Fantastic day today. Captured a large, strong tornado in 4K raw with my F5. Got some really great footage of the tornado to add to the previous days lightning and storm structure footage. Can’t wait to get into the edit suite to start grading and cutting this footage. Still have 2 more days of storm chasing to go!
Here’s a short sample of some lightning I shot today with my F5. Can’t wait until I get a bit more time to edit together the incredible footage I’ve obtained in 4K raw of severe thunderstorms.
The PMW-F5 and F55 are fantastic cameras. If you have the AXS-R5 raw recorder the dynamic range is amazing. In addition because there is no gamma applied to the raw material you can be very free with where you set middle grey. Really the key to getting good raw is simply not to over expose the highlights. Provided nothing is clipped, it should grade well. One issue though is that there is no way to show 14 stops of dynamic range in a pleasing way with current display or viewfinder technologies and at the moment the only exposure tool you have built in to the F5/F55 cameras are zebras.
My experience over many shoots with the camera is that if you set zebras to 100% and don’t use a LUT (so your monitoring using S-Log2) and expose so that your just starting to see zebra 2 (100%) on your highlights, you will in most cases have 2 stops or more of overexposure headroom in the raw material. Thats fine and quite useable, but shoot like this and the viewfinder images will look very flat and in most cases over exposed. The problem is that S-Log 2’s designed white point is only 59% and middle grey is 32%. If your exposing so your highlights are at 100%, then white is likely to be much higher than than the designed level, which also means middle grey and your entire mid range will be excessively high. This then pushes those mids into the more compressed part of the curve, squashing them all together and making the scene look extremely flat. This also has an impact on the ability to focus correctly as best focus is less obvious with a low contrast image. As a result of the over exposed look it’s often tempting to stop down a little, but this is then wasting a lot of available raw data.
So, what can you do? Well you can add a LUT. The F5 and F55 have 3 LUTS available. The LUTS are based either on REC709 (P1) or Hypergamma (P2 and P3). These will add more contrast to the VF image, but they show considerably less dynamic range than S-Log2. My experience with using these LUT’s is that on every shoot I have done so far, most of my raw material has typically had at least 3 stops of un-used headroom. Now I could simply overexpose a little to make better use of that headroom, but I hate looking into the viewfinder and seeing an overexposed image.
Why is it so important to use that extra range? It’s important because if you record at a higher level the signal to noise ratio is better and after grading you will have less noise in the finished production.
Firmware release 1.13 added a new feature to the F5 and F55, EI Gain. EI or Exposure Index gain allows you to change the ISO of the LUT output. It has NO effect on the raw recordings, it ONLY affects the Look Up Tables. So if you have the LUT’s turned on, you can now reduce the gain on the Viewfinder, HDSDI outputs as well as the SxS recordings (see this post for more on the EI gain). By using EI gain and an ISO lower than the cameras native ISO I can reduce the brightness of the view in the viewfinder. In addition the zebras measure the signal AFTER the application of the LUT or EI gain. So if you expose using a LUT and zebra 2 just showing on your highlights and then turn on the EI gain and set it to 800 on an F5 (native 2000ISO) or 640 on an F55 (native 1250ISO) and adjust your exposure so that zebra 2 is once agin just showing you will be opening your aperture by 1.5 (F5) or 1 (F55) stop. As a result the raw recordings will be 1.5/1 stop brighter.
In order to establish for my own benefit which was the best EI gain setting to use I spent a morning trying different settings. What I wanted to find was a reliable way to expose at a good high level to minimise noise but still have a little headroom in reserve. I wanted to use a LUT so that I have a nice high contrast image to help with focus. I chose to concentrate on the P3 LUT as this uses hypergamma with a grey point at 40% so the mid range should not look underexposed and contrast would be quite normal looking.
When using EI ISO 800 and exposing the clouds in the scene so that zebras were just showing on the very brightest parts of the clouds the image below is what the scene looked like when viewed both in the viewfinder and when opened up in Resolve. Also below is the same frame from the raw footage both before and after grading. You can click on any of the images to see a larger view.
