Welcome to our community

Be a part of something great, join today!

  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Grading RED footage...

the camera only records at the one fixed "real" ISO anyway.

IMHO I don't think that the notion of a "native" 320 ISO has to do with a *fixed recording* on the capturing / camera / sensor side. As far as my understanding goes using 320 ISO has to be interpreted on the software side for the Red camera to be more meaningful. I.e., in software 320 ISO just means that no gain (unity gain) or minimal gain in case it does, will be applied to whatever raw data has already been recorded. Any number higher than 320 in software shall provide greater or more positive gain to that raw data (which has the effect of moving histogram to right) and any number less than 320 in software provides negative gain (which has the effect of moving histogram to left). But, the raw data remains intact.

Basically, what I am trying to say is that if you optically expose to 500 ISO using metering (i.e., changed aperture size, exposure time, etc.) then that does not mean that the camera magically transformed the raw data soon to be recorded to 320 ISO natively, or internally. It recorded what ever was coming out of the sensor that corresponded to 500 ISO metering as determined by optics, and hence the raw data recorded is based upon 500 ISO and not some native 320 ISO data. In my understanding if you use 320 ISO in software, after the optics were exposed to 500 ISO metering, then the software shall do nothing/minimal changes to that data. Any number higher or lower than 320 in software shall have the effect as described above, and in my previous posts.

But, again, I could be all wrong :ohmy:
 
I've been wondering how to set the .r3d files to be essentially 0 gain in post. I've been looking at ISO 100 as being zero gain, but who knows. Like I said in my other post about my first test, I need a good bit of time to work this all out more for myself....

Going by what others say isn't my thing, I gotta see it myself.
 
why is the "development" of the raw image referred to as "grading"? am I missing something?
 
why is the "development" of the raw image referred to as "grading"? am I missing something?

I agree with what I think you're saying. Here when we are discussing exporting from Redcine I have been calling it "developing" and we track it as such. Then after the "developed" files (usually DPX) are on a finishing machine I have been referring to that portion of the job as "grading".

Of course Redcine is being used as the final grading product for some people so there will always be different terms thrown about.
 
if you have been disappointed in your footage, you can always go back and regrade the RAW. You might be surprised what you have.
I'm kinda confused by this recent flurry of posts from Jim. What if we have been pretty pleased with our RED footage up until this point and we know how to expose our footage correctly? Does build 16 actually offer any image improvements within the camera itself, or merely "help" users attain better images that may not have a grasp on how to do this already? Don't get me wrong, it sounds like the new setup will be very advantageous for a lot of shooters - but with all the hype behind "build 16," I'm wondering what exactly to expect coming down the pipeline. On a recent post, Jim stated that "[with] some simple tips, your camera will increase in range because you will be getting the most out of it." That doesn't sound to me like the camera is actually performing better with build 16, but rather that it's easier for users to hit the "sweet spot" and protect the best range possible when they are shooting. Again, that's good news - but I just want to know what we should all really be expecting from this build.
 
Hey Chris, how did your shoot go? Are you experimenting on the post end of things or still shooting?

http://www.reduser.net/forum/showthread.php?t=14069

As for card tricks, I did a few high speed shots of card springs, hard to frame up though, its always been a tough shot to pull off and have lit really nice. I once did a still photo of it years ago... dozens of tries for one good frame:

http://www.dpchallenge.com/image.php?IMAGE_ID=252437

Too bad I shot that at high ISO and didn't own any super bright lights at the time. Still a cool shot. I love how the cards are back lit and glow. Hard getting the light to strike them right.

I didn't have red cards on me though... @#$% Oh well, some other time, with more time.
 
There are three important stages to getting the best possible image.

1. Expose correctly and knowing the sweet spot.

2. The conversion of the sensor data to raw.

3. the conversion of raw to another format.

I am limited in my knowledge of code but i imagine that at point 2 would be an important point in terms of programming and then point three as well.

For me, the ideal is to maintain as much data integrity throughout the process so I would do a test shot of ideal exposure, under exposed and over exposed and bring them into red alert. Find out how much info is in the highlights and when the noise arrises and then depending on your particular shot make a judgement on which way you want to go. Adjust your exposure and then shoot. It is probably better to export a tiff to photoshop for more exactness.

If you are shooting lone or don't have the time, you are better off under exposing if you have highlights that are clipping. Again it's a judgement on what highlights are acceptable to your subject matter.

