Welcome to our community

Be a part of something great, join today!

Log in
  • 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 :)

Determining ISO...

David Mullen ASC said:
Maybe it's just me... but I shot eight features in HD without any scopes or waveforms and they came out fine, exposure-wise. I've spent the past several years trying to stop being so anal-retentive about exposures. Sure, I don't want to make mistakes, but there's going to be a limit to how much I effort I want to spend thinking about exposure. My natural tendency is to be over-prepared and over-precise, so I'm trying to be much looser about it, do more by eye, etc.
Click to expand...

This is somewhat freeing, I must admit, and it is good to get this "relax, kid, have some fun" advise.

But is this equivalent to a master AC saying, "I'm trying not to measure distances anymore"? lol.
 
richcam1 said:
The RED has no Knee so it is more unforgiving is this regard so, test how she handles the highlights.
Rich
Click to expand...

IMHO, in this regard the RED is not more unforgiving than a F900R for instance. This is where colour granding comes into place. The white roll-off has to be adjusted in post. Although a F900R has a knee funktion white roll-off has to be helped in post as well IMO. RED is not WYSWYG on set. With LUTs that might change in the future.

Hans

(For a given contrast in a scene it's vital to know how many "headroom" one needs for a nice white roll-off that will be created later. This of course goes straight against the blacks or even dark mids wich will become noisier. Noise in the blacks again can be fought by crushing them or brighten them with light (both ways decrease the contrast) - which leads to the finding that usable latitude is key which again is limited by the sensor and of course the used bit depth.)
 
Dylan,

I think this is very positive for you. It seems to me that the best (and perhaps only) way to do this is not through on-set scopes and waveforms on the live picture. But through experience. Trial and error. Knowing how the camera reacts in different situations.

I think the best way is (initially) to continue testing and start to get the feeling. Then, on set, when you are pretty much set have the data tech record a quick test and open it up in REDCine. You could take a quick look at the histogram in there, and the image itself. If you like what you see, tell Guspie that you're ready to roll.

I bet that after a while, you'll have had the camera in so many of the 'typical' situations that you'll feel 110% confident rolling without that final REDCine check. And you'll be one of the few DPs that will. I think this camera will have a personality of its own, that needs experience to know how it will react.

After 6 months of shooting with it, it'll be rare you'll encounter a situation that you won't instinctively know what to do with it, IMO. You'll be seeing what the image looks like on your monitor (and EVF soon) and then you'll know how it looked when 'finished'. Experience will be the key. I see this a huge positive for people like you who are not shying away from this new technology, but rather are all over it.
 
"I understand the history of the 18% card, but I agree more with Ansel Adams in that: find out what your capture media can do and center your scene into that window. The way you handle highlights is personal tast/art."

I love the spin offs that this thread has taken. Yes, Chris, I agree. It will be good to have some quality time with the RED and then be able to "roll with it".

This is what shooting with 500ASA film stock is for me. It is a relaxed place, where the meter sits sad and lonely in the AC's bag.

So...to swing it back to what I was originally asking...which, I think, I now have my answer...

There is no way to "accurately" monitor in real time the signal that is being recorded?

Rich, your quote above is exactly what I am talking about.

Here is what the test I did back in December showed me. I "centered" that scale and called it "normal" based on what the 720 output to a waveform told me. When we put it in RED CINE and looked at it, that "normal" shot now sat at the lower half of the RED CINE histogram. That tells me that what is "white" (say 109) out of the 720 output, was actually more like gray?

I can figure out my own system to deal with this, but I am just curious to know if there is a more accurate way of monitoring the signal live - without the REDCINE step...

Sorry if I sound like a broken record!

Dylan Macleod, CSC
Cinematographer
Toronto, Canada
www.dylanmacleod.com
 
Dylan, I know what you are saying, and I agree with you! It is a bit challenging to expose RED at first.

Obviously, test test test until your comfortable.

A good combo is to use the stoplight in the histogram (keep it just barely turned off) and a scope on the preview. Maybe when the dual link outputs are enabled a scope will give us dead nuts on signal monitoring.
 
I think were just seeing a big new learning curve. I started working with RAW cinema back in like May of 2007, and it took me MONTHS to really understand what I was doing. Learning what LUTs are for footage, for your monitor, etc.. it was all new to me. The older digital cine cameras were masking a lot of the work that really goes into making a digital image because of the built in processing, sure you had control over it to some degree... but I don't think many people, myself included, understood what was really happening.