As you can see using LUT P3 and 800 EI ISO (PMW-F5) and zebra 2 just showing on the brightest parts of the clouds my raw footage is recorded at a level roughly 1.5 stops brighter than it would have been if I had not used EI gain. But even at this level there is no clipping anywhere in the scene, so I still have some extra head room. So what happens if I expose one more stop brighter?
So, as you can see above even with zebras over all of the brighter clouds and the exposure at +1 stop over where the zebras were just appearing on the brightest parts of the clouds there was no clipping. So I still have some headroom left, so I went 1 stop brighter again. The image in the viewfinder is now seriously over exposed.
The lower of the 3 images above is very telling. Now there is some clipping, you can see it on the waveform. It’s only on the very brightest clouds, but I have no reached the limit of my exposure headroom.
Based on these tests I feel very comfortable exposing my F5 in raw by using LUT P3 with EI gain at 800 and having zebra 2 starting to appear on my highlights. That would result in about 1.5 stops of headroom. If you are shooting a flat scene you could even go to 640 ISO which would give you one safe stop over the first appearance of zebra 2. On the F55 this would equate to using EI 640 with LUT P3 and having a little over 1.5 stops of headroom over the onset of zebras or EI 400 giving about 1 stop of headroom.
My recommendation having carried out these tests would be to make use of the lower EI gain settings to brighten your recorded image. This will result in cleaner, lower noise footage and also allow you to “see” a little deeper into the shadows in the grade. How low you go will depend on how much headroom you want, but even if you use 640 on the F5 or 400 on the F55 you should still have enough headroom above the onset of zebra 2 to stay out of clipping.
This came up as a question in response to the post about my prototype lens adapter. The adapter is based around an electronic Canon EF mount and the question was, what do I think about DSLR zooms?
There is a lot of variation between lenses when it comes to sharpness, contrast and distortions. A zoom will always be a compromise compared to a prime lens. DSLR lenses are designed to work with 24MP sensors. A 4K camera only has around 9MP, so your working well within the design limits of the lens even at 4K. While a dedicated PL mount zoom like an Angenieux Optimo will most likely out perform a similar DSLR zoom. The difference at like for like apertures will not be huge when using smaller zoom ratios (say 4x). But 10x and 14x zooms make more compromises in image quality, perhaps a bit of corner softness or more CA and these imperfections will be better or worse at different focal lengths and apertures. At the end of the day zooms are compromises but for many shoots it may simply be that it is only by accepting some small compromises that you will get the shots you want. Take my storm chasing shoots. I could use primes and get better image quality, but when you only have 90 seconds to get a shot there simply isn’t time to swap lenses, so if you end up with a wide on the camera when a long lens is what is really needed, your just not going to get the shot. Using a zoom means I will get the shot. It might not be the very best quality possible but it will look good. It is going to be better than I could get with an HD camera and a very slightly compromised shot is better than no shot at all.
If the budget would allow I would have a couple of cameras with different prime lenses ready to go. Or I would use a big, heavy and expensive PL zoom and have an assistant or team tasked solely with getting the tripod set up and ready asap. But my budget isn’t that big. I could spend weeks out storm chasing before I get a decent shot, so anything I can do to minimise costs is important.
It’s all about checks and balances. It is a compromise, but a necessary one. It’s not a huge compromise as I suspect the end viewer is not going to look at the shot and say “why is that so soft” unless they have a side by side, like for like shot to compare. DSLR zooms are not that bad! So yes, using a DSLR zoom is not going to deliver quality to match that of a similar dedicated PL zoom in most cases, but the difference is likely to be so small that the end viewer will never notice and thats a compromise I’m prepared to accept in order to get a portable camera that shoots 4K with a 14x zoom lens.
What about DSLR primes and why have I chosen the Canon Mount?
