both figures are custom sculpted (DAZ morphs & ZBrush), re-rigged and one is weightmapped, the other not.

Must have something to do with normals, as it happens on the raw object file, too. I tried to smooth it out and do some other tricks I have, but it was persistant.

Doesn't happen on "vanilla" David 3 for example, but of course this is way beyond my tolerance threshold for "Poser Peculiarities".

Especially as there was no other performance improvement or flaw fix whatsoever for me. (Tested render speed several times)

Well, SR1 is supposed to fix standard V4's exploding butt, but I didn't bother checking after discovering this. If I really feel the insatiable need to use V4, I either use V4WM or one of my own custom V4s anyway.

:-(

exploding butt thing?

wha?

Oh please explain as I am about to upgrade. if there is serious issues I may not.

Huh.  That looks a bit like the Grouping Tool bug which the Pro Pack had when it was first released.  IIRC, Poser was dropping the last listed polygon in a geomCustom group.  How many polys are affected?    If you omit the line feed at the last line of an obj file, Poser will similarly drop the last listed polygon.  Wondering if smoothing is somehow similarly skipping out on it process too soon.

PoserPro-2014 computes magnets differently than previous versions.

As V4 uses a lot of build in JointSmootherMagnets (JSM) to help her bending, this causes her buttocks to enlarge instead of compress when you bend her legs.

So, bluntly said, PP-2014 breaks V4. (As well as all other figures using magnets for rigging)

SR1 gives you an additional magnet checkmark for the cr2, so you can disable the new magnet behaviour on a per figure base.

But it also introduces the mesh shading errors on certain figures as described above.

It goes away when the figure is switched to the new Unimesh skinning.

BUT, this of course is still no solution as the new Unimesh skinning method breaks ColorCurvators PoseMorphLoader, which I almost use on a daily basis.

(See screenshot).

And if it breaks that, who knows what other essential tools the new skinning might break, too.

The new skinning also changes the mesh appearance slightly, and doesn't work for all older meshes.

It's great to have an alternative skinning method which allows subdivision, but this doesn't mean the old skinning method can be effectively made unuseable.

It needs to be fully functional, especially given the fact that a lot of people still use older figures in lack of a viable set of new figures.

OK, I worked through all the above, and as you know I am also an extreme morpher.
But?
After more magnet horror then I can cope with, I NEVER-EVER use a magnet in a figure.

Magnets belong in a hardware store. Not in a cr2.
And, oh, no, nobody will change my opinion on this one any more.

So my question is;
What happens with a cr2 of YOUR figure when all magnets are removed.

Do a test; delete al magnets, save that cr2, and retest.

On a side note, I have not seen this in any figure I tested so far with PP2014 SR1.

Vilters, we completely agree on magnets: They are great tools for sculpting but should not be used for rigging.

There are no magnets in any of my re-rigged figures.

My figures are hybrids made out of David3's body and V3RR's head to save polygons.

The body mesh geometry is identical to David 3, but has been of course heavily morphed over the years. No different grouping, just re-rigging.

Older figures use capsule shaped falloff zone, newer are weightmapped.

There are no geometry errors. I checked each "problematic" vertice in Wings3D.

I'll send you a PM,so you can check out the shading error for yourself.

That is OK, I am looking forward to check this.

And very pleased that we are on the same magnet level.

I" ll be home in a hr or so, and check imediately.
I have several Poser versions up to PP2014 SR1,  and other apps installations to compare and troubleshoot.
tony-cross-1959@hotmail.be

I'm finding that the SR-1 has done strange things to rendering. If I set up a "final" render in Poser 2014 (at 1600 x 1600 pixels) and let it go all night long, it doesn't get beyond calculating indirect light. (At least, that's what the UI tells me.)  But if I send the render to queue, it will work.

Odd, isn't it?

Another bit of info:

The normals show up perfectly fine in Wings3D.

(There are also no errors to be seen in it's preview mode)

Problem found.
Will send a PM after some extra checks.

Something was changed in SR1, but this issue is not an SR1 problem.

