[brodie_geers]

So I've got some Dosch HDRI's that I'm using for my environment.

If I just use the same map for all the channels the background channel looks far too...bright? I'm not entirely sure what this indicates or how to properly fix it. I did a test by applying a gamma correction in PS of .45 and then it looks much more like I'd expect. But IF that's the solution, should I only do it for the Background channel, or should I apply that to the other channels as well?





-Brodie

[Half Life]

Why not just adjust the exposure/EV number?

But yes, I often have to manipulate the intensity of the HDRI (particularly Dosch) in the environment settings.

Best,
Jason.

[brodie_geers]

It doesn't seem, to me, to be an exposure issue. Regardless of the exposure, the image has the "washed out" look of an image whose gamma is too high. I could start tweaking it until it looks better, but it seems like there should be a definitive way to go about fixing this. One of the best things about Maxwell is that you don't have to worry about this Linear Workflow mumbo jumbo.

My guess is that the HDRI has a 1.0 gamma and what I'm seeing in the background (and reflection?) channels at render time is also coming out at 1.0 gamma rather than being shifted to 2.2? I'm grasping at straws.

-Brodie

[gotoxy av-media]

never use cheap dosch-stuff!

munch

[tom]

When you load HDR image directly in viewer, what do you see?

Maxwell display should match the same with the following viewer:
http://ict.debevec.org/~debevec/FiatLux/hdrview/

[brodie_geers]

Above is the image converted to 8 bits, so this is what it looks like in your viewer (and Photoshop...and Maxwell).

I just remembered that Maxwell comes with some HDRI's, a couple of which are from the SkiesV2 pack that I got from dosch. Opening the ones in the Maxwell directory and comparing them to the ones from Dosch provided some insight. The ones in the Maxwell directory seem to have already been gamma corrected (adjusting my dosch version with an Exposure/Gamma adjustment and adjusting the Gamma by .45 (or adjusting the MW version with a 2.2 adj) makes the images identical).

So Tom, tell me if I'm on the right track here. The Dosch HDRI's have a baked in Gamma of 2.2 (for reasons I can't fathom), but HDRI's going into MW should have a gamma of 1.0. So I need to gamma correct ALL of the HDRI's from Dosch before placing them into MW. To say it another way, every channel in MW should have an HDRI that I've applied a gamma correction of .45 to in order to get them back to Gamma 1.0. And this seems to be exactly what NL did to the dosch images before shipping them out with the software.

Sound about right?

Anyone got a snazzy way to apply a .45 gamma correction to about 300 files? I guess PS can do that with a batch process right?

-Brodie

[brodie_geers]

Slight caveat. As it turns out the high res version of the HDRI (each HDRI has 9 images which correspond to different mapping types and resolution sizes) seems to have the correct 1.0 gamma. This is similar to another issue I posted awhile back with another Dosch pack which had the same issue although I never quite figured out why. Glad I'm finally (I think) getting this figured out.

-Brodie

[tom]

Yes, all your words are precisely correct. And, I really don't know why Dosch decides to ship them with gamma applied. As you see, in every other application their images look the same. So, it's surely not Maxwell. On the other hand, a lot of other HDR image sources provide them with linear gamma. Because; PS, Maxwell, HDR Shop etc already applies gamma 2.2 when displaying them (not when rendering of course). I hope this makes sense.

-edit:
Here's a useful discussion
http://www.hdrlabs.com/cgi-bin/forum/Ya ... 1218917875

[brodie_geers]

Tom, thanks. That link was indeed informative.

-Brodie