I’ve been shooting RAW video with my Canon 5D Mark-III for a long term project and it’s simply otherworldly.
If you haven’t heard of this yet, the Magic Lantern firmware hack for Canon cameras allows you to shoot 24 1080p raw pictures per second now. Think of it as a continuous 24fps burst mode of raw pictures.
You batch convert those raw pictures into TIFF files and then change the image sequences into super-duper high quality video. Your Canon DSLR can now hold a candle to the RED camera or Black Magic Cinema camera. (Minus the professional workflow of course).
You have the same quality, dynamic range, colour control and exposure latitude as normal RAW still photos but with video!
The only downsides are file size (90MB/sec!), the need for expensive 1000X CF cards and processing time.
A hidden downside is the lack of timecode and metadata for your processed video.
Here you can see how the .mov file created by the raw2dng app has a file creation date different than the original video. I shot the RAW video at 8:42pm on June 20th but the ProRes video file wasn’t created until July 4th so it will be out of order in Final Cut Pro.
If you import a video file created by raw2dng into FCPX, the video lacks timecode and therefor will be arranged by the wrong time in your event browser. FCPX is all about metadata and you don’t want the video times to be wrong when working on a long-term project.
I’ve written a simple Applescript that will batch convert folders with TIFF files into 1080p24 ProRes422HQ video. It is included in the .zip file at the bottom of this post. If you just use the .mov files created by raw2dng then skip to Step #3.
The file creation date of our video file is different than the original RAW file.
Run my other Applescript which changes the file creation date of the new .mov files to match the .RAW files of the same name. This works for batch folders. You can find the script at the bottom of this post.
Now the .mov file creation date matches the original .RAW file. The video file still lacks timecode though.
Get the QTchange app for Mac. The demo allows you to do 8 files at a time or you can buy it for only $25 which is well priced. QTchange will add a timecode to the video files based on the file creation date that we set earlier.
Now when we import our converted video files the date and timecode are correct!
I’ve put the Applescripts used in this tutorial into a zip file you can download here. Feel free to share with credit.
The TIFF to ProRes422HQ script will need to be modified before you run it on your computer. Launch the AppleScript Editor app on your mac and open the file. Change the file path of the “.set” settings file to match the location on your hard drive. There is also a small Applescript to create a new settings file if you don’t want ProRes.
If you can remember exactly where you were standing when you shot an image you can just use Google Maps to find the GPS position afterwards.
Go to http://maps.google.com and zoom into the area you took a photo. Then Right Click (CNTR-Click on a Mac) at the exact spot you want and then select “What’s Here?”. The GPS Longitude and Latitude will appear in the search box.
Method #2: Just take a picture with your smartphone at the same location as your D-SLR.
Upload the image to http://regex.info/exif.cgi and it will tell you the embedded GPS info.
Once you have this Info you can map it with Google Maps.
You can then embed the gps location into your D-SLR picture by using exiftool.
Add “-overwrite_original” to the command or else it will create a backup copy of the image.
You can find exiftool for Mac or PC or Linux here. Also try googling “Exiftool GUI” if you want a graphical user interface.
Method #3: Sync a GPS route from your iPhone/Android to a batch of images.
This is the method I have been using lately and I like it. Mainly because I find the app to be very reliable.
Simply launch the EveryTrail app and run it for the day that you are taking pictures. At the end of the day you can sync all your photos with GPS Photo Linker.
How does it work?
Basically the app creates a “.gpx” file which contains data like this: Date, Time, Lat, Long, Date, Time, Lat, Long, Date, Time, Lat, Long, etc.
GPS Photo Linker looks at the time each of your photos was taken and correlates it to the closest GPS point.
EveryTrail.com is a cool online trail mapping site. It has a lot of cool features but we are only interested in creating creating a GPS route of our walk.
You can make a GPS route of your trip and then upload a .GPX file to the web to download and sync with your pictures.
Here’s the process:
Download and install the app. I think you need to buy Pro version because the Free version doesn’t let you upload the .GPX files.
Setup an Account in the App.
Go into settings and change your GPS precision.
Click “Start Tracking” under “My Map”
When you are done, Select “Pause” and then “Finish”
Upload the Trip to EveryTrail.com
Now go to EveryTrail.com and log in with the username and password you setup.
Click on “My Tracks”
Scroll down the page and you’ll find a link to “Download GPX for your GPS”
Now launch GPS Photo Linker or GpicSync and sync your photos with the .gpx file. The program will automatically embed the GPS longitude and latitude in the photo’s EXIF info.
