In this modern age where almost any look can be created in post I find that there is still something extremely satisfying about creating as much of the final look of your content in camera as possible. And one thing that can make a huge difference is optical filtration. As camera resolution continues to increase one type of filter that I find particularly useful is the diffusion filter. Diffusion filters can help take the digital edge off an electronic camera. They do this by causing some of the light passing through the filter to scatter which has a softening and contrast reducing effect, especially around highlights. By using different materials to scatter the light the effect can be coloured or modified for different looks. A little bit of diffusion can really help to tame difficult highlights.
For a recent workshop where we had a scene that was designed to give the feel of an old Edwardian study, I decided I wanted to create a very romantic look. So I after playing with a couple of different diffusion filters I settled on a Formatt Hitech 1/4 Soft Gold filter. This filter is really nice for this type of shot as it adds a warm golden glow to high contrast areas. It also makes skin tones look a little richer.
If you compare the first image which was shot without the filter and the other images that were shot with the filter, while I am happy with the shot without the filter, I really do feel that the filter transforms the shot into something that looks more romantic and has an “older” feel to it. Perhaps the 1/8th version of the filter might have been a better choice for a less strong effect, but this really is a filter I like a lot.
The scene without any additional filtration (click on the image to enlarge it).A shot from the scene with the Formatt Hitech 1/4 soft gold filter (click on the image to enlarge it).Another shot from the scene with the Formatt Hitech 1/4 soft gold filter (click on the image to enlarge it).
The extra glow around the candles really enhances the sense of the candles being a part of the lighting while the softening of the highlights on the actors faces helps to make the images look more organic and less digital.
In this shot the only light was from the candles and the 1/4 Soft Gold filter makes the scene feel very warm and cosy.
Well, I was recently asked if I could come up with a rig to do the same using Sony cameras for an upcoming blockbuster feature with an A-list director being shot by a top DP. This kind of challenge is something I enjoy immensely, so how could I not accept the challenge! I had some insight into how Hoyte Van Hoytema did it but I had none of the fine details and often its the fine details that make all the difference. And this was no exception. I discovered many small things that need to be just right if this process is to work well. There are a lot of things that can trip you up badly.
So a frantic couple of weeks ensued as I tried to learn everything I could about infrared photography and video and how it could be used to improve traditional day for night shooting. I don’t claim any originality in the process, but there is a lot of information missing about how it was actually done in Nope. I have shot with infrared before, so it wasn’t all new, but I had never used it this way before.
As I did a lot of 3D work when 3D was really big around 15 years ago, including designing award winning 3D rigs, I knew how to combine two cameras on the same optical axis. Even better I still had a suitable 3D rig, so at least that part of the equation was going to be easy (or at least that’s what I thought).
Building a “Test Mule”.
The next challenge was to create a low cost “test mule” camera before even considering what adaptations might be needed for a full blown digital cinema camera. To start with this needed to be cheap, but it also needed to be full frame and capable of taking a wide range of cinema lenses and sensitive to both visible and infrared light. So, I took an old A7S that had been gathering dust for a while, dismantled it and removed the infrared filter from the sensor.
A7S being modified for infrared (full spectrum).Panavised and Infrared sensitive A7S with Panavision Primo lens.
As the DP wanted to test the process with Panavision lenses the camera was fitted with a PV70 mount and then collimated in it’s now heavily modified state (collimation has some interesting challenges when working with the very different wavelength of infrared light compared to visible). Now I could start to experiment, pairing the now infrared sensitive A7S with a second camera on the 3D rig. We soon found issues with this setup, but it allowed me to take the testing to the next stage before committing to modifying a more expensive camera for infrared.
This testing was needed to determine exactly what range of infrared light would produce the best results. The range of infrared you use is determined by filters added to the camera to cut the visible light and only pass certain parts of the infrared spectrum. There are many options, different filters work in slightly different ways. And not only do you need to test the infrared filters but you also need to consider how different neutral density filters might behave if you need to reduce the IR and visible light. Once I narrowed down the range of filters I wanted to test the next challenge was find very high quality filters that could either be fitted inside the camera body behind the lens or that were big enough (120mm +) for the Panavision lenses that were being considered for the film.
