Everything Else

The cost of being a freelancer.

2 Comments

You spend weeks juggling dates, turn down other significant jobs to work with an important client to keep them happy. Book flights, pay out lots on advanced expenses for a 10 day job and all is set. Then at the last minute the 10 day job becomes a 4 day job, new flights have to be booked etc and all your planning goes out the window. I don’t think some clients realise how much it can cost a freelancer to have work cut or cancelled. It’s not just the loss of earnings from the cut job but also the loss of business with other clients and other additional costs. It takes time to re-book flights and hotels etc.

Earlier in the year the plan had been: Fly long haul from London to Singapore (client A), fly from Singapore to LA for a job for client B, then while in the US go storm chasing, fly home. One round trip ticket, 6 days for client A, 4 days for client B no dead time. Cost of ticket split between clients. Everyone happy.

But client A decides they can get better value for money by extending my trip, so I work with client A to make that happen, but this means I have to cancel client B. So now I’m all set to fly to Singapore for 9 days for client A. Then fly home. Then fly to USA for storm chasing a few days later. This involves the expense of two round trip long haul tickets. Client A happy, client B not so happy (but understands the situation). Everything is confirmed, set in stone. Flights are booked, hotels booked. Dates blocked out in my diary so when people try to book me on those dates I have to turn them away. June is always very busy for me.

But now last minute, client A has decided they no longer want to extend the trip, so I will fly home 5 days earlier than booked (IF I can get flights). There is now dead time in between Singapore and Storm chasing that I cannot fill, time when I could have been working for client B or client C, D and E who were turned away because I thought I was going to be busy.

What’s really annoying is that Client A KNEW that I was going to have to cancel another important job to help them with the extra days in Singapore.

So all in all I’ve gone from 10 days of work to 4, I’m having to pay out for long haul flights that I could have got covered by Client B, all because I decided to work with client A to help them out. When I quoted client A for the 10 day job I gave them a 10 day discounted rate. Now they have asked me to send in a new quote for the job and have indicated they are expecting the same 10 day rate when now the job is only 4 days!

Not only all that but I used an upgrade voucher that took a year to earn on the original Singapore flight bookings, but as it’s so late in the day now I will loose that voucher when I re-book, and there is still the big question as to whether I can actually get seats and how much extra they will cost booking just two weeks ahead instead of two months ahead.

Pissed off. Glad I got that off my chest.

Review, Uncategorized

And now for something completely different: The Cinemartin NEXT (part1).

Leave a comment
Shooting with the Cinemartin NEXT
Shooting with the Cinemartin NEXT

So many of us will have used an external video recorder before and I’m sure everyone reading this has used a computer at some point. The Cinemartin NEXT is an interesting fusion of a highly portable video recorder with a super compact high end computer. Designed to go on the back of a camera the NEXT can record in beautiful uncompressed 4K 444 RGB in 10 bit at up to 30fps with an update/upgrade coming later that will allow 4K at up to 60fps via dual 6G HDSDI.

It can also record compressed and all the way down to SD resolution. But what makes the NEXT different is that it is also a very powerful PC. The sample unit I had was running windows but you can also install the majority of operating systems on it. There are some clever tools installed as standard including the capture applications and a very clever background renderer that will transcode whatever it is you are shooting to ProRes or H265. Internal storage goes all the way up to 1TB and it’s very, very fast. Memory up to 32GB.

With so much computing power on tap you can do all kinds of things. For a fast turn around job like news you can shoot directly on to the unit and then edit the footage as soon as you have finished shooting. There’s no need to transfer footage from the camera to the NEXT, it’s already on there. It can run pretty much any edit software, I used Premiere CC and editing on it was quick and easy. As it has a direct HDSDI output (as well as 4K ready HDMI outputs) you don’t even need to render out your finished edit to play it back. In most cases you could cut your news story and play it out directly from the timeline to the uplink truck. If you don’t have an uplink truck then being a computer you can render out your story as a file to upload over the internet.

