RAW S-Log Clip to Download

[Peter G. Johnson]

True! It is the only clip that you can download for free. Before I purchased the s-log upgrade and began recording to an external recorder (ProPres422, 10Bit, 220Mbps etc) I searched painstakingly for hours to find an example, so well done, Douglas.

As it turns out, your clip looks pretty much exactly what my clips look like before grading. I don't use any measurements to calculate the correct exposure. If it looks alright through the view finder, then I assume it should be alright. Sometimes I get it wrong, but most of the time the finished product looks great.

I am however always mindful of what Alister says about over exposure, but on location I'd go crazy if I kept worrying about that.

Alister reminds me of the teacher who will always push you to get the better results. Nothing wrong with a good teacher.

[Douglas Villalba]

Originally Posted by Peter G. Johnson

Alister reminds me of the teacher who will always push you to get the better results. Nothing wrong with a good teacher.

Nothing wrong with that.
I read daily what other people say about things that interest me including Alister Chapman, Andy, Timur Sivan, Philip Bloom, etc. That is why I bought the F3 for a feature I directed, filmed and edited last October ("The Awakened") that required shooting with very little lighting. The camera was a heaven sent. Even before S-Log I was able to shoot a whole movie inside an small house with 7' sealing. Not a lot of room to hide lights there.

[Doug Jensen]

Originally Posted by Peter G. Johnson

True! It is the only clip that you can download for free. Before I purchased the s-log upgrade and began recording to an external recorder (ProPres422, 10Bit, 220Mbps etc) I searched painstakingly for hours to find an example, so well done, Douglas. .

Of course, you could have saved yourself hours of painstaking searching and just bought 126 clips (50GB of raw unadulterated footage) for about a hundred bucks. I don't know what your time is worth to you, but $100 doesn't even cover an hour of my time. :-)

PMW-F3 S-LOG & External Recorders Test Drive

[Peter G. Johnson]

By the time I got serious with this type of craft it was too late for a career change, so I need to do the painstaking stuff to avoid spending more money.

[Leonard Levy]

Alister,
Thanks for the explanation. I have a bunch of tests I've run as well, but i tend to be too damn lazy to post them. In fact sometimes to even look at them as carefully as I did when shooting. I think the first S-log I ever shot may have been overexposed and that would explain why I was not happy with the grading.

I for one agree completely about using the viewfinder readout and or waveform & for setting exposure, but I guess everybody has there own ways of doing things.

[Alister Chapman]

There are clips and S-Log frame grabs on my bog for download. They have been there since October last year.


Canon C300 and Sony F3 footage to download. | XDCAM-USER.COM

More Codec and Gamma Tests. | XDCAM-USER.COM

Convergent design Gemini S-Log frame grabs. | XDCAM-USER.COM

[Douglas Villalba]

Originally Posted by Alister Chapman

There are clips and S-Log frame grabs on my bog for download. They have been there since October last year.


Canon C300 and Sony F3 footage to download. | XDCAM-USER.COM

More Codec and Gamma Tests. | XDCAM-USER.COM

Convergent design Gemini S-Log frame grabs. | XDCAM-USER.COM

Hello Alister,

I downloaded the folder F3-C300 Clips with the F3 recorded to the internal recorder at 1080p/25, 8 bit 4:2:0. Is there a 10 bit I can download so I can grade?

[Alister Chapman]

No, 8 bit only but the only difference in the grade would only be possible additional banding and a bit more noise. In terms of the look you can create it will be essentially the same.

The 8 bit file has been downloaded about 300 times and I simply can't afford the bandwidth for full 10 bit RGB files. That's why I normally provide frame grabs or single DPX frames. The grading is then identical to working with a clip, just limited to a single frame. I did have a 3 second long RGB DPX file on the site, but after over 100 downloads (45Gb) and only one paypal donation I decided enough was enough and removed the clip.

[Douglas Villalba]

Originally Posted by Alister Chapman

No, 8 bit only but the only difference in the grade would only be possible additional banding and a bit more noise. In terms of the look you can create it will be essentially the same.

The 8 bit file has been downloaded about 300 times and I simply can't afford the bandwidth for full 10 bit RGB files. That's why I normally provide frame grabs or single DPX frames. The grading is then identical to working with a clip, just limited to a single frame. I did have a 3 second long RGB DPX file on the site, but after over 100 downloads (45Gb) and only one paypal donation I decided enough was enough and removed the clip.

Got it.

The clip for download is in the 4-5 f-stops (10%-87% waveform). Do you have any single frame DPX frame with a range over 0-100%?

[Bruce Schultz]

Originally Posted by Alister Chapman

It's only when you go to a camera with 12 or 14 bit linear recording, like the F65 that you can really afford to place your exposure range wherever you want it.

And Sony recommends with the F65 to expose Caucasian faces at 33% not 38%. My tests with that camera bear this out. When you get down in exposure that much (33%) you can't see the faces all that well so having a good LUT working on a monitor goes a long way towards nailing exposure - you just can't use the monitor as a final arbiter of it.

