selos69 Posted May 27, 2008 Share Posted May 27, 2008 Hi Guys, I understnad EXR files are open source and all, but Im not sure exactly what that means. What exactly are the benefits of saving as an 'open source' file over a standard tiff or png? I was reading (vismasters?) that you can use it to control the exposure between environments, i.e. In an external scene to make the interior brighter and vice versa? Cheers! Link to comment Share on other sites More sharing options...
SandmanNinja Posted May 27, 2008 Share Posted May 27, 2008 Just a quick note to an Aussie mate... As far as I know, EXRs are like HDRs - 32-bit and they have values greater than "1" or "255" (whatever the scale is), so they "emit" light. So you can use them just like an HDRI file. I usually split my HDRs into two areas - HDRI - Environment and HDRI - Lighting. I, too, await more cogent replies. Link to comment Share on other sites More sharing options...
Trevor Tizard Posted May 27, 2008 Share Posted May 27, 2008 http://en.wikipedia.org/wiki/OpenEXR Link to comment Share on other sites More sharing options...
shaneis Posted May 27, 2008 Share Posted May 27, 2008 Hi Guys, I understnad EXR files are open source and all, but Im not sure exactly what that means. It means that the source code for the file format it publicly available, so if you feel like developing plugins, software etc for EXR, you can do so. What exactly are the benefits of saving as an 'open source' file over a standard tiff or png? I was reading (vismasters?) that you can use it to control the exposure between environments, i.e. In an external scene to make the interior brighter and vice versa? Cheers!openEXR was developed by the good folks at ILM. One of the main day-to-day benefits of openEXR over other high dynamic range formats is that the file size is much smaller. To loosely answer your question about PNG or 8-bit TIF, they are low dynamic range, in other words, limited in how much tonal data can be stored in the file compared to EXR, for example. Think 256 possible tonal values for a pixel as opposed to 4,294,967,296. Hmm, better! This also explains why blacks and whites are clipped/ clamped in 8-bit files, and why banding artifacts appear in grey tones. All this is hard to tell when looking at a monitor - 16,777,216 combinations/ iterations (remember the "16 million colours" option for CRT monitors?) or converted to just over 10 million on a top-notch LCD with a 12-bit LUT... this is why CRT's are still preferred for colour work. As a side note, float data, calibrated monitors, gamma-corrected images/ cameras, RAW colour-clip...all this linear data comes together in the simple-yet-messy-knot that is LWF. Long story short - don't use anything less than 16-bit while you're working, preferably 32-bit, always keep an original in 32-bit and convert a copy of the original image to 8-bit when you're finished. EXR's are great for this 'cause they're smaller on-disc. ...As far as I know, EXRs are like HDRs - 32-bit and they have values greater than "1" or "255" (whatever the scale is), so they "emit" light... They have values outside the 0-255 range in both +ve and -ve directions... "whiter than white, and blacker than black", perhaps? Or not unlike the visible spectrum compared to the entire electromagnetic spectrum. For more info on openEXR, check http://www.openexr.com/ and for more info on floating-point, 32-bit vs 16-bit vs 8-bit et al, Wikipedia will answer all your questions and more. S. Link to comment Share on other sites More sharing options...
AJLynn Posted May 27, 2008 Share Posted May 27, 2008 There was a very good presentation by some Neoscape guys at a Boston user group meeting last month. (Any Neoscape guys listening? The web site has videos of the presentations on particles but not the ones on EXR...) It had nothing to do with it being open source, which doesn't really factor into daily use, but they demonstrated how to use Photoshop to see the quality of the color information in an image, and how bad JPEG is, how PNG is better, 16-bit is better than 8-bit and 32-bit is best of them all. It lets you work with the exposure in ways than non-high-dynamic-range formats don't - because of the ability to do super white, for example, if you've got parts of the image that are overexposed that would just be white in a normal image, in HDR you can drop the exposure and see the detail. Also, in my experience when you start messing with things like levels, in an 8 bit format you lose quality very quickly, so I don't work in anything less than 16 bit these days. Link to comment Share on other sites More sharing options...
Nic H Posted May 27, 2008 Share Posted May 27, 2008 You can also embed channel info into exrs, zdepth, specular, reflction, ao etc You need the ProEXR plugin to use these channels in photshop/after effects however, but Shake and Fusion support it out of the box. Its really handy for embedding all your passes into one file. Link to comment Share on other sites More sharing options...
Guest percydaman Posted May 27, 2008 Share Posted May 27, 2008 full float format with pretty good compression. Link to comment Share on other sites More sharing options...
shaneis Posted May 27, 2008 Share Posted May 27, 2008 You can also embed channel info into exrs, zdepth, specular, reflction, ao etc You need the ProEXR plugin to use these channels in photshop/after effects however, but Shake and Fusion support it out of the box. Its really handy for embedding all your passes into one file. If you run OS X or a Linux distro, Cinepaint also supports embedded channels in openEXR. Blender (all OS) supports same. http://www.cinepaint.org/ http://www.blender.org/ S. Link to comment Share on other sites More sharing options...
selos69 Posted May 28, 2008 Author Share Posted May 28, 2008 Cool, Im understanding it now. Thanks guys! 1 more question though, and the reason I couldn't be bothered using EXR's till now is how do I apply filters that have been dissallowed in photoshop? Link to comment Share on other sites More sharing options...
AJLynn Posted May 28, 2008 Share Posted May 28, 2008 You'll need to do all your 32-bit stuff then convert to 16-bit before continuing. Link to comment Share on other sites More sharing options...
Crazy Homeless Guy Posted May 28, 2008 Share Posted May 28, 2008 In reality, the adjustments you can do in 32bit with Photoshop are rather disappointing. I have been using EXR for a couple of years, but as of last week I decided to start rendering to 16bit Tiff files. Link to comment Share on other sites More sharing options...
John Dollus Posted May 28, 2008 Share Posted May 28, 2008 I agree w/Travis. If you are not using ProExr to extract the extra channel info and are using Photoshop, you might as well be using half float Tif files. They are smaller and you get the same results. Combined with the Vray render elements, it is a more space-concious way. Link to comment Share on other sites More sharing options...
Guest percydaman Posted May 28, 2008 Share Posted May 28, 2008 I agree w/Travis. If you are not using ProExr to extract the extra channel info and are using Photoshop, you might as well be using half float Tif files. They are smaller and you get the same results. Combined with the Vray render elements, it is a more space-concious way. I know a good reason to use 32bit exr's. Continuity with your other office? LOL Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now