This is where the image performance gap gets even narrower. A high quality DSLR prime can perform just as well as many much more expensive PL mount lenses. The difference here is more about the usability of the lens. Some DSLR lenses can be tiny and this makes them fiddly to use. They are all All sorts of sizes, so swapping lenses may mean swapping Matte boxes or follow focus positions etc. Talking of focus, very often the focus travel on a DSLR lens is very, very short so focussing is fiddly. If the lens has an aperture ring it will probably have click stops making smooth aperture changes mid shot difficult. My prime lenses are de-clicked or never had clicks in the first place (like the Samyang Cine Primes). It’s not so much the issue of requiring a finer step than the one stop click, but more the ability to pull aperture during the shot. It’s not something I need to do often, but if I suddenly find I need to do it, I want a smooth aperture change. That being said, one of the issues with using Canon EF lenses with their electronic iris is that they operate in 1/8th stop steps and this is visible in any footage. Ultimately I am still committed to using the Canon mount lenses simply because there are so many to choose from and they focus in the right direction unlike Nikon lenses which focus back to front. For primes I’m using the excellent and fully manual Samyang T1.5 Cine Primes. I really like these lenses and they produce beautiful images at a fraction of the price of a PL mount lens. My zoom selection is a bit of a mish-mash. One thing about having a Canon mount on the camera is that I can still use Nikon lenses if I fit the lens with a low cost Nikon to Canon adapter ring. If you do this you can only use lenses with an actual iris ring, so generally these are slightly older lenses, but for example I have a nice Sigma 24-70mm f2.8 with a manual iris ring (and it focusses the RIGHT way, like most Sigmas but unlike most Nikon mount lenses). In addition I have a 70-300mm f4 Nikon mount Sigma as well as an Old Tokina 28-70mm f2.6 (lovely lens, a little soft but very nice warm colour). One thing I have found is that most of the Nikon to Canon adapter rings are little bit on the thin side. This prevents any zooms from being Parfocal as it puts the back focus out. Most of the adpaters are made in two parts and it’s quite easy to take the front and back parts apart and add shims made out of of thin plastic sheet or even card between the two halves to correct the back focus distance.
So there you have it. Overall DSLR lenses are not a huge compromise. Of course I would love to own a flight case full of good quality PL mount, 4K ready, glass. Perhaps one day I will, but it’s a serious investment. Currently I use DSLR lenses for my own projects and then hire in better glass where the budget will allow. For any commercials or features this normally means renting in a set of Ultra Primes or similar. I am keeping a close eye on the developments from Zunow. I like their 16-28mm f2.8 and the prototype PL primes I saw at NAB look very good. I also like the look of the Zeiss 15.5 to 45 light weight zoom. Then of course there is the excellent Fujinon 19-90mm Cabrio servo zoom, but these are all big bucks. Hopefully I’ll get some nice big projects to work on this year that will allow me to invest in some top end lenses.
I’m running an in depth and extensive workshop on Sony’s PMW-F5 and F55 cameras in LA on the 4th of June at AbelCine. This workshop is for DP’s and DiT’s and will look at how to setup and configure the cameras in their various modes. Lens choices and lens options, rigs and accessories as well as exposure levels with standard/Hypergammas, Log and RAW.
More Details below. You can book a place HERE.
The Sony PMW-F5 and PMW-F55 both offer a Super 35mm, CMOS image sensor, wide dynamic range, high sensitivity, and an extensive variety of internal recording options. DoP and DIT, Alister Chapman, has put together a course on how to use these powerful tools in a variety of ways, including Raw recording, Log Recording, and Video with Scene Files.
This workshop covers the many functions of the F5 and F55, including detailed instruction on camera operation, lens choice, production techniques, and aftermarket accessories. We will discuss the similarities and differences of the these two cameras as well as the AXS-R5 Raw Recorder, viewfinders, lenses and 4K monitors. Emphasis is on developing an understanding of the camera sensor, internal recording options, lens choice, and menu systems.