I have a suggestion for Douglas, why not do another test with the same setup, but this time shoot with your light meter style and another identical shot using the simple viewfinder middle box percentage at 38% on a grey card? I would like to download those two and compare them in Resolve using the Log settings.

BTW, Resolve Lite is a free download and will do up to 2 nodes of HD 1920x1080 without watermarks.

[Doug Jensen]

Bruce, why don't you do that test and tell us what you find out? :-) I'm finished doing any testing for public consumption. I have already come to my own conclusions, best practices, and methods and I have no interest in explaining and/or defending my findings any further.

And not only that, I completely agree with Alister that using a light meter is not nearly or easy or fool-proof as the camera's own built-in exposure tools. Sure, my light meter is a good tool for some purposes, but is a lousy alternative to newer technology for setting the overall exposure. The F3 ain't no film camera.

And not only that, if you exceute both of the methods correctly that you propose testing, the results would be the same, wouldn't they? I am a firm believer that there can only be one correct exposure, regardless of what method(s) you use to find it.

[Bruce Schultz]

Roger all of that Mr, Jensen, but I was directing that suggestion to Mr. Villalba who started this thread - you both have similar first names.

I agree with Alister and you that reflective metering via the camera is the very best method of determining the exact light level the sensor is seeing, but as a long time DP & HD DIT I have shepherded many a film DP into the gaping maw of digital motion photography and can understand how abandoning a light meter is like throwing one oar overboard in the middle of a vast ocean.

I too have no need or necessity to conduct additional tests of processes that I know are accurate and true.

[Douglas Villalba]

Originally Posted by Bruce Schultz

And Sony recommends with the F65 to expose Caucasian faces at 33% not 38%. My tests with that camera bear this out. When you get down in exposure that much (33%) you can't see the faces all that well so having a good LUT working on a monitor goes a long way towards nailing exposure - you just can't use the monitor as a final arbiter of it.

I have a suggestion for Douglas, why not do another test with the same setup, but this time shoot with your light meter style and another identical shot using the simple viewfinder middle box percentage at 38% on a grey card? I would like to download those two and compare them in Resolve using the Log settings.

BTW, Resolve Lite is a free download and will do up to 2 nodes of HD 1920x1080 without watermarks.

I would think that the Douglas is me, the one that posted my test.

First you have to understand the test.

The test is to determine what happens to the surrounding areas that may be over or under exposed when you expose your subject using what ever method you use.

Example that comes to mind right now:

In an establishing shot you have inside a building someone spying through a window (- 6.5 - f 0.7). Across the street two actors under the shade of a building. Behind the two actors a car with gunmen is approaching (direct sun light)(+ 6.5 - f 64). Where do I want my pivoting point? what f-stop?

Look Bruce regardless of how I came up with my exposure you have 14 1 f-stop Bracketing clips. You can easily see the max under & over. If I had used a rear lit gray scale card and change f-stops like I did to determine actual latitude I had to come to the same conclusion.

If you need any ProRes HQ original clip, let me know which one and I will upload it so you can open it in Resolve. I've got resolve light, but I know how to use Apple Color better.

Good references:

A Simple Explanation of F-Stop Numbers
Sunny 16 rule - Wikipedia, the free encyclopedia
Accurate Exposure with Your Meter

[Douglas Villalba]

Originally Posted by Bruce Schultz

And Sony recommends with the F65 to expose Caucasian faces at 33% not 38%. My tests with that camera bear this out. When you get down in exposure that much (33%) you can't see the faces all that well so having a good LUT working on a monitor goes a long way towards nailing exposure - you just can't use the monitor as a final arbiter of it.

I have a suggestion for Douglas, why not do another test with the same setup, but this time shoot with your light meter style and another identical shot using the simple viewfinder middle box percentage at 38% on a grey card? I would like to download those two and compare them in Resolve using the Log settings.

BTW, Resolve Lite is a free download and will do up to 2 nodes of HD 1920x1080 without watermarks.

My previous answer was started long before you reply, but it took me a long time to finish. Sorry.

[Alister Chapman]

Originally Posted by Douglas Villalba

The clip for download is in the 4-5 f-stops (10%-87% waveform).

That's not correct this is log and 10-87% equates to 8 to 9 stops.

The S-Log curve is as follows:

0-10% 3 stops (If trying to use 8 bit this means each stop represented by only 8 grey shades, 10 bit is 30 grey shades per stop)
0-20% 5.5 stops
0-30% 7 stops
Middle Grey 38%
0-40% 7.5 stops
0-50% 8.5 stops
0-60% 9.5 stops
0-70% 10 stops
0-80% 11 stops
0-90% 12 stops
0-100% 12.6 stops
0-104% 13 stops

Arguably the first stop above black is going to be quite noisy so is unlikely to be of any significant use in a real world image, so measured dynamic range is 13 stops but latitude 12 stops.