Expect PM soon.
Best regards
Tony

Quote - Problem found.
Will send a PM after some extra checks.

Something was changed in SR1, but this issue is not an SR1 problem.

Expect PM soon.
Best regards
Tony

PM ? we want to know to.

Yeah. Please share the solution.

No solution, yet.

Vilters noticed that the mesh is unwelded and not perfectly symmetrical, but I don't think that's the root of the current shader problem.

Remember, PP-2012 and vanilla PP-2014 had no problems with the figures. Only SR1 found a few vertices it refused to digest.

All 3rd gen DAZ meshes unweld when exported from Poser, but they re-weld fine once you turn them back into a figure.

(And before someone says it, if you import the original DAZ geometry into Wings, it gets split into hundreds of tiny groups scattered all over the mesh, so I need to use the exported file to preserve the original grouping.)

And the slight asymmetry is caused by rounding errors between ZBrush, Wings3D, PoseMorphLoader and the Poser object ex/importer. Poor thing must have been ex- and imported a few hundred times in the last ten years.

Never bothered to fix it as real humans aren't perfectly symmetric, either.

(Actually several faces I sculpted I made intentionally asymmetric to add a bit of personality.

:-)

 Joe

So this does not actually RUIN your meshes so much as introduce a rendering error.

I hope there is a fix for you. I don't ever recall seeing this issue in any figures I tested.

From what you are saying-it is unlikely others will see this-unless they use your export/import workflow.

That said- I hope Vitters can get your concerns raised on the Mantis bug tracking board.

Eric

Quote - No solution, yet.

Vilters noticed that the mesh is unwelded and not perfectly symmetrical, but I don't think that's the root of the current shader problem.

Remember, PP-2012 and vanilla PP-2014 had no problems with the figures. Only SR1 found a few vertices it refused to digest.

All 3rd gen DAZ meshes unweld when exported from Poser, but they re-weld fine once you turn them back into a figure.

(And before someone says it, if you import the original DAZ geometry into Wings, it gets split into hundreds of tiny groups scattered all over the mesh, so I need to use the exported file to preserve the original grouping.)

And the slight asymmetry is caused by rounding errors between ZBrush, Wings3D, PoseMorphLoader and the Poser object ex/importer. Poor thing must have been ex- and imported a few hundred times in the last ten years.

Never bothered to fix it as real humans aren't perfectly symmetric, either.

(Actually several faces I sculpted I made intentionally asymmetric to add a bit of personality.

:-)

 

Hmm... in my experience I don't use any import options in Poser (under normal circumstances) and I only export with "Include existing groups in polygon groups" (again under normal circumstance) and I don't have any issues.  If a modeler is splitting the groups, that's something else, of course.

There are parts of the Gen 3 meshes that aren't welded to start with.  Could be the eyeballs for all I know, I haven't investigated what actually would get welded.  But if I export A3's mesh with the settings I posted, I get the same number or unwelded verts as the vanilla A3 mesh.  If Poser was splitting them out there, it would be a difference and a large one.

I'm also not saying that nothing ever got botched from an export from Poser either.  I am confident it happens (but also in modelers too).

Anyways, I'm hoping for a resolution to the anomalies in your figures, JP, I do lurk around and I realize you have put a lot of work into them. :)

I choose "Include Body Part Names in Polygon Groups" for one and "Include Existing Groups in Polygon Groups" for the other.

As some smoothing with the MorphBrush quickly reveals, the mesh is split.

But once it's turned back into a figure, Poser welds it all back together.

Welding it during export or import, or inbetween in a modeller, will change the vertex order, and so make all previous morphs unuseable.

If I need to work on an actual welded mesh (Like for making JCMs), I use PoseMorphLoader.

Poser never had a problem with that workflow, so, no, I'm pretty sure the welding/unwelding really isn't part of the problem.

Okay, I just tried and there was no split with the morph brush...

I brought David in (no import options - blDavid.obj) and played with the morph brush, smooth and such, no splits.