Cinemartin NEXT connectivity.
Cinemartin NEXT connectivity.

There are 4 USB3 connectors and 2 ethernet connectors so you can connect to the internet either directly via ethernet or by using a wifi or 3G/4G wireless modem.

Power consumption is a very modest 15 to 29w so it will happily run of a D-Tap from your camera batteries for 3 or 4 hours. If your in a hotel room or studio you can plug in up to 2 external computer monitors (as well as an HDSDI/HDMI monitor) to expand your desktop beyond the built in high brightness touch screen display.

In the next installment I’ll take an even more in depth look at the NEXT from Cinemartin. Including some image quality comparisons.

Uncategorized

Storm Chasing and Film Making Workshop, 1 place left!

Leave a comment
Me shooting a tornado with the PMW-F5 and AXS-R5 on my Miller Solo tripod.
Me shooting a tornado with the PMW-F5 and AXS-R5 on my Miller Solo tripod.

As usual I am running a combined film making workshop and tornado chasing tour. This year the trip runs from June 16th to June 25th. There is one place still open if anyone is interested.

What do you get? For a start over a week to pick my brain and learn about shooting run and gun. I will be shooting with a PXW-FS7, but you are welcome to bring whatever camera you wish. As there are only 3 guests on this trip you can be assured of plenty of one to one time.

The trip will start and end in Denver, Colorado. Where we will go in between depends on the weather. Each day I will make a detailed weather forecast and we will go to the area most likely to produce severe storms. Do note that I do not wish to get as close as possible to a tornado. I want to get photogenic footage of the storms and sometimes this means hanging back a little to find the best position taking in to account light and the motion of the storm. It should be an exciting and adventurous experience, but I am not doing this as a thrill seeking event and I don’t wish to put myself or anyone else in danger.

You can expect to see impressive Supercell thunderstorms that can produce incredible lightning displays, giant hail and damaging straight line winds. If we are luck we may even see a tornado or two. We may spend a lot of time in the car, last year we drove over 1,300 miles in 10 days. 300 to 400 miles a day is not uncommon, sometimes a lot more. In June the storm systems tend to move slowly giving some of the best video opportunities. We should end up in the Northern Great Plains, so Colorado, Nebraska, Montana and South Dakota are normal, but we could also end up down in Texas, Oklahoma and Kansas.

The cost is $1,500 USD plus hotels at cost (allow up to $1,500 for hotels). For more details please drop me a message using the contact form.

PMW-F55, PXW-FS7, Scene Files and Luts

WYSIWYG LUT’s to play with.

5 Comments

Here are a couple of high dynamic range WYSIWYG LUTs to play with. These are for the F5/F55/FS7. The camera should be set to SGamut3.cine/S-log3 and the EI should be set to the base EI (2000EI F5/F55 and 1250 EI on the F55).

WYS-ACALW1 will give almost the full dynamic range of the camera with lots of highlight roll off. It’s created to capture an extremely large dynamic range to help cope with very bright scenes such as sunny exteriors. Skin tones should be around 55-65% for the best results so zebras set to around 60%.

WYS-ACHGC1 will give a 12.5 stop dynamic range. It has more contrast than ALW1 but a bit less dynamic range. The colour palette is based on a Canon type look. Skin tones should be around 60-70% so zebras set around 65%.

Let me know what you think.

Alisters-WYSIWYG-LUTS1

If you find the LUT’S useful, please consider buying me a beer or a coffee.


Type



 

A7, PXW-FS7, Scene Files and Luts

S-Log2 to Canon LUT’s for the FS7/F5/F55/A7s

3 Comments

More LUT’s for you to play with!

These LUT’s are primarily designed for the PMW-F5 and PXW-FS7 but they should also work with the PMW-F55 and A7s. You need to shoot in SGamut/S-Log2.

