Not all HDRI's are the same. They are made with areas of light "baked" into the dome image but some are more intense than others. You can go to Iray preview and play with the Dome rotation controls and see how the different rotations affects your scene. make sure you try all the axis.
As for your spots, they won't have an effect on the dome, other than laying a shadow down from your model. For your model, you just need to find the light settings that work for your scene. A general rule of thumb I use is to increase lumens by a factor of 10, until you notice a big change. Then you can use the intensity slider to fine tune that. I also usually start at 100,000 Lumens.
So, for example, you put 50,000 Lumens in and test, but the light barely shows... try 500,000 lumens next and repeat. Once you see the big jump in light, than play with the settings in smaller amounts and use the intensity slider to fine tune it. Once you have the spotlight set just the way you like, select it and make new ones by Create New Spotlight --> Copy Selected Items. Then you'll get a clone with all the same settings.
I notice your spotlight geometry is set to point (default). Try using Disc and set the width and height to values between 75-100. This creates a wider area of light projection, which softens the shadows. A point geometry will create sharper edges on your shadows.
Another thing you can experiment with is to render the scene twice. Do one pass with the Dome on to set the model in the environment, with the base shadow. Then you can turn the dome off and ground off and just shoot the model alone on a transparent background. When you get the model looking good, save it and than use an image editing program like GIMP or Photoshop to layer the two renders together.
Lighting is an art, and it takes a lot of time to learn well. I'm no where near as good as many of the Devs on here with mine yet. I'd suggest reading up on photography tips for lighting, especially the basic "3 point Lighting System"
And remember, HDRI's are a sort of "Quick & Dirty" means of lighting a scene and providing a background. All the best DAZ renders are done with actual scenes and proper lighting and some take hours to set up properly and render.
Not all HDRI's are the same. They are made with areas of light "baked" into the dome image but some are more intense than others. You can go to Iray preview and play with the Dome rotation controls and see how the different rotations affects your scene. make sure you try all the axis.
As for your spots, they won't have an effect on the dome, other than laying a shadow down from your model. For your model, you just need to find the light settings that work for your scene. A general rule of thumb I use is to increase lumens by a factor of 10, until you notice a big change. Then you can use the intensity slider to fine tune that. I also usually start at 100,000 Lumens.
So, for example, you put 50,000 Lumens in and test, but the light barely shows... try 500,000 lumens next and repeat. Once you see the big jump in light, than play with the settings in smaller amounts and use the intensity slider to fine tune it. Once you have the spotlight set just the way you like, select it and make new ones by Create New Spotlight --> Copy Selected Items. Then you'll get a clone with all the same settings.
I notice your spotlight geometry is set to point (default). Try using Disc and set the width and height to values between 75-100. This creates a wider area of light projection, which softens the shadows. A point geometry will create sharper edges on your shadows.
Another thing you can experiment with is to render the scene twice. Do one pass with the Dome on to set the model in the environment, with the base shadow. Then you can turn the dome off and ground off and just shoot the model alone on a transparent background. When you get the model looking good, save it and than use an image editing program like GIMP or Photoshop to layer the two renders together.
Lighting is an art, and it takes a lot of time to learn well. I'm no where near as good as many of the Devs on here with mine yet. I'd suggest reading up on photography tips for lighting, especially the basic "3 point Lighting System"
And remember, HDRI's are a sort of "Quick & Dirty" means of lighting a scene and providing a background. All the best DAZ renders are done with actual scenes and proper lighting and some take hours to set up properly and render.
Thanks for the tips. I just asked since in this video the lights they put were lighting up the hdri even after reducing the intensity to 0.
You must be registered to see the links
2) if you use photoshop to merge the two layers, how will you get the shadow of the model since in the first render we are only rendering the background?
Thanks for the tips. I just asked since in this video the lights they put were lighting up the hdri even after reducing the intensity to 0.
You must be registered to see the links
2) if you use photoshop to merge the two layers, how will you get the shadow of the model since in the first render we are only rendering the background?
I never really tried to change a dome lighting, so I imagine it just acts like a .. well .. a dome! haha. But it's a big dome, and the scene is at, or near the zero point, so I think you would need some pretty bright lighting to actually reach the dome surface. But then again, DAZ does some weird stuff with domes.
For your second question ... there's really two ways to do this. For an HDRI, I think the easiest way would be to render the model alone with Ground ON, but dome off. That would generate a fake shadow on a transparent alpha layer that will layer well in PS. The dome is still providing lighting, just not the background image. So you can rotate the dome to adjust where your shadows will fall. Just make sure you test render it or check Iray preview to make sure the ground is lined up properly. If AUTO doesn't line it up properly, turn Ground Position Mode to MANUAL and then change the Ground Origin Y axis up or down until the shadow surface is lined up properly. With a single figure and no props, AUTO should work fine. Usually Manual is only needed when some other stuff is poking down below ground level.
I did a test render. Just a basic G8F on one of the default HDRI's in DAZ.
