Open plea to those who created incredibly useful (and free) python scripts in years past: I think a lot of us would be willing to pay for them if you updated them for P9 / PP2012. I checked the scripts in Free Stuff and see that most of them have not been updated in years. Which didn't matter until now; they worked fine in P8 / PP2010.

I suspect the OP can't get Light Emitting to work because he/she has IDL turned off.

That's the tradeoff. If you want to use Light Emitters, you have to be willing to accept the longer renders with IDL. Otherwise, boosting the ambient value of a material will cause JUST that material to be bright, with no illumination affecting anything around it.

Someone please tell me I'm wrong.

Quote - Once you load the pose in PP2010 and save, you can bring the scene back in PP2012. The Python code is used only in applying materials, it is not executed after that

I'm doing that a lot these days. I have some indispensible scripts whose authors don't seem to be in any hurry to update (they were free, so what's the motivation?) So I use them in PP2010 and reopen the scene in PP2012. Fortunately 2010 can read 2012 scenes, it just says something like "Newer version than expected, will try to read anyway." And it does. 

I wonder if this is possible: A Poser plug-in that allows it to read older python scripts along with newer ones.

I wouldn't go so far as to call that poster a troll, but I'm looking for a bit more than a hallelujah and SM's list of improvements. There are very nice hallelujah threads already. This one is intended for links to professional reviews, to help people who are still deciding whether to buy -- or are seeking unbiased views from someone other than a paying customer who really wants to feel it is wonderful. Significant excerpts from such reviews, when they are written, would be welcome, too.

I, too, found the two images very close, but I admit I do not have the eye of a connoisseur. I also didn't find the "review" (just a very brief blog comment about SSS) extremely helpful. But that's all there is right now! Hopefully the 3D media is taking a long, comprehensive look at it.

Specifically I want to know if the new versions make it worthwhile to keep SSS, IDL, hardware shading, hardware shadows in preview, etc, turned on. If the results are worth the increased overhead and rendering time.

For example, I found the shadows in preview to bog down Poser's responsiveness once you get beyond a very simple scene, and that hardware shading can produce some odd-looking results in preview even while it lets you see some node-based things previously only visible upon rendering. I want to read how a real pro optimizes such  things, or do they just turn them off in the interest of facility. 

Also I'd like to see a list of fixes and tweaks that were made in the service releases for P8/ PP2010 but were gone again in the initial release of P9 / PP2012. It's onerous to search through numerous threads here and at RDNA (some of them with hundreds of replies) to find such info. Onerous for Smith Micro, too. They may not plow through all those threads, but I bet they read reviews of their product.

In PP2012 loaded a complicated scene from PP2010 and turned on hardware shading. The resulting shadows in preview were terrible: blocky and solid black. And it slowed down the program's response considerably. Not worth it in this case.

I haven't yet built a big scene from scratch in PP2012 to know if the preview shadows remain accurate, and if they bog down the works, as more and more shadow-casting objects are added.

The problem turned out to be n-gons, that is, polygons with more than 4 sides. Poser likes 4-sided polys best, 3-sided ones most of the time (unless they are long and narrow), and n-gons not very much at all.

When adding new geometry to a model, i.e. by combining props, n-gons in any of the parts will cause polys to be deleted from the new parts. This doesn't happen 100% of the time, but often enough that it pays to make sure there are no n-gons before exporting anything from a 3D modeler for use in Poser. Too bad, because n-gons allow for more interesting and efficient modeling. N-gons are often created when beveling and smoothing corners in 3rd-party apps.

Subdividing every n-gon can create an ugly patchwork mesh that is more difficilt to work with, say with Poser's Groupiing Tool. But it must be done so Poser can handle it.

Another wonderful undocumented "feature" of Poser.

I tried adding one prop at a time instead of two. It did not matter.

I re-imported the combined prop with no options checked, including weld identical vertices. No difference. 

Note that I can combine these props with a different, less complex prop without problems. When I try to combine them with a different, equally complex prop, the same vanishing-polys thing happens. When I say complex, I mean less than 50,000 vertices, which is really not all that complex.

I'm starting to think the only solution might be to do it as a figure, with all new added pieces being body parts. I don't really like that solution because I try to have as few figures in my scenes as possible. More than a couple figures and Poser becomes slow and jerky.

The poly's aren't just invisible or reversed, they are gone, missing, deleted. I have verified this with the Group tool. 

A strange thing: When I export the group of props as an OBJ, and I open that OBJ in a modeling app, all the polys are there. It is when it is imported back into Poser that some of the polys are deleted. (Not an entire former individual prop, but pieces of it.) Even when resaved as a new OBJ in the modeling app, the same thing happens when pulled back into Poser.

Somehow it is the combining of props in Poser that is screwing up the polys. And it happens when the construction has become pretty complex (but not always). Also, upon trying to reproduce it while merging the same group of props, sometimes different polys are removed, but still in the same former individual props. Rebuilding those props from scratch and using them instead doesn't solve the problem; once the overall structure reaches a certain level or type of complexity, almost anything that is merged into it will have missing polys. I am not sure what that state is, or how to avoid it.

The OBJ file is almost 2 MB, or I would have posted it. I was hoping someone else had experienced this and figured out what was causing it.

Do you lose your material zones that way?

The thing I don't like about the Morph Tool in P8 / PP2010 is that it creates a Custom Morph dial in every single body part, from every fingertip to every tongue joint. I like the fact that it works across body parts, but I wish it would generate morph dials only in the parts actually touched by the tool. Automatically adding every body part to a custom morph seems unnecessary and wasteful.

Correct. It was repaired after the last report and before StaceyG looked. Thanks.

Two threads on this now, originally posted in the wee hours. Problem remains at 10:40 a.m. eastern. Hopefully somebody at Rendo checks the site today.

Same here. Nothing in the sale is discounted. Hey, all the steps but one were performed!