There are two types of LUT. The AC-Canon-Log versions will output Canon C-Log with Canon wide range colorspace while the AC-Canon-Look versions output using Canons wide dynamic range gamma. The C-Log versions should be a close match to a C300 shooting C-Log but the footage will need some grading as it is still log. The Canon Look versions will have better contrast and need less grading but have less dynamic range.

I don’t have a C300 to hand to test these, so I am not sure how close they are and it will vary a little depending on whether you use an FS7, A7s or F5/F55.

Click on the links below to download the LUT sets. PLEASE DO NOT HOST THESE ELSEWHERE OR DISTRIBUTE THESE ELSEWHERE OR VIA ANY MEANS OTHER THAN A LINK TO THIS PAGE.

Alisters S-log2 to Canon Luts

If you find the LUT’S useful, please consider buying me a beer or a coffee.


Type



PMW-F55, PXW-FS7, Scene Files and Luts

LUT set for S-Log2 in Custom Mode for F5/F55/FS7

3 Comments

Quite a lot of people like to use S-Log2 or S-Log3 in custom mode. I’m not a fan of this method myself, I prefer to use CineEI, but for some people using S-Log2 or S-Log3 in custom works for them. If you use S-Log2/3 in custom mode then you are working with 709 color space, so if you want to use a LUT in post you need a LUT designed for this combination. So I have created two sets, one set for S-Log2 and one set for S-Log3. The LUT’s include over and under exposure compensation and you can download them here.

NORMAL: Indicates correct Slog2 exposure, middle grey at 32% and white at 59%, for SLog3 use middle grey 41%, white 61%.

1OVER etc indicates that the LUT will compensate for footage one stop over exposed.

If you are exposing the Slog2 so that skin tones are in the 60-70% region you will most likely need to use the 2 or 3OVER LUT’s.

Click on the links below to download the LUT sets. PLEASE DO NOT HOST THESE ELSEWHERE OR DISTRIBUTE THESE ELSEWHERE OR VIA ANY MEANS OTHER THAN A LINK TO THIS PAGE.

Alisters Custom mode SLog2 to 709(800) LUTs

Alisters Custom Mode Slog3+709 to 709(800) Luts

If you find the LUT’S useful, please consider buying me a beer or a coffee.


Type



Technology, Workflow

Why is log sometimes hard to grade?

7 Comments

This comes up a lot. People shoot in log, take it in to the grading suite or worse still the edit suite, try to grade it and are less than happy with the end result. Some people really struggle to make log look good.

Why is this? Well we normally view our footage on a TV, monitor or computer screen that uses a gamma curve that follows what is known as a “power law” curve. While this isn’t actually a true linear type of curve, it most certainly is not a log curve. Rec-709 is a “power law” curve.

The key thing about this when trying to understand why log can be tricky to grade is that in the real world, the world that we see, as you go up in brightness for each stop brighter you go, there is twice as much light. A power law gamma such as 709 follows this fairly closely as each brighter stop recorded uses a lot more data than the previous. But log is quite different, to save space, log uses more or less the same amount of data for each stop, with the exception of the darkest stops that have very little data anyway. So conventional gamma = much more data per brighter stop, log gamma = same data for each stop.

So time to sit down somewhere quiet before trying to follow this crude explanation. It’s not totally scientifically accurate, but I hope you will get the point and I hope you will see why trying to grade Log in a conventional edit suite might not be the best thing to try to do.

Lets consider a  scene where the brightness might be represented by some values and we record this scene with a convention gamma curve. The values recorded might go something like this, each additional stop being double the previous:

CONVENTIONAL RECORDING:  1  –  2  –  4  –  8  –  16

Then in post production we decide it’s a bit dark so we increase the gain by a factor of two to make the image brighter, the output becomes:

CONVENTIONAL AFTER 2x GAIN:  2  –  4  –  8  –  16  –  32

Notice that the number sequence uses the same numbers but they get bigger, doubling for each stop.  In an image this would equate to a brighter picture with the same contrast.

