The Pro’s and Con’s of 12 bit linear raw or recording raw to S-Log.

This came up in the comments today and it’s something that I get asked about quite a lot.

Sony’s high end cameras, designed for raw – F5, F55, F65 all use 16 bit linear data. This linear data contains an impressively large amount of picture information across the entire range from the darkest shadows to the brightest highlights. This huge amount of data gives footage that can be pushed and pulled in post all over the place. 16 bit raw gives you 65,536 discreet values.

The FS7 and FS5 use 12 bit linear raw. 12 bit data gives you 4096 discreet values, a small fraction of what 16 bit has. This presents a problem as to record 14 stops with linear data you need more than 12 bits.

Not Enough Code Values.

There just aren’t enough code values with only 12 bits (which is why no one else does it). So Sony “cheat” for want of a better word and use data rounding to make it work. This reduces the tonal information in the shadows.  On it’s own this isn’t a huge problem, just make sure you expose brightly to avoid trying to pull to much info out of the shadows and definitely don’t use it for low light. On high key scenes 12 bit raw is very nice indeed, this is where it excels. On low key scenes it can be very grainy, noisy and shadows often look coarse and lack smooth textures. Expose nice and bright and you will get great highly gradable footage.

Transcoding can add to the problems.

Where you really do run into problems is if you take 12 bit raw (with it’s reduced shadow data) and convert that to 10 bit log (which has reduced highlight data relative to the scene you are shooting).

What you end up with is 10 bit log with reduced shadow data compared to a straight 10 bit log recording. If you compare the direct 10 bit S-log from an FS7 (or F5/F55) to 10 bit S-log derived from 12 bit raw from an FS5, the FS7 picture will have a touch more shadow information while the highlights from both will be similar. So the 10 bit direct log recording from an FS7 will typically be slightly better than the raw derived log from an FS5 in the shadows.

If 4K S-Log is really important to you – get an FS7, F5 or F55.

So I’d much rather have an FS7 (F5 or F55) if I want to shoot UHD or 4K S-log. But, if you only have an FS5, the raw to log workflow will outperform the limited 8 bit UHD log, so it is still beneficial for FS5 owners to shoot raw and convert it to S-Log with an external recorder. Better still record raw, then you really will have a better image.

Raw with the FS7.

On the FS7 the benefits of recording 12 bit raw over 10 bit S-Log are less clear. For bright, well exposed scenes the 12 bit raw will  have an edge. But the files you will create if using an Atomos or Convergent Design recorder are huge compared to the compressed internal recording and that needs to be considered. For low key or under exposed scenes there is no benefit to shooting 12 bit raw.

On the FS7 it is not a good idea to take the 12 bit raw output and record it as 10 bit S-Log on an external recorder. While you may have a less compressed codec, you will be compromising the shadows compared to the cameras own internal 10 bit log recordings.  In most cases you would be better off simply taking the HDMI output and recording that as it avoids the 12 bit linear shadow bottleneck.

Again though – exposing nice and bright is the key to a good result. Get the data up into the brighter parts of the recording and the raw can be fantastic.

Internal and external log brightness shifts.

When you record S-Log internally on the Sony cameras the recordings use full range data levels to maximise the codec performance. You can use data range (which exceeds the normal video range) as it is assumed the data will be graded and as part of this process restored to video range data for viewing. However when recording on an external recorder the recordings sometimes use full video range rather than data range or if it’s data range don’t have the right metadata. This shouldn’t be a huge problem if the grading software behaves itself and treats each type of content correctly, shifting each to one unified range, but sadly this is rarely the case (especially with Adobe). So not only do the internal and externally recorded images come out with different brightness and contrast, but also LUT’s designed for one don’t work the same with the other. It’s a bit of a minefield to be honest and one of the reason why I prefer to always grade with dedicated grading software like resolve which handles the levels conversions properly (most of the time at least).

Advertisements

Guide to Cine EI – Still Current.

Just a reminder that my guide to shooting with Cine EI for the PMW-F5 and F55 cameras is still just as valid today as it was when I wrote it back in 2013. There have been a few tweaks to the cameras menu here and there, but the principles and basic operation have not changed.

So if you are new to Cine-EI please take a look at the guide. It takes you through how to shoot with Cine EI, which LUT’s to use and how to expose them.

Cine-EI Mode when recording S-Log2/3 and raw on the F5 and F55.

ND filters for the Solar Eclipse.

I know many of my readers are planning on travelling to the US and the path of totality to shoot the solar eclipse on the 21st of August.

A couple of words of caution.