Overview
Part 1: Camera Functionality
Part 2: Practical Appilcation
One of the most common questions I get asked is “which gamma curve should I use?”.
Well it’s not an easy one to answer because it will depend on many things. There is no one-fits-all gamma curve. Different gamma curves offer different contrast and dynamic ranges.
So why not just use the gamma curve with the greatest dynamic range, maybe log? Log and S-Log are also gamma curves but even if you have Log or S-Log it’s not always going to be the best gamma to use. You see the problem is this: You have a limited size recording bucket into which you must fit all your data. Your data bucket, codec or recording medium will also effect your gamma choice.
If your shooting and recording with an 8 bit camera, anything that uses AVCHD or Mpeg 2 (including XDCAM), then you have 235 bits of data to record your signal. A 10 bit camera or 10 bit external recorder does a bit better with around 940 bits of data, but even so, it’s a limited size data bucket. The more dynamic range you try to record, the less data you will be using to record each stop. Lets take an 8 bit camera for example, try to record 8 stops and that’s about 30 bits per stop. Try to extend that dynamic range out to 11 stops and now you only have about 21 bits per stop. It’s not quite as simple as this as the more advanced gamma curves like hypergammas, cinegammas and S-Log all allocate more data to the mid range and less to highlights, but the greater the dynamic range you try to capture, the less recorded information there will be for each stop.
In a perfect world you would choose the gamma you use to match each scene you shoot. If shooting in a studio where you can control the lighting then it makes a lot of sense to use a standard gamma (no knee or knee off) with a range of up to 7 stops and then light your scene to suit. That way you are maximising the data per stop. Not only will this look good straight out of the camera, but it will also grade well provided your not over exposed.
However the real world is not always contained in a 7 stop range, so you often need to use a gamma with a greater dynamic range. If your going direct to air or will not be grading then the first consideration will be a standard gamma (Rec709 for HD) with a knee. The knee adds compression to just the highlights and extends the over-exposure range by up to 2 or 3 stops depending on the dynamic range of the camera. The problem with the knee is that because it’s either on or off, compressed or not compressed it can look quite electronic and it’s one of the dead giveaways of video over film.
If you don’t like the look of the knee yet still need a greater dynamic range, then there are the various extended range gammas like Cinegamma, Hypergamma or Cinestyle. These extend the dynamic range by compressing highlights, but unlike the knee, the amount of compression starts gradually and get progressively greater. This tends to look more film like than the on/off knee as it tends to roll off highlights much more gently. But, to get this gentle roll-off the compression starts lower in the exposure range so you have to be very careful not to over expose your mid-range as this can push faces and skin tones etc into the compressed part of the curve and things won’t look good. Another consideration is that as you are now moving away from the gamma used for display in most TV’s and monitors the pictures will be a little flat so a slight grade often helps with these extended gammas.
Finally we come to log gammas like S-Log, C-Log etc. These are a long way from display gamma, so will need to be graded to like right. In addition they are adding a lot of compression (log compression) to the image so exposure becomes super critical. Normally you’ll find the specified recording levels for middle grey and white to be much lower with log gammas than conventional gammas. White with S-Log for example should only be exposed at 68%. The reason for this is the extreme amount of mid to highlight compression, so your mid range needs to be recorded lower to keep it out of the heavily compressed part of the log gamma curve. Skin tones with log are often in the 40 – 50% range compared to the 60-70% range commonly used with standard gammas. Log curves do normally provide the very best dynamic range (apart from raw), but they will need grading and ideally you want to grade log footage in a dedicated grading package that supports log corrections. If you grade log in your edit suite using linear (normal gamma) effects your end results won’t be as good as they could be. The other thing with log is now your recording anything up to 13 or 14 stops of dynamic range. With an 8 bit codec that’s only 17 – 18 bits per stop, which really isn’t a lot, so for log really you want to be recording with a very high quality 10 bit codec and possibly an external recorder. Remember with a standard gamma your over 30 bits per stop, now were looking at almost half that with log!