Now lets consider recording in log. Log uses the same amount of data per stop, so the recorded levels for exactly the same scene would be something like this:

LOG RECORDING (“2” for each stop):  1  –  2  –  4  –  6  –  8.

If in post production if we add a factor of two gain adjustment we will get the same brightness as our uncorrected conventional recording, both reach 16, but look at the middle numbers they are different, the CONTRAST will be different.

LOG AFTER 2x GAIN:  2  –  4  –  8  –  12  –  16.

It gets even worse if we want to make the log footage as bright as the corrected conventional footage. To make the log image equally bright to the corrected conventional footage we have to use  4x gain. Then we get:

LOG AFTER 4x GAIN:  4  –  8  –  16  –  24  –  32

So now we have the same final brightness for both the corrected conventional and corrected log footage but the contrast is very different. The darks and mids from the log have become brighter than they should be, compare this to the conventional after 2x gain. The contrast has changed. This is the problem with log. Applying simple gain adjustments to log footage results in both a contrast and brightness change.

So when grading log footage you will typically need to make separate corrections to the low middle and high range. You want a lift control to adjust the blacks and deep shadows, a mid point level shift for the mid range and a high end level shift. You don’t want to use gain as not only will it make the picture brighter and darker but it will also make it more or less contrasty.

One way to grade log is to use a curve tool to alter the shape of the gamma curve, pulling down the blacks while stretching out the whites. In DaVinci Resolve you have a set of log grading color wheels as well as the conventional primary color wheels. Another way to grade log is to apply a LUT to it and then grade in more conventional 709 space, although arguably any grading is best done prior to the LUT.

Possibly the best way is to use ACES. The Academy of Motion Pictures workflow takes your footage, whether log or linear and converts it to true linear within the software. Then all corrections take place in linear space where it is much more intuitive before finally be output from ACES with a film curve applied.

cameras, cinematography, lenses, PMW-F55, PXW-FS5, PXW-FS7, Shooting Tips, Technology

More info on CMOS sensor grid artefacts.

Leave a comment

Cameras with bayer CMOS sensors can in certain circumstances suffer from an image artefact that appears as a grid pattern across the image. The actual artefact is normally the result of red and blue pixels that are brighter than they should be which gives a magenta type flare effect. However sometimes re-scaling an image containing this artefact can result in what looks like a grid type pattern as some pixels may be dropped or added together during the re scaling and this makes the artefact show up as a grip superimposed over the image.

Grid type artefact.
Grid type artefact.

The cause of this artefact is most likely off-axis light somehow falling on the sensor. This off axis light could come from an internal reflection within the camera or the lens. It’s known that with the F5/F55 and FS7 cameras that a very strong light source that is just out of shot, just above or below the image frame can in some circumstances with some lenses result in this artefact. But this problem can occur with almost any CMOS Bayer camera, it’s not just a Sony problem.

The cure is actually very simple, use a flag or lens hood to prevent off axis light from entering the lens. This is best practice anyway.

So what’s going on, why does it happen?sony-grid-artefact-explained

When white light falls on a bayer sensor it passes through color filters before hitting the pixel that measures the light level. The color filters are slightly above the pixels. For white light the amount of light that passes through each color filter is different.  I don’t know the actual ratios of the different colors, it will vary from sensor to sensor, but green is the predominant color with red and blue being considerably lower, I’ve used some made up values to illustrate what is going on, these are not the true values, but should illustrate the point.

In the illustration above when the blue pixel see’s 10%, green see 70% and red 20%, after processing the output would be white. If the light falling on the sensor is on axis, ie coming directly, straight through the lens then everything is fine.

But if somehow the light falls on the sensor off axis at an oblique angle then it is possible that the light that passes through the blue filter may fall on the green pixel, or the light from the green filter may fall on the red pixel etc. So instead of nice white light the sensor pixels would think they are seeing light with an unusually high red and blue component. If you viewed the image pixel for pixel it would have very bright red pixels, bright blue pixels and dark green pixels. When combined together instead of white you would get Pink or Blue. This is the kind of pattern that can result in the grid type artefact seen on many CMOS bayer sensors when there are problems with off axis light.

