Page 1 of 2

Maxwell sky as HDRI different v2.5

Posted: Sun Dec 19, 2010 7:59 am
by Richard
Previously one could output maxwell's sky from studio and then use this as HDRI as IBL with intensity of 1 and it would match almost to a T! With 2.5 this same workflow results in SUPER over bright result.

Current example with 2.5 for physical sky only iso100 / shutter 2800 - IBL needs iso75 / shutter 16000+ to get near same results.

Can anyone else report same issue?

Re: Maxwell sky as HDRI different v2.5

Posted: Sun Dec 19, 2010 8:32 am
by Richard
Hmmm? actually it seems when internal emitters are turned on!

Re: Maxwell sky as HDRI different v2.5

Posted: Sun Dec 19, 2010 8:51 am
by Richard
Hmmm? Nope worked it out loading the HDRI into MXCL to check the output they are actually all white! None of the sky settings transfered to the HDRI!

Checked on windows 32 and 64 including testing those from the packaged preset skys and can confirm ALL render out as white HDRI!

Re: Maxwell sky as HDRI different v2.5

Posted: Sun Dec 19, 2010 2:08 pm
by Mihai
I'm getting the same lighting from Physical Sky and turned into HDR. Are you using a camera view when creating and rendering with the HDR? It looks white when loaded in Maxwell but lower the exposure and you should see it looks like the physical sky.

Re: Maxwell sky as HDRI different v2.5

Posted: Sun Dec 19, 2010 11:00 pm
by Richard
Hmmm? Thanks Mihai.

Yeah your right, it does have some info there! but seems needs to be adjusted before bringing back in as IBL.

So what mate do you think I could be doing different? It was just a straight swap back in when I've used this method previously in 1.7 now the results are rather unpredictable, in fact seemingly unusable without lots of tweaking which is negating any time saved using this method to speed the results. Weird!

Re: Maxwell sky as HDRI different v2.5

Posted: Sun Dec 19, 2010 11:24 pm
by Richard
Mihai wrote:It looks white when loaded in Maxwell but lower the exposure and you should see it looks like the physical sky.
I get the same background result if I play around with exposure but loose the illumination level matching the background - shadows gone more sky dome like!

Re: Maxwell sky as HDRI different v2.5

Posted: Mon Dec 20, 2010 5:13 pm
by numerobis
for me the illumination seems to be the same at intensity 1,0 (i didn't measure it, but it looks like it is).
the hdr opened in PS is bright white, but if it is "calibrated" at intensity 1,0 then it has to be like that i think.

Re: Maxwell sky as HDRI different v2.5

Posted: Tue Dec 21, 2010 3:00 am
by Richard
Yeah I don't know what I'm doing wrong! If I use the HDRI from the save out I don't get anywhere near the same results. With 1.7 it was just save out and slot back in - no setting no tweaking. Now it would seem the you have to open the saved out image in MXCL and tweak it there first then start the whole testing tweaking process between background intensity and other channels. I've ditched 1.7 off my system unfortunately otherwise I'd do a comparison test.

Re: Maxwell sky as HDRI different v2.5

Posted: Tue Dec 21, 2010 3:55 pm
by brodie_geers
Here is what I did

Open Studio
Create Camera
Render Image
Save Physical Sky to HDR
Switch Lighting to IBL
Put HDR in Background slot
Make sure "Same as Background" is checked for all channels
Render

And I'm getting identical results. Maybe you could do a pack n go, zip the folder, and upload it to dropbox (http://db.tt/xpTYcu5) or something so we can compare results?

-Brodie

Re: Maxwell sky as HDRI different v2.5

Posted: Wed Dec 22, 2010 4:58 am
by Richard
Hmmm? Mate you shouldn't have to render the image!

Thats weird though - Mihai also notes the white resulting HDRI.

I'll try again with a naked studio scene!

Re: Maxwell sky as HDRI different v2.5

Posted: Wed Dec 22, 2010 6:59 am
by Richard
AHHHHHH! Good prompt Brodie! Narrowed down the issue! And I'm not an idiot or doing anything wrong! It's a maybe minor (though JD probably wouldn't say that) SU/Plugin/Studio bug!!

Here is what I did!:

1. Open SU file export sang / physical sky with default camera to render,
2. Open MXS file with Studio - test with Fire "Perfect"
3. Now save sky as HDRI slot back in as IBL - test again "Way over bright"
4. Create camera in Studio / physical sky - test with Fire "Perfect"
5. Now save sky as HDRI slot back in as IBL - test again "Perfect"!

All tested with the Tom clear sky preset - indeed shows an issue that sky saved with SU camera imports (and clones of) save out overbright (too intense) HDRI!

To note the SU camera operates correctly on both HDRI & Physical sky saved out from Studio created camera.

Re: Maxwell sky as HDRI different v2.5

Posted: Wed Dec 22, 2010 1:39 pm
by Mihai
That's a weird one. What are the camera settings for the camera exported from SU and the new camera you create in Studio? Fstop, SS, ISO? Is auto exposure turned on for the SU camera?

Re: Maxwell sky as HDRI different v2.5

Posted: Wed Dec 22, 2010 2:28 pm
by Richard
Yeah I thought weird too mate!

Remember these cameras don't match, this I imagine should not change the created out and back in effect of the HDRI unless the settings are change between tests. There are certainly differences between the cameras but again I would imagine this should not effect one camera and not the other. It is pretty easy to replicate!

SU (settings as imported - SU out of box > Plugin default):

FStop: 5.6 | SS: 522.45 | ISO: 100

Studio (default, create camera):

FStop: 8.0 | SS: 500.00 | ISO: 100

Re: Maxwell sky as HDRI different v2.5

Posted: Wed Dec 22, 2010 3:22 pm
by Richard
Mihai

JD responded on the SU plug sub forum:
Check your step 4; I would guess that your two cameras have different EVs.

What I see here is that EV=15 is the point where the HDR will match when replacing the Physical Sky it was created from. If, however, you move your camera's EV to 17, then the HDR look darker than the Physical Sky did, and vice versa -- if you change EV to 13, the HDR will look brighter. In your example, you are comparing HDR outputs from cameras with EV=14 and EV=15; the HDR from the EV=14 camera is brighter than the HDR created using the EV=15 camera you created in step 4.

So the moral of the story would be: when creating HDRs in Studio, use a camera where EV=15. I have no idea whether this would be a bug, or if it's just how things work.
Mate This could well be the issue - would then this be a bug as I would imagine it incorrect considering the output HDRI should match the sky using the camera for orientation only?

Re: Maxwell sky as HDRI different v2.5

Posted: Wed Dec 22, 2010 11:40 pm
by Mihai
Richard wrote: 3. Now save sky as HDRI slot back in as IBL - test again "Way over bright"
If you used the same camera settings between the Phy Sky and HDR created from that same camera and they don't match, then something is wrong. Could you PM me a simple version of this scene? Just the camera and one object would be enough. Leave the camera directly exported from SU and the Phy Sky settings you used.