First, make sure you never view the sun directly or through an optical viewing device such as a DSLR camera, telescope or lens without covering the lens with a dedicated solar viewing filter.

DON’T try to use sun glasses, welding goggles or conventional ND filters.

Most regular photo ND filters only cut the visible spectrum. Many will pass IR (Infra Red) without much attenuation. When shooting or looking at the sun it’s not just the brightness but also the IR that can damage your eyes or the sensor as this is where most of the heat energy is. So a 16 stop ND might well reduce the brightness of the sun to a useable level for a video camera but may not reduce the damaging IR and heat. This could lead to damage to the cameras own ND filters or the sensor itself.

To look at the sun with your eye’s you MUST use a proper solar filter or solar viewing glasses. This will normally be a silvered reflective filter or film. The silver reflective coating reflects away the harmful and damaging infra red as well as reducing the brightness. If the filters you are using are of the film type, check for pin holes in the film before using them.

On a video camera it is much safer to use IRND filters rather than conventional ND filters if you can or add a good quality IR cut filter in front of the ND filters.  Don’t skimp on the filters you use.

Make sure you have enough ND for the camera to use an aperture wider than f11. If you are having to stop down to f11 or f16 you will not be getting the sharpest possible image due to diffraction effects. On cameras with smaller sensors such as 2/3″ or 1/2″ you really want to have the aperture more open than f8 for the best results, so that means you’ll typically need 16 stops of ND!!

Try to avoid screw on filters. Once the moon completely covers the sun you will want to quickly remove the heavy ND filters. Then once the sun comes out from behind the moon you will need to replace the filters very quickly. So use something that allows you to do this quickly and easily. A mattebox with slide in filter trays or magnetically attached filters are a good way to do this.

If shooting with a very long lens to get the sun to fill the frame you will be surprised by how fast the sun will move through the frame. So you will need a really good fluid head to allow you to slowly track the sun without jarring or bumping the shot.

When is a DP not a DP?

A post came up in my Facebook feed the other day. It started out with something along the lines of:

“As a brand new DP, what equipment should I buy, then how do I find clients and gain experience”?

This isn’t the first time I’ve seen these kinds of posts and I see many others from “DP’s” that are seeking help with basic skills.

Here’s the thing – the term DP, “Director of Photography” has become totally meaningless. Anybody and everybody with a camera seems to regard themselves as a DP. Once upon a time it took years of experience to work your way up the ladder before you could call yourself a DP, but today owning a camera of some sort appears to be the only qualification required.

There’s two aspects to this.

1: Real, experienced, skilled DP’s no longer want to call themselves DP because the term no longer separates someone with real skills and experience  from someone without either. Real, time served DP’s are almost embarrassed to associate themselves with the new “I’ve got a camera so I’m a DP” DP’s with little or no real world experience.

2: Production companies and employers are somewhat sceptical of anyone that calls themselves a DP because that person could be someone with zero experience or 20 years of experience… who knows!

As a result the term “DP” has really become quite worthless and meaningless which is a great shame.

When I started working as a cameraman in the late 80’s there were several levels of camera operator. There was the basic cameraman, someone that you would normally consider to be a competent camera operator that could focus and expose properly. Compose a shot nicely and shoot a range of different shots that an editor could cut together to tell the story.

Next up was the “Lighting Cameraman”. Typically a lighting cameraman would have had at least a few of years of experience as a professional cameraman and then in addition be capable of self lighting interviews, corporate videos or smaller drama scenes in a pleasing manner using a variety of lights. A lighting cameraman would also have a deeper understanding of contrast ratios, colour balance and the use of filters and gels.

Above this was the “Cinematographer”. A cinematographer would have all of the skills of a lighting cameraman, but would generally be someone working in drama or on narrative based productions rather than factual productions. As a result a cinematographer would normally also have a good understanding of a wide range of different grip and support equipment as well as being familiar with a wider range of types of lights compared to a lighting cameraman. A cinematographer would typically work with a gaffer or electrician when lighting rather than doing it all himself as in the case of a lighting cameraman.

Then comes the Director of Photography. It’s interesting to note that Vittorio Storaro doesn’t like this term because he believes there can be only ever be one director on a shoot, never 2.

A DP or DoP used to mean an extremely skilled and experienced cinematographer that was in charge of a camera crew or camera crews for a larger production. The DP often does not actually operate the cameras, instead the DP instructs and guides the camera operators on what to do. The DP is the “director” of the camera department and as with most supervisory roles the most experienced and skilled person in the camera department.