By the way, I do realize meters are calibrated for 18% gray, I'm just taking a new approach in how I calibrate cameras and interpret my meter.

As for how high up you should expose the Red One, there is a lot of debate surrounding this. My personal view is that you need to think like this when you approach a RAW camera:

"I have two ranges, or two film stocks basically, one is linear and has "X" amount of dynamic range. The other film stock is the LUT I will apply to my footage, it has "Y" amount of range, which is less than "X", it's response to tones is also different and these characteristics must be considered when lighting the set."

In others words, I'm saying that you should probably develop a post LUT first that will be applied to your RAW footage. Treat post LUTs as the digital equivalent of a print film stock, and your RAW files as a digital equivalent of a negative film stock.

Once you enter this mind set, your work will start looking more like you envisioned it. This also means that, depending on the highlight characteristics of your post LUT, you won't always be exposing as high up as you normally would if you wish to avoid more complex secondary color correction, which also means more time, and time = money.

You can set you meter and interpret anyway you like as well. I personally work from highlights down, I sorta like to think of it as carving an image, you start at the top (highlights) and work your way down into the wood (shadows). But, you can start in the middle and work outward if you like, I've done both, I just prefer a modified Ansel Adams zone system approach when shooting digitally.

Hope that helps!
 
If you are having a hard time monitoring the Red One... You might want to just take a trial and error approach to figuring out your ISO, shoot several exposures, take notes, etc... Then open the files in post, look at them and compare your notes. Not the fastest way, but you are bound to eventually figure it out.
 
Cardmaverick said:
If you are having a hard time monitoring the Red One... You might want to just take a trial and error approach to figuring out your ISO, shoot several exposures, take notes, etc... Then open the files in post, look at them and compare your notes. Not the fastest way, but you are bound to eventually figure it out.
Click to expand...

Hey Chris

This is exactly what I am going to do with my test next week. Start with the output from the 720 and waveform as a "baseline" then cross reference it with REDCINE to be able to better interpret what the 720 output is telling me.

I am down with LUT's and RAW and will figure out how to rate the camera.

Thanks for the insight though.

Back to my original mantra;

"Is there a way to accurately and precisely monitor the signal that is being recorded "live""?

Thanks.

Dylan Macleod, CSC
Cinematographer
Toronto, Canada
www.dylanmacleod.com
 
dmacleod said:
"Is there a way to accurately and precisely monitor the signal that is being recorded "live""?

Thanks.

Dylan Macleod, CSC
Cinematographer
Toronto, Canada
www.dylanmacleod.com
Click to expand...

If you keep your ASA setting at the nominal setting, which you must decide for yourself (between 250 and 320), then the monitoring system is supposed to be accurate.

Emmanuel
 
Humm... Perhaps this issue with live monitoring is one of the "surprises" that will be found in one of the newer builds getting ready to be released.

I can't imagine they don't know about this issue.

Wait... I'm curious, since I don't have a Red camera, but I do have an Andromeda, I had a similar issue once with regard to monitoring live output on a histogram... Maybe the issue is more to do with the histogram measuring remapped tones or something. I believe the histogram you are looking at in Redcine is probably set up for looking at the footages linear data with no signal processing applied in a higher bit depth, where as the histogram for the live camera output is more a histogram representation of footage after a LUT has been applied in a lower bit depth. Not so sure about the bit depth, but if you are measuring data after signal processing, this would certainly explain why the histograms don't match up.

I'm just speculating though.
 
The monitor out has one big advantage, it's the worst your image will ever look, so, regardless of your ISO rating it's not a bad place to start. If you like what you see, you will invariably have more to work with in post. Personally I make the determination at the beginning of a project what the need and mood of the project dictates. The Red One gives you many options here, 250-320 is great when you want super clean practically noiseless images, but ISO 500 and 640 can be very beautiful for a more textured look.
I chose 640 for my first music video because I specifically thought 320 was too clean for a nighttime club/street feel. Like anything you need to season to taste.
 
interesting...

interesting...

"Wait... I'm curious, since I don't have a Red camera, but I do have an Andromeda, I had a similar issue once with regard to monitoring live output on a histogram... Maybe the issue is more to do with the histogram measuring remapped tones or something. I believe the histogram you are looking at in Redcine is probably set up for looking at the footages linear data with no signal processing applied in a higher bit depth, where as the histogram for the live camera output is more a histogram representation of footage after a LUT has been applied in a lower bit depth. Not so sure about the bit depth, but if you are measuring data after signal processing. "

I think that is exactly it.

