XfcpX – From Canon XF100 to FCPX – a working flow!

Please note: Canon has finally realeased their plugin. If you are still here for the XfcpX tool, please check out version 2 of it here: XfcpX 2

So, you are one of those happy owner’s of the Canon XF100? And you really want to work with Final Cut Pro X now that you’ve paid for it? You have banged your head at the wall, cursing at Apple, Canon and your dog over the fact that you could not import your footage in a manner that made any sense? If that’s you, read on – I may just have the solution you are looking for…and it’s Fast, Free and Lossless (Yeah, that’s what I said!)

UPDATE January 2012: Sorry all, I can see some of you had unanswered comments. My blog was spam-attacked, leaving me with some 8000 comments to sort through before getting to yours. I’ve installed a better spam protection now, so if your comment/question got lost in the clean-up please repost.

So, I too, have been in your situation. I’ve searched the internet high and low for a working solution that made sense. Sure, there are options – you could roundrobbin’ your files through FCP7… Well, that’s a no-go for me; even though I have my trusty FCP7 still installed, it’s on a another boot drive as I did not want to take the chance of messing up my production system. And booting to another OS to ingest my video, is not really an option for me, when I think in workflow terms. So, there’s the other option of the “Best video converter for mac” crap you see spamming all over the place. Is it Aunsoft? Foxreal? Whatever… are they even different products or just rebranding of the same code? Either way, they are $35 and reports claim that they are unstable, hard to do batch stuff in and they bloat the files… The bloating I could live with, disk space is cheap, but they do introduce a transcoding step – and we all know what that means: Loss of quality.
Also, the lack of any kind of workflow support just makes it a non-viable option for me. Daniel Houghton did a nice job of creating an Automator script to help with batch transcoding using Adobe Media Encoder. But I am not too fond of solutions that are too automated. I want options. And Adobe Media Encoder is more money out the window, unless you preown it ;-)

Considering that Canon claims their XF Pro codec is a “real” mpeg2 stream, wrapped in the MXF container, I am not about to accept a transcode of my master material, unless absolutely nessecary. It turns out, I don’t have to ;-)
So, after some investigation, edu-ma-cation and a few buckets of coffee, I came up with an idea of my own – and seeing as it does not look like Canon is going to step up to the plate in the near future, I took the liberty of going ahead and implementing the damn thing as well.

The proverbial tree does not grow into the skies though – read on…

The Good

So, is this the Holy Grail? The ultimate tool? Nope. Does it work? Absolutely yes. And it’s fast as well. Oh, and did I mention, NO transcoding. Basically, I am rewrapping the original mpeg2 stream in a .mov container that FCPX will play nice with – after which you can have FCPX transcode it if needed. But you have your original yuv422p/50Mbps footage going straight into the belly of FCPX – yummy yummy ;-) It’s all possible only due to the brilliant work of the guys doing the whole FFmpeg open source project. So yeah, don’t thank me – thank them… they did the hard work; my little utility is simply wrapping it in a nice UI that will enable it to be used in a decent workflow for us happy XF100 owners ;-)

The Bad

So yeah, I’m not really a Mac programmer. Neither am I a Cocoa specialist or anything fancy. I am also on a limited time-budget. So this is not a pretty program, very Mac’ish or really a nice package or anything like that. Update – I just repackaged and precompiled the whole thing… the installer should be alot easier now

…And the Ugly

And while we’re on the subject of time-budget – Any time spent on stuff like this, is time not spent making money from my customers. This means, that this program is as-is. Don’t expect support, don’t expect bugfixes and don’t expect it to get improved in regards to functionality. However, just below this paragraph I am providing you with a donate link so you can send me a little token of appriciation if you feel like it. Now, let me elaborate on this: Please, do not in any way feel obliged to pay anything for this little utility. I am dead serious – if you use it, and don’t feel like paying there’ll be no hard feelings from my part – I pinky swear; I am just happy you can use it to be productive. I made the code for myself – and it does not cost me a dime (or almost nothing) to make it available to you, so enjoy ;-)
But, if there’s enough funds comming in from this little Paypal link, I may find that I can afford to spend some more time on this program. While there will be bugfixes to whatever annoys me the most, a little funding can go a long way to make the product better. And it can get alot better – stuff like a nice installer, individual file conversion instead of just whole cards, previews of the video files, more options to split sound and video, maybe even a true Log&Transfer kinda thing with setting I’s and O’s on individual clips, better handling of metadata (big one for me) etc. etc. I make no promises though – Java programming for the Mac is not where I get my income from, and a man’s gotta eat you know ;-)