A strange new species has been popping up on lawns across Alberta in the last 28 days. Journal videographer Ryan Jackson got to the bottom of the outbreak with this whimsical parody video. Video by Ryan Jackson, edmontonjournal.com
So after doing my last election video I wanted to do something a little different and a little funny for the Edmonton Journal. I came up with the idea to do a funny parody video a while ago but I wasn’t sure if this was the election to use the idea. Luckily my awesome wife pitched the idea too my awesome boss Kerry during Journal bowling night and Kerry liked the idea.
Before I would formally start the project through, I wanted to show that I had a plan. So Journal photography intern and Loyalist College graduate Megan Voss and I created a storyboard for the video.
With any video, the more planning and preparation you do at the beginning the less work you will have to do at the end.
My logic was that on election day there would be a huge audience in “election mode” and hungry for content. Presenting something different, funny and “get out and vote!” ‘ish would be heavily shared. Especially on Facebook where the audience leans towards comedy and activism. I turned out to be right as this was one of the most successful and most watched videos we’ve had in a single day!
I also made the video generic enough that we can use it for future elections and it could also be relevant for any election anywhere really. It also opens the door for future Planet Alberta videos on Alberta animals like the Jacked-Up Pickup Truck.
Here's Megan Voss cutting out my drawings and rough script of the storyboard.
I sketched out all of my shots and pasted them on the wall with a rough script. This was to assure my bosses that I had a plan for what to do and also get get feedback and suggestions for improvement to the script.
In order to get all the signs, Megan and I created a large Google Map with dozens of campaign headquarters for dozens of candidates. Then we just drove all over town for two and a half days to get all the signs. 76 in total! Getting the signs actually took longer than shooting the actual film!
I woke up at 4:30 am one morning and couldn't sleep so I got up and went out to shoot signs. It worked out because the light was perfect for this landscape.
Cutting out the baby signs.
Here's all of the signs we made for the time lapse scenes. I lucked out because this guy has a PDF of his sign on his website.
Here's Megan Voss pouring the dirt for the baby sign to grow out of.
Used three Alienbees Einstein studio strobes to light the baby sign. Why? Because the sun was very inconstant and I needed every frame of the time lapse too look the same. Wanna know the secret to BBC Planet Earth's amazing plant time lapses? They aren't actually shot in nature! They are all done in the studio with strobes and then superimposed into the nature shots after the fact. Tears...
Shot the baby sign being born with a 1D Mark IV and a 100mm Macro.
Used lots of fishing lines to bring the signs to life. "The Family is Complete"
Here's how I shot the opening scene with the globe similar (but not similar at all) to the opening of BBC's Planet Earth. Just a Canon video light with a cinefoil snoot. Panasonic GH2 with a Canon 100mm macro lens.
Here I'm filming my awesome trophy wife Ashe in curlers for the opening shot. I was originally going to find an old lady for this shot and then Ashe volunteered to do the shot which made life much easier.
Here I am photographing all the signs on my neighbours lawn. I didn't want to use my own front lawn for more than one shot.
Here I'm shooting the signs "sprouting" along the sidewalk. I used my 1D Mark-IV with a 300mm and a 2X converter = 780mm lens.
Chris and Nathan hiding behind the fence. I got the idea for this shot one night while trying to fall asleep.
Here's how I shot the weed wacker scene. The tree trunk was in the shadows so I used two reflectors to bounce light into the shade. The camera was down low with a 50mm f1.2 lens which is equivalent to a 100mm on a GH2.
I rarely get photos of myself working. Thanks Megan Voss!
We simply taped the fishing line to the back of the signs. I was super impressed with how well this worked. I actually bought 6-pound-test line which was much thinner and probably wouldn't have shown up on film but I kinda liked the heavier 20-pount-test line because it was stronger and it looked funny that you could see the line on screen.
Jeff's and Chris and Nathan brining life to the signs with fishing line.
This is the funniest behind the scenes photo. I got a shot of Bowie the Beagle lifting his leg "just to be safe" but the shot never ended up in the final cut because I didn't want to imply that that the dog actually peed on the signs. Only that the dog was a "threat". Pretty awesome that Lucy could lift his leg with fishing line. This dog is going to be a star I tells you. A star!
“99% Preparation, 1% Pressing a Button” Ryan Jackson, staff photographer with the Edmonton Journal tells the stories behind his photos and gives tips and advice to students. Failure is always an opportunity to overcome obstacles and win. Sometimes people burn down bridges to see if you can swim across. Recorded in Victoria, BC at the Canadian University Press Conference on Jan. 12, 2012.
