Shooting in log?

[Stu Maschwitz]

The Rec709 space has no inherent limitation—as I showed, you can freely underexpose a Rec709 image. Then it's all about a) whether the noise and quantizing start to bum you out in the shadows and b) whether the dark image you're looking at makes any creative sense to you while you're making it.

The limitation I discussed in my post is what happens when you expose per the standard Rec709 rules of thumb.

-Stu

 

[donatello b]

"IF it can handle being underexposed! I must say, Paul pretty much had the fear of god put in him from the camera's owner about underexposing."

i'm the owner of the camera ... i don't recall putting the fear of god about underexposing? over exposing YES ( however, if shooting 2k then i would put fear on not under exposing 2k on RED103 ) .. in general, i recommend do not use lower ISO then 320 , do not over expose, choose where you want to clip ... also they did quick camera test the day before the shoot ( don't know if they reviewed the test)

i was at the shoot and took some meter reading notes ... i don't have the notes in front of me as i'm out of town ..
from what i recall ( Paul would be more accurate) .. the test was ISO 320 , shutter 1/40 ... when i took incident readings with my minolta V meter ( calibrated 10/07 at light metric in LA)- in general i found the T stop they were using was within 1/3 of stop i was reading ( as in they were setting T stop 1/3 over - also i was using 1/48 which will be some of that difference) ... on some shots i had camera take a Red spot reading on the white walls - at this time i don't remember the exact reading on the shot shown - but in general the white wall average was in the low 50'sIRE ( say the white area behind her hair ) .... the windows & frame area were blown out many stops ....

 

[Stu Maschwitz]

In addition you can always "crush" the noise in the shadows using a sigmoidal function of the following form...

You can't crush them if all your skin tones and gray cards are there!

But I'm totally gonna go play with that sigmoidal function anyway, looks fun.

-Stu

 

[Stu Maschwitz]

i'm the owner of the camera ... i don't recall putting the fear of god about underexposing?

Sorry Donatello, didn't mean to misattribute!

-Stu

 

[Dj Joofa]

The Rec709 space has no inherent limitation—as I showed, you can freely underexpose a Rec709 image. Then it's all about a) whether the noise and quantizing start to bum you out in the shadows and b) whether the dark image you're looking at makes any creative sense to you while you're making it.

The limitation I discussed in my post is what happens when you expose per the standard Rec709 rules of thumb.

-Stu

I think then you might rework your film out LUTs by altering the white point from whatever is currently being used now. Adjusting your film out LUT white point so that it corresponds to your Rec 709 display shall correspondingly drive your film out to the right setting.

Kindly try the sigmoidal functions to see if you can reduce noise in shadows without effecting highlights. I have used them to much advantage.

 

[Dj Joofa]

You can't crush them if all your skin tones and gray cards are there!

But I'm totally gonna go play with that sigmoidal function anyway, looks fun.

-Stu

Thats why I suggested that you play around with the "a" parameter of the sigmoidal function to experiment if you can retain your skin tones. I am not saying that you would be able to, but lets see what happens.

 

[SF Geek]

When you guys get all of this sorted, please tell me what it all means because I sure have the fear of underexposure in me. You can use my camera if you'd like Stu. I'm always up for learning more things.

 

[donatello b]

i haven't come up with a conclusion on where to set zone V (18% grey) yet.
i don't have a monitor that i would trust to make the call ...
i have been looking at RGB readings ... i usually take about 2 weeks of testing ( shooting real images ) and then another week studying the results.. and at some point i'll do a film out & may spend some $$ on a top CC system ...
right now i'm not going to take the 3 weeks for test because IMO firmware 15 will void my test and then i have another 3 weeks and i have to say i'm a bit tired of all the past testing ... so i'll just make images till mar 15 using what i know so far ...

from what i recall on the Moses/Paul shoot zone V varied from 36-43ire ( when i checked Red spot meter )
in general i'm looking to find where to place zone V and then be able to have a +4 and -5 of good image ...i hope to be able to squeeze out a little more on brights (+1 ) with a IF i do X i live with Y ... and maybe another 1- 1 1/2 on bottom end with "if" (acceptable repeatable noise) ...

 

[Stu Maschwitz]

i haven't come up with a conclusion on where to set zone V (18% grey) yet.
i don't have a monitor that i would trust to make the call ...

Well, no monitor that you attach to the camera will help you make that call. Not until we get new monitoring options anyway.