I exported that obj out (tmpExport.obj) only "Include existing groups in polygon groups" then imported that.  Still same thing - used pull, used smooth.  I didn't get any splits.

Ah, yes, the original blDavid.obj. That of course is welded.

But as I said, if I use that in Wings, it's split into a gazillion groups.

So, my "work-object" is always a geometry file I export from a fully zeroed figure out of Poser. Which Poser splits during the export.

(I shouldn't have said "default David geometry" as that suggests I was using the original blDavid.obj)

(Except for the PML files, which are exported welded without grouping)

Also, a lot of my figures are custom made hybrids, so there is no original pre-welded DAZ object file to start with.

Yeah and you are right, if you re-weld them at this point, the vertex count will change, sabotaging the morph targets.  I don't even think trying to bring the uvs across from the old to your new version, if welded, would even work.  Then when you go back in Wings, it sounds like that's over with anyways.

Poser is a harsh mistress.  :-)

You don't do everything exaclty as she wants it, everything goes wahoonie shaped.

But usually especially Wings3D plays very well with Poser. If a mesh looks good in Wings3D, it will look good in Poser, too.

BTW, before there was PML and ZBrush and the MorphBrush, I did most morphing with magnets. Sometimes several dozends at a time. Lol.

And I morphed a lot in Wings, despite the split mesh. The secret is to "stitch" the mesh back together without actually welding it by scaling the vertices.

Unbelieveable tedious to stitch Vicky's or David's hand or head back on if they were split, but it worked.

Modifying a mesh is so incredibly easy now.

Does that happen with Victoria4 and Michael4 too?

I've never really messed with DAZ's 3rd 4th generation unimeshes meshes.
I'm use to SubD meshes. like Genesis.Rex,Rox's etc etc.

I remember from the old days that V1 Possetta had a vertice that was not welded togeather think it's around her gen's between her legs.
Jump to genesis there's a few vertices not welded think there in one of the forearms.
or maybe it's unwelded edges. I just know the mesh is off.

I found this out when I had genisis in zBrush posed and symmetry didn't work.
I don't have 3rd 4th generation unimeshes meshes loaded.
u could check any mesh with zBrushes symmetry.

But unwelded vertices would not cause me a lot of grief with game style meshes.
But unwelded vertices would cause SubD meshes a lot of grief in any app.
or unwelded edges.

I remember it was easy to have stacked edges in LW 7 if you didn't watch what you where doing.
LW 7 was what 3rd generation was finished in.

Poser Pro 14 with SubD's changed the game.

There's wing,hex ,blender all free maybe one would do or show what the other don't 

Hi, I had  opened a bug for this but I think it got lost. Or I didnt promote the bug enough.

There is a bug in the python API for the new skinning, I think.

Posers export function is also affected.

The functions that tell vertex positions have now a different behaviour when asked for world coordinates. This affects weld vertices.  In classic skinning both vertices involved in a weld receive the same world coordinate (=after weld) but in the unimesh skinning, you receive the before-weld (unwelded) coordinates. This is the deviation that my exporter detects.

In classic skin its all good so I suppose its better to switch back to classic skin when using PML (can you switch back and forth without loosing data?)

this is seriously making me rethink purchasing pp2014 right now.

.<!!

"In classic skin its all good so I suppose its better to switch back to classic skin when using PML (can you switch back and forth without loosing data?)"

That's what I'm afraid of.

If I have to switch to "Unimesh" skinning, what will that do to my cr2's ?

Will it mess with the other tools I regularly use to modify my figures ?

If I have to constantly switch back whenever I use PML, (Which I do use very, very often), will that "degrade" my figure over time ?

It took 13 years to develop that group of figures. I definitely won't give it up or start new just because Poser decides they aren't good enough anymore.

But I'm still hopeful this can be fixed one way or the other.

Quote - this is seriously making me rethink purchasing pp2014 right now.

.<!!