Shooting flat: There is a lot of talk about shooting flat. Some of this comes from people that have seen high dynamic range images from cameras with S-Log or similar which do look very flat. You see, the bigger the captured dynamic range the flatter the images will look. Consider this: On a TV, with a camera with a 6 stop range, the brightest thing the camera can capture will appear as white and the darkest as black. There will be 5 stops between white and black. Now shoot the same scene with a camera with a 12 stop range and show it on the same TV. Again the brightest is white and black is black, but the original 6 stops that the first camera was able to capture are now only being shown using half of the available brightness range of the TV as the new camera is capturing 12 stops in total, so the first 6 stops will now have only half the maximum display contrast. The pictures would look flatter. If a camera truly has greater dynamic range then in general you will get a flatter looking image, but it’s also possible to get a flat looking picture by raising the black level or reducing the white level. In this case the picture looks flat, but in reality has no more dynamic range than the original. Be very careful of modified gammas said to give a flat look and greater dynamic range from cameras that otherwise don’t have great DR. Often these flat gammas don’t increase the true dynamic range, they just make a flat picture with raised blacks which results in less data being assigned to the mid range and as a result less pleasing finished images.
So the key points to consider are:
Where you can control your lighting, consider using standard gamma.
The bigger the dynamic range you try to capture, the less information per stop you will be recording.
The further you deviate from standard gamma, the more likely the need to grade the footage.
The bigger the dynamic range, the more compressed the gamma curve, the more critical accurate mid range exposure becomes.
Flat isn’t always better.
Sorry folks for the delay in announcing this, work got in the way. The winners in the My Horizons film contest are:
1st Prize: Bob Wall FW Focus Productions and “Choices”
2nd Place: Anna Possberg “My Ice World”
3rd Place: Boris Watz “Bridge in Tunnel”.
I was particularly impressed by the use of real people to tell the story in “Choices”. The subject is not the most interesting of topics but the film is well shot and edited and this means it is enjoyable to watch.
So here’s a little something I have been working on. This was born out of necessity and the reality of shooting 4K ENG style with a large sensor camera. Like many I needed a zoom lens with a good zoom range. 10x zoom was my minimum. While you can get some very nice PL zoom lenses, for example the Fujinon Cabrio 19-90mm this is only a 4.7x zoom. There are higher ratio zooms like the beautiful Angenieux 24 to 290, but this is a massive lens requiring a massive tripod and not at all suitable for ENG or handheld shooting. Sony F3 and F5/F55 users do have the option of the Sony SCL-Z18X140 which is an 18mm to 240mm servo zoom. It’s a good lens, not fast (f3.5 – f6.3) but useful, but quite a lot of money for such slow lens that is clearly based on DSLR optics. So, what about a DSLR lens? Well that’s what I started to look at, there’s plenty of choice, including the Tamron 18-270 a nice and useful 15x zoom range (just be aware that the Tamron’s focus back to front compared to most broadcast and film lenses). There are also the Canon and Sigma 18-250mm lenses.
Anyway… I’m trying to standardise on one DSLR mount and that’s Canon, so that means I need electronic aperture control for many of my lenses. Why Canon and not Nikon with it’s manual iris? Well almost all Nikon lenses focus back to front which is a real PITA. I purchased an MTF Effect Canon mount and control box, this gives me a Canon mount and now I have electronic iris control plus optical image stabilisation if the lens has it. But the electronics engineer in me started to wonder if it was also possible to add a servo to the lens to create a power zoom. After several abortive attempts, lots of swearing and not an insignificant spend on motors and gearboxes I finally got it working, a servo motor to drive an off-the-shelf DSLR zoom lens, or a PL mount zoom lens. Then I wondered about combining the zoom rocker with an iris control. This is where things got really interesting as I was delving into new territory for me and that’s micro controllers. In addition Canon don’t publish details of the language and signals needed to control their lenses. Almost no-one has completely cracked all the codes needed to drive the lens, so I had to reveres engineer the protocol. After a few weeks of tinkering late into the evenings (sorry to my wife and daughter!) I finally had the major codes and could program a controller to drive the lens.