It’s such a shame that these job titles are now largely meaningless. In my day you had to earn the right to call yourself a lighting cameraman, cinematographer or DP and that took years. It was highly unusual to find cameramen under the age of 20 as you would normally have need to have worked as an assistant first. Most lighting cameramen and cinematographers were in their 30’s because it took time to gain the skills and experience that producers expected.

 

 

Version 9 Firmware for the Sony PMW-F5 and PMW-F55 Released.

Looks like I will have a busy day today upgrading cameras. As well as the re-release of Version 4 for the PXW-FS5, Sony have also released version 9 for the PMW-F5 and PMW-F55 cameras.

Version 9 adds some additional high speed frame rates when recording with the R7 recorder. It also adds extra parallel recording functions when using the CBK-WA100 Wireless Adapter and interestingly also adds Long GoP recording (XAVC-L) when shooting at 29.97fps and 59.94fps (sadly no 24,25 or 50fps Long GoP).

The new firmware can be downloaded from here:

PMW-F5

https://www.sony.co.uk/pro/support/software/SET_BPE-SS-1038/210

PMW-F55

https://www.sony.co.uk/pro/support/software/SET_BPE-SS-1039/200

FROM SONY: IMPORTANT NOTE:

Please install Version 9 ONLY if your F5 or F55 camera has been successfully updated to Firmware Version 8 or higher. Otherwise, it is very important to note that for the F5 and F55 cameras with the serial numbers range listed in the Release Notes document, the user cannot perform the Upgrade to Version 9 and should contact the Sony Service.

NEW FEATURES:

1) Frame rates for 4K and high frame rate recordings added.
72, 75, 90, 96, and 100 FPS have been added to the available values in “Frame Rate” when the AXS-R7 is attached to the PMW-F55.

2) Compatible with the “Parallel Rec” mode with CBK-WA100 attached.
The “Parallel Rec” mode enables synchronization with the same file name between the XAVC Proxy recording by using the wireless adapter  CBK-WA100.

3) XAVC HD Long added (when the system frequency is set to 29.97 or 59.94).

PXW-FS5 Firmware Version 4.02 Released.

Sony have just released firmware version 4.02 for the PXW-FS5. This firmware fixes the bugs found by Sony in the initial release of the version 4 firmware and includes the new Hybrid Log Gamma picture profile No. 10 along with a change to the cameras base ISO rating. I note that there is no mention of the problems with HLG clips in Adobe Premiere, so this will require further testing to see if this has been fixed.

The firmware can be downloaded from here:

https://www.sony.co.uk/pro/support/software/SW_122115_PSG/50

From Sony:

Ver4.02 (Functionally, it is the same as the Ver.4.00.)

V4.02 fixes the following issue:
1.      Video image may be recorded with short delay of 2 or 3 frames of audio in other recording modes than AVCHD.
2.      When choosing [HLG1],[HLG2] or [HLG3] in the PictureProfile and CENTER SCAN in the CAMERA/PAINT menu, rebooting the camera may cause brightness and color shift.

Ver4.00(For your information)
1.    Support for High Dynamic Range (HDR) by shooting in Hybrid Log-Gamma** (HLG) standard
2.    Support for continuous 120fps High Frame Rate (HFR) recording in 1080p with CBKZ-FS5HFR (sold separately)
3.    Option to change the minimum ISO sensitivity number to ISO 2000 from ISO 3200 when recording S-Log2/S-Log3

Downloadable User Guides For The PXW-FS7 and FS7M2

quick-guide Downloadable User Guides For The PXW-FS7 and FS7M2

I was recently asked by Sony to write a user guide for the PXW-FS7 and FS7M2. Well it’s now complete and available for free download from Sony. The guide does not replace the manual but should act as a useful point of reference for those unfamiliar with the cameras. It should also help guide you through the use of the CineEI mode or change the various gammas settings in custom mode to suit different types of scene.

There are sections on exposure tools and controls, the variable ND filter, exposure tools and controls. Custom mode paint settings, Cine EI and LUT’s and additional information on the various shooting modes and functions.

There are two versions of the guide. One is an ePub book that can be displayed and read by may book reader programs such as iBooks and the other is an interactive PDF formatted for use on a mobile phone or tablet.

Click here for the ePub version and Click here for the PDF version.

What is HLG and what is it supposed to be used for?

While we wait for Sony to re-release the version 4 firmware for the FS5 I thought I would briefly take a look at what HLG is and what it’s designed to do as there seems to be a lot of confusion.