Kinda like the Viper in the sense that maybe you don't want to see the "RAW" image. So you apply an LUT to get something "useful" to look at.

I certainly would like the option of knowing the "whole story" though. Maybe in another build?

I don't have my RED either, but will be testing with my friends #98X body next week.

All good information to have in the back pocket.

Dylan Macleod, CSC
Cinematographer
Toronto, Canada
www.dylanmacleod.com
 
I've been thinking some more about this....

Like I mentioned earlier about applying LUTs in post and what not, I have a feeling that the big missing piece with regard to this whole output monitoring issue is this:

User definable viewing LUTs

Bingo! Thats it. I bet if you could load a linear LUT into the camera for viewing, you would be seeing the "real" picture with regard to how things are really being recorded. This all makes sense now. If you are shooting for a certain LUT that will be applied in post, it makes sense to measure your data after the LUT has been applied, but still be able to monitor the RAW data in its natural pre-processed state.
 
With regard to assigning the camera higher sensitivity ISOs, you are really just setting yourself up for under exposure of the RAW data. I'm not sure thats the best idea in the world. I also think its confusing people. It might be best to assign the camera just two ISOs really, one for a proper reproduction of 18% gray, and another for guys like me who base off of highlights. From that point on, just say your underexposing, not increasing sensitivity, because you really aren't.

It's just me and my opinion though, but I really do think the community as a whole would be less confused if this approach were taken.
 
To each his own, when shooting at 500 or 640 you can almost treat your highlights like you would with film knowing you have at least 1.5 stops of headroom there. That's plenty to create a nice rolloff even if you do get some clipping. Noise is not your enemy if you can learn to use it, control it and tame it. It can and has become another tool in your visual arsenal.
 
I think you've got to set aside the notion that how the RAW image comes up in REDCINE when processed into 10-bit LOG RGB, that it looks darkish, is somehow indicative that you should be rating the RED camera much, much slower to somehow make it look "correct" when it gets processed.

It would be better to think of it in terms of what rating gives you a color-corrected image that you like. What matters is the final results, not how the RAW file comes up automatically.
 
Exactly David, thats why I keep telling people to develop a LUT they like for a project prior to shooting. The only reason the RAW linear image looks dark is because of the monitor though, most monitors aren't setup to display linear images, so you need a LUT just for viewing the RAW data correctly, which is something you can choose in Redcine.

Ideally you would be lighting and exposing for the end "print" LUT, but at the same time, a good DP will also be aware that there is more information they can use if they need to via secondary color correction if they need to say, cool down a highlight, or perhaps bring back more shadow detail.

I'm not sure what you mean about rating slower based off a "darkish" image though.
 
David Mullen ASC said:
Which suggests: 250 ASA = +5 stops over, +5 stops under

Though in practical terms, detail existing at 5-stops overexposed at 250 ASA doesn't seem right, neither does 4-stops over at 125 ASA. My brief testing at 320 ASA showed clipping occuring above the 3-stops overexposed range. But then, that's not based on a Stouffer's wedge chart...
Click to expand...


David, I meant that 320ASA set in the camera corresponds to 125-150 ISO in real life.
By setting 320ASA, the camera gives you 4 stops over and 6 under. So maybe you want to go to a faster ASA setting so we can have the +-5stops balance.
The problem is the more you set your 18%grey to the right the more you get into the noisy sensor area.
The last 4 stops are really noisy and have some compression artifacts.
 
David Mullen ASC said:
I think you've got to set aside the notion that how the RAW image comes up in REDCINE when processed into 10-bit LOG RGB, that it looks darkish, is somehow indicative that you should be rating the RED camera much, much slower to somehow make it look "correct" when it gets processed.

It would be better to think of it in terms of what rating gives you a color-corrected image that you like. What matters is the final results, not how the RAW file comes up automatically.
Click to expand...

This makes sense. In testing negative over the years, sometimes underexposing, then printing up, will yield a color saturation that is more desirable than rating it as reccomended.

Because I am at the "nuts and bolts" testing stage with RED, I am really trying to discover the tonal range. I am used to seeing the waveform tell me that the image is clipped and adjusting accordingly. Now I know, this is not the case with RED.

Dylan Macleod, CSC
Cinematographer
Toronto, Canada
www.dylanmacleod.com
 
You must log in or register to reply here.
Back
Top