1 stop...explain?

[metal0130]

I am curious as well, I would assume the stops are digital versions of a cameras F-stop, but I am not sure how different stops can be used in the same file.

edit: I am guessing here while I too wait for an answer... but maybe the low stops are used in the under exposed areas of the image and in the more bright areas, a lower stop would be used to bring out more detail... again, just guessing..

[JamesMK]

Well, this is just how HDRI works by default so to speak.

You snap the original photos using a method called 'bracketing', which is basically just a fancy way of saying you take several photos of the same thing (using a locked down tripod) with different f-stops, then combine those photos into a single HDRI file. The individual photos only have a low dynamic range (typically 8 bit/channel, commonly referred to as LDR in this context), but when combined (using HDRShop for instance) you let the software calculate the full dynamic range and write the result into a HDR file (typically 32 bit float/channel).

The different stops capture different ranges, so on one end of the scale the dark areas will be readable, while the brighter areas are just blown out. On the other end of the scale, only the brightest stuff is seen, while the rest is close to pitch black. Combine all those and you get your high dynamic range. Your shaders can use the full range, or they can apply exposure and/or gamma adjustments to access specific areas of the HDR.

This is all just plain vanilla HDR, in other words: no special Double Negative trickery, just how these things work.

[coldfuse[ofX]]

Ah wonderful,that is really well explained. Yea i didn't think it was trickary just never really heard of it, especially in terms of HDRI, but i guess thats because i haven't really looked into HDRI lighting.

Ok thanks alot JamesMK and Metal0130.

[coldfuse[ofX]]

Ah wonderful,that is really well explained. Yea i didn't think it was trickary just never really heard of it, especially in terms of HDRI, but i guess thats because i haven't really looked into HDRI lighting.

Ok thanks alot JamesMK and Metal0130.

[jeremybirn]

One more thing to mention: when photographers talk about the exposure going up or down by a stop, they don't always mean an f-stop change to the aperture. Sometimes they mean an equivelent amount of exposure change from another adjustment, such as from changing the shutter speed of the camera.

If you are shooting a bracketed set of images to make HDRI, then you want each individual image to have the same focus and depth of field, so that they can all be assembled together without weird fringes and artifacts. If you changed the f-stop, then you would be changing the depth of field, and making images that didn't all 100% line-up. So the person who was speaking probably meant going up or down a stop by changing the shutter speed, adding neutral density filters, or some other means that doesn't affect the aperture and DOF.

-jeremy

[coldfuse[ofX]]

Thanks jeremybirn, one thing i wanted to ask then, can changing the f-stop be applied to HDRI lighting? or by the explination of DOF changing create a problem for it?! The only reason i ask is because i have heard people try and explain the technique with changing the stop...meaning F-stop. So i was just wondering if it was bad teaching, or that another method of HDRI lighting for different effects can be to change the DOF rather then the exposure.

[ExP]

coldfuse[ofX], you're confusing a few things here.
DOF and exposure are directly related in photography terms. What they mean when saying "changing the f-stop on an HDRI image" is just that. After you create your HDRI file (with the techniques that JamesMK and jeremybirn explained) you have the ability to stop down your exposure of the image, making it darker, or stop up (increasing the f-stop) making the image darker.

Just think at an HDRI image as a file that stores all the possible light intensities of the real life subject (well almost all of it). Now, in post, just as you would in real life, you can freely change the exposure of your "CG camera" and display the image either darker or brighter, just like you would change the f-stop on your real camera.

[jeremybirn]

']Thanks jeremybirn, one thing i wanted to ask then, can changing the f-stop be applied to HDRI lighting? or by the explination of DOF changing create a problem for it?! The only reason i ask is because i have heard people try and explain the technique with changing the stop...meaning F-stop. So i was just wondering if it was bad teaching, or that another method of HDRI lighting for different effects can be to change the DOF rather then the exposure.

I can't tell from your question whether you mean they are using a real camera and shooting a bracketed set of images, or if you are just talking about going up or down to different stops in digitally adjusting the exposure of an HDRI file.

If you mean they are going out with a real camera and shooting a series of images that they intend to assemble in HDRShop (or with Photoshop CS2's Assemble to HDR script), then it is a bad idea to change the camera's aperture setting (f-stop) and it would be better to start with shutter speed changes. Ignoring this advice might still work in some cases (there isn't always a visible difference in DOF at every f-stop) but it creates a real risk that some parts of your shot might not line-up properly after parts of the shot fall out of focus at the wider apertures.

-jeremy

[coldfuse[ofX]]

coldfuse[ofX], you're confusing a few things here.
DOF and exposure are directly related in photography terms. What they mean when saying "changing the f-stop on an HDRI image" is just that. After you create your HDRI file (with the techniques that JamesMK and jeremybirn explained) you have the ability to stop down your exposure of the image, making it darker, or stop up (increasing the f-stop) making the image darker.

Just think at an HDRI image as a file that stores all the possible light intensities of the real life subject (well almost all of it). Now, in post, just as you would in real life, you can freely change the exposure of your "CG camera" and display the image either darker or brighter, just like you would change the f-stop on your real camera.

Yea i was reading up on F-STOPS and shutter speed's and saw that there was a correlation between the two. I was just wondering from the comment Jermybrin made.

If you are shooting a bracketed set of images to make HDRI, then you want each individual image to have the same focus and depth of field, so that they can all be assembled together without weird fringes and artifacts. If you changed the f-stop, then you would be changing the depth of field, and making images that didn't all 100% line-up.

When he remarked that when photographers were talking about changing the stop of a camera, it didn't directly mean, to change the F-STOP as only doing this will relate to DOF, however adjust your shutter speed and F-STOP for exposer. So i was wondering, if just affecting the DOF would create something different...and still work, but i guess not.

In relation to talking about taking HDRI photo's the lecture was talking about taking photos from a 8M lens which would take about a 180 degree veiw, inthwich they would take 360 degrees three times (while addjusting the exposure and shutter speed multiple times) with about a 15 percent overlap and then stitch together.

Does anyone still use the crome ball barring? Being told that at this moment you could only go up to a certain amount of resolution i think it was 9k, but don't hold me to that, where as taking it from the camera and stitching it up resolution could be even higher.

[playmesumch00ns]

Yep taking multiple angles then stitching a panorama is the preferred method here as well, but I've used chrome ball setups in the past with some success. Ultimately it depends on what you want to do with the image. I know dneg do lots of crazy environment stitching stuff so maybe they need that extra resolution, but for lighting a cg character you can get away with a 2kx1k pano quite happily.

Regarding stops, one stop is a brightness change of 2. So a stop down makes the image half as bright and a stop up makes the image twice as bright (this is the reverse of how it actually works on a camera, but anyway).

Stopping up and down is best done using shutter timing, as depth of field caused by a small apertures can lead to fringing, and ND-filters are never really neutral (though I haven't seen tested whether this leads to noticeable issues). Of course matters are complicated when you're taking an hdri in a studio setup and have timed flashes, so it's not a hard-and-fast rule.

[coldfuse[ofX]]

Guys Thanks so much for the explination, you have all answered my question and more some. I really appreciate it.

I think pretty much now, exeryone has covered what i needed to know and hopefully it may give others a little understanding into HDRI, thanks guys.

[CGTalk Moderation]

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.