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 :)

EPIC Native ISO

If you don't have a very good reason, is in a very controlled studio setting or have very good knowledge of what you do, there is no reason whatsover not to shoot the Epic at 800 or above, as far as I have seen.

David is right, thoigh. Everytqhing is. Alibrated at 320...

But on the Epic, that implies thastuff really get overexposed when they look lile they ate

I also tend to disagree with David on the exposure to therigjt theory on the R1mx. While that is mathematically and theoretically right, the mx, more than the m, breKs and overdrives a bit on the R1 when it approaches clip.

Not so on the Epic...

Shoot 800 and you havethat 1.25 stop highlight prote tion which is allmost enpugh, when the sun suddenly breaks in the middle of a shot.

The real good thing aboit Epiv vs R1 mx, though - is that overexposure looks good on it...
Real beautifull it is..

Lol
But I guess that was not your question...
 
I've been leaving the Epic at 320 iso during daytime exteriors, as this also corresponds to what it looks like when "RAW" is toggled. Have I been in error?
 
I've been leaving the Epic at 320 iso during daytime exteriors, as this also corresponds to what it looks like when "RAW" is toggled. Have I been in error?

You are in no way "wrong".

Otoh, you also have no headroom on a cam tha does not create much noise in the extreme lowlights.

But everthing is still
calibrated at 320, so you are good.."
 
I've been leaving the Epic at 320 iso during daytime exteriors, as this also corresponds to what it looks like when "RAW" is toggled. Have I been in error?

It's just another legitimate way of working. There's nothing to stop us underexposing at ISO320, if we want to protect unexpected/uncontrolled highlights. It'd get the exact same results as using ISO800 to bias towards underexposure.

I quite like setting ISO320 on the Red MX so that I can use the red clipping as a kind of true RAW clipping "zebra" for exposure.
 
I've had beautiful images at 800, broad daylight, no fill for the shadows. I get plenty of detail in the shadows and the highlights look like film.

I'll go to up to 1280 for night exterior with negligible noise for the under exposed elements.
 
This may be a dumb question but is the ISO really Raw? Wouldnt the colorist set the ISO and be able to save highlights that were blown out at 320 or is setting the iso right from the start on camera the only way to do this? Thanks for your response!
 
800ISO is the factory recommended rating. You can veer from that either direction as per your own testing but running ISO equal or lower to "digital negative sensitivity" is questionable at best. The RAW sensor sensitivity does not equal developed sensitivity nor should it.
 
I've been leaving the Epic at 320 iso during daytime exteriors, as this also corresponds to what it looks like when "RAW" is toggled. Have I been in error?

It's not "wrong". You should just know that you will have more range in the highlights if you rate 800, without in my opinion sacrificing anything including increased noise. You should really shoot the greyscale test I described in an earlier post. I've shot this test and graphed the results with both the M and MX sensor and learned a lot from the exercise. Day exteriors benefit greatly from increased range in the highlight tones.
 
Well put Brook.

I haven't landed on where I'm going to rate the Epic yet. I may end up rating the EI higher then 800 (Possibly 1000). Since it records 5k, and the deliverables could be in 4k or smaller, some of the noise that might be there in the 5k image will get hidden in the down sample to 4k or smaller, so I might be able to get away with more.

Here is the way I see the EI of the MX chip- in the Red One MX, for sure, (Still testing with the Epic). The "native" EI of the chip is 320. A simple way to check is to turn the camera to 320 and then switch to RAW view, not much changes. (The RedColor view has it's own LUT which does make things "brighter", but not by much on the Red One MX.) This is even more clear on the Epic, as almost nothing changes. The catch being at least on the Red One MX, the sensor is VERY, VERY, VERY clean and there is a lot of room for under exposure. At EI 320, the sensor is not very balanced in terms of DR. There is MUCH more under exposure latitude then over exposure latitude. By bumping it up to EI 800 the over and under exposure latitude is much more balanced, and is most similar to how I am used to working with my light meter. And since I know the sensor is VERY clean, I can comfortably "underexpose" it and know that I have solid highlight protection so that it rolls of nicely, and I have clean details in the shadows. So even though the "native" EI is 320, the effective working EI for myself ends up being around EI 800.

As the Red One MX has the same chip as the EPIC, my hunch is that it is the same here too. Although I'm still testing and playing around with it to see where I am comfortable rating it. I think I can "get away" with more on the EPIC due to the increase of resolution, and also the processing in camera of the RAW data feels cleaner to be then on the Red One MX (that may just be me) - so I think I can push it harder then the Red One MX if I need to.