Here's the model done with Dome OFF and Ground ON. I rotated the dome to project her shadow to the right, so it lines up with the shadow from the single spot light I am using to highlight her upper body, since this HDRI has a lot of shadow areas.
Now, here are the two images layered together. The shadow is there now. However, that is the shadow from the HDRI mainly. The spot does very little. That's because inside an HDRI scene, it's pretty hard to overcome all that light in a 360 degree dome around your model. To get a spot light bright enough to cast a decent shadow would wash out your model. That's why it's easier to do this in Post work.
Now, here's a nice trick you can do in your editor. To make the shadow darker, just lay another layer on top! Easy! The model won't change, since she's a fixed image, but the shadow is a transparent alpha layer and when you stack them up, they get darker. Here's two layers on top of the HDRI...
This is all quick and dirty. Less than 20 mins to compose and render and another 10 in PS. I've been coding all day so it was nice to play in DAZ again for a bit
I've got a crappy video card BTW (still using an GTX 1050ti) and usually I end doing CPU rendering. But even my crappy vid card was able to render the HDRI in 10 seconds and the G8F reached 100% in about 3 mins. - When I rendered them separately. If I did this as a one-shot with everything showing at once, my render time would probably triple. Maybe a little less since it's an HDRI. For complex scenes, I need 2+ hours for a nice render in soft lighting. But if I do this layering, I can sometimes cut that down to 30 mins or so, between the background and foreground. It helps that i know Photoshop pretty well.
I never really tried to change a dome lighting, so I imagine it just acts like a .. well .. a dome! haha. But it's a big dome, and the scene is at, or near the zero point, so I think you would need some pretty bright lighting to actually reach the dome surface. But then again, DAZ does some weird stuff with domes.
For your second question ... there's really two ways to do this. For an HDRI, I think the easiest way would be to render the model alone with Ground ON, but dome off. That would generate a fake shadow on a transparent alpha layer that will layer well in PS. The dome is still providing lighting, just not the background image. So you can rotate the dome to adjust where your shadows will fall. Just make sure you test render it or check Iray preview to make sure the ground is lined up properly. If AUTO doesn't line it up properly, turn Ground Position Mode to MANUAL and then change the Ground Origin Y axis up or down until the shadow surface is lined up properly. With a single figure and no props, AUTO should work fine. Usually Manual is only needed when some other stuff is poking down below ground level.
I did a test render. Just a basic G8F on one of the default HDRI's in DAZ.
Here's the model done with Dome OFF and Ground ON. I rotated the dome to project her shadow to the right, so it lines up with the shadow from the single spot light I am using to highlight her upper body, since this HDRI has a lot of shadow areas.
Now, here are the two images layered together. The shadow is there now. However, that is the shadow from the HDRI mainly. The spot does very little. That's because inside an HDRI scene, it's pretty hard to overcome all that light in a 360 degree dome around your model. To get a spot light bright enough to cast a decent shadow would wash out your model. That's why it's easier to do this in Post work.
Now, here's a nice trick you can do in your editor. To make the shadow darker, just lay another layer on top! Easy! The model won't change, since she's a fixed image, but the shadow is a transparent alpha layer and when you stack them up, they get darker. Here's two layers on top of the HDRI...
This is all quick and dirty. Less than 20 mins to compose and render and another 10 in PS. I've been coding all day so it was nice to play in DAZ again for a bit
I've got a crappy video card BTW (still using an GTX 1050ti) and usually I end doing CPU rendering. But even my crappy vid card was able to render the HDRI in 10 seconds and the G8F reached 100% in about 3 mins. - When I rendered them separately. If I did this as a one-shot with everything showing at once, my render time would probably triple. Maybe a little less since it's an HDRI. For complex scenes, I need 2+ hours for a nice render in soft lighting. But if I do this layering, I can sometimes cut that down to 30 mins or so, between the background and foreground. It helps that i know Photoshop pretty well.
Wow thanks for the tips haha. I recently got myself a 4090 to get started in the daz business to create a VN based on my story. I am a writer with coding background but never had the gpu to render things myself. So here's hoping I could get good at this
Wow thanks for the tips haha. I recently got myself a 4090 to get started in the daz business to create a VN based on my story. I am a writer with coding background but never had the gpu to render things myself. So here's hoping I could get good at this
Edit: Forgot to answering the second way to do what you asked...
You could also render the image on the HDRI normally, the first time - then do another render, like above, of the model alone with the ground shadow and layer that on top. This way really wouldn't make any difference in this situation with an HDRI, but in a normal scene with environment and props and no HDRI, it would be the best way to get proper shadows.
You do the first render to get the complex shadows on all the various surfaces in your scene. Then on your second render, you hide the scene and leave just the model so you can get nice lighting and effects on that. This has the benefit of you being able to shorten the first render time by a lot, since you are not really concerned with how well the actual model looks... just the background and shadows. On the second render, that's where you capture the detail and layer it on top,