Once I had some filters to play with (I had 15 different IR filters) the next step was to start test shooting. I cheated here a bit. For some of the initial testing I used a pair of zoom lenses as I was pairing the A7S with several different cameras for the visible spectrum. The scan areas of the different sensors in the A7S and the visible light cameras were typically very slightly different sizes. So, a zoom lens was used to provide the same field of view from both cameras so that both could be more easily optically aligned on the 3D rig. You can get away with this, but it makes more work for post production as the distortions in each lens will be different and need correcting. For the film I knew we would need identical scan sizes and matched lenses, but that could come later once we knew how much camera modification would be needed. To start with I just needed to find out what filtration would be needed.
At this point I shot around 100 different filter and exposure tests that I then started to compare in post production. When you get it all just right the sky in the infrared image becomes very dark, almost black and highlights become very “peaky”. If you use the luminance from the infrared camera with its black sky and peaky highlights and then add in a bit of colour and textural detail from the visible camera it can create a pretty convincing day for night look. Because you have a near normal visible light exposure you can fine tune the mix of infrared and visible in post production to alter the brightness and colour of the final composite shot giving you a wide range of control over the day for night look.
So – now I know how to do it, the next step was to take it from the test mule to a pair of matching cinema quality cameras and lenses for a full scale test shoot. When you have two cameras on a 3D rig the whole setup can get very heavy, very fast. Therefore the obvious camera to adapt was a Sony Venice 2 with the 8K sensor as this can be made very compact by using the Rialto unit to split the sensor from the camera body – In fact one of the very first uses of Rialto was for 3D shooting on Avatar – The Way of Water.
With a bit of help from Panavision we adapted a Panavised Venice 2, making it full spectrum and then adding a carefully picked (based on my testing) special infrared filter into the cameras optical path. This camera was configured using a Rialto housing to keep it compact and light so that when placed on the 3D rig with the visible light Venice the weight remained manageable. The lenses used were Panavision PV70 Primo’s (if you want to use these lenses for infrared – speak to me first, there are some things you need to know).
3D rig with an Infrared capable Venice Rialto and normal Venice 2 with Panavision Primo lenses.
And then with the DP in attendance, with smoke and fog machines, lights and grip we tested. For the first few shot we had scattered clouds but soon the rain came and then it poured down for the rest of the day. Probably the worst possible weather conditions for a day for night shoot. But that’s what we had and of course for the film itself there will be no guarantee of perfect weather.
Testing the complete day for night IR rig.
Testing how smoke behaves in infrared. Different types of smoke and haze and different types of lights behave very differently in infrared.
The large scale tests gave us an opportunity to test things like how different types of smoke and haze behave in infrared and also to take a look at interactions with different types of light sources. With the right lights you can do some very interesting things when you are capturing both visible light and infrared opening up a whole new world of possibilities for creating unique looks in camera.
From there the footage went to the production companies post production facilities to produce dailies for the DP to view before being presented to the studios post production people. Once they understood the process and were happy with it there was a screening for the director along with a number of other tests for lighting and lenses.
Along the way I have learnt an immense amount about this process and how it works. What filters to use and when, how to adapt different cameras, how different lenses behave in the infrared spectrum (not all lenses can be used). Collimating adapted cameras for infrared is interesting as many of the usual test rigs will produce misleading or confusing results. I’ve also identified several other ways that a dual camera setup can be used to enhance shooing night scenes, both day for night and as well as at night, especially for effects heavy projects.
At the time of writing it looks like most of the night scenes in this film will be shot at night, they have the budget and time to do this. But the director and DP have indicated that there are some scenes where they do wish to use the process (or a variation of it), but they are still figuring out some other details that will affect that decision.
Whether it gets used for this film or not I am now developing a purpose designed rig for day for night with infrared as I believe it will become a popular way to shoot night scenes. My cameras of choice for this will be a pair of Venice cameras. But other cameras can be used provided one can be adapted for IR and both can be synchronised together. I will have a pair of Sony F55’s, one modified for IR available for lower budget productions and a kit to reversibly adapt a Sony Venice. If you need a rig for day for night and someone that knows exactly how to do it, do get in touch!