HLG stands for Hybrid Log Gamma. It is one of the gamma curves used for DISTRIBUTION of HDR content to HDR TV’s that support the HLG standard. It was never meant to be used for capture, it was specifically designed for delivery.

As the name suggests HLG is a hybrid gamma curve. It is a hybrid of Rec-709 and Log. But before you get all excited by the log part, the log used by HLG is only a small part of the curve and it is very agressive – it crams a very big dynamic range into a very small space – This means that if you take it into post production and start to fiddle around with it there is a very high probability of problems with banding and other similar artefacts becoming apparent.

The version of HLG in the FS5 firmware follows the BBC HLG standard (there is another NHK standard). From black to around 70% the curve is very similar to Rec 709, so from 0 to 70% you get quite reasonable contrast. Around 70% the curve transitions to a log type gamma allowing a dynamic range much greater than 709 to be squeezed into a conventional codec. The benefit this brings is that on a conventional Rec-709 TV the picture doesn’t look wrong. It looks like a very slightly darker than normal, only slightly flat mid range, but the highlights are quite flat and  washed out. For the average home TV viewer watching on a 709 TV the picture looks OK, maybe not the best image ever seen, but certainly acceptable.

However feed this same signal to an HDR TV that supports HLG and the magic starts to happen. IF the TV supports HLG (and currently only a fairly small proportion of HDR TV’s support HLG. Most use PQ/ST2084) then the HLG capable HDR TV will take the compressed log highlight range and stretch it out to give a greater dynamic range display. The fact that the signal gets stretched out means that the quality of the codec used is critical. HLG was designed for 10 bit distribution using HEVC, it was never meant to be used with 8 bit codecs, so be very, very careful if using it in UHD with the FS5 as this is only 8 bit.

So, HLG’s big party trick is that it produces an acceptable looking image on a Rec-709 TV, but also gives an HDR image on an HDR TV. So one signal can be used for both HDR and SDR giving what might be called backwards compatibility with regular SDR TV’s. But it is worth noting that on a 709 TV HLG images don’t look as good as images specifically shot or graded for 709. It is a bit of a compromise.

What about the dynamic range? High end HDR TV’s can currently show about 10 stops. Lower cost HDR TV’s may only be able to show 8 stops (compared to the 6 stops of a 709 TV). There is no point in feeding a 14 stop signal to a 10 stop TV, it won’t look the best. From what I’ve seen of the HLG curves in the FS5 they allow for a maximum of around 10 to 11 stops, about the same as the cinegammas. HLG can be used for much greater ranges, but as yet there are no TV’s that can take advantage of this and it will be a long tome before there are. So for now, the recorded range is a deliberately limited so you don’t see stuff in the viewfinder that will never be seen on todays HDR TV’s.  As a result the curves don’t use the full recording range of the camera. This means they are not using the recording data in a particularly efficient way, a lot of data is unused and wasted. But this is necessary to make the curves directly compatible with an HLG display.

What about grading them? My advice – don’t try to grade HLG footage. There are three problems. The first is that the gamma is very different in the low/mid range compared to the highlights. This means that in post the shadows and mid range will respond to corrections and adjustments very differently to the high range. That makes grading tricky as you need to apply separate correction to the midrange and highlights.

The second problem is that the is a very large highlight range squeezed into a very small recording range. It should look OK when viewed directly with no adjustment. But if you try stretching that out to make the highlights brighter (remember they never reach 100% as recorded) or to make them more contrasty, there is a higher probability of seeing banding artefacts than with any other gamma in the camera.

The third issue is simply that the limited recording range means you have fewer code values per stop than regular Rec-709, the cinegammas or S-Log2. HLG is the least best choice for grading in the FS5.

Next problem is color. Most HDR TV’s want Rec-2020 color. Most conventional monitors want Rec-709 color. Feed Rec-2020 into a 709 monitor and the colors look flat and the hues are all over the place, especially skin tones. Some highly saturated colors on the edge of the color gamut may pop out more than others and this looks odd.

Feed 709 into a 2020 TV and it will look super saturated and once again the color hues will be wrong. Also don’t fool yourself into thinking that by recording Rec2020 you are actually capturing more. The FS5 sensor is designed for 709. The color filters on the sensor do work a little beyond 709, but nowhere near what’s needed to actually “see” the full 2020 color space. So if you set the FS5 to 2020 what you are capturing is only marginally greater than 709. All you really have is the 709  with the hues shifted and saturation reduced so color looks right on a 2020 monitor or TV.

So really, unless you are actually feeding an Rec 2100 (HLG + 2020) TV, there is no point in using 2020 color as this require you to grade the footage to get the colors to look right on most normal TV’s and monitors. As already discussed, HLG is far from ideal for grading, so better to shot 709 if that’s what your audience will be using.

