Daz 3D is part of 
Connect
DAZ Productions, Inc.
224 S 200 W, Suite #250
Salt Lake City, UT 84101
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.Daz 3D is part of
Connect
DAZ Productions, Inc.
224 S 200 W, Suite #250
Salt Lake City, UT 84101
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.
Comments
Thank you very much!
Thank you for the script and the patient explanations. But just to clarify: Do you mean to scale of the size of the emissive object? - Or did you mean to "scale" some other value?
Still no luck for me, its hanging on the apply fix button, but I appreciate you trying to help me get it to work
Thanks for this but I can't get it to work. It's more than likely me but I've followed everything mentioned here (I think) but still no go. Would someone please be kind enough to help me, I'm kinda new to this and chances are I've stuffed something up.
It worked for me, however the idea of moving your scale down means that you will typically get a harsher shadow compared to when it is larger.
The way I got it working was, in fact, mixing your script with what KindredArts stated in https://www.daz3d.com/forums/discussion/551301/ghost-lights-what-s-happening-and-what-s-next#latest and not changing the scaling at all.
However, instead of going to the max like what was required with KindredArt's solution, because of the way your script turns on the two new options, I had a lot more control over the cutout opacity, and therefore not as much Luminance.
Luminance: 100000.0
Units: cd/cm^2
Cutout Opacity: .10
This gave me the same lighting, not seeing the black reflection anymore, and gave the smoother shadows that I would typically see.
Hi, i made some tests now with the fix and i must say it has no effect.
i loaded a ghostlight, selected it and loaded the script.
The script shows every gl in the scene, i click on "Apply fix" and nothing happens and the script window doesn't close itself. Is this correct?
Tested in DS 4.20 beta
- normal gl has no effect in lighting
- gl with fix makes the scene darker and adds a strange shadow into the scene. This shadow only appears through the fixed gl, but not by the normal non fixed one.
- i was not able to make the gl work with the fix too.
I meant Scale the size of the emissive object. Bear in mind the use of different luminance units, Lumen is amount of light irradiated no matter how big or small light is, and the other units refer to strength of light emitted.
For those of us with lots of Iray scenes and scene subsets, this is going to be major pain to deal with. I use "ghosted" emissive lights in nearly every scene created since Iray was introduced into DS many years ago. Hundreds.
There should be a settings selection somewhere in DS that permits the user to select how this is handled so that the defect is transparent to the user. We have been told "You need to bump up the luminanace value." and "Daz has not changed this, nVidia has.". Are we supposed to go through every scene created and re-used, search for "ghosted" emissive lights, and determine how to "bump up the luminanace value"?!? I don't care iwho is at fault for this debacle--DAZ or nVidia--DAZ should be providing a solution within DS. Period!
I noticed this problem immediately after rendering a scene with 4.20. And now I have wasted half a day trying track down the issue and find a workaround. I typically don't spend any time on this forum unless I have to track down a solution. And usually, it's a solution to a an issue that DAZ injected into a revision (e.g. the file-save defect with Dome and Scene plus no Environment Map is still not fixed!). This tells me a lot about how DAZ views it's end users. It is simply not right for DAZ to place the burden of dealing with this on the end user. Can you imagine buying a new car from GM and them telling you that you are responsible for any defects in parts that are supplied to GM by 2nd or 3rd tier parts suppliers? Anyone who thinks DS is free is not paying attention. We pay for it every time we buy content. In fact, I have self-illiminating propsets--puchased from the DAZ store--that include "ghosted" emissive primitives provided for lighting the scene.
I suggest that every single person affected by this submit a ticket. I, for one, intend to ask for a rollback version of DS until there is a real fix to this issue.
Your're right. Script does not inform neither closeswhen the task ends, but it adds the new special Iray properties in the parameters tab. If the object does not show this new properties is an indication something went wrong, in that case Log File might give a clue.
I agree. In fact one particular set I've used a few times includes ghost lights at windows that illuminate the interiors of room to simulate both day and night. I believe it's called Euro Apartment by Roguery? Anyway, that's one such set that'll be hit hard by this 'bug fix'... which doesn't seem like a bug fix to me. In fact, after reading a forum entry by KindredArts, I did some experimenting and found that I can take an emissive surface, reduce the opacity to 0.0000001 and ramp up the luminance to a much higher value than before, and this 'ghost light' works exactly the same as it did before - gives me light, and does not show up in reflections. So what was fixed? Or was it just 'patched' by making the end user THINK it stopped working.
Truth be told I was careful only to install the Public Release of 4.20 rather than the General Release. I'm actually still using the 4.15.0.2 general release for my mainstream work and intend to continue to do so. I don't consider the old way of creating ghost lights as exploiting a 'bug' at all. It was a feature, and a very useful one!
It breaks all the scenes from PAs!!!
ok. Thanks.
it would be good to write this information somewhere in the script.
The dark background disappears with "Enable Iray Matte = off", but how do i make the Ghostlight working now?
Currently the luminance in my testrender is set to 1200, but there is no difference between GL = on and GL = off.
It is still not working. Even setting luminance to 1200000000.0 (kcd) does not make any changes.
:-(
jag11,
Thank you very much for providing this workaround. For what it is worth, my testing shows it to be the best one I have yet seen on the forum (here I am again today on the forum instead of working on my backlog, frustrating).
