MX IR TEST...

[Dylan Macleod CSC]

My assistant Brent J.Craig wanted to do an infrared test after the comparison test we shoot last week. We only had a 2K tungsten light (full spot) to test with. But you can see the vast improvements in the new sensor!

Both shots have a Rosco IR held over the top 1/4 of the frame. It is noticable on the Mysterium sensor - but invisible on the MX.

Both shots were done simaultaneously with an ND 1.2.

Nice work RED!

(EDIT:I have reposted a new image that was processed through REDCINE-X later in this thread. That image is much more accurate in terms of a comparison)

 

[Hans von Sonntag]

Thanks Dylan,

Looks like IR-contamniation is not an issue anymore. Great.

Hans

 

[Chris Parker]

very good illustration of this. thanks dylan and brent.

 

[Paul Leeming]

Great example Dylan, thanks for sharing it!

 

[Jason Diamond]

Great example! thanks for doing that.

 

[Ryan E. Walters]

Thanks for doing this test and posting the results- but I am having trouble interpreting the image- why is the jacket white in one shot and brown in the other? I thought these were taken at the same time under the same conditions ...

 

[Dylan Macleod CSC]

Thanks for doing this test and posting the results- but I am having trouble interpreting the image- why is the jacket white in one shot and brown in the other? I thought these were taken at the same time under the same conditions ...

They were shot at the same time with identical ISO's (I think 320), T-stops and ND's.

There are going to be inherent differences in the two sensors - but that isn't what this portion of the test was about - it is more about looking at IR.

I opened the files up in REDALERT (I know - old school). Used Camera RGB color space and REDSPACE gamma. Then output TIFF's and converted to JPEG.

 

[Ryan E. Walters]

Thanks for the info. That is quite a color shift between the two. Was the same brand of ND used for each one?

 

[Tom Lowe]

Epic win.

 

[Dylan Macleod CSC]

Thanks for the info. That is quite a color shift between the two. Was the same brand of ND used for each one?

Same brand. Same lenses (Cookes) - same everything - except sensor.
Pretty impressive!

 

[David Battistella]

Color Science?

Same?

 

[Dylan Macleod CSC]

Color Science?

Same?

REDALERT - not playing with new Color Science. Just looking at IR in this test.

 

[Miltos Pilalitos]

We only had a 2K tungsten light (full spot) to test with.

Do you remember who the manufacturer of the light is? Is it normal for a tungsten unit to produce so much IR radiation? :scared:

 

[Mitch Gross]

Do you remember who the manufacturer of the light is? Is it normal for a tungsten unit to produce so much IR radiation? :scared:

IR is heat. 90% of the energy out of a tungsten bulb is heat.

 

[Miltos Pilalitos]

IR is heat. 90% of the energy out of a tungsten bulb is heat.

Well yeah, all electromagnetic radiation can produce heat but somehow i was under the misconception that light fixtures for cinema applications were filtering the IR spectrum out.

I guess they don't. :smiley:

 

[Dylan Macleod CSC]

Well yeah, all electromagnetic radiation can produce heat but somehow i was under the misconception that light fixtures for cinema applications were filtering the IR spectrum out.

I guess they don't. :smiley:

It was an ARRI 2K open face. But I don't think it would make a difference. I've been able to reproduce this with a variety of tungsten fixtures.

Lets keep the eye on the prize though - new sensor - NO WORRIES!!!

 

[Miltos Pilalitos]

Lets keep the eye on the prize though - new sensor - NO WORRIES!!!

Damn! A few weeks ago i spent a fortune for Tiffen Hot mirror filters! I guess i will use them as ND filters now...:crying:

 

[Ryan E. Walters]

These are very impressive results- I wonder what the new contamination level is? Although once you get past 1.2 color shifts really start to become more of a problem ...

Thanks again for providing us with the test results.

 

[Roberto Lequeux]

I don't understand. With a hot mirror held over the top 1/4 of the frame why is the jacket still white-pink? And the original color is green, right?

 

[Dylan Macleod CSC]

I don't understand. With a hot mirror held over the top 1/4 of the frame why is the jacket still white-pink? And the original color is green, right?

We probably shouldn't have even put the IR filter in at all. It complicates the comparison of the two sensors.

The top part of the Mysterium sensor (that is filtered) looks like the IR is pretty much gone - the soft filter edge makes the distinguishing line a little blurred - but if you put your hand over the bottom part of the Mysterium image and only look at the top part - which is getting filtered - the blacks start to look a little more neutral.

In the MX sensor - the line between filtered and unfiltered is invisible. The IR filter isn't doing anything. It doesn't need to.

The lamp was spotted in as well - so there is a bit of a vignette. Not ideal - but just look at the overall difference in the two sensors.

MX rules!