-Stu

 

[Stu Maschwitz]

Here's a new post to give you some basis for any future tests you might do while Rec709 monitoring is our only option:

http://prolost.blogspot.com/2008/02/digital-cinema-dynamic-range-short.html

This will be my recommended bracketing range to Paul when he shoots round two of the test next week.

I will be providing the fancy grading system.

-Stu

 

[reality]

Whatcha smokin' Stu?

I used to hold you in very high regards but now I'm not so sure.

 

[mezmo]

It's that high grade 'reality' weed that kicks in just when you start
to use this camera for filmout.
Mezmo

 

[David Mullen ASC]

Your blog post says:
"Shooting with RED One and monitoring Rec709, if you put 18% gray at 45% IRE, you'll capture 2 stops over 18% gray, plus some headroom."

But the Rec 709 output is controlled by what ASA rating you set the camera to, even though that's only metadata. So if you set the camera to 1000 ASA, for example, you'd still find on Rec 709 monitoring signal the same clip point, but on the RAW files, you would have underexposed the footage and thus gained, let's say, a stop and a half or two stops more information in the highlights, at the expense of information in the shadows. Not to mention more noise overall once you correct to normal brightness.

This is what I discovered as part of the LART testing, that the ASA settings essentially just boost overall 720P signal as you rate the camera faster, but of course, has no affect on the RAW recording other than the fact that you are underexposing more and more as you rate the camera faster. I expected when I rated the camera faster at 1000 or 2000 ASA, that I'd see more highlight detail in the Rec 709 signal, but I didn't -- but in the RAW files, there as a lot more highlight detail in the shots rated at higher ASA's because they were underexposed more.

This is why using waveforms and false color as an exposure tool only works then the ASA rating selected is close to the true sensitivity of the system so that your 720P signal is in the ballbark of what's getting recorded in RAW.

 

[donatello b]

"my recommended bracketing range to Paul when he shoots round two of the test next week"

define next week ?
monday , wednesday , saturday ?
you going to leave us hanging till then !

SFGeek has offered his camera , i have 103 .. one of us can either be at Paul's or your place Monday AM ( or we shoot at both places) ... we have the camera's available , windows are easy to find ... you must have some subjects around the office ? or we just take a walk on the campus to find a subject = we could see results by monday eve ....

 

[Stu Maschwitz]

But the Rec 709 output is controlled by what ASA rating you set the camera to, even though that's only metadata. So if you set the camera to 1000 ASA, for example, you'd still find on Rec 709 monitoring signal the same clip point, but on the RAW files, you would have underexposed the footage and thus gained, let's say, a stop and a half or two stops more information in the highlights, at the expense of information in the shadows. Not to mention more noise overall once you correct to normal brightness.

Aha! I was wondering abut that. Very interesting indeed.

I don't think I'd be comfortable using that mechanism for encouraging underexposure though, because you would lose the ability to see exactly where you are clipping, no?

Thanks for the revelation David,

-Stu

 

[Ariana]

That's what I understood from Deanan's comments.

I am wondering why you're advising people on exposure without fully understanding the camera yourself. I haven't had alot of time on the camera and was hoping for more clarity on exposure for RED.

 

[Stu Maschwitz]

I am wondering why you're advising people on exposure without fully understanding the camera yourself. I haven't had alot of time on the camera and was hoping for more clarity on exposure for RED.

I'm in the same boat, and this conversation is helping to provide that clarity. I'm sorry if you see it as me dispensing uninformed advice. I intended my blog posts to be nothing more than an opening up of my personal experimentation and thought process about this stuff. I'd hoped I made it pretty clear that this was the case, but again, my apologies if I somehow inadvertently claimed to be an expert on the camera.

But I am grateful for the conversations that have sprouted up around my recent postings and find them incredibly valuable.

-Stu

 

[Ariana]

I know myself and others value the information on Prolost. Thank you for blogging. Because people have alot of respect for your blog, they expect (at least I do) that the info is very well researched and faultless. Stu, you've set the bar quite high for yourself with your previous blogs.

 

[Stu Maschwitz]

Because people have alot of respect for your blog, they expect (at least I do) that the info is very well researched and faultless...

Thanks for the kind words Ariana. As you are seeing, this research does not always happen behind closed doors!

-Stu

 

[I Bloom]

Aha! I was wondering abut that. Very interesting indeed.

I don't think I'd be comfortable using that mechanism for encouraging underexposure though, because you would lose the ability to see exactly where you are clipping, no?

Thanks for the revelation David,

-Stu

Stu,
Thanks for the great post on Prolost and this great conversation. Really digging into this is what I've been waiting for.

IBloom