I noticed that the script errors-out if the scene contains a Null (log: "Error in script execution"). Unfortunately, there are still a lot of content providers that use a Null instead of a Group (not to mention all of the propsets purchased in the DAZ store that include Nulls). I have attached the scene that I have been using for testing. I really, really hope that you will continue to work on this until DAZ builds a solution into DS (if they ever do, not holding my breath).
I have attached the scene that I used for testing. Just to give you an idea of how important this issue is to me, I used a variation on this light in just about every Iray scene I created. Hundreds.
> I noticed that the script errors-out if the scene contains a Null (log: "Error in script execution").
yes. Same problem here. If the scene contains a NULL, and error appears.
I hate to do this but there is one other issue with this solution. Using the emissive object with the opacity set to 1.0 (which it must now be in order to emit), the object will generate a shadow. Anyone know if there a workaround for this? If not, this would be a deal-breaker for me.
I set the opacity to 1 and refraction index to 1 and the refraction weight to 1. (after applying the jag11 script to add the new parameters.)
Just as an FYI, in doing some experaments I found that ghost lights made this way will still work the same in older verstions of DS
Thanks for providing the scene. I fixed the script. Download from here and I'll update link at the original post.
HTH
barbult,
Thank you. Progress. I probably should know this regarding refaction index and refraction weight. In your testing, do you see the "the light mesh in reflections"? Did you find a workaround for this, as well?
Attached is a simple scene I used for testing. It includes a plane with a reflective surface (Polished Aluminum Shader) in order to accomdate this test case. The attached image shows the resulting render.
My computer is tied up rendering, but I'll try your scene when I can.
Thanks.
I used the newer version of the script, but it has still no effect, even when you enlarge the luminance much
Ok, I loaded your scene. As it is, I see the dark plane, too. jag11 posted the solution: Change your emissive surface luminance Units to Lumens (lm) and scale your plane to a small value. The light output from lumens does not diminish when the size of the source (your emissive plane) is reduced. Your plane has a negative scale number. I changed it to 0.01% and set the lumens to 100000 (nothing magic about that number - pick what is appropriate for your scene usage). You have Two Sided Light on. Why? I don't see any difference when I turn that off in this scene.
With the lumens and scaling solution, I don't know if refraction is still needed. I haven't experimented with that since reading jag11's insight on lumens and scaling. It's a learning experience for all of us.
Edit: By the way, that is a pretty neat scene that you created. I like it!
It is far more natural to work with lumens and easier to manage in our minds.
No surprise a Spotlight inside Daz Studio is 1500 lumens. An advantage I see working in lumens is lower Exposure values.
As for the two sides: oversight. The vast majority of emissive lights I use everyday are one-sided. This is was a case of quick-let's-get-a-test-scene-together. I have attached a new version that uses a one-sided emissive light plus it implements the lumen suggestesion by jag11. However, By setting the scale to 0.01, it appears to me that we are essentially creating a form of 2D point light. Typically, I want the scaled up geometry of the emissive object. I have also attached a couple of renders that illustrate why this not going to be the desired solution in many cases.
The first render is of a scene that I have been working on for some time and for which I need to use in order to incorporate a render into a deliverable soon. The render was done in January (prior to DS 4.2). It utilizes 4 emissive planes (there is also a point backlight plus a distant light to simulate the sun). One emissive plane is above the G8F figure. Another is just in front of the middle mirror and is used to illuminate the entire figure front. There are also 2 small emissive planes pointing to the figure head. One to figure head front and one to figure head left. I need the geometry of the mirror light in order to properly illuminate the figure. If I set the scaling to 0.01, it will not work properly.
For the second render, I used DS 4.2. I applied the jag11 script to all of the emissive lights but I only used the scale=0.01 method on the two lights pointing to figure head. It took about a half hour to tweek it down to where I could live with it. But the two images are very different.
I never set out to implement complex lighting. I do the minimum I think necessary in order to get the image I'm looking for. I already had lighting that worked before 4.2. I am not happy with the prospect of re-working scene lighting when it was fine before 4.2.
One render has illumination outside the building at ground level. The other does not. What caused that?
Unfortunately I'm getting the same result. As of 4.20, none of the ghost lights emit light in renders, the script doesn't fix them, and as per previous suggestions, i attempted cranking up the luminance (tried one at 700,000), also changed the luminance units to lm, no dice. Nothing seems to pick them up, in fact the only light that seems to render - at all - are distant sun lights.
For me , it work perfect with the actual 4.20 version.
Thanks for the script.
Thanks for the script. The problem with Ghost Lights was the only thing that was preventing me from downloading DS 4.20.
Cheers,
Alex.
I don't know. There is no additional lighting at ground level. I was curious about that myself. I have no way to test it until I revert back to an earlier version of DS (which I intend to do on Monday).
Actual Ghost Lights (as sold in the DAZ store) do work in 4.2. In order to test, simply multiply your Luminance value by 1000000. You should see the scene lighting up similar to your last render. However, if your Ghost Light is in the camers' field of view, you will likely see lots of fireflies. If you change the Luminance Units to lm and then scale the light to 0.001, you will have a very different light than the one you started with. I strongly suggest that everyone affected by this submit a ticket. I know it's a pain but, in my experience, that is what actually gets DAZ's attention.