I’m afraid I can’t show you the test results, that content is private and belongs to the production. The 3D rig is being modified as you don’t need the ability to shoot with the cameras optically separated, removing the moving parts will make the rig more stable and easier to calibrate. Plus a new type of beam splitter mirror with better infrared transmission properties is on the way. As soon as I get an opportunity to shoot a new batch of test content with the adapted rig I will share it here.
What do the zebras measure when shooting S-Log3 using the CineEI modes in the FX3 and FX30?
The convention for zebras with the majority of cameras is that zebras are a viewfinder applied measurement. As such they almost always measure the “viewfinder” image. As the LCD on the FX series cameras is in effect the viewfinder, the zebras measure what you see on that screen. So, when you have a LUT on, the zebras measure the LUT, not the S-Log3.
Common ways to use the zebras include measuring skin tones, which for the default s709 LUT will be somewhere in the region of 60% depending on the face brightness. You could also use the LUT’s to measure the brightness of a white card or white piece of paper which should be around around 81% for a proper white card or 83% for white paper.
You could also use Zebras to indicate when you are close to clipping Depending on the LUT that you are using the peak LUT output will typically be at 100%, so a common usage would be to have Zebra 2 (which measures from the zebra point and everything above) set to a touch below 100 to act as a clipping indicator. BUT it must also be remembered that depending on the Exposure Index in most cases the LUT will have a lower highlight range than the S-Log3 recordings. So, when your highlights hit 100% on the LUT there may still be available headroom in the S-Log3 recordings. If you end up backing off your exposure every time the LUT clips you may be missing out on the full recording range and un-necessarily bringing the mids and shadows down. So, my preference is to measure the exposure of a white card or skin tones and to get the mid range and shadows right, rather than obsessing over small amounts clipping.
The s709 LUT does fit the full highlight range of the S-Log3 into it’s output. But as there is only a tiny difference between +5 and +6 stops (approx 1.5%) it is very difficult to determine what is clipped and what is 1 stop below clipping. +4 stops above middle grey is output at 93% and +6 above middle grey is at 98% so it becomes very difficult to see what is really going on in the highlights via the LUT when the top 2 stops are crammed into just 5% of the recording range. Zebra 2 set to 95% (for example) would appear over 1.5 stops below clipping, even if set to 97% zebras will show almost a full stop below clip.
It is one of the frustrations of the FX3/FX30 that there is no way to monitor via the LUT and measure the S-Log3 at the same time.
Over the last 2 weeks I have been shooting some tests for a major feature film. The tests involved a special process that involves the use of Infrared light and shooting outdoors.
On the test day we had some fairly bright light levels to deal with. So as you would normally do we added some ND filtration to reduce the light levels. Most of the equipment for the shoot was on hire from Panavision, the main cameras being Panavised Sony Venices with PV70 mounts and Panavison lenses. But for reasons I can’t go into yet, we were unable to use the Venice internal ND filters, so we had to use external ND’s.
The first ND’s we used were circular Tiffen IRND’s that were the correct size for the PV lenses. But much to my surprise these made very little difference to the amount of IR reaching the camera. For our application they were absolutely no good. Fortunately, I had a set of Formatt Hitech IRND’s in my camera bag and when we tried these we got an equal visible and infrared cut. So, the Tiffen’s were put back in their boxes and the Formatt filters used instead.
Back at Panavision we did some further testing and found that both the Tiffen and Schnieder IRND’s that we tested had very little IR cut. But the Formatt Hitech and Panavision IRND’s that we tested cut the IR by a very similar amount to the visible light. In addition we were able to test the Venice built in ND filters and found that these too did a very good job at cutting both IR and visible light by similar amounts.
So, my recommendation is – if you are ever concerned about infrared light contaminating your images use a Venice 2 with it’s built in ND’s, Panavision or Formatt Hitech IRND’s.
For some reason many people now believe that the only way you can shoot with S-Log3 is by “over exposing” and very often by as much as almost 2 stops (1.7 stops is often quoted).
