Welcome to our community

Be a part of something great, join today!

Register 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 :)

Wrapping my mind around MX..

J

Jesse Lee Cairnie

Member
Joined
Jan 29, 2010
Messages
16
Reaction score
0
Points
0
I am shooting on the MX for the first time this coming week.. And I have been absorbing so much info from all the great threads here..

But I am still trying to wrap my mind around the MX sensor so that I can translate creatively all the technical benefits of all this overwhelming amount of technology..

Please bear with me as I ramble on some basics.. Feel free to guide me to previous threads on the topic..

As I understand it 800asa is the "sweetspot" as it has the greatest highlight detail to signal noise ratio providing the greatest DR and thus the "thickest negative"..

This is achieved by monitoring and shooting in 800asa in Redcolor and using the exposure tools in camera to prevent highlight and shadow clipping.. Thus light meters get set to 800asa.. Then in Redcine-x processing the footage at full Debayer at 800asa to whichever format desired.. And apply sharpening to taste in post..

If all of this is the case..

To get this result.. I am to treat the MX as 800asa even in Day Ext.. Using ND infront of the lens to adjust shooting Asa.. But always having the camera set to 800asa in Redcolor giving me the "thickest negative?"

Translating this in terms of RAW..

So does this mean that shooting, viewing and metering for RAW at 320asa does not allow the Redcolor science to take advantage of the Highlight detail and DR.. Effectively lowering both and thus having a "thinner negative?"

If Asa is simply metadata.. does this mean that it is the color science in Redcine-x that is determining the Highlight detail and DR?

So in terms of old school film terms.. One is rating "320asa" stock at "800asa" and printing up?

And thus it is the effective "printing up" that gives the MX its strongest image? In a sense the digital version of rating celluloid film 2/3 of a stop over?

Am I getting all this correct?

Cheers!
 
You have one thing backwards, an overexposed negative is "thicker" and an underexposed negative is "thinner" so rating the camera at 320 ASA instead of 800 ASA would be getting more exposure and thus a "thicker" negative... but at the expense of overexposure detail.

800 ASA is a sweet spot between noise and clipping, but that doesn't mean you can't give the image more exposure IF you watch out for clipping, i.e. a scene without bright highlights in it could be exposed more, or use a lower ASA rating, etc.

If you had to "print up" a negative, i.e. use lower printer light values to compensate for a thin negative, then you would have had defacto rated the film faster. Not sure I'd call that "rating a film 2/3 stop over", I'd say "rating a film 2/3 stop faster or higher".

Here's the thing with digital: above the clip point, there is no information, period. But in the noise floor, there is information, it's just not considered usable generally and ultimately it gets crushed into black. So if the overall system (camera, sensor, recording, etc.) is less noisy, then more of that information at the bottom falls into a usable range. So because the M-X image is less noisy and more of the bottom is usable, that allows you to bias the exposure to protect more of the highlights, i.e. rate the camera faster. You can think of that as underexposing the image -- or not... you could also think of it as exposing the image correctly for the most practical dynamic range. It's a bit of semantics, but I don't think Red wants to say that you should underexpose the image for the best results, they are saying that 800 ASA is a good starting point for shooting scenes of typically wide dynamic range and having a decent amount of information in both the highlights and the shadows for later color-correction.

However, if there aren't any hot highlights to worry about, then you might as well "expose to the right" (of the histogram) and thus push more information into the lower noise range.
 
Last edited:
Thanks for posting Jesse. I think this is a very common question among RedUsers.

David, I was thinking about this today actually. How low do you think it is safe to push the ASA when shooting scenes with no hot highlights? Do you ever see any reason to shoot at 100 ASA? Thanks.
 
I think there is a law of diminishing returns when it comes to overexposing a sensor to reduce noise; at some point, the noise is probably as about as low as you'd ever need it to be anyway. So I don't really see a need to go below 400 ASA with the M-X sensor if you need less noise, maybe 250 or 320 ASA would be pushing it but OK. But 100 ASA just seems too dangerous in terms of clipping problems. Plus sometimes I think that overly clean, overexposed digital photography starts to look too plasticy for my tastes.
 
Real Subject Contrast Ratio vs. test charts

Real Subject Contrast Ratio vs. test charts

Clint Lealos said:
How low do you think it is safe to push the ASA when shooting scenes with no hot highlights?
Click to expand...

Every subject has its own contrast ratio, and how much of that subject you want to push over "white clip" depends on how much fill light you have (in order to get the actor's faces out of the noise etc.)

In older color movies the used very heavy fill lighting so that much of the luma (shadow) contast was removed with spot lighting anything dark and flagging off anything bright, that made the colors look brighter since the ratio of luma (shadow) to chroma contrast was leveled with fill lighting (the color film was printed for higher than life contrast, so lighting was used to reduce shadow contrast range).

