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 :)

#1 Problem...

...So in Build 16, we wanted to align everything (as much as possible). Now... they all line up. What looks right on your LCD, is what you see when you open your file in RED Alert! (and REDCINE)...

Jim

So Build 16 is already supported by REDCINE and RED ALERT ? !!!
 
Now that BUILD 16 ALPHA has been out for a week or so is there a guesstimate when some of us might be able to expect to get a beta build? Would be helpful for those of us obsessively checking the site to know if it the wait time is a few days, a few hours or a few weeks off.
 
studiodrome, check post #27 of this thread.
 
My crystal ball has been pretty cloudy... missing dates seems to be what we are best at.
Jim

You got that right, Jim. If it were up to a camera company other than yours, the date when a product like the RED One would be brought to market would definitely be a few years into the future.

In that light, you REALLY have missed the date by delivering it now. Shame on you!

:wink:

PS: If your next camera is to be called RED "Epic", I believe the RED One should be known as RED "Catalyst."
 
I heard rumors that build 16 will have better low light capability and that it will have the ability to control when the fan kicks in for those crews using it in multiple camera shoots and one camera needs to reload while the other keeps shooting. How are these updates coming along?
 
Please stop confusing people, simply explain it! ISO,RAWdata,histogram,"False Color"

Please stop confusing people, simply explain it! ISO,RAWdata,histogram,"False Color"

with Build 15 and before... was missed exposure. We have seen so much footage that had to be fixed in post from this that it motivated us to fix this in Build 16. ...
Those that used "False Color" and got to know the system have had the best results. ...
Jim

Since weeks i read the confused posts about ISO Settings, RAW Data etc. For sure the questions are often not exact, but the answers boost the confusion. This is my first post, i work with a RED One since 3 month, i studied photography and computer sciences, and it would be a pleasure to get answers on my simple questions.

1. I think you still confuse people, when you now give a range of ISO 320-500. There is in truth only one ISO that is the ISO of the built in chip, that has exactly one gradation at the sensitivity (ISO) the sensor (chip) has. So set the ISO to 320 (i hope this is definitely the true sensitivity) and the rest should and can be done in the post processing (yes, there are possibilities to change the gain for a sensor, but it is not done here). YES/NO ?

2. The RED One (Build 15) confused people when they changed ISO, cause every information about clipping, histogram (including "False Color") shown on the display with an ISO set to other than 320 is wrong. Wrong, cause everything shown on the display does not match to the RAW data. YES/NO?

I think everybody will agree, the most important thing is not the correct brightness of the image on the display, the most important thing is, to know what the RAW data will be like. It is the RAW data that will be used for the postprocessing, nothing else. So it is perfect and the best thing ever to work with RED, cause with "False Color" and the histogram set RED One to ISO 320, you know exactly what you get for the postprocessing, this gives a boost of information i can work with on the film set, get the light perfect for using the gradation of the sensor perfect.

3. So ISO can be used to change the look on the display, but nothing else. To get the correct information about the clipping on RAW data, correct histogram information, correct "False Color" information, ISO 320 has to be set. The most important thing is to know exactly what will be the RAW data like. YES/NO?

An other "ISO" (= "brightness") can and has to be set up at postproduction, then with better results, cause you know exactly (with the information given by the histogram and "False Color" when you shoot the pictures) what RAW data is available and what it will look like when contrast and brightness will be changed in the postprocessing.

4. The change that will be done with Built 16 should be, that only the image on the display will get brighter or darker when you change the ISO settings, but the histogram or "False Color" will and should not change, cause the RAW data also does not change and the clipping and "False Color" information has to (for sure) correspond with the RAW data. You need to know what the RAW data looks like, not what the histogram of the brighter image on the display looks like.

I posted this content to help other people understand what is going on and how the workflow should be, that it is most important to be able to read the histogram and "False Color" to get best RAW data, and not to play with ISO settings to change the metadata to get a e.g. brighter image on the display. I hope i could help people to understand what is going on and reduced the confusion. Everything should be clear now.

Sam
 
The native ISO is 320. But you need to know you can push it to a higher ASA so that you can figure it into your exposure. If your ASA is "", then you are allowed to do "" with your f/stop and shutter speed.

Jim has said several times that the native ASA of the camera was ASA 320 - ASA 500 can look good as well, but ASA 320 is the native. Just like if you were to shoot on film you can push film and you can push digital. If you were to push film 2 stops in processing, then on the shoot you'd pretend like your film was 2 stops more sensitive. So then you'd figure your film at that ISO and then figure out what you have to work with for your F/Stop and shutter speed.

Build 15 Jim said that the monitor didn't match what was recorded - that the monitor was darker than what was really recorded. Build 16 made it so the monitor reflected what was being recorded. Knowing the monitor would be a bit darker, they encouraged people to help judge exposure not by the monitor brightness, but by the false color reader and by the histogram. Now with build 16, the monitor shows the correct brightness, and you still have your false color and histogram at your disposal for tools to help confirm and shoot with confidence.
 