Ryan Jackson, multimedia photojournalist with the Edmonton Journal speaks on his “I Was There” music video, the need for multimedia journalists, trends in online video, trends in technology, and tips for student journalists to experiment with video at their papers. “Will cats save Journalism?…. I don’t have all the answers.” Recorded in Victoria, BC on Jan, 14, 2012.
You can find the GoogleDoc presentation at docs.google.com/present/view?id=dhf88d7p_549m8mxppf9
and more on my blog at punkoryan.com/training
When my friends Kelly and Tracey told me they wanted to do trash the dress photos after their wedding I was super excited. Kelly is a microbiologist PhD candidate at the University of Saskatchewan so we decided to do something that was fun in his lab. This photo isn't photoshopped yet. Just straight out of camera.
Kelly's brother mixing up food colouring in water.
Tracey with the coloured beakers. We used dry ice to make them seem to be boiling.
Here you can see my light setup. A large Octobox to the top left, a softbox to the back right, a ringflash on the camera and there is another flash behind the shelf on the left.
I used an Eye-Fi card tethered to my iPad with the ShutterSnitch App so I could see my images on a bigger screen. This comes in very handy when setting up photos like this as it involves moving a lot of little elements in the frame here and there.
Well Tracey's dress is ruined.... mission accomplished!
After the lab we went down to the South Saskatchewan River and did some cool photos. For light I just had an Alienbees Einstein with a 48-inch Octobox. I used the new Lithium Vegabond battery pack for power.
I used Pocketwizard Flex5's with the AC3 Zone controller so that I could dial the power of the Alienbees Einstein up and down. This is a 1.5 second exposure, F2.8. ISO2500 on a Canon 1D Mark-IV.
It was 32-degrees Celsius that day so the water was still pretty warm.
6-seconds, F2.8, ISO640 to get the stars. Alienbees Einstein turned all the way down to 2.5 w/s.
Shows like Mythbusters are really good at taking something that only lasts a second like an explosion and dragging it out over a long period of time. Multiple cameras and slow motion help. But you can’t fill an hour of television with just that. You need anticipation, prediction and reaction to the event to really tell a story and make it interesting.
Here’s how I approach this video.
#1. I need A-Roll. A person of authority explaining what is happening today.
#2. I need B-roll to go with each thing that the interviewee mentions
Keywords: Horses, Humans, Media, Race
#3. I need to talk to the people who will be in the race. What are their predictions?
#4. The spectacle will only last a few seconds so I know I need more than one camera.
Camera 1: My Canon XH-A1 – This cameras has 20X zoom so I will put it at the end of the track.
Camera 2: Canon HV30 – I’ll put this camera at the beginning of the track and frame it on the starting gates. It can record 63 minutes to a tape so I just set it up and push record. Just leave it running.
Camera 3: Canon SD960 IS point-and-shoot camera that shoots 720p video. I set it up on a mini-tripod half way down the track and push record. Just leave it going.
Viewers decide if they want to keep watching a video in the first 10-seconds so I show the first 3-seconds of the race first to tease the viewer and hint that there is something cool coming. You better stay and watch!
I immediately go to by A-roll explaining what is happening today.
He's talking about horses so I show the only two horses that are on the track. Lame shot but it was all I had.
He talks about runners so I show the two runners that I see walking up.
I interview one runner and ask him for his predictions for the race.
This is a media spectacle so I show the other media there. Remember "say it and then show it". The media is mentioned in the interview so I show the media.
I interview the other runner and ask him for his predictions for the race.
Now I show the race. I have three cameras. One at the gates.
A second camera (a Canon Point-and-shoot on a mini-tripod) midway down the track to show the horse....
...followed by the runners....
Then my third camera is at the finish line. With my three cameras I was able to get a tight shot of the gates, a side view and an overall view.
Reaction. Emotion. High-Fives!
Followup reaction interview #1.
Followup reaction interview #2.
Ask the Jockey what he thinks. The end!
I get back to my car and ingest all the video footage into my laptop.
In Final Cut Pro:
I choose the A-roll “what is happening today”
I choose predictions “what do you think will happen?”
I choose the 3-second clip from Camera 2 and 3.
I choose the 12-second clip of the entire race from Camera 1. (yes the race only lasted 12-seconds!)
I choose the reaction clips “how was that for you?”