When Sony introduced the original A7S, the FS5, F5, F55 and FS7 shooting S-Log3 with these cameras was a little tricky because the sensors were quite noisy when used at the relatively high base ISO’s of these cameras. When exposed according to Sony’s recommendation of 41% for middle grey and 61% for a white card the end result would be fairly noisy unless you added a good amount of post production noise reduction. As a result of this I typically recommended exposing these particular cameras between 1 and 2 stops brighter than the base level. If using the F5 or FS7 I would normally use 800EI which would lead to an exposure +1.3 stops brighter than base. This worked well with these cameras to help control the noise, but did mean a 1.3 stop loss of highlight range. In other examples I used to recommend exposing a white card at white at 70% which would equate to an exposure a touch over 1 stop brighter than the base level.
With the introduction of the original Venice camera and then the FX9 we got a new generation of much lower noise sensors with dual base ISO’s. It soon became clear to me that these new cameras didn’t normally need to be exposed more brightly than the Sony recommended levels when using their low base ISO’s and even at their high base ISO’s you can typically get perfectly acceptable results without shooting brighter, although sometimes a small amount of over exposure or a touch of noise reduction in pots might be beneficial. No longer needing to expose more brightly brought with it a useful increase in the usable highlight range, something the earlier cameras could struggle with.
Then the A7S3, FX6 and FX3 came along and again at the lower of their base ISO’s I don’t feel it is necessary to shoot extra bright. However at the 12,800 high base ISO there is a fair bit more noise. So I will typically shoot between 1 and 2 stops brighter at the high base ISO to help deal with the extra noise. On the FX6 and FX3 this normally means using between 6400 and 3200 EI depending on the scene being shot.
Even though I and many others no longer advocate the use of extra bright exposures at the lower base ISO’s with these newer cameras it really does surprise me how many people believe it is still necessary to shoot up to 2 stops over. It’s really important to understand that shooting S-Log3 up to 2 stops over isn’t normal. It was just a way to get around the noise in the previous cameras and in most cases it is not necessary with the newer cameras.
Not having to shoot brighter means that you can now use the Viewfinder Display Gamma Assist function in the A7S3, A1 or the FX9 (for those times you can’t use a LUT) to judge your exposure with confidence that if it looks right, it most likely will be right. It also means that there is no longer any need to worry about offset LUT’s or trying to correct exposure in post before applying a LUT.
Of course, you can still expose brighter if you wish. Exposing brighter may still be beneficial in scenes with very large shadow areas or if you will be doing a lot of effects work. Or perhaps simply want an ultra low noise end result. But you shouldn’t be terrified of image noise. A little bit of noise is after all perfectly normal.
And one last thing: I don’t like the use of the term “over exposing” to describe shooting a bit brighter to help eliminate noise. If you have deliberately chosen to use a low EI value to obtain a brighter exposure or have decided to expose 1 stop brighter because you feel this will get you the end result you desire this is not (in my opinion) “over exposure”. Over exposure generally means an exposure that is too bright, perhaps a mistake. But when you deliberately shoot a bit brighter because this gets you to where you want to be this isn’t a mistake and it isn’t excessive, it is in fact the correct exposure choice.
I’m just putting this here in case it is of use to someone stuck with an Invalid Licence message when starting a new installation of Sony’s Catalyst Browse or Prepare software. Typically you get this error if you are migrating the operating system to anew computer via a backup or have uninstalled and then re-installed one of the Catalyst products.
On a Mac
Press and hold Shift + Command (?) while launching the application from the application folder.
You should then be prompted to log in to your Sony Creative Software account, after which the activation should proceed as normal.
If that does not correct this issue, you may need to remove the licence files from you system.
Locate this folder:
/Users/Shared/
This folder contains files that corresponding to the Catalyst products installed on your system and will have the extension .LICENSE. Delete the .license files and retry the online registration process.
Please note that the .license files are hidden. To make them visible, press Shift + Command (?) + Period (.)
On a Windows PC.
Press and hold Shift + the Windows key while launching the application.
This should open a window prompting you to log in to your Sony Creative Software account to activate your software as normal.
If that does not correct this issue, you may need to manually remove the licence files from your system.
Locate the following folder:
C:\ProgramData\Sony\
This folder contains files that correspond with the Catalyst products you have installed on your system and they will have the extension .LICENSE. Delete the .license files and retry the online registration process.
