OLPFS NIGHT TEST

[David Battistella]

Looking at the r3d it kind feels like the STH was selected when the LLO was in and vice versa.

The LLO shot is teal colored like the STH OLPF,

the STH shot is RED colored like the LLO OLPF.

Is there a chance they could have been accidentally reversed? Just covering all the bases.

I have both OLPF's so maybe I can run a quick test next week.



Battistella

 

[steve green]

Changing the OLPFs was very easy and I would feel comfortable doing it during a shoot, but definately away from the set. It was a quick switch, just get everything ready beforehand.. I would even feel comfortable doing the switch, blinded in an old Film changing bag if it's a particularly dusty environment. The rubbery OLPF housing acts like a spring pushing it out a little after you unscrew it. You just need to pay attention to the super tiny and stubby screws.. (you lose one and it won't sit flat) If you got big meaty fingers, forget it... The little wrench provided needs to be triple the length, and I almost wish it was straight instead of L shaped, because it's cumbersome to screw back on. But, it was probably made like that so that it can fit inside the OLPF box and be used as a backup. I would suggest using a proper wrench, which I ended up doing. And I also wish Red just gave you an extra set of screws with each OLPF purchase.

How about captive screws, that seems to be the best way at this point to make the switch doable on set.

 

[Mark L. Pederson]

you also need to black shade for each olpf

Nope. Black shade only cares about sensor temp and shutter speed. It doesn't care about any filters you put in front of the sensor.

 

[Elsie N]

Dust and time issues are real concerns. It's not really practical to spend that much time and try to find a dust free environment on a non-hobbiest professional set. You might be able to pull off a switch at lunch, if you're lucky, but I would rather just have lunch. If you want both OLPFs then you should rent two cameras, or else pick one. Rare is the situation where switching would actually make sense. My typical sets run at about $1750 per shooting hour, all told, and many out there run at far higher rates than that. Renting a Dragon here costs $850 - plus there is the stress factor and the real chance you will get dust in there. If you rent two then you can switch instantly, and even have two lenses pre-readied. People should be smart about this.

A truly professional set will have a "clean tent" to make a change if necessary. Or they will have two cameras (as you suggest) and as I have (and will possibly set up with the different OLPFs if I ever become disenchanted with the STH version for some situations.)

My answer was meant to convey that there will seldom be perfect conditions for use of any item. Thinkers will find a work-around, or else will forgo the activity if the risk-reward ratio is too high.

 

[Patrick Kendall]

See attached images... When you cycle through them, the left 1/3 of the image seems to have a pinkish tone...

Oye! hopefully it's a cold temperature issue. I brought it from a warm apartment and fired up the camera and shot straight away out in the 25 degree temperature with the STOLPF on... I took it up with Red to see what they say. Thanks for the heads up.

 

[Patrick Kendall]

Looking at the r3d it kind feels like the STH was selected when the LLO was in and vice versa.

The LLO shot is teal colored like the STH OLPF,

the STH shot is RED colored like the LLO OLPF.

Is there a chance they could have been accidentally reversed? Just covering all the bases.

Battistella

David, the STOLPF was installed by RED and they set it to ST for sure. And when I switched to LLOLPF, I definately switched it in the menu. I just double checked again today as I still have the LLOLPF installed. But I remember Jarred saying that OLPF selection could be adjusted in Redcine-X incase someone forgets to do the switch on set..

 

[David Battistella]

Cool.

Just checking. Thx. I like to try cover all possibilities, even if they seem banal.

I don't see a way to change OLPF in the current Beta FW build, but maybe someone at RED could try swapping the OLPF header on the clips?

Battistella

 

[Graeme Nattress]

Under daylight or tungsten conditions, swapping OLPF and keep the white balance the same (say when you shoot a chart or grey target) the visual appearance stays pretty darn constant. At least that's exactly what I see when I look at the charts shot under the various OLPFs that I test and calibrate for in-camera use.

I don't think, looking at the image data in the raw, that the OLPFs have been mis-assigned in camera. The STH data looks typical of that OLPF, as does the data from the LLO ones.

What I think could be happening is an interaction with the street-lighting. Was it sodium light for the street lights?

Graeme

 

[Gavin Greenwalt]

I don't think, looking at the image data in the raw, that the OLPFs have been mis-assigned in camera. The STH data looks typical of that OLPF, as does the data from the LLO ones.

Graeme

I sure hope that's not the case. I don't care how clean the image is if they're Matrix Green in night scenes.

 

[Brian Merlen]

i think its the street lights, your not in a controlled environment. Phill Holland has shown many examples in a controlled environment that do not present this. to further support my theory look at the wide shot (first one). the far away most bright looking light that is facing straight into the lens is Red then Green. I don't necessarily think it was that street light, but maybe? what i really bet is there is a closer one not in the shot throwing light into the glass causing a tint, and without a matte box and flags to stop it, the street light changed the color cast...

was there any closer street lights in that wide shot then the far away (but noticeably bright one in the shot shooting straight into the lens)...

