Grading to the rescue.

farss wrote on 4/30/2012, 11:47 PM
I was asked to shooting something last weekend. A $0 job down by the harbour, anything to get away from the desk. I thought, hm, that Tiffen Polariser has been sitting in the kit bag for years and never been used, time to give it a work out.

It was late afternoon on very cloudy autumn day so I have no idea why I thought a polariser was going to achieve anything. Even sillier the only way to rotate this filter on my EX1 was by putting my finger on the glass and pushing it around. Waste of time really, it made no difference but couldn't be bothered taking it off and the show was about to start. After the show shot some more footage just tp make all the hassle of getting to the place maybe worthwhile.

Got back home, took the pola off and oh my, that's a lot of grease on it, hope that didn't.....

Looking at the footage I think it did, yuck, washed out and soft. Good thing there was no money on the line but even so, what an idiot. Never try anything new unless you can tolerate an epic fail. I thought maybe a bit of tweaking would help, added Vegas's Color Corrector and tried upping the Saturation, yuck. Tried upping the gain, now, wow, we're cooking. Dialled down Offset, kept an eye on the waveform monitor and the Asus ProArt monito, did some more tweaking and BINGO.

I think if I was shooting set pieces under controlled lighting I'd explore diffusion in front of the lens in depth, it really does seem to take the edge off video. I dial Detail well down in my camera but it still doesn't give the same result.



Grazie wrote on 5/1/2012, 12:51 AM
Nice comparison Bob.

OK, was that the failure was that you were expecting MORE Sat. with the Pola? Is that it? Or was it that you had gotten grease from your fingers over the the front of the twisty?

I use a Formatt Linear Pola, and even with that I need to have the Sun at a particular angle, and then I get greater saturation.

The other side of the coin is that maybe your Tif was acting like a momentary ND and THAT allowed you to get more digital info captured, and THAT gave you more latitude to do more colour grading.

I've got a director's cut of Pirates of the Carib. And part of that does show how bleak and washed-out the clips looked PRIOR to Colour Grading. BUT, all the latitude was captured.


farss wrote on 5/1/2012, 3:11 AM
Yes, I had thought, overcast day, it's going to look flat and then of course I knew with the sun very low on the horizon and behind the camera a pola was going to do nothing. I even tested that before I put the thing on the camera. Why did I do that then, don't know.

Certainly the Tif pola was costing me over 1 stop of light, in fact I was back to ND1 and I've had days down here where with ND2 I'm shooting at f16 which is not good at all, very soft images. Also there was several layers of clouds that were moving around so the light levels were changing a lot.

I'm inclined to think the diffusion from the grease had something to do with what I ended up with. The other thing is to fit the pola I had to take the B&W 416 IR cut filter off the camera and that's why the reds really pop, the original EX1 does red very, very well, too well at times without that filter.

When I first looked at the footage, the waveform monitor was reading from 30% to 70%, with some of the shots that's not much to work with and yet I could stretch it out to 0% to 100% and it didn't fall apart and it really should have, wierd.

"I've got a director's cut of Pirates of the Carib......"

That's a riddle that had me puzzled for a long time. Thing to remember is you are seeing the image on a monitor that is incapable of displaying the data that's been recorded correctly, that data contains probably 4 stops or more of dynamic range and when that range is flattened down to fit onto what a monitor can display, it looks flat. I've watched people at trade shows work with that kind of footage and it is jaw dropping. You can push the dials around every which way and get it to look anyway you want and it doesn't fall apart. Very hard to do that with 8bit video from the cameras I can afford.

To be honest I'm not at all certain what I ended up with from my shoot was all that good even after the grading but it was something new, something I didn't expect, something to think about and maybe try to take further.


JasonATL wrote on 5/1/2012, 6:56 AM
This is almost exactly how just about everything I shoot nowadays looks. For what I do, I prefer the latitude to dial in the look in post rather than baking it in in-camera.

Agreed, nice demonstration.

This looked like nearly everything I do (in terms of pre- and post- um, post). I shoot flat on my DSLR and have adjusted my EX1 to more flat profiles (using the CINE-based picture profiles). When contrast (or stretching the offset and gain) is added, it goes from mushy to sharp and saturation comes back. More importantly, I still have plenty of shadow detail to play with, as blacks are not crushed.

I'm not saying everyone should shoot this way - and I certainly understand the utility of not having to grade everything. I'm just saying that it works for me.

What I find most interesting is that the flat image resulted not from your picture profile, but from a filter.
paul_w wrote on 5/1/2012, 7:52 AM
Good demo Bob, thanks for posting. Kinda like the 'lost zeppelin acoustic tapes' feel of the music too.
I had a similar deal recently using an experimental mirror lens fixed on the FS100. I think what is happening here is rather than a flat profile log curve for wide dynamic range, its more a matter of grease causing diffraction and washing out the blacks. My experience with a mirror lens also gave washed out blacks, as expected from an f11 1000mm mirror.
Contrast to the rescue! With some added saturation in places. It helped me a lot too with my tests.
But hitting the grain level is always the crunch point. Sometimes, i like grain so i dont mind hitting it. Its subjective i think.

amendegw wrote on 5/1/2012, 8:40 AM
I love these demos and resultant video postings (that's the way I learn).