Please make some donations

So, without further ado:


Yeah, so in short – this is what you are going to do:
– Check Prereqs
– Install Java (If needed)
– Install XfcpX
– In…joy! :-)

Check Prereqs

This is developed and tested under Lion – I see no reason for it not to work under Snow Leopard for instance… but I have no clue really. If you try it out, please let me know ;-) Anyways, make sure your Mac is fully up-to-date by clicking the apple icon in the top left of your screen and select “Software Update..” – if you see any packages inthere, install them all unless you have a good reason not to. In order for this guide to work, you should as a minimum have the App Store installed… but there’s other goodies to be had, such as the ProRes codecs that only comes as an update in there for FCPX owners.(But of course, you already did install those – right?)

Install Java

If you don’t already have Java installed, go to http://support.apple.com/kb/DL1360 and download the latest version of Java – install it. Note: You probably already have Java and can skip this step – but this is the link in case you don’t ;-)

Install XfcpX

Download XfcpX, open the zip file and double click the installer – if all goes well, you should be able to find the utility in your Applications folder (Or use spotlight).

So, what’s going on here?

The utility is fairly simple. I made it in a way that made sense for my workflow. It assumes you used Canon’s XF Utility to backup the cards from your camera. These are your Master files, and should never be touched. I am importing all my cards using the XF utility to a single folder called “Master Files”. What I am saying is, that the Master Folder in XfcpX should basically point to the same directory as your backup location option inside the XF utility ;-)
Once you’ve located this, the right side “Cards to Process” should populate with the cards you have backed up. (They are listed by the date name that XF util uses to name each folder). The Destination Folder in the bottom is up to you. Some place all their converted files in one folder, and import them to FCP from there. I prefer to create a folder structure for each project and place the media there, something like /Media/Customer_name/Project_Name/ – the reason is, that if you use the “use folder names for meta data” thingie inside FCPX, you have prepopulated some keywords/metadata already…always nice !. XfcpX will use the destination folder, to create a single folder for each card processed (same name as XF Util used) and will place all the processed .mov files from the card inside this folder. When you have set the master folder and the destination folder, select the cards you want to process and click “1..2..3..GO!” ;-)
When a card has been fully processed, it will be moved to the Previous Cards list in the left side to clear up some clutter. If you need to reprocess it, you can just click it back to the process list on the right and select it for processing.
As with many of these things, you should not tamper with the folder structure/names of the cards that XF utility backed up – I am dependent on them ;-)

Final note

That’s it… you should now have full quality copies of your video, wrapped in a FCPX friendly .mov container ready to import into FCPX and let it do it’s thang!
You can, in your Event, just select “Import files” and it will eat it all up – but an even better option, if you ask me, is to say “Import from camera”, then click “Open Archive..” and browse to the Destination folder. The reason for this, is that it will display the import window where you can select individual clips with in’s and out’s to import ;-)
Depending on your system, you may experience a greenish flicker on the video when you play it back inside FCPX – don’t be scared, it’s not a “real” problem. The video is fine, will render fine and all that – it’s due to FCPX’s need for speed… if you go to the Preferences in FCPX and click “Playback”, select “Higher Quality” instead of “Better performance” and the problem goes away. This has nothing to do with the rewrapped files, just a performance issue in general – other formats does this too inside FCPX. If you can live with it, rest assured it will render just fine and it’s only a playback issue inside FCPX’s preview(Although, color correction might be abit more tricky :p ).

There you have it – now, forget about the tedious technical stuff and go be creative… shoo, come on, out in the real world and make some footage!

Please make some donations


******************* Known limitations *******************
– XfcpX will not handle Interlaced footage (the 50i, 60i modes etc.). That is to say, it will handle it, but you will get the raw interlaced footage into FCPX with all the artifacts you’d expect. This is due to the fact that I am not converting it. I could probably make an option to convert interlaced footage, but there would have to be a high demand for that. In my mind, Interlaced footage belongs in the past and should be shunned. The 3 key points of this software is to fit in a nice productive workflow, be fast and leave the original footage intact. Converting files, to handle interlaced footage would go against the latter 2.
– On some systems, the resulting .mov files wont play in QuickTime. This has to do with the videos being tagged, improperly, with a FourCC of m2v1 instead of m2v. It’s just the tag though – and the files should play fine, once inside FCPX. It’s on the to-be-fixed list.