Native ISO is a VERY tricky / touchy subject, as too many people just read the headlines and do not bother with the details and implications of what that means. I highly suggest that you do some testing of your own, figure out what your on set working method is, and what your post workflow is going to be, and then develop a rating based off of that. Something that is KEY to working with the Red RAW data is what / how you process it in post. I've seen some MX footage shot at EI 320 that was processed nicely, and EI 800 footage that was processed horribly. But they both were shot correctly. The is the same when working with film. Just because Kodak or Fuji says their film is EI 500, or 250, that doesn't mean that it is EI 500 or EI 250- how you shoot it and then develop it can significantly change that rating depending on your needs. Treat RAW data the same way.

The raw view is misleading and of no real utility on MX
 
I see ISO "only" as an interpretation (LUT) of sensor reading.

I dont think we can say that a sensor has a "native" unique ISO.

As long as you have sufficient sensitivity for a scene, it's OK.

...and, also, as long as the noise level is low enough, then again it's OK.

I'd say "recommended ISO" over "native ISO"

...bur i may be totally off track :)



Antoine

No, I agree with you - ISO is just a preview LUT.
 
Thanks guys, I'm starting with ISO 800 and 5600K as my entry point (as they are defaults in RC-X too). This might change based on given conditions and as I learn how to use the EPIC more efficiently but in the end it all starts making more sense during post when you see whether you really exposed things right or not.

My thanks to both the 320 and the 800 crowd ;)
 
This may be a dumb question but is the ISO really Raw? Wouldnt the colorist set the ISO and be able to save highlights that were blown out at 320 or is setting the iso right from the start on camera the only way to do this? Thanks for your response!

Basically if the RAW file is clipped it is clipped. When a pixel goes to white on a CMOS sensor there is no information to recover. period. Film was a bit different in this regard becuase it protects teh highlights better.

So rating at 800 pretty much ensures that you are not going to clip highlights (unless you want to and there is a given number of pixels in just about any scene that will roll off that way.


For me, (and I will revist this on the EPIC) going back to the R!-M sensor, I found it really performed much better with MORE light and with DAYLIGHT. It was messy in tungsten light and RED for a while reccomended that tungsten be shifted toward teh blue spectrum to better deal with the noisy blue channel that would happen in the R1 M. Then the MX cam out with the advertized rating of 800 ISO (there was talk of rating it at 500 ISO for a while and in the end 800 was the "factory default".

In futher testing with the R1-MX i still rated the MX at 320ISO while watching hihglights and the image very carefully. I found this gave me the widest histograms and allowed me to expose to the right (as RED reccomends as well). In extremely bright outdoor situations You can rate at 500 for about a stop of highlight protection in scenes where it is harder to control light.

So picking a rating is giving you highlight protection. That is it.

I think the sensor produces better images when I rate it at 320ISO then if I want a "brighter image" I wind up the ISO in post. If I rate at ISO 800 and need to "boost" then I am looking to move to ISO 1000 or ISO1280 and that is not a place I want to comfortably go. Rating at 320 lets me know when I am "pumping" the image to 400, 500, or 800 ISO (acceptable amounts of noise for me)

So the rating really comes down to how much noise you personally like but my lightmeter at 320ISO and the camera at 320ISO and the stop seem to match up nicely. Monitor RAW and you are good to go.

There is much less room for error this way because if you blow out, it's gone, whereas if you blow out at 800, it is not gone, so 800ISO is a nice safe rating.

David
 
Okay Gotcha so even ISO as a "RAW" setting its not really raw because if you shoot at 500 and clip I can go into redcinex and bring it down to 320 but once they are clipped theyre clipped the highlights will be lost, so ISO isnt really a raw value if the metadata only contains information from the chosen iso value set on camera.
 
Yes, ISO is really raw - it's just metadata indicating rendering intent. However, based on you, the human operator setting an ISO you will adjust the exposure coming into the camera via aperture or ND and that part cannot be totally undone later. If there is no clipping you may have made the scene too dark and add noise, but you can still brighten it up. If you've over-exposed you may have unnecessarily clipped and see it when you darken it down.

Graeme
 
It's not "wrong". You should just know that you will have more range in the highlights if you rate 800, without in my opinion sacrificing anything including increased noise. You should really shoot the greyscale test I described in an earlier post. I've shot this test and graphed the results with both the M and MX sensor and learned a lot from the exercise. Day exteriors benefit greatly from increased range in the highlight tones.

True!

The headroom will have to be reclaimed in post if needed. RedlogFiolm does a good job for that. Lowering the ISO/FLUT/Exposure settings may be needed. I think ISO is the most logical thing to lower towards 320.
 
I find it good to leave ISO as setting rendering intent, and then if necessary bias the exposure for grading with the FLUT.

Graeme
 
The raw view is misleading and of no real utility on MX

I disagree- it depends on your working method. It may be useless to you. But I have found it very handy on the Red One MX. I have found that when in RAW view all the other meters respond more accurately, and I can feel more comfortable about my light meter readings. But agin- you have ot know your post workflow and how the image is going to be developed. And if RAW view is left on, it can really scare a client if that client knows little about what RAW view really is ...
 
Back
Top