It's just a little bug ya might not ever see sr2 will probably fix

I have no regrets getting Poser Pro 14

and would strongly advice to anyone to get Poser Pro 14

"It's just a little bug ya might not ever see sr2 will probably fix"

Sorry if I disagree, but this "little bug" affects ALL MY FIGURES.

As I neither plan to start doing postwork nor rendering clothed figures only, this basically makes Poser unuseable for me.

Actually it gets even worse, as even Vanilla PP-2014 introduced a rendering error THAT WAS NOT THERE BEFORE.

Even though it is much smaller than the one intrduced by SR1, it is equally totally unacceptable. I go to great lenghts so that my figures look as realistic and render as perfect as possible. I spend weeks rigging and creating JCMs so they bend natural.

I won't accept any flaw that was introduced to push subdivided low res meshes.

This flaw has to go 100% away.

PP-2012 render. PERFECT thighs.

Thighs suddenly show "pimples".

The thighs are perfectly smooth again.

But now ColorCuvators PML stops working and what will the new skinning do to other tools I use to maky my figures ?

I'm not yet sure what really happens if PML is used with new skinning.  With a bit luck the check is just overreacting. But its not easy to check. Many parts of the program are using the functions in question. My theory is that the figure skinning doesnt change the content of the figure, at least not morph-wise. You could make the morphs on a classic-skinned copy of your figures, and use the injection pose writer (and the new injection pose loader) script to copy the morph data into the figure with the unimesh skinning. Then you do not have to switch.

I used the figure Sydney G2 and am using the G2Fem_Shirt with her.

I have tried this 2 ways. The first was when I conformed the shirt to her before posing her and I had the same results.

Then I tried to pose her first, loaded the shirt then conformed the shirt to her.

The shirt explodes on both tries! I have attached images for review.

Quote - Does that happen with Victoria4 and Michael4 too?

No it does not.

It happens on custom figures that were created/modified in a specific way.  That doesn't make an excuse for anything in either direction, that is just what the issue is.  It is important and officially being looked into.

Quote - OK I am having a broken issue too since I ran Poser 2014 SR1 update this morning.

I used the figure Sydney G2 and am using the G2Fem_Shirt with her.

I have tried this 2 ways. The first was when I conformed the shirt to her before posing her and I had the same results.

Then I tried to pose her first, loaded the shirt then conformed the shirt to her.

The shirt explodes on both tries! I have attached images for review.

EDIT: After research we found that the problem is with the cloth and NOT with PP2014 SR1!!

Disregard or moderators please delete.

 Also-it did not RUIN the figures-they were still perfectly usable-in case someone is wondering if the title suggests that the models themselves were damaged....

Quote - > Quote - Does that happen with Victoria4 and Michael4 too?

No it does not.

It happens on custom figures that were created/modified in a specific way.  That doesn't make an excuse for anything in either direction, that is just what the issue is.  It is important and officially being looked into.

With each new version or new app I half to adapt. some times a little some times alot.
Meshes might get motified ,none ,a little or a lot.
Never seen a app that I couldn't kick it's ass thou.Never will.

Quote - I'm not yet sure what really happens if PML is used with new skinning.  With a bit luck the check is just overreacting. But its not easy to check. Many parts of the program are using the functions in question. My theory is that the figure skinning doesnt change the content of the figure, at least not morph-wise. You could make the morphs on a classic-skinned copy of your figures, and use the injection pose writer (and the new injection pose loader) script to copy the morph data into the figure with the unimesh skinning. Then you do not have to switch.

Hi Col,

I've been using a similar flow for PML in 2014 for Roxie (actually it was part of a flow I developed in the early days of PML so we didn't have to worry about .CR2 issues).

• Revert the figure to classic skinning.
• Do my PML work
• If I like the results, export the figure as a .obj use the "as morph" and "include body part names in polygon groups"
• Reload the .obj as morph with Netherwork's impressive "Morph Imp"

I'm probably being paranoid, but the same flow allows me consolidate multiple morphing sessions as well, which let's me create a "clean" morph.

Hopefully SM can get you fixed up so the "unimesh" skinning method doesn't present a barrier.

-K