 

[D Fuller]

i think its the street lights, your not in a controlled environment. Phill Holland has shown many examples in a controlled environment that do not present this. to further support my theory look at the wide shot (first one). the far away most bright looking light that is facing straight into the lens is Red then Green. I don't necessarily think it was that street light, but maybe? what i really bet is there is a closer one not in the shot throwing light into the glass causing a tint, and without a matte box and flags to stop it, the street light changed the color cast...

was there any closer street lights in that wide shot then the far away (but noticeably bright one in the shot shooting straight into the lens)...

I think Brian has hit the nail mostly square on the head. I see four traffic lights facing the camera -- red in one shot and green in the other, it's a one-way street, so there would not likely be a light facing the opposite way, but there is lots of color coming from the sources we can plainly see in the shot. And given that the streetlights in the shot are most certainly a discontinuous spectrum, plus the white balance having been done who-knows-what mix of sources, I wouldn't get too worked up about what you see here. There are way too many variables to know what the OLPFs are responsible for here.

Even the ability to see detail in the shadows will be affected by the color of light the sensor is seeing, regardless of the OLPF installed.

 

[David Battistella]

I don't think that explains away the other two grabs though.

it might be true of the WS, what was really needed was to get a color checker underneath one of those street lights on each OLPF to see how this kind of light affects color reproduction on each OLPF. In my city the type of light changes from street to street depending when it was added and where it is.

Under tungsten, and daylight I'd say both OLPFs would be hard to differentiate, but it's hard to know without a reference.

The r3ds when given the same look have fairly different histograms.

Would be be good to have a grey card on both just to wb them both and see how they match then.

Battistella

 

[Brian Merlen]

Even in the tights I suspect that color cast is a reflection off street lights. to further strengthen the theory you can even see a red light in one of the tight shots... it just seems too convenient that were seeing red/green issues while having red green lights in the area, all bouncing off of every thing.... I'd bet on it that its a light issue...

ps how long did you roll? if you rolled for a while, scrub the clip, see if it changes? or like suggested repeat it with grey card/charts in the shot too...but i really am pretty sure its just reflections both in the wides or tights...i bet matte box with flags would help the situation as well, get rid of any of the extraneous lights (that i still think is the culprit)...


ps with the one way street theory, even if not faced opposite exactly, what about the street corner that had a light at the intersection...(possibly)... i am betting there are numerous lights in the area, and even in the wide shot look at how many things have a color cast...getting a close up on any of them still would have the cast, those things project far and wide with this sensitive a sensor... i mean hell even the review of the new varicam i read they can't even have a tally light on it at ISO5000 without it ruining there shots...im guessing a red at 1600 could have the similar issues off even very little light..

 

[Patrick Kendall]

i think its the street lights, your not in a controlled environment. Phill Holland has shown many examples in a controlled environment that do not present this. to further support my theory look at the wide shot (first one). the far away most bright looking light that is facing straight into the lens is Red then Green. I don't necessarily think it was that street light, but maybe? what i really bet is there is a closer one not in the shot throwing light into the glass causing a tint, and without a matte box and flags to stop it, the street light changed the color cast...

was there any closer street lights in that wide shot then the far away (but noticeably bright one in the shot shooting straight into the lens)...

Yes Brian, the wide shot was shot about 20 feet from the mouth of a one way street in front of my apt. No traffic lights were pointing from behind me. the other two shots were shot on Central Park West, which is two ways. And yes Graeme, it looks to be Sodium Street lights. In the wide and police box shot, a large Sodium lamp was overhead. As I wanted to do a quick test and compare illumination capabilities of OLPFs.

Also, after talking with Red about the pink band issue, it seems like my camera may not have been at the proper operating temperature in the first test which was with the STOLPF.

 

[Adam Montville]

LLO OLPF came in today. Pointing it out the window for tests. Seriously impressed!

 

[Brian Merlen]

is it really that pronounced or is the monitor skewing your phone to expose for the red and not the outside? that looks like an insane difference lol!!

 

[Adam Montville]

is it really that pronounced or is the monitor skewing your phone to expose for the red and not the outside? that looks like an insane difference lol!!

It was exposing for the monitor, but it was still night time out. Not "pitch black", but the camera phone still wouldn't be able to see anything.

 

[Brian Merlen]

looks pretty amazing! i haven't tried my LLO yet but looking forward to it

 

[Matt Ryan]

I have to say I will be using the LLO OLPF for pretty much everything I shoot. Used it on an incredibly low budget music video with a tiny lighting arsenal. I had to light a prison set with 3 750w leko lights (we even shot slow mo!). That is it. One coming through the back window, one for key, and the other for bounce to bring up as much of the cell as possible. Shot entirely at 800ISO and wide open at t1.3 on super speeds. Lots of hand held, so 180 degree shutter was a must. Have to say the results are pretty "clean," especially for 3200k lighting.

Dragoncolor-RG4 straight out of camera:

Here's the Dragoncolor-Redlogfilm image at 5000ISO to see what info is there in the low end:

Pretty blown away with the results. Sorry for the jpeg compression. I've shot plenty of material on the skintone olpf but after being burnt by noise even at 400ISO, I really like the cleanliness of the LLO olpf (no noise reduction needed). Lots more useful info there in the low end. On the other hand, the skintone highlight olpf is equally amazing but with the highlights. Just watch your exposure carefully to avoid a noisy low end.

 

[Brandon J.F.]

Awesome shot, Matt. The LLO is my weapon of choice as well.