Don’t let the hype and fanfares that have surrounded this update cloud your vision. HLG is certainly very useful if you plan to directly feed HDR to a TV that supports HLG. But if you plan on creating HDR content that will be viewed on both HLG TV’s and the more common PQ/ST2084 TV’s then HLG is NOT what you want. You would be far – far better off shooting with S-Log and then grading your footage to these two very different HDR standards. If you try to convert HLG to PQ it is not going to look nearly as good as if you start with S-Log.

Exposure levels: If you want to get footage that works both with an HLG HDR TV and a SDR 709 TV then you need to expose carefully. A small bit of over exposure wont hurt the image when you view it on a 709 TV or monitor, so it will look OK in the viewfinder. But on an HDR TV any over exposure could result in skin tones that look much too bright and an image that is unpleasantly bright. As a guide you should expose diffuse 90% white (a white card or white piece of paper) at no more than 75%. Skin tones should be around 55 to 60%. You should not expose HLG as brightly as you do Rec-709.

Sure you can shoot with HLG for non HDR applications. You will get some slightly flat looking footage with rolled off highlights. If that’s the image you want then I’m not going to stop you shooting that way. If that’s what you want I suggest you consider the Cinegamma as these capture a similar DR also have a nice highlight roll off (when exposed correctly) and do use the full recording range.

Whatever you do make sure you understand what HLG was designed for. Make sure you understand the post production limitations and above all else understand that it absolutely is not a substitute for S-log.

Sony suspends release of PXW-FS5 firmware version 4.

I don’t think I have ever seen this before.

FS5-1024x537 Sony suspends release of PXW-FS5 firmware version 4.
Sony halts the roll-out of
the version 4 PXW-FS5 firmware update.

Just a few days after the release of firmware version 4.0 for the PXW-FS5 and Sony have taken it down from their websites and suspended the release until some time in early August.

So far no explanation has been given and understandably those that have already applied the update to their cameras are somewhat concerned as there is no way to roll back the firmware.

Clearly something is up, but I don’t know what. To halt the firmware release it must be something of some significance, but I don’t know what that might be (Update: One possibility is that it could be due to the issue with playback in Adobe Premiere).

I have sent emails to my contacts in Sony, but have yet to receive a reply. I know some of them are on holiday. If I hear anything I can share I will let you know.

In the mean time… I updated my FS5 on the 20th. I have been using it every day since then and so far nothing bad or unusual has happened. Nor have I heard any reports of anything that may be of concern. So I would not panic if you have already done the update. I doubt it does any harm to the camera.

 

 

PXW-FS5, Version 4.0 and base ISO – BEWARE if you use ISO!!

The new version 4.0 firmware for the PXW-FS5 brings a new lower base ISO range to the camera. This very slightly reduces noise levels in the pictures. If you use “gain” in dB to indicate your gain level, then you shouldn’t have any problems, +6dB is still +6dB and will be twice as noisy as 0dB. However if you use ISO to indicate your gain level then be aware that as the base sensitivity is now lower, if you use the same ISO with version 4 as you did with version 3 you will be adding more gain than before.

Version 3 ISO  in black, version 4 ISO in Blue

Standard 1000 ISO – 800 ISO
Still 800 ISO- 640 ISO
Cinegamma 1  800 ISO – 640 ISO
Cinegamma 2  640 ISO – 500 ISO
Cinegamma 3  1000 ISO – 800 ISO
Cinegamma 4  1000 ISO – 800 ISO
ITU709 1000 ISO – 800 ISO
ITU709(800) 3200 ISO – 2000 ISO
S-Log2 3200 ISO – 3200/2000 ISO
S-Log3 3200 ISO- 3200/2000 ISO

At 0dB or the base ISO these small changes (a little under 3dB) won’t make much difference because the noise levels are pretty low in either case. But at higher gain levels the difference is more noticeable.

For example if you  often used Cinegamma 1 at 3200 ISO with Version 3 you would be adding 12dB gain and the pictures would be approx 4x noisier than the base ISO.

With Version 4, 3200 ISO with Cinegamma 1 is an extra 15dB gain and you will have pictures approx 6 time noisier than the base ISO.

Having said that, because 0dB in version 4 is now a little less noisy than in version 3, 3200 ISO in V3 looks quite similar to 3200 ISO in version 4 even though you are adding a bit more gain.

Camera setup, reviews, tutorials and information for pro camcorder users from Alister Chapman.