I need to do some more work to balance the illumination of the window vs the light coming through, but at least that solves the underlying problem.

TransparentBsdf is good, glass and translucent let color through but too much scatter so it's basically just a big blur, which is probably as it should be. The above is mixing principledbsdf for the glass itself with transparent for the light passing through, and then I boosted the exposure in postFX.

Wait, I think I've got it. It appears to be that the glass has shadows turned off and that also somehow allows light to pass through but prevents it from getting colored (not sure why). Let me do some more tests.

Hmmm.

It's not the image map either, when I switch to just using blue:

That's like this:

Right, I wasn't actually using glass, translucent, transparent, and principled all at once. In the image it's just the principled that is connected but I also tried all three of the others in its place. I left them all in there to show how I had them set up in terms of colors and normals. Just imagine plugging any one of them in to the cycles surface node.

Oh, also, the saturation is cranked up on the HSV node in case there was too much white in the image. The result is the same without the HSV node though.

As a halfway measure, maybe RO could give vendors an easy way to promote an image that used their item. You can already click on the gallery link and see what other people are doing with a product (which is great, I use it), but maybe a way for a vendor to say "this image is great and really shows off my work", highlight it, and give the artist some kind of reward (e.g., could even just be points).

But personally at this point, I basically just assume I will want to rework the shaders on the majority of what I buy. It's just part of setting up a scene.

Perhaps I misconstrued how Ken's tool works. I purchased but have not yet installed Dawn2.

I would just note that making textures for existing clothing, making poses, making facial expressions, making characters -- these are all easier than making clothing. And the skills you would develop making textures, including making promo images and submitting to the RO marketplace, are things you would need to make clothes. So that may be a more sensible place for people to start.

Regarding facial expressions, there is a little new trickiness there in LF and Dawn because of the combination of morphs and face chips. When you save an expression you want to do it in a way that doesn't erase a character; that could maybe use a tutorial or something somewhere, or at least an explanation of best practices. (Even better would be functionality in Poser that makes it easy to separate character morphs from expressions.)

I just got a email from PoserSoftware highlighting it, so maybe now that they are using it as a selling point, there will be some attention to the underlying functionality....

:(

For thumbnails, about half of my HDRIs have .jpg thumbnails with the same names in the same folder. And that's basically how thumbnails in the library work too. I would be satisfied on that front if it displayed that non-HDRI thumbnail image and not the real thing.

That looks pretty cool! But I have a couple of wishlist items for it!

I have a whole bunch of HDRI images. When I go to pick them, it would be super helpful to see some thumbnails so I have a general idea of what I'm getting. Often I have to try a few before I get one I like.

Second, since you can rotate the Y axis and you can also rotate the camera, it's really hard to know what part of the image you are looking at. Poser 13 has a dome that can be used to show your HDRI background, but it doesn't sync if you manually rotate the HDRI image. It would be incredibly helpful if your script could set up that dome and the XYZ controls handled both the HDRI background and that preview. If you add that, I will buy it for sure!

I really like the change for that reason. What I don't like is that when you open an old scene, all those checkboxes are checked so you have kind of destruction settings that you might not notice.

I have not noticed the progressive slowdown you're reporting (I'm not using a Mac though, maybe that's relevant?), even when I do multiple animations in the same scene. My #1 issue with animation is that I don't have a good way to insert frames between existing frames. Select and drag does not work right for all figures, some morphs do not carry over properly. Copy-paste does not work right if you have clothing that is set to auto scale to a figure that is not 100%, and in any case it's tedious if you just want to insert one frame in the middle. There might be a solution I'm overlooking but I haven't stumbled across it yet.

Graphics card is a GeForce RTX 4090 with 24GB. Render settings give me a choice of Nvidia GeForce, OptiXNividia GeForce, and the CPU. Which one should I pick?

It's possible to put a plane in front of your light, and give it colored transparency, so you get a projected image. But getting the shader and the light intensity and the relative positioning all correct at once is very tricky when any one of those things ruins the whole effect and you don't know which one you've got wrong. Also, you need to make an enclosure around the light and plane so there's no leakage to the side, and there isn't a hollow tube primitive. So you need to make one side of the enclosure transparent which requires a trans map.

It would be nice if there was a prop in the library with a complete gobo setup, with all the correct spacing and all you had to do was load your image. I've struggled with it a few times and never been satisfied.

The bat signal is the easy version, because there's no color and the light bleed around the outside is intentional.

I put a work-in-progress image in that other thread but here's a completed one. This is a 20-image story, the first 12 are up in my gallery already https://www.renderosity.com/users/ChromeStar/gallery , I'm posting the rest 3 per day. This particular image is #14, so sneak peek for tomorrow.