Attached Link: http://host1.bondware.com/~syydr/download.ez?download_section_id=31&Form.sess_id=671634&Form.sess_k

Here is an interesting test which I can get Poser 5 to freeze 100% of the time when switching light sets - I would really appreciate any and all feedback from people who perform this test, especially if anyone can get a successful render without a freeze. This information will really help us in tracking down this bug and properly reporting it to Curious Labs. If you want to perform this test, follow these instructions: This will require that you have installed two free light sets from RDNA "Creed Lights 01" and "Pastels" which are available in their Free Lights download section (see link) 1) Launch Poser 5 to the factory state with the default lights and Don in his whites. 2) Switch lights to the DNA Creed Lights 01 - (first one on the list "Bark at the moon) and render the scene. 3) Switch lights to the DNA Pastels (first one on the list - Pastel 001) - and try a render. On my system, it freezes 100% of the time even after a fresh reboot. The funny thing is that on the surface it looks like both DNA light sets use Light 1, Light 2, Light 3 etc for their names. The difference between them is that the DNA creed light has no Lights 2 or 3, but it has a Light 1, Light 4 and Light 5. The Pastel set uses Lights 1 thru 4 but has no Light 5. Also the lights in one set (with the same names in the other) must change from inf to spot or vice versa, and maybe this is what is causing the problem. Anyway, I'm still tracking it down, and any feedback others provide will be very helpful. Thanks, Ron

P.S. I'm not suggesting in any way that the light sets are defective. In fact, I know that they are NOT defective because I have carefully evaluated them. They do, however present certain conditions to Poser 5 which leads to a freeze.

It's a known issue. There's a clean lights python script in Free Stuff that will knock out all the lights except one. Run that script each time before switching to a new light set and you should be fine. Cheers!

which pastel lights..there are global and spot sets ???

Thanks, Crescent, but we already know that it is a "Known" issue. But if it were that well known, then why was it not fixed in SR3? Clearly Curious Labs did not get enough HELPFUL feedback to locate the exact problem, and so far nobody has been able to document the bug with a REPEATABLE test. Complaints don't help - nor do dismissive statements. This is a valid test that actually might provide the necessary data to Curious Labs to see that it gets fixed in the next release. Don't participate if you don't want to, but please don't try to discourage others from doing so. Also, based on other information we have gathered, this test might reveal the underlying cause of MANY existing bugs in Poser 5 which are due to the processes of UPDATING and have nothing to do with lights per se. Work-arounds are fine, but they should not be an acceptable permanent alternative to a fix. Thank you for your contribution.

Ron...it locked up for sure.. hit render...loading textures in the render box...and nothing...

hmmmm..... well, it locked up the first time, but I've been trying over and over, shutting down P5, restarting, etc, and it is rendering fine everytime now....

The Pastel set from RNDA used in this test is called "RDNA Pastel Spotlights" and it is located on the second page of their Light Downloads section.

Yes, one of the characteristics of this challenging problem is that after you get P5 to successfully accept the swap, it will not repeat during the same session and even during some subsequent sessions. Re-launch Poser and start the test from the beginning - it should freeze if you have waited for awhile between sessions, but may not as long as the P5 kernel is still in memory (a trick used to make subsequent launches of P5 faster). If you wait an hour or so and try again the problem is likely to recur.

I wasn't trying to be dismissive - I thought you were looking for testing to find a workaround for the problem. I'd heard rumor that the lock-ups had something to do with the shadow mapping feature, but with the python workaround, I didn't bother investigating it further. Cheers!

Good approach. Wish I could help. <>