Now it’s all combined into a single box. A zoom rocker and servo motor to drive the zoom ring. A thumb wheel on the hand grip that operates the iris and a second small box with a knob to control focus, plus image stabilisation control. This means you can fit handles to the ring and have the focus control wherever you want. I have also developed a miniaturised lens control box without the zoom servo and rocker. This is the size of a match box and includes a thumbwheel/knob for iris and a port where you can plug in the remote focus control.
The next phase is to add Bluetooth control. This should be relatively straight forward now I know the lens protocols. I can already add bluetooth to the micro controller, so I just need an App to talk to it. Anyone out there good at writing iPhone Apps? I’d really like to get full remote control of zoom, iris and focus on an iPhone.
When will this become a product? Well I am working with a well known manufacturer in the UK at the moment to see if we can make this an affordable product. If that deal doesn’t work out then I’ll go the Kick-Starter route and do it myself. It will probably end up as two different products. An all-in-one box with zoom rocker and Canon lens controls and a second unit that just has the zoom rocker and servo for use on PL or Nikon lenses.
There’s only one place left on my storm chasing workshop if you fancy a workshop with a difference and a bit of an adventure. https://www.xdcam-user.com/2013/05/storm-chasing-workshop-and-adventure-may-23rd-to-may-30th-come-join-me/
I have been given the opportunity to place an order for a minimum of 5 Zunow 16-28mm f2.8 FZ mount lenses. The list price for this lens is EUR 7650 ($11K) but if we can order a minimum of 5 it will be EUR 3,800 ($5K approx). If as a group we buy 10 the price would be even lower at EUR 3,600 ($4.5K approx). Prices exclude any taxes or shipping (from the EU).
Sony FZ mount for PMW-F3, PMW-F5 and PMW-F55
• Focal Length: 16-28 mm
• Minimal optical distance: 1.1 ft/0.3m
• IRIS: F2.8-22
• Front Diameter: 98mm
• Filter Screw 95mm P1.0
• Weight: 1.9 Kg/4.19 lbs
Made in Japan by renown lens manufacturer Zunow the lens is a beautiful low distortion wide angle vary-focal lens. At 16mm the wide angle views this lens has to offer are beautiful. Designed for 4K it is sharp and clear out into the corners. Flare is minimal and contrast excellent. At f2.8 for a 16mm lens this is respectfully fast. At 24mm there is near zero geometric distortion making the lens particularly well suited to industrial and structural imagery where pin cushion or barrel distortion can be a significant problem. Even at 16mm the distortion is minimal. The iris is an 8 blade iris, the bokeh is very pleasing. It is not designed as a zoom lens, the focus shifts slightly as you change focal length. What it is, is a single wide angle lens that can provide a variety of focal lengths without the need to swap lenses while maintaining prime lens image quality.
The lens has large and easy to read markings for aperture, focal length and focus distance. The focus markings are in both metric and imperial. The rear element is baffled for 16:9 to reduce flare and reflections within the camera body. The construction is all metal, mostly black anodised aluminium alloy.
So how will the group buy work? First use the contact form to contact me if you are interested. I will have a sample lens available at some of the workshops I’m running in the UK and I hope to take one with me to Cinegear at the end of May. I wish to close the deal during the week commencing June 10th. Payment for the lenses will be made to the European Distributor who will then ship out the lenses. Buyers will be responsible for all taxes, duties and shipping charges. These are in addition to the group buy price.
Below are some frame grabs from the lens, shot using a PMW-F5 with the frame grabs pulled from the raw clips in Adobe Premier. Aperture was between f4 and f5.6. Click on the thumbnail to view a larger version and then click on the link above the image for the full 4K frame. Please consider this is a very wide angle lens when looking at the images.