I think the simple way to figure this is not to confuse yourself by thinking of all the ASA's.

Red One is ASA 320. Shoot it like that. But if you need to push it (like you'd do with film) then you have some room to do that.
 
Red One is ASA 320. Shoot it like that.

For me - being a computer guy having zero experience with film - what you are saying here has been what worked best. I simply light for the histogram I see on the LCD (camera at ISO 320) because it tells me how good I'm using the CMOS.

1) If I have too much light and see clipping on the histogram, I can always reduce it with ND filters and closing down the f-stop to the point I get the desired depth-of-field.

2) If I have too little light, I have to light the scene until I like the look, get a good coverage of the histogram and open up the f-stop as much as possible and still getting some depth-of-field.

2a) If I cannot get enough light into the scene I have to live with a sub-optimal coverage of the histogram (and therefore the CMOS) - and that will get me noise when I raise the levels up in post.

To change the ISO to something else than 320 in the camera was deadly until now, because it made the histogram display incorrect information. But even in the future, I will never change that ISO to something else than 320 - simply because it would break my way of working. I am not thinking film, I'm thinking RAW data and filling that ISO 320 histogram as good as possible.


But if you need to push it (like you'd do with film) then you have some room to do that.

Could you confirm if I get the term "push" right. For me, push means my 2a) from above: I cannot get enough light into the scene, so I have to live with only part of the histogram being populated - which means I will have to crank up the levels in post - which then means I will get more noise.

I always thought you could "push" film because it is analog. So it is forgiving if you do not hit the real film ISO. But you cannot "push" something digital - like a CMOS sensor. So until I am missing something, introducing the term of "push" for the RED is misleading - it should be called "utilizing headroom that was calculated in or crank it up in post".


If it was just for me, I would remove any ISO settings from the camera, and call it "Monitoring Brightness Boost" - that simply allows me to still see something on the monitor when the scene is quite dark.

That will not happen, I know. But at least I hope Build 16 will never ever change the histogram again and let it simply display what is recorded by the sensor - regardless of any other settings.

My 0.02c
Mike
 
No confusing, answers to helb other people understanding

No confusing, answers to helb other people understanding

Build 15 Jim said that the monitor didn't match what was recorded - that the monitor was darker than what was really recorded.

I think that still continues confusing people.
The monitor is darker if (and only if) the ISO is set lower than 320. Lower ISO means less sensitivity and so the the monitor shows a darker image, but for sure as explained above, RAW data is not changed.
So if ISO is set higher than ISO 320 the image on the monitor will be brighter than the RAW data.

Just like if you were to shoot on film you can push film and you can push digital.
Yes, of course, you can push digital for stops in Photoshop, RED cine, or Premiere, but not the transistors on the sensor. You change the material of film or the development process to change gradation or sensitivity, but you do not change the sensor, neither you exchange the transistors on it nor you have a different sensor for other ISOs. So you always work on the RAW data and push stops (if needed) after the shot in the postprocessing and need to know what the RAW data looks like, where exactly is too less light or too much light for the sensor.
 
Could you confirm if I get the term "push" right. For me, push means my 2a) from above: I cannot get enough light into the scene, so I have to live with only part of the histogram being populated - which means I will have to crank up the levels in post - which then means I will get more noise.

Yes that is correct. Usually when you'd shoot film, when you do that it would increase your contrast. If you underexpose and then process longer. With film, your exposure determines your black level and your processing determines your white level. So if you under expose, that will crush your blacks and most likely pull your midtones down too. Then when you go to process, if you process longer, then your highlight can be pulled up midtones if you process normal, or if you push process, then you can pull your whites back up to a normal white level.

The same idea still applies to digital. Only the difference is, film dies quicker on the under exposure, digital dies quicker on the over exposure.

I always thought you could "push" film because it is analog. So it is forgiving if you do not hit the real film ISO. But you cannot "push" something digital - like a CMOS sensor. So until I am missing something, introducing the term of "push" for the RED is misleading - it should be called "utilizing headroom that was calculated in or crank it up in post".

Yes, film is more forgiving in regards to holding more information when it is over exposed.

I understand what you are saying about not calling it a push. Because technically that is correct. But we put labels on things to understand them, and if push is the label that works for the DP's who've used film for years, or those of us who are used to thinking of that concept, why not call it a push? If that is what makes things click for understanding things - especially if you will get the same result and it's just a matter of a label. Should we still say CUT, Print that!, Speed? on a movie set that is shot digital? Would it be technically correct no, but the thought process still applies.
 
To further my last post to Mike

I understand completely what you are saying about calling something an ISO as opposed to gain. I worked at a TV station for a number of years, and also shot still photography as hobby on film.