@paul_w, I see you used my heat wave FX at the 01:26 mark (I knew I should have patented that [grin]). I also liked the birds.


System Model: Alienware Area-51m R2
System: Windows 11 Home
Processor: Intel(R) Core(TM) i7-10700K CPU @ 3.80GHz, 3792 Mhz, 8 Core(s), 16 Logical Processor(s)
Installed Memory: 64.0 GB
Display Adapter: NVIDIA GeForce RTX 2070 Super (8GB), Nvidia Studio Driver 527.56 Dec 2022)
Overclock Off

Display: 1920x1080 144 hertz
Storage (12TB Total):
OS Drive: PM981a NVMe SAMSUNG 2048GB
Data Drive1: Samsung SSD 970 EVO Plus 2TB
Data Drive2: Samsung SSD 870 QVO 8TB

USB: Thunderbolt 3 (USB Type-C) port Supports USB 3.2 Gen 2, DisplayPort 1.2, Thunderbolt 3

Canon R5
Canon R3
Sony A9

paul_w wrote on 5/1/2012, 9:02 AM
@Jerry, lol - agreed, great to see what people are doing on here, with explinations added.

Serena wrote on 5/2/2012, 1:20 AM
Bob, certainly looks a lot better. But is the original soft, or just low contrast? Both original and graded clips look the same to me in terms of resolution, but increasing the contrast makes things look sharper. I think your finger grease created a soft contrast filter rather than diffusion, scattering light into the shadows. I don't have your clips to work with but roughly fiddling with one of my own (shot on cine4) much the same result is achieved using levels. Not identical, but similar. Of course normally I use curves as first grade, but then black levels are low to start with.
farss wrote on 5/2/2012, 3:27 AM
"Bob, certainly looks a lot better. But is the original soft, or just low contrast?"

Throw in low contrast lighting as well and I'm not 100% certain. Perhaps a measure of both. It was certainly an eye opener to see how apparent resolution increased as I increased contrast. I'd seen it done before but when you do it for yourself with footage you've shot that's a different experience :)

I have shot some footage using Cinegamma, best shot was a gent with his arm around a 40W fluro as he talked to the audience. Tack sharp, nice roll off from the highlights etc but that really looked "digital".

I'm kind of inspired to buy / borrow some Promist filters to take this further.

Richard Jones wrote on 5/2/2012, 5:22 AM

I'm impressed with the improvement you got and must remember your Gain + Offset trick. Thank you.

In the old days of film photography (still and cine) we sometimkes used to get soft focus either by stretching a fine nylon stocking over the front of the lens or by using a UV or similar clear filter smeared with fine grease obtained by running the tip of your finger along the side of your nose and then wiping this carefully over the filter (honestly! But then I'm so old I can remember all these old film and still photography tricks of yesteryear!) By the way, this latter technique did not lead to any significant light loss or exposure correction,

If you watch an older film and look at the close ups of the faces of some of the stars you will notice how soft focus has been used to try to hide any slight defects in their complexions. Often it's far too obvious but I do know that some of the old directors used to use the stocking trick. The better the stockings you buy for your wife/girl friend the finer the mesh and the finer the mesh the less obvious the result!

We also used the same terchniques in the dark room although here the diffusion went in a different direction (white into black rather than black into white --- or was it the other way round?) The result was in fact very similar, save to the closest inspection by the most expert eye.


ushere wrote on 5/2/2012, 7:28 AM
we used to use fine stockings stretch BEHIND the lens....
Serena wrote on 5/2/2012, 7:44 AM
Cine4 is my default and the graded output doesn't look digital. Art Adams and Mitch Gross do likewise, so I feel pretty confident with that default setting. Detail "off".
farss wrote on 5/2/2012, 7:55 AM
Thank you Richard, it's nice when someone assumes I'm not old enought to remember all those tricks :)
One I recall in the darkroom was blowing smoke in front of the enlarger. if you think that's a serious health risk here's one I learned just recently.

You take a B&W print and hang it in a sealed glass tank. In the bottom you boil mercury. I'm not certain what this does exactly, perhaps the mercury binds to the silver?

One I did try was to make a very high contrast positive and negative film, offset them and print through them.

farss wrote on 5/2/2012, 7:57 AM
"Detail "off". "

Thanks I'll try that, I set it to -20 and wind coring up to 30.

apit34356 wrote on 5/2/2012, 11:37 AM
thanks Bob for sharing. I forgot how much I enjoyed reading the vegas forums.