While working with raw has many advantages before the down conversion, anything changed at step two means a permenant change (you can always come back), however you still will be able to color grade the footage. So too much fiddling in red alert will make it harder for you further down the line. It is best at this stage and for further color correcting IMHO to just maintain the data integrity unless you are severly down converting. That way you have more of the balance to play with. And while raw is great for the perfectionists in us, you would be surprised at how much detail can be maintained in other formats when down converted and especially the trade off is more in the favour of maintaining your data integrity from what i have done in my testing. However, you should just test this for yourself and see what feels right for you.
A very simple curve or even none at all and no touching the exposure or iso unless maybe bringing them down a touch if there's too much over exposed areas.

The reasoning behind this is if you want to really be able to deal with noise and pulling detail then you need as much info as possible with a fully fledged color correction program. It needs more work than curves. Different zones need their own masks and other tools are needed to deal with things like noise.

This is all very easy to test for yourself with a few hours to spare. And you should do tests. The name of the game for me is to find the best trade off and that has so many variables and even changes from shot to shot. That's what good tools provide you with. The ability to confront different circumstances with the best compromise.

And imaging is always about compromise wether you talk film, digital, painting or whatever. We can all appreciate just how amazing the human functioning of sensor perception is and the goal of the artist that seeks to emulate it.

2rrkt53.jpg
 
I'm kinda confused by this recent flurry of posts from Jim. What if we have been pretty pleased with our RED footage up until this point and we know how to expose our footage correctly? Does build 16 actually offer any image improvements within the camera itself, or merely "help" users attain better images that may not have a grasp on how to do this already? Don't get me wrong, it sounds like the new setup will be very advantageous for a lot of shooters - but with all the hype behind "build 16," I'm wondering what exactly to expect coming down the pipeline. On a recent post, Jim stated that "[with] some simple tips, your camera will increase in range because you will be getting the most out of it." That doesn't sound to me like the camera is actually performing better with build 16, but rather that it's easier for users to hit the "sweet spot" and protect the best range possible when they are shooting. Again, that's good news - but I just want to know what we should all really be expecting from this build.

Sounds like this to my ears, but i think i also heard that the noise level has been improved. that would give better "usable" DR.
Plus, the compression is supposed to have been improved also.
 
Is Jim really really big downstairs ?
 
I'm kinda confused by this recent flurry of posts from Jim. What if we have been pretty pleased with our RED footage up until this point and we know how to expose our footage correctly? Does build 16 actually offer any image improvements within the camera itself, or merely "help" users attain better images that may not have a grasp on how to do this already? Don't get me wrong, it sounds like the new setup will be very advantageous for a lot of shooters - but with all the hype behind "build 16," I'm wondering what exactly to expect coming down the pipeline. On a recent post, Jim stated that "[with] some simple tips, your camera will increase in range because you will be getting the most out of it." That doesn't sound to me like the camera is actually performing better with build 16, but rather that it's easier for users to hit the "sweet spot" and protect the best range possible when they are shooting. Again, that's good news - but I just want to know what we should all really be expecting from this build.

Yes... and yes. There is less noise in Build 16. And a better quality noise.

And yes... it will make it much easier for everyone to expose and grade properly.

In addition, if you have a current way you work, you won't lose that option... just less noise.

And... there are some new features.

Jim
 
What if we have been pretty pleased with our RED footage up until this point and we know how to expose our footage correctly?

Great point, Häakon! what if, indeed... what if we've taken a look at those recent, amazing wildlife stills on that very thread... and what if, that agreed, we do miss... a sleep-mode rather than care fussing about allegedly clipped whites/crushed black and "shortcomings" of the red one, where we would be better call them "limitations" or "already excellent technical specs"... Jim don't get mad at me, and remember we've been told the feature was thought of, but "no promise" that it would ever be enabled.

Denis (sleep-mode weirdo)
 
Great point, Häakon! what if, indeed... what if we've taken a look at those recent, amazing wildlife stills on that very thread... and what if, that agreed, we do miss... a sleep-mode rather than care fussing about allegedly clipped whites/crushed black and "shortcomings" of the red one, where we would be better call them "limitations" or "already excellent technical specs"... Jim don't get mad at me, and remember we've been told the feature was thought of, but "no promise" that it would ever be enabled.

Denis (sleep-mode weirdo)

Not sure the point here... we have enabled features, are enabling more features, and will continue to enable features. Sorry if we aren't making your schedule. Doing the best we can.

jim
 
is BUILD 16 looking pretty close for a beta release?

Getting close... we are making some changes and improvements based on Alpha testing.

Jim
 
Won't help you too much...

Jim

9_xxx.jpg
 
Looks really nice Jim :-)

Graeme
 
Back
Top