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 :)

ASA - Exposure

Tom Akinleminu

New member
Joined
Aug 11, 2008
Messages
4
Reaction score
0
Points
0
I read somewhere that the "mystery"-sensor has an ASA-sensibility equivalent to something like 320 ASA.
So I plan to adjust my light meter to 320, for next week's shot with a RED camera. The possibilities of shifting the ASA-values to 500, 1000, 4000 or even 8000 ASA is confusing to me. It is kind of tempting to set the camera to 8000 ASA and shoot my night scenes with no lamps, but it also seems fishy..
does anyone have experience with outdoor night scenes with RED, and light and postpro? I mean, if I put the camera to 500 or 1000 ASA, am I really shooting 500 or 1000 ASA, or is it only pretending to be 500-1000, but in fact remains 320 heavily underexposed?

thank you!
tom

ps: and is there a RED-manual / operational guide available for download somewhere?
 
if I put the camera to 500 or 1000 ASA, am I really shooting 500 or 1000 ASA, or is it only pretending to be 500-1000, but in fact remains 320 heavily underexposed?

ISO is just metadata so the more you push it the more you're underexposing. The real question is how far can you underexpose and still get an image you like out of RedCine or RedAlert?

Those apps have noise reduction built in to some degree so they are part of the equation. You need to shoot a test and then render all the way out at different ISO's.

You need some kind of real exposure, the camera doesn't see in the dark.
 
Hi Tom - welcome to the boards.

There is an enormous volume of discussion on exposure here - and 'correct' methods and parameters have changed over time with the different builds. So get in touch with your Red Technician and s/he should be able to word you up.

In the meantime, here's a couple of quick pointers from Jannard:
http://www.reduser.net/forum/showthread.php?t=14461
http://www.reduser.net/forum/showthread.php?t=14760

I also found MacGregor's investigations helpful (although the actual tests relate to previous builds, much of the info is still relevant):
http://www.reduser.net/forum/showthread.php?t=12863

Be a bit careful of the user comments in this thread - read the first post and last post before the rest of it, that should get you properly aligned.

I haven't been around the forum for the last month or two so can't say if there's more recent information. I assume that there is, would be grateful if others could point me in the right direction.

RED TECH RED TECH RED TECH is your best and only friend. And tests to confirm that he ain't a lying bastard is a good idea.
 
experience so far, excluding Build 16:

the camera is about 320ASA. Don't know how the software can affect true sensitivity.
That means if you rate the sensor up to 8000ASA it will be a digital enhancement in the end. Think about it. Doesn't alter the sensor. It can only capture the light it captures.
 
Something to keep in mind is that all electronic sensor systems and film stocks have their own sensitivity range, represented by this ISO / ASA value. Typical "video" cameras are no different. RED is not a "video" camera in a typical sense because it is 35mm, it is more of a "cinema" camera. But it also records data as the unprocessed RAW info from the sensor. This leaves the actual exposure level in terms of setting the mid grey point and high and low ranges (and apparent ISO / ASA value) as something to be done in post, when "developing" your footage.

In a typical "video" camera that does not record RAW data, the image is processed on-board. For example, a Sony F900 may have a native ASA rating of its sensor of 250 for example. I don't know what it truly is off the top of my head, but let's just say that hypothetically it's ASA 250. When you set ASA level on that camera, you are affecting the way the camera processes and sets mid points for the internal gamma and exposure curve that is applied to the sensor data. You are not actually affecting the physical ASA or sensitivity of the sensor. But once you set this value and record your footage, it's encoded to HDCAM format that way, no going back, the original sensor data is gone.

With RED, what you have when you take your mag away from the camera is the RAW sensor data. No processing has been done on that data, only compression. So no setting of levels, no curves applied, no ASA modifications, no sharpening, etc.. All of this is done in post when you "develop" the footage and you choose your exposure / ASA level. You choose your amount of sharpening, what color space to assign the RAW color data into, etc.. Leaving the ASA as a "soft" value that gets set in post is just one inherent trait of shooting RAW and that translates to one more level of freedom that shooting RAW gives the artist.
 
