Welcome to our community

Be a part of something great, join today!

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 :)

is this true ? what do you think?

True. All silicon is more sensitive to IR and near red than film used to be (without special chemical additions).

While RED (and others) add some IR filtration in front of the sensor, you can't make that fully ND proof. Any filter that blocks IR and near red perfectly, will also block some visible red. This could lead to problems, with wides in particular, which manifest as greenish corners. To avoid this, all manufacturers only go so far with internal filtration. If you add heavy ND to any silicon-based camera, you should add some IR blocking. IR will make your image go softer, since it doesn't focus on the same plane. Plus, some materials used in clothes (synthetics in particular) are strong IR reflectors.

I have seen black nylon jackets go aubergine color under such conditions, and that can be a color grading nightmare!
 
Uli, I've seen the same, LOL!

But do either of you have first hand knowledge of this specifically with the Alexa? I thought I've heard Arri say that the Alexa has a full strength IR blocking filter in front of the sensor, so no further IR filtration is necessary.


P.S. I found my own answer ;). Arri says the Alexa is good w/ no IR cut upto and including ND .9 (three stops). After that, IR cut should be used.

This answer can be found under 3. Exposure Questions:

http://www.arri.com/camera/digital_cameras/learn/alexa_faq.html?no_cache=1
 
Supports my own observations with Alexa, just like RED. You get away without additional IR filtration up to 0.9 with the current sensor (MX). This is a very simplified explanation though.
 
So what is the explanation behind RED not being as bright at 800iso compared to other cameras? This is something I have noticed even from just light metering a scene. It seems that you have to push the 1/4 tones pretty far up with the mid tones following along to match the brightness of other cameras at the same exposure settings. Although 3/4 tones to shadow seemed to be around where they should be based on metering and comparisons. ??
 
Any log is already an interpretation of linear RAW. S-Log is even darker than REDLogFilm, while C-Log is lighter.
This is a very simplified description, though.
 
Hmmm.... I have to disagree with these common results to some extent. I think the EPIC's biggest fault is not that it's less sensitive or that it doesn't have the same great color reproduction. Both of those statements are untrue. However, they are true within the context of these tests when comparing video right out of the camera. EPIC video, or anything shot with any RED camera, needs a helping hand in post. There is also a fundamentally different approach to how RED interprets its respective color and luma curves at ISO 800 (or any other EI) compared to how ARRI or SONY are doing it. Doesn't make it wrong or right, it's just different. And the beauty of shooting RAW is that we have complete control over how all those properties are interpreted if we don't like it. Comparing color reproduction on a RED camera vs. the same on a camera shooting a baked format is an apple and orange comparison. I do agree though that the default color science on the RED's still isn't as pleasing as what ARRI and Sony are doing out of the box. But in post I can match the look of the Alexa any day and do it all the time. I believe Mark Toia posted some Alexa/EPIC matching samples around here somewhere I think it was him. Even ARRI-RAW doesn't give us a fraction of the level of control we have with RED. That's a double-edge sword in some ways -- it's awesome to have the level of control we do with the R3D workflow, but for the uninitiated it can be overwhelming and it's easy to mess things up until you get the hang of it. I've found that while correcting R3D or tweaking in post, less is usually more. A few very subtle changes are all that are needed to make the clients happy.

The real sticking point I have is with the Tiffen filter claims. I've found otherwise. The Tiffens always get full marks and seem wonderful in my studio/lab tests. In real-world shooting, not so much. When I'm in the desert in the middle of July, with large rocks or sand dunes radiating IR as if they were some form of energy beacons, I consistently have better results with the Formatt and Pancro filters. The Tiffens do seem to hold up better at low to moderate IR levels and they have more neutral or clear color properties vs the Formatt IRND's, which are a bit too purple. Simple WB corrects the color cast, but in extreme IR environments, I've found the Tiffens fall short. I would say that 65% or more of my shooting is outdoor, EFP, lifestyle type shooting for corporate clients. IR filtration and heavy ND are a way of life for me, and much of it is in very intense IR and UV, high-alpine environments, desert heat, etc..

Maybe what I'm trying to say is the Tiffen filters look great (as if they're not even there) while blocking relatively low amounts of IR. The Formatt and Pancro filters do a much better job at blocking more IR, especially in intense environments, but are not as color-neutral.
 
Jeff Kilgroe said:
Hmmm.... I have to disagree with these common results to some extent. I think the EPIC's biggest fault is not that it's less sensitive or that it doesn't have the same great color reproduction. Both of those statements are untrue. However, they are true within the context of these tests when comparing video right out of the camera. EPIC video, or anything shot with any RED camera, needs a helping hand in post. There is also a fundamentally different approach to how RED interprets its respective color and luma curves at ISO 800 (or any other EI) compared to how ARRI or SONY are doing it. Doesn't make it wrong or right, it's just different.
Click to expand...

