Scarlet and infra red polution

[Andrew Salter]

Hi l have seen it mentioned a problem regarding infra red pollution with scarlet, and blacks not being true blacks ! does this happen all the time. If not! why and when does it occur, l thought this would have been filtered out in the capture process, as this does not seem to happen with other single chip cameras, FS100 F3 or the new Canon C300 though those cameras are video not raw.

 

[Andrew Allsbury]

As I understand it this only happens when using certain ND filters, and is worsened when stacking certain ND filters. There are several IRND filters available that sound like they don't have the issue. Much more expensive though...

 

[Joel Arvidsson]

It was a very nice video here with test of differnt nd filters. You might want to have a look at that. Did find the link thoe. Somebody else might chime in.

 

[Andrew Salter]

Ok that makes sense, has any one tried the new variable Heliopan filters or can recommend a variable ND filter that does not cause infrared problems

 

[stu aull]

Well there is a damn good thread on this in one of the EPIC/Red Discussion forums:

http://www.reduser.net/forum/showthread.php?67617-Neutral-Density-Filters-Advice

It seems to be a very worthy discussion that all Scarlet users should be aware of....

Apparently ALL digital-style cameras are much more sensitive to IR, NOT just RED. (Tho I understand the _original_ ie NON-MX_ RED ONE sensor was particularly sensitive.)
Where this really comes apparent is when we start to use a lot of ND on our glass, necessary because of the great sensitivity (native 800ISO) of our sensors - most notably in daylight. Here's what mitigates this: as we start stacking NDs (or stronger NDs) onto the lens, it DOES cut down VISIBLE light, but NOT IR! So as we ND-up, the RATIO of IR increases to the point IR starts affecting color, notably in the blacks.

As I recall, from a great test done on THIS blog;
http://www.ryanewalters.com/Videos/Tests/tests.html ("Epic, IR tests")
once you get past about .9 ND (which I'll admit I have NO idea what amount that really is, but presumably a factor indicated ON the filter?) you start to see this pollution problem.

So the solution? We need to start incorporating IR filtration when using ND.
And guess what? to make it even MORE complicated/annoying, some IR filters themselves actually introduce subtle tints into the images! The above mentioned blog actually did a test to see which filter brands did the best/worst.

AND - these IR filters are not cheap at $2-300...

We have lots to learn!

Stu Aull
Alaska

 

[Trevor Meeks]

I'll be purchasing the Schneider Platinum IRND filters http://www.filmtools.com/schneider-platinum-irnd-4x5-650.html

 

[stu aull]

For us non-Matte box users in Canon-ville:
the same Schneider Platinum IRNDs in 77mm are here:

http://www.filmtools.com/schneider-platinum-irnd-77mm.html

Stu Aull
Alaska

 

[Terry VerHaar]

Well there is a damn good thread on this in one of the EPIC/Red Discussion forums:

http://www.reduser.net/forum/showthread.php?67617-Neutral-Density-Filters-Advice

It seems to be a very worthy discussion that all Scarlet users should be aware of....

Apparently ALL digital-style cameras are much more sensitive to IR, NOT just RED. (Tho I understand the _original_ ie NON-MX_ RED ONE sensor was particularly sensitive.)
Where this really comes apparent is when we start to use a lot of ND on our glass, necessary because of the great sensitivity (native 800ISO) of our sensors - most notably in daylight. Here's what mitigates this: as we start stacking NDs (or stronger NDs) onto the lens, it DOES cut down VISIBLE light, but NOT IR! So as we ND-up, the RATIO of IR increases to the point IR starts affecting color, notably in the blacks.

As I recall, from a great test done on THIS blog;
http://www.ryanewalters.com/Videos/Tests/tests.html ("Epic, IR tests")
once you get past about .9 ND (which I'll admit I have NO idea what amount that really is, but presumably a factor indicated ON the filter?) you start to see this pollution problem.

So the solution? We need to start incorporating IR filtration when using ND.
And guess what? to make it even MORE complicated/annoying, some IR filters themselves actually introduce subtle tints into the images! The above mentioned blog actually did a test to see which filter brands did the best/worst.

AND - these IR filters are not cheap at $2-300...

We have lots to learn!

Stu Aull
Alaska

You explained the problem correctly - when using a lot of ND, the ratio of visible to IR goes up. And .9 ND is 3 stops.

 

[Zeb B]

You explained the problem correctly - when using a lot of ND, the ratio of visible to IR goes up. And .9 ND is 3 stops.

I believe each .3ND is equal to a stop. So .6 is 2 stops, .9 = 3stops etc

 

[Daniel Kunz]

I'm wondering about adding a IR-Cut filter BEHIND the lens inside the CANON Mount, so that i can use whatever ND filter i want.
http://www.edmundoptics.com/products/displayproduct.cfm?productid=1524

 

[D Fuller]

I'm wondering about adding a IR-Cut filter BEHIND the lens inside the CANON Mount, so that i can use whatever ND filter i want.
http://www.edmundoptics.com/products/displayproduct.cfm?productid=1524

Interesting idea. The problem with this solution would be that you would have to remove the lens every time you wanted to take it off, and you would have to change it when you change lenses - or have several filters.

A similar idea is to get a "Hot Mirror" IR filter and put it first in the matte box. Then you can use whatever ND you like, and stack NDs if you need to . With IRNDs you cannot stack them without inducing color shifts.