This is the beggining of a series of posts I’m going to make on shooting video.
As a way of teaching what to shoot when making videos I will explain my thought process and why I captured each shot in a few different videos.
For starters there is my Under the Big Top Circus Camp video I shot back in August 27, 2008.
Here is my formula for something like this.
#1. Get my A-Roll (someone explaining the story)
#2. Get a secondary A-Roll (someone else explaining story)
#3. B-roll B-roll B-roll. You need 10X the B-roll than your A-roll. Visuals to backup what the A-roll is saying.
Think of it like this. The A-roll stimulates the communication side of your brain. B-roll stimulates the visual side of your brain. If either side is boring then the video fails.
My process for a video like this:
I get to the event and immediately find someone of authority that I can ask to “tell me what is happening today.”
Frame the person nicely using Rule of thirds. Looking into the frame.
-Would you please spell your name for me? (pieces of paper get lost. get them to spell it out on tape)
-Please tell me your name and title.
-Tell me what is happening today AND what is the funnest thing about it?
-Tell me what the kids say about it.
-Tell me why it is important.
-Anything else you would like to add?
-Please spell your name one more time. (has saved by but several times. can’t hurt to get it twice incase the tape drops out of something goes wrong the first time)
-Do you have a cell phone you could be reached at if I have any other questions?
Okay now I have my main interview. I’m safe. I just need to talk to one or two kids to ask them about the camp and I’ll be golden.
Bam! That’s it for A-Roll. #1. Get a person of authority to talk about what’s happening today #2. Get one or two other people to talk about what’s happening today
Now I just need some interesting B-roll a.k.a. “visual sequences” a.k.a. “coverup shots” to go with my A-Roll.
During the interview I payed attention to keywords:
Physical theatre. Puppetry. Magic. Concentrate. Professionals teaching. Happy kids. Tight-rope walking. Juggling. Stilts. Confidence. Skills.
A simple rule to remember for video. Say it and show it. If your interviewee talks about something then you have to SHOW or ILLUSTRATE that idea.
I know I need shots to go with each of those keywords in the interview.
Two things to remember for B-Roll
A great rule to remember when shooting B-roll is wide-medium-tight. You can triple your B-roll by shooting a whole scene, then come in tighter on part of the scene, then really tight on the most interesting part.
Now instead of one shot that you can only use for 3-seconds you can go 3-seconds wide, 3-seconds medium, 3-seconds tight.
Not only have you tripled your B-roll but your B-roll is also more interesting!
Don’t just stand in one place and get a one-minute clip of a person doing something. That is boring.
Instead get short, tight clips of each portion of what they are doing and build a sequence of events. MUCH more interesting.
I break down individual shots below. Watch the video again and look for each of these elements.
Tight Shots from various angles are important. I start the video off with this frame because it is tight and interesting. The visual side of your brain is immediately interested. Always start your video off with the most interesting shot you have. People decide if they are going to stay on a video in the first 10-seconds so it must be impactful.
A wide overall shot of all the kids. It is good to start out with a wide shot to show perspective and give the viewer an idea of where they are. If you shoot everything wide-medium-tight, wide-medum-tight, wide-medium-tight you will always triple your B-roll and make it more visually interesting. Bring your viewer into the scene.
Now I go tight on an interesting element of the previous overall. Notice I have my lens set to wide-angle and I simply moved my camera low and close to the cylinder. It is always better to "zoom with your feet" than to simply use your zoom lens. You will have less shaky video as well it makes for a more visually interesting frame.
A "peak moment" B-roll shot where the kids throw their wands in the air. This becomes a generic b-roll shot that I can use to coverup the a-roll.
Nice tight shot of a kid. Tight faces and eyes show emotion. It never hurts to have a few tight faces for use as B-roll. They show emotion in the video. If you are always wide-wide-wide-wide then viewers never feel engaged in the story as much. This frame grab was actually front page main art of the Edmonton Journal newspaper the next day.
Okay here I show the instructor and frame it with the kids feet. I could have shot this wide and shown the whole kid. But do I need the whole kid? No. It is more about the instructor and keeping just the feet makes it more interesting.
Outdoor scene setter. This is a B-roll shot that you can always get and always use. Simply show the viewer where you are. You can grab this on your way there or your way out. Can never hurt to get a shot of the building and it always comes in handy for B-roll in your video.
A tighter shot of the same building. Signs are always useful B-roll. If you see a sign, shoot it!