It is different.
It is also frustrating at times. Particularly when I try to explain to a DP why the camera is reading completely different than his light meter. Not something I have to do when shooting Alexa. I'm not kicking the EPIC in the shins here. Just saying it would be nice if RED cameras reacted at 800 the same way light meters and other cameras react at 800.
 
Uli Plank said:
We tested Epic, Alexa and F3 against each other under identical conditions at the Ludwigsburg Film Academy last year. The complete results were published in Digital Production (Munich).

Very short excerpt:
The Epic is the least sensitive and has the highest resolution.
F3 and Alexa are more sensitive. Alexa and Epic have much better colors then F3.

Yes, the Tiffen Hot Mirrors are the best if you shoot with 0.9 ND or more.

Yes, Abel Cine is a very competent company.

Yes, we own one Epic and two Scarlets instead of one Alexa. And some Hot Mirrors.
Click to expand...


Thanks for the input Uli. Is there a way we can check out your published findings? Sounds like a fascinating read!
 
They were published by "Digital Production" (Munich) in German in their 01/2012 issue. It's in German and still available as PDF.
 
The RED cameras are darker than other cameras when set to the same exposure index. This is not rocket science, it is a simple and easily repeatable observation. It is true for the RAW files as well. At our AbelCine EXPO all I did was point it out.

Likewise, the RED cameras see colors differently than other cameras. Not "better" or "worse" but different. This is particularly true for IR issues, as I was attempting to make clear with my demonstration. The point was that each camera had particular needs, which is why one cannot ask "Which is the best IR filter?" but instead "Which is the best IR filter for my camera?" or "Which is the best IR filter for most uses in most cameras, if I were only to buy one set of filters?" These are different questions and the ones I was trying to provide information on with the EXPO demonstration.

We'll have an edited version of this comparison test on our blog in a few days. In the meantime the complete EXPO can be viewed at this link:

http://new.livestream.com/abelcine/FiltersExpo
 
Jeff Kilgroe said:
Hmmm.... I have to disagree with these common results to some extent. I think the EPIC's biggest fault is not that it's less sensitive or that it doesn't have the same great color reproduction. Both of those statements are untrue. However, they are true within the context of these tests when comparing video right out of the camera. EPIC video, or anything shot with any RED camera, needs a helping hand in post. There is also a fundamentally different approach to how RED interprets its respective color and luma curves at ISO 800 (or any other EI) compared to how ARRI or SONY are doing it. Doesn't make it wrong or right, it's just different. And the beauty of shooting RAW is that we have complete control over how all those properties are interpreted if we don't like it. Comparing color reproduction on a RED camera vs. the same on a camera shooting a baked format is an apple and orange comparison. I do agree though that the default color science on the RED's still isn't as pleasing as what ARRI and Sony are doing out of the box. But in post I can match the look of the Alexa any day and do it all the time. I believe Mark Toia posted some Alexa/EPIC matching samples around here somewhere I think it was him. Even ARRI-RAW doesn't give us a fraction of the level of control we have with RED. That's a double-edge sword in some ways -- it's awesome to have the level of control we do with the R3D workflow, but for the uninitiated it can be overwhelming and it's easy to mess things up until you get the hang of it. I've found that while correcting R3D or tweaking in post, less is usually more. A few very subtle changes are all that are needed to make the clients happy.

The real sticking point I have is with the Tiffen filter claims. I've found otherwise. The Tiffens always get full marks and seem wonderful in my studio/lab tests. In real-world shooting, not so much. When I'm in the desert in the middle of July, with large rocks or sand dunes radiating IR as if they were some form of energy beacons, I consistently have better results with the Formatt and Pancro filters. The Tiffens do seem to hold up better at low to moderate IR levels and they have more neutral or clear color properties vs the Formatt IRND's, which are a bit too purple. Simple WB corrects the color cast, but in extreme IR environments, I've found the Tiffens fall short. I would say that 65% or more of my shooting is outdoor, EFP, lifestyle type shooting for corporate clients. IR filtration and heavy ND are a way of life for me, and much of it is in very intense IR and UV, high-alpine environments, desert heat, etc..

Maybe what I'm trying to say is the Tiffen filters look great (as if they're not even there) while blocking relatively low amounts of IR. The Formatt and Pancro filters do a much better job at blocking more IR, especially in intense environments, but are not as color-neutral.
Click to expand...

What is arri raw missing in terms of colors that redcode has? Just curious. I love the redocode for the low resolution proxies an such but as a raw format they do not really differ that much do they?
 
My biggest concern about these ND-ing testing is why they used TUNGSTENS for tests? Why not HMI's? In a good sake who uses tungstens these days? HMI's and Fluorescent with some LED's as well have zero issues with IR pollution - cause they have almost zero IR in the effective spectrum. I think this specific prespond of the Misterium-X sensor to nd/irnd filtration is purely overestimated. I can easily find a lots of weakness of other cameras comparing to RED - for example... resolution :)
 
You must log in or register to reply here.
Back
Top