On the DVCAM, in a field shoot, I bring along a calibrated monitor, judge my colors and see if it's exposed right. I then iris up and down to get the exposure I want. Then as a last resort, up the gain. I use a light meter to get my contrast ratios the way I want. Ok, I want my key at "" foot candles, and then I want my fill at "" foot candles. In that regards you don't think so much as balancing an equation, you look at your histograms and see what looks good.

Film is an artistic equation. Because back when they didn't have that instant feedback of a monitor or histograms, there was a system built so a day or two later when they saw their prints, they'd know with confidence when they shot them, they'd turn out right.

So say I am shooting with ASA 400 film. I have an f/stop of f/8 and a shutter speed of 1/125. Let's say that is the proper exposure. Then I decide I want to have a shallower depth of field. So that means I need a larger aperture. So I open up 2 stops. So now I am shooting at f/4. Now I am 2 stops over exposed. So I know I need to eat 2 stops somewhere in this equation. Pull processing I've decided isn't an option. I like my film stock and it's characteristics, so I'm stuck with that, that can't change. I've already chosen my aperture, and I don't want to change my motion flow with the shutter speed. So my only option is to put in 2 stops worth neutral density filters. My equation is now balanced again.

In digital 6dB of gain = 1 stop? I think? It's been a long time. You could go about talking "add 1 stop of gain" etc. If that's what works for you, then great! But if you are working on a team who doesn't think that way natively, who thinks in ISO's rather than dB's, a road block for them would be trying to conceive that what a stop is in dB.
 
One more thing to help out:

Stop being a unit of measurement.

There is a 1 "stop" difference between each of these states that is completely equal. There is reciprocity between them.

Shutter Speeds
Lets light in longer (more exposure time)1/30th - 1/60th - 1/125th - 1/250th - 1/500th Lets light in shorter (less exposure time)

F/stop
Lets more light in f1.4 - f2.8 - f4 - f5.6 - f8 - f16 - f32 - f64 Lets less light in

ISO
Less sensitive (more exposure time needed) 50 - 100 - 200 - 400 - 800 - 1600 - 3200 More sensitive (less exposure time needed)
 
Hey Jim,
Just a quick note. We are producing a documentary on your first Oakley Factory Pilot Stu Thomsen! We are shooting from June 28-July8th and Stu would love to have you in the film if you are interested in being interviewed.
our latest film is @ www.JoeKidonaStingray.com
Steve at Oakley is involved, but we would love an interview too.. Please let me know.

Thanks!

John Swarr
Bang Pictures, inc
john@bangpictures.com

Found this... a Bob Haro drawing.

Jim

9_stu.jpg
 
True ISO - Build 15

True ISO - Build 15

In pre-16 builds, when you had correct exposure (really correct), the monitor looked darker than it should have. Many got to understand this and adjust for it. If not, the tendency was to open up the exposure until the mids on the monitor looked right. That led to over-exposure. Conversely, some users got comfortable with the "pull it up out of the dirt" concept, thinking that erring on the side of under-exposure could always be saved in post with no chance of clipping. While a little of this was OK, a lot was not (as with most things...). The disparity between an exposure meter, the image on the LCD, the histogram and "Stop Lights" made for tricky business. Those that used "False Color" and got to know the system have had the best results.
Jim

So, I'm three weeks into a feature with the camera, after only a small amount of prior experience with it. This and other quotes about the mismatch among metering tools is causing me some concern. I've seen some dailies, which all look good, but still I want to make sure I'm doing the right thing here.

I have been using the 320 ASA setting for 90% of my shooting, and using the RGB histogram as my primary exposure tool, trying to spread them as well as possible without hitting the stop light indicators. (For low key scenes that I know won't clip, I've been using 160 ASA to give the people at video village an image that looks a bit more like I expect it to after color correction.)

As I said, this method appears to be working for me, but I need to make sure that the histogram @ 320 is giving me accurate information. I've read through the forums as best I can, but I came onto this project late, and have had to start shooting with only minimal prep.

The other thing that has concerned me is that during my camera test day, using false color, the camera set at 320ASA seems to respond more like a 125-160ASA system. That is to say, when spot metered, a gray card turned green in false color only when the aperture was opened to a setting corresponding to an ASA in that range rather than at 320 ASA.

I apologize if this has been discussed clearly somewhere that I haven't been able to find, but any further info would be greatly appreciated.

Thanks!

Chris Freilich
Cinematographer
Currently shooting RED feature "Achchamundu! Achchamundu!"
 
Chris, sounds like you have it about right.

If you're setting exposure based on the histogram at ISO320 you're doing the best possible service to your image and shouldn't worry. Build 16 is going to really revamp this anyway so all is good!

HTH

Paul
 
Back
Top