The ProgramData folder is a hidden folder. If you can’t see it, you will need to adjust your Folder Options to allow hidden files, folders, and drives to be displayed.
These are mainly stability releases that fix some minor bugs, but if you have an FX3 on the original version 1 firmware then this version adds the CineEI mode and LUTs. It is a major update that is well worth having.
Before attempting to update the camera you should insert a fully charged battery
The FX3 is updated via a computer application. While there is a Mac application it is a complete pain in the rear end to get it to work and I would urge you to find a windows PC to do the update, it is far simpler and far more likely to be successful. The good news is that once you have updated to version 2.02 future updates can be done by uploading the update file to an SD card and initiating the update from the camera like the FX30.
The FX30 is updated by placing the downloaded BODYDATA.DAT file on to an SD card that was previously formatted in the camera. Then place the card in the camera and go down to the SETUP – SETUP OPTION – VERSION page of the menu. Here you should see the cameras current firmware version plus a “SOFTWARE UPDATE” button. Press (select) the software update button.
On the next page it will say “Update ?” and show the old firmware version and the new firmware version. Then just below this is a box where it says “Please follow these precautions until the very end”.
What isn’t clear at this step is that you need to scroll down inside that box and read the full list of precautions before the camera will allow you to do the update.
Scroll down until you get to “This update may take several minutes, device automatically reboots when complete”. If you don’t scroll down and just press the “Execute” button you get a large popup telling you to “Follow the precautions to the very end” and pressing “OK” simply takes you will go back to the previous page. So do make sure you scroll down through the full list of precautions before you press execute. Once the update starts the screen will go blank, the only clue that the update is happening will be the slow flash of the media LED on the back of the camera. The update takes about 10 minutes to complete and the camera will reboot when it’s done.
Many people wish to bake in the cameras Exposure Index settings when shooting using CineEI in order to avoid having to make an exposure correction in post production (given that the cameras are ISO invariant when shooting Log in reality it makes vey little difference whether you add gain in the camera or in post production – gain is gain). On cameras such as the FS7, FX6 or FX9 one way to do this is by baking in the built in S-Log3 LUT. To avoid confusion – that is using the CineEI mode with the “S-Log3” LUT enabled and in the LUT settings “Internal recording” set to ON so that you are recording the “S-Log3” LUT.
While this will bake in the EI change, this technique comes with many issues. For a start, just as when you use S-Log3 in custom mode and alter the ISO, whenever you move away from the cameras base ISO you loose dynamic range. When you bake in a LUT and change the EI, you are in effect changing the ISO and there will be a corresponding loss of dynamic range. When you bake in a LUT this loss of dynamic range is exacerbated by a reduced or altered recording range.
At lower EI’s the available recording range shrinks as the LUT is made darker and at the same time upper recoding level of the LUT is reduced. At 200 EI the recording range only gets to around 78%. At the bottom end the shadows are crushed and shadow information lost by the range reduction. This then causes a post production issue because LUT’s designed for the normal S-Log3 input range of 0-94% will now be applied to recordings with a much reduced range and after application of a LUT in post the final output won’t get to 100% without further complex grading where the image will need to be stretched more than normal and this degrades quality.
At high EI’s the LUT becomes brighter but the clip point remains the same. So for each stop you go up, 1 stop of highlights just disappears beyond the LUT’s hard clip point and can’t be ever recorded. Again in post this can cause issues because when you apply a normal S-Log3 LUT the heavy clipping in the recording causes the highlights to look very heavily clipped (because they are). Again, for the best results you will need to grade your footage to allow for this.
So, in practice the idea of baking in the S-Log3 LUT to eliminate the need to do any post production corrections doesn’t work because the addition of the S-Log3 LUT introduces new limitations that will need to be corrected if you want good looking images. Plus adding the S-Log3 LUT in camera and then adding another LUT on top in post is never going to deliver the best results due to the way LUT’s divide the image into brightness zones.
And – if you are baking in the S-Log3 LUT, then really this is no longer EI as there is now no longer an offset between the exposure and the recording, you are simply recording at a higher/lower ISO.