thanks + new question

thanks + new question

hey all, thank you very much for your very helpful replies.
now i got a second question (might belong to another thread/forum/place, but what the hey): i shot some short shots with the red, just in the rental house, to practice getting the stuff through the various RED-applications and into Final Cut. I have a MacBook (not a MacBook Pro). We will not be editing on this silly machine, don't worry, but we had planned on using it on the set. RedRushes works fine, Red Alert also seems to work (meaning i get an image and i can work on it and render it, but its all very slow). RedCine doesn't work! I get no image. What's up? I'm reckoning that my MacBook (Processor 2GHz Intel Core Duo, Memory 1GB 667 MHz DDR2 SDRAM) is too slow or small or something, but how come the other applications work and RedCine doesn't? RedCine is the one I was planning on using for exporting halfway colorcorrected footage for editing in FinalCut.
Once again, I thank you all in advance.
tom
 
1280x1024?

1280x1024?

hey all, thank you very much for your very helpful replies.
now i got a second question (might belong to another thread/forum/place, but what the hey): i shot some short shots with the red, just in the rental house, to practice getting the stuff through the various RED-applications and into Final Cut. I have a MacBook (not a MacBook Pro). We will not be editing on this silly machine, don't worry, but we had planned on using it on the set. RedRushes works fine, Red Alert also seems to work (meaning i get an image and i can work on it and render it, but its all very slow). RedCine doesn't work! I get no image. What's up? I'm reckoning that my MacBook (Processor 2GHz Intel Core Duo, Memory 1GB 667 MHz DDR2 SDRAM) is too slow or small or something, but how come the other applications work and RedCine doesn't? RedCine is the one I was planning on using for exporting halfway colorcorrected footage for editing in FinalCut.
Once again, I thank you all in advance.
tom

I loaded REDCINE on my brother's computer to convert some R3D files into 4K REDLog TIF, but the display was messed up at 1024x768, so I changed the desktop to 1280x1024 and the display worked. The display of the preview motion was also a problem unless you reset the display mode to the lowest resolution, seems like the program needs a little more work to make sure that it does an init of all values, anyway it did work after changing the resolution to a higher one and fiddeling with the display settings each time it is run.

The TIF output converted just fine for use in my edit system, and the results look quite good if the color matrix, levels, gamma, and curves are adjusted right for each shot.

About the speed, I think you can get away with EI 800 if you shoot 4K edit the uncompressed TIF files to make CC frames, then reduce the 2K frames the right way to have frames for editing.

In a film recorder that uses full aperture you will have less than 1000 lines in the image if you shoot 2:1, because the exposed part is not 2048 wide due to the soundtrack area.

I think people should use the abbreviation EI (Exposure Index) rather than ASA now for Digital Cameras, the old ASA standard had built in overexposure and was a standard for film stock S shaped curve, no digital camera can conform to the way film stocks were ASA rated. Film stocks are now EI (or ISO) rated since the speeds no longer conform to the way the ASA was measured, that is no underexposure safety, so a old rating of ASA 100 becomes a new rating of about EI 200 for the same film, makes the film sound faster but you just get thin negatives.

Here is a link from Kodak about EI vs. ISO,

http://www.kodak.com/global/en/professional/support/techPubs/cis185/cis185.jhtml
 
f/2.0 @800 @ 4K or f/1.2@400@ 2K?

f/2.0 @800 @ 4K or f/1.2@400@ 2K?

Can't shoot 4K, only got S16mm lenses. T1.3. Could get 35mm lenses for than one night, but they ain't Highspeed..


Humm, what would look better f/2.0@800@4K>2K or f/1.2@320@2K?

From a sharpness and not having the lens soft, the 4K would win, the 2K sensor is smaller, so would the 4K down res to 2K from 800 have less noise than the 2K at 320? Maybe? REDCODE acts as a noise filter, so noise artifacts show up at different brightness levels, it would depend on your lighting.

You might shoot a test before, the main factor is how you reduce the 4K down to 2K, if you use bad software the noise will go up rathen than down.

I am sure there are some lenses around, maybe someone here or CML could drop by with some, not rent, just hang out while you use their lenses?

If you get a DSLR mount you can use SLR lenses that do not cost much, and are not too much different in resolution (for 4K) by the time the images makes its way through the OLPF and micro lens array... The OLPF resoultion is 1/2 in the 2K image, so a 4K OLPF image down res to 2K will be higher resolution than shooting a 2K OLPF image, give or take.
 
Back
Top