Main A-Roll interview. Get there, ask for a person of authority to talk to. Move them close to a window with nice light. Choose an interesting but-not-too-distracting background. Frame them using the rule-of-thirds and looking into the frame. Lock your tripod. Put a mic on them. Ask them to "Tell me your name and title. Tell me what's happening today. What's the most important thing? Why does this matter? Anything else you would like to add? Spell your name one last time. A phone number if I have any questions? Thank you"
Wide shot of all the kids. I've shown the viewer where we are. If I stay on this frame though it will get boring.
Bam! Go tight on a kid. Get him reacting to something.
Bam. Go tight (or medium) on two other kids interacting and reacting with something. I've made three interesting shots by showing the room wide and then showing some tighter elements from within the room.
Wide shot of kids in the room.
Tighter shot of one kid. Shoot from down low to eliminate distracting elements and focus on him.
Bam! Show what he is doing! I could have just kept the camera on the whole kid for 5-seconds but instead I show the whole kid, show the top of him, show what he is doing. More interesting than a locked-off wide shot of the whole kid.
Tight face. Faces show emotion and connect the viewer with the story. When your interview subject in the A-roll says "the kids are happy" you have a face to show that.
Another interview for my A-roll. Frame him using the rule of thirds. Try and find a quiet, clean background that is not-too-distracting. Lock the tripod. Move to the side and ask "Whats your name and title? Tell me what is happening today? What is the best thing about it? Why do you come here? Anything else you would like to add? How do you spell your name and how may I contact you if I have any other questions? Thank you."
Sequence sequence sequence. Don't just shoot a wide-angle of the kid walking across the tight rope. Show them getting on, tight shot of the top of her balancing....
Tight shot (with my wide angle lens) of her feet. If I had just showed her whole body it would be boring. If I shoot her wide-medium-tight and get a sequence of her getting on, balancing and walking then I have more interesting shots.
Golden Moments. This is more of an editing than shooting decision but sometimes it is good to just show a few seconds of natural sound and natural action. Here the teacher and the puppet are talking and high five eachother. It's cute. You don't need A-roll talking over it. You can just show this cute moment and give the viewer a break from the interview to experience what it was like to be there.
The kid surprised me when he picked up his puppet and started talking with it. Sometimes kids are really lame to talk to and other times they amaze you with golden moments.
Sequence 1 of 7. I could have stayed back 10-feet and shot a boring wide-shot of the instructor and the kid but instead I went tight on his hands....
Sequence 2 of 7. Shot down low to clean the background and emphasize these two subjects interacting.
Sequence 3 of 7. Get a tight shot of him tightening the straps.
Sequence 4 of 7. Back up and get a shot of the instructor helping the kid up.
Sequence 5 of 7. Wide shot showing perspective and them interacting.
Sequence 6 of 7. Detail shot. Get down low and show the feet and stilts.
Sequence 7 of 7. Reaction reaction reaction! I would have liked to have shown the kid's reaction but the instructor was actually more excited than the kid so I showed that.
Third A-roll interview. I had the last two people facing left so I get this kid to face right. Switch it up and make it more interesting. There weren't any interesting backgrounds around so I just moved him into the hallway (where it is also quieter) and use a clean white wall. Rule of thirds. Looking into the frame.
Sequence. Show the instructor talking to the kid.
Show what the kid is doing....
Show him doing his thing. If I had just stayed back 20-feet and got a long clip of him doing this it would be boring. Going in tight each element makes it more visually interesting and you can build a sequence that tells a story.
Use your zoom to compress the image. A super-wide image of the whole classroom would be boring because you aren't focusing in on anyone. Going medium-tght on just a few kids makes them larger and more interesting.
Tight shot of a kid practicing. You've got to pick certain people and focus in on them or else your video is general shots of general people. Find characters for your video and focus on them.
Another overall shot. Here I kept the camera low and locked in one place and just let the kids walk out of the frame. Starting out people always move the camera to follow the subject. Instead keep the camera steady and let the subject move in and out of the frame. It is more interesting.
Shoot from up high. Just like with still photography you don't want to suffer from "5-foot 9 syndrome" and shoot everything from the same perspective. Shooting up high or down low can clean your backgrounds and make it more interesting.
Peak action shot of kids practicing for generic B-roll. I just stood in one place and waited for the kids to do something rather than constantly moving the camera around looking for action.
My end shot. Just showing the kids applauding. Seemed like a relevant ending.
Nameplate. It is important to have branding when your videos are published in multiple platforms like YouTube, Vimeo, your website, etc.