This is a very rare problem and only occurs in certain circumstances. But when it does occur it can spoil an otherwise good shot. It happens more with full frame lenses than with lenses designed for super 35mm or APSC and wide angles tend to be the biggest offenders as their wide Field of View  (FoV) allows light to enter the optical path at acute angles. It’s a problem with DSLR lenses designed for large 4:3 shaped sensors rather than the various wide screen format that we shoot video in today. All that extra light above and below the desired widescreen frame, if it isn’t prevented from entering the lens has to go somewhere. Unfortunately once it enters the cameras optical path it can be reflected off things like the very edge of the optical low pass filter, the ND filters or the face of the sensor itself.

The cure is very simple and should be standard practice anyway. Use a sun shade, matte box or other flag to prevent light from out of the frame entering the lens. This will prevent this problem from happening and it will also reduce flare and maximise contrast. Those expensive matte boxes that we all like to dress up our cameras with really can help when used and adjusted correctly.

I have found that adding a simple mask in front of the lens or using a matte box such as any of the Vocas matte boxes with eyebrows will eliminate the issue. Many matte boxes will have the ability to be fitted with a 16:9 or 2.40:1 mask ( also know as Mattes hence the name Matte Box) ahead of the filter trays. It’s one of the key reason why Matte Boxes were developed.

Note the clamp inside the hood for holding a mask in front of the filters on this Vocas MB216 Matte Box. Not also how the Matte Box’s aperture is 16:9 rather than square to help cut out of frame light.
Arri Matte Box with Matte selection.

You should also try to make sure the size of the matte box you use is appropriate to the FOV of the lenses that you are using. An excessively large Matte Box isn’t going to cut as much light as a correctly sized one.  I made a number of screw on masks for my lenses by taking a clear glass or UV filter and adding a couple of strips of black electrical tape to the rear of the filter to produce a mask for the top and bottom of the lens. With zoom lenses if you make this mask such that it can’t be seen in the shot at the wide end the mask is effective throughout the entire zoom range.

f5-f55-mask

Many cinema lenses include a mask for 17:9 or a similar wide screen aperture inside the lens.

 

Uncategorized

Canon Launches C300 Mark II.

1 Comment
The new Canon C300 Mark II.
The new Canon C300 Mark II.

In the run up to NAB the rumour mill has been working hard and one of the cameras expected to be seen was an update to the Canon C300. Well, here it is, the C300 Mark II.

Externally it’s very similar to the original C300 but slightly larger and heavier overall. The key change to the camera body is ability to change the lens mount between EF, EF Lock and PL mounts. Other headline additions to the Mark II are a new sensor and a new codec that allow the camera to shoot in HD, UHD and 4K. I’m not going to go through all of the details here, for those you can take a look at the Canon web site.

NEW XF-AVC CODEC.

The new codec XF-AVC is very important to this camera and Canon as it adds the ability to go beyond the HD limitations of Mpeg2. It’s based on Mpeg4/AVC. The C300 Mark II can record UHD and 4K at up to 410Mb/s, 10 bit 422 and in HD can even record 12 bit 444. Now it is a little unclear in the released information what frame rates this does allow. But it appears that in 4K you are limited to 30fps but HD/2K goes to 60fps. The camera records to a pair of CFast 2.0 cards.

There is also a Long GoP version of XF-AVC for use at 2K/HD. This has a bit rate of 50Mb/s up to 60fps and in addition there are also Proxy version at 25Mb/s and 35Mb/s for recording compact HD files on SD cards within the camera.

Unfortunately this does mean yet another codec family to be added to everyone’s editing software, but Canon claim that support will be in place by the time the camera is released around September. You know I really wish ALL the camera manufacturers would get together and use one or two common codecs rather than each manufacturer having their own codec. Just imagine the chaos if every car new type of car used a different type of fuel!