You need to think less about light meters and gray cards for real world shooting, than where in relationship to mid-tone to the actors faces fall, you cannot open the iris to compensate for an underexposed actors face because you will push other parts of the frame over white clip, you need to set the iris and ND to hold the important highlights, the add fill lighting where it is needed and flag off things that are to light in order to have what is important in the frame look visable on all end use display devices. Since the subject matter can be light to dark, colored or monochrome, your eyes and mind are needed to make judgments about where you want those tones in the finished image, you cannot just "grade it later" if you have the ratios wrong when shooting, unless you plan on using "keys" and "travailing mattes".

The camera ISO and K values just set the monitoring LUT it seems, so to be able to have enough data to work with later it makes sense to set the LUT in the center of the usable range, you can then light within that range and see that you are holding enough between clip and noise to use later, if you need to you can change the ISO to see the highlights or shadows better on the monitor, but you should NOT adjust the iris or ND, just the fill lights or flags as you do not want to upset the exposure for the other parts of the frame that are already adjusted right.
 
Ah yes.. I meant "thicker" and "thinner" in regards to quality of information in the negative.. Which does seem to read a bit backwards.. Whoops..

In regards to ASA.. Would you say shooting Day Ext.. It would be best to shoot at 800 and use ND to best use the highlight detail and DR?

I seem to be stuck on this idea as it seems counter intuitive to shoot high ASA for Day Ext.. But the Highlight detail and DR at 800 seems like a logical thing to do?

Would I be wrong to use that as a conceptual starting point or am I missing something?
 
Jesse Lee Cairnie said:
In regards to ASA.. Would you say shooting Day Ext.. It would be best to shoot at 800 and use ND to best use the highlight detail and DR?

I seem to be stuck on this idea as it seems counter intuitive to shoot high ASA for Day Ext.. But the Highlight detail and DR at 800 seems like a logical thing to do?

Would I be wrong to use that as a conceptual starting point or am I missing something?
Click to expand...

Well, just to make things a bit easier, I shot the M-X Red One at 500 ASA in daytime for "The Chicago Code" pilot, just because of all the ND needed for 800 ASA... it didn't affect my highlight range too much. But yes, get used to using heavy ND filters.
 
I agreed with David.Just finished shooting "The Man With The Iron Fists",all 5 MX rated in 500 ASA in day,night,interior or exterior,even day for night,and i think its my sweet spot for this camera.
BTW the NextLab from FotoKem is awesome,the H.264 dailies in the ipad perfectly match the on set monitors

Regards
Chan Chi Ying
DP HK
 
Everything gets easier to understand if you empty your brain of "film-thinking".

1. Don't ever shoot ISO 100-160. Highlight protection is very limited. The sensor/system doesn't like it at all.

2. Try to stay in the ISO 400-1280 range. ISO 800 is the sweet spot for highlight protection and noise. If you go outside this range, higher ISO (like 2000) is better than going lower (like ISO 250).

Use NDs to keep you in this range when necessary.

This is nothing new to what has already been posted and explained. Just a simple reminder that this is what RED officially recommends for MX.

Jim
 
We have been contemplating removing ISOs under 320 altogether...

Jim
 
Jannard said:
We have been contemplating removing ISOs under 320 altogether...

Jim
Click to expand...

Which I think would have been a very de-confusing move!

Those insisting on blowing their highlights would have the FLUT option, but it would be more obvious that they were on their own.

For post, though, the sub 320 settings are very usefull..
 
I know the EPIC does everything, but just now I thought wouldn't it be
wonderful if at any ISO setting below 320 a small ND filter moved into place
inside the camera. I'm sure the reasons to omit internal ND filters far out-way
the benefits. :)
 
Gunleik, I'm curious about this. I've rarely gotten ISO 250 to work, and anything under that just lowers the whole ceiling, no longer revealing the details in the highlights, just making them grayer. At that point, Levels (lowering the Output White) or even a brightness control could do the same thing.

But, I am always learning, so may have missed something (?).

Phil
 
Phil Bates said:
Gunleik, I'm curious about this. I've rarely gotten ISO 250 to work, and anything under that just lowers the whole ceiling, no longer revealing the details in the highlights, just making them grayer. At that point, Levels (lowering the Output White) or even a brightness control could do the same thing.

But, I am always learning, so may have missed something (?).

Phil
Click to expand...

Agreed that 200-250 generally is a good ballpark pointer as to what you may have to do...

When I am done with the exposure bracketing test, there are some examples where I find it easier to go lower, but those are by no means what you would hope to be normal.

320 seems to be a very good post startingpoint for 90% , independent of gamma...
 
Thank you guys.. I think my mind has been wrapped and I'm feeling pretty good going into this..

Cheers
 
Jannard said:
We have been contemplating removing ISOs under 320 altogether...

Jim
Click to expand...

I think I'll frame this part of the thread and put it up in the office to help guide some people I've worked with.

Seriously, while I can't support removing features, it certainly shows that either some education needs to be done, or some other feature improved.

To what extent is this documented in the manual? Maybe that might be enough.
 
You must log in or register to reply here.
Back
Top