To be honest, not many of us have to charge extremely cold batteries, but on my Northern Lights trips my batteries often get very cold and often I will charge them while they are still very cold. I’ve always known that this isn’t health for the battery, but what i didn’t know was that actually it can be quite dangerous.
It turns out that if you charge a lithium battery that is very cold (below 0c/32f) it will appear to charge more or less as expected. But, when the cells are very cold metallic lithium gets deposited on the anode of the cell. If you repeatedly charge the battery at low temperatures the lithium will continue to build up on the anode causing a safety risk. This metallic lithium can cause the cell to become less stable and more prone to bursting into flames if the battery is over charged, gets hot or is shocked such as through being dropped or crushed.
So – all you Aurora chasers and others that shoot in very cold conditions – let your batteries warm up before you charge them. It won’t be obvious that charging them is causing harm and the last thing you want is a battery suddenly bursting into flames on a job some time down the road because you’ve dropped or bumped it.
Sony’s Cinema Line cameras all have the ability to record metadata from inertia and gyroscope sensors about the way the camera moves while shooting. This metadata can then be used to stabilise your footage in post production. The stabilisation that this can provide is normally very good and tends to look a lot more natural than using post production stabilisation that looks at the footage and tries to hold it steady. However, until recently the only way to make use of this metadata was via Sony’s Catalyst Browse software.
Now however an Open Source project known as GyroFlow has made it possible to use the Sony metadata in FCP-X and DaVinci Resolve via an OpenFX plugin and a FCP-X plug-in. In addition there is a standalone GyroFlow application that can stabilise the footage and then export a stabilised version of the clip.
GyroFlow is a collaborative Open Source project, so different developers are working on different aspects and plug-ins, so it is a bit more disjointed than a lot of commercial products. But, it is free and it will get better, so why not give it a try. The main website for the project is here: http://gyroflow.xyz/
This website uses cookies to improve your experience. We'll assume you're ok with this, but you can opt-out if you wish. Read More
Necessary cookies help make a website usable by enabling basic functions like page navigation and access to secure areas of the website. The website cannot function properly without these cookies.
Name
Domain
Purpose
Expiry
Type
wpl_user_preference
www.xdcam-user.com
WP GDPR Cookie Consent Preferences
1 year
HTTP
YSC
youtube.com
YouTube session cookie.
54 years
HTTP
Marketing cookies are used to track visitors across websites. The intention is to display ads that are relevant and engaging for the individual user and thereby more valuable for publishers and third party advertisers.
Name
Domain
Purpose
Expiry
Type
VISITOR_INFO1_LIVE
youtube.com
YouTube cookie.
6 months
HTTP
Analytics cookies help website owners to understand how visitors interact with websites by collecting and reporting information anonymously.
Name
Domain
Purpose
Expiry
Type
__utma
xdcam-user.com
Google Analytics long-term user and session tracking identifier.
2 years
HTTP
__utmc
xdcam-user.com
Legacy Google Analytics short-term technical cookie used along with __utmb to determine new users sessions.
54 years
HTTP
__utmz
xdcam-user.com
Google Analytics campaign and traffic source tracking cookie.
6 months
HTTP
__utmt
xdcam-user.com
Google Analytics technical cookie used to throttle request rate.
Session
HTTP
__utmb
xdcam-user.com
Google Analytics short-term functional cookie used to determine new users and sessions.
Session
HTTP
Preference cookies enable a website to remember information that changes the way the website behaves or looks, like your preferred language or the region that you are in.
Name
Domain
Purpose
Expiry
Type
__cf_bm
onesignal.com
Generic CloudFlare functional cookie.
Session
HTTP
NID
translate-pa.googleapis.com
Google unique id for preferences.
6 months
HTTP
Unclassified cookies are cookies that we are in the process of classifying, together with the providers of individual cookies.
Name
Domain
Purpose
Expiry
Type
_ir
api.pinterest.com
---
Session
---
Cookies are small text files that can be used by websites to make a user's experience more efficient. The law states that we can store cookies on your device if they are strictly necessary for the operation of this site. For all other types of cookies we need your permission. This site uses different types of cookies. Some cookies are placed by third party services that appear on our pages.