Compared the the F5/F55/FS7 this new codec is very similar to XAVC. It’s interesting to note though that the C300 Mark II at the time of launch doesn’t appear to have any 4K frame rates faster than 30fps. It can record at up to 120fps in it’s slow motion mode, but this is limited to 2K/HD and uses center crop of the sensor, so all your lenses become much longer lenses. The plus to this is that it will minimise any aliasing artefacts, something that can sometimes be an issue on the FS7, although the FS7 will get center crop functionality in a future firmware update. Of course the F5/F55 have the benefit of both full frame and crop modes as well as the ability to change the optical filter depending on the mode you wish to use. In addition the FS7/F5/F55 can all go up to 180fps internally and 240fps in raw.

The C300 Mark II can also provide a simultaneous raw output to feed to an external raw recorder such as the Convergent Design Odyssey or Atomos Shogun.

NEW SENSOR, NEW LOG.

The sensor on the C300 Mark II is also new. It is a 9.84 Mega Pixel sensor with 8.85 active pixels (each pixel has two photosites) which Canon claim can provide up to 15 stops of dynamic range, which is very impressive. To record this Canon have a new log curve Canon Log 2 and the native ISO is 800, with the best dynamic range from 800 ISO and up. According to Canon above 800 ISO the dynamic range remains a constant 15 stops. It will be interesting to see what actually happens with the dynamic range at high gain levels as most cameras see a drop off in dynamic range above the native ISO. I’m not sure how you can increase the gain and have a constant dynamic range in a fixed recording range as sensor noise is always a limiting factor. It will be interesting to see this in the real world. From what I can tell from some of the web clips about the camera Canon Log 2 has a fairly low mid grey point around 36% with 8.7 stops below mid grey and 6.3 above. With 8.7 stops crammed into a pretty small range it will be interesting to see just what can be squeezed out of the shadows. I’m sure it will be good, but the question is just how useable is it?


As well as a new log curve there is also a wide range of gamuts including 709, 2020, P3 and “film gamut”. Some of these gamuts are huge. Film Gamut appears to extend beyond the visible spectrum according to the chart in the video above. I have some doubts as to whether the camera can actually fill them (in the same way that an FS7 or F5 cannot fill the included SGamut and SGamut3 gamuts, only the F55 can fill them). Again this will be an area to look at closely once the camera is launched, but the options are certainly good looking.

With this wide range of gammas and gamuts, LUT’s will be important and the C300 Mark II does have LUT’s but I can’t find any information beyond the facts that it has LUT’s and that the LUT’s can be output over the SDI and HDMI outputs as well as baked in.

One area where Canon do have some very clever technology is in autofocus. The C300 uses a technology called Dual Pixel AF and this is also included and improved upon on the C300 Mark II. You can do some clever things like move the autofocus target area and change the response times of the AF (see the video above for more details). It’s a clever system and I’m sure many shooters will find it helpful when shooting 4K.

All in all the C300 Mark II does look like a very interesting and capable camera. I’ll be sure to check it out at NAB. It’s obviously going to go up against the cheaper Sony FS7 as well as the F5. It has some very nice features. I think not being able to shoot above 30fps in 4K is a bit limiting and you only have a single, new type codec so there is no legacy codec support. Lets hope Canon get good XF-AVC support supported quickly.

I didn’t really get on with the ergonomics of the original C300, that’s a personal thing, for me as a traditional video shooter I just don’t like the top heavy layout, but I know many shooters that love the layout. In particular it’s well suited to those from a DSLR background.

It won’t be available until September and the price is set to be around $16k or £11K. I’m sure it will be very popular so get your pre-orders in now. I will have to get hold of one to figure out the best way to use the new Canon Log 2 curve and the cameras LUT system as I’m sure many people will want some in depth workflow and exposure help with this camera.