******************* Updates below *******************
25-08-2011 – (V1.1c):
– Fixed a bug introduced with the NTSC Audio fix ;-) It resulted in just the Left mono channel being mapped to a stereo channel, leaving right mono sound out of the file. This means any video you converted with the previous version, only have left sound in it – reprocessing of the files is needed, sorry :(

23-08-2011 – (V1.1b):
– Fixes a bug with that produces really small(~400Kb) and useless .mov files. (Due to faulty timecodes for audio in NTSC version of the cam)

22-08-2011 – (V1.1a):
– New version up. Handles directories with spaces now ;-)
– Fixed “Chance to hang when processing large MXF’s”

60 Portraits in 6 hours

Time for an english blog post! Now, why is that? Did I suddenly go all international ? No, not really – I did, however, create a BTS video of a recent assignment. A video which I figure may have a wider appeal than just the local suspects, so I chose to annotate it in english – and this video/assignment is what this blog post is about. So there – english it is :-)

I was approached to shoot the entire staff of the GBS department – a subsidiary division of IBM in Denmark. They wanted a new profile so they initiated a little project to boost their presence, and for this they needed new profile pictures of each employee. The key points were that they wanted something “business like”, something fresh, something coherent that signaled that they were all part of the same team – and they did not mind moving away from the traditional portrait on a white background.

There were practical limits though – I couldn’t invite all 60 people to my studio, so Mohammad had to come to the mountain. The mountain in this case, being a generic and boring office space, with white walls, fluorescent lightings and where the most prominent piece of decor was a whiteboard ;-) In my scouting session, I did however notice the office space had blinds to cover the windows – a semi dark grey kind. Considering the traditional 8-bar logo of IBM my brain started working something out. I figured I could tone down the background, making it all a blue’ish tone to, if not emulate the logo, then at least, hopefully, create a coherent look for the portraits that made sense in some way.

The day arose and I arrived early with an assistant, setting up in the office space. The lighting plan was all mapped out in my head – and to my delight, it worked out like a charm just as I imagined it would – save a few tweaks etc. of course. My contact person had arranged for a “flowing approach” where my little photo studio was open for business for something like 2 x 3 hours during the day, and people could just drop in and get the picture done as they saw fit. That worked out fine, at the busiest points I had a queue of maximum 4 people – noone waiting more than 10 minutes. I, on the other hand, had plenty of time to drink coffee in between – with planning, this could’ve been executed in something like 2-3 hours; but with this many people you cannot expect everyone being able to fit their schedule to yours – so all in all, it was a sensible approach that worked out well.

The shot it self, was very limited – in the sense that all the portraits had to be reasonably alike to achieve the goal of coherence. In order to keep the blinds filling the background, but not get the window panes in the frame, my camera was pretty much locked down except for the height, in which I had some leeway. That, unfortunately, meant that I had to order my poor subjects around, asking them to move slightly around to position themselves correctly in the frame – they were good sports about that though. I was shooting at 200mm @ f/11, enabling me to keep focus on them relatively easily, getting a pleasing amount of blur on the blinds and totally eliminating texture from the outside. Having the outside as a background, over a timespan of 6 hours was a challenge. I did count on it, so I made sure I had enough room on both sides of the shutterspeed to both keep a sharp image (Yes, the flash freezes movement, but having ambient light in the exposure can introduce ghosting on too slow speeds – an artifact I am not too fond of) as well as staying within sync speed of the flashes. There were *some* differences in the background brightness throughout the day, but nothing I cannot live with.

For the lighting part, I chose a big key light, gelled with a full CTO in an octobox. I moved it abit more to the side, to light primarily only one side of the face, letting the other fall into semi-deep shadows. Balancing for the CTO in camera, enabled the background to fall into the blue shades I was after. Then I added an ungelled kicker behind on the shadow side to prevent the dark part from falling into the background. I made an effort of “overdoing” this a bit, giving the kicker room to draw attention to itself. It gave the impression of light falling from the background outside (same temp) and helped modelling the face in 3D. All in all, this lighting gave me the edgy, contrasty look I was going for. Finally, I flagged off the Key so it would not touch the background (very small area @ 200mm, so easy to flag off) and added an ungelled hairlight for good measure ;-)

A breakdown of each light:

Exposing for the background

Exposing for the background
No lights, with a rough WB of 3000K'ish, notice the blueish tones on the blinds

Key Light

Key Light
No diffusion material, notice how the blinds turn back to neutral, since the color-gelled Key now illuminates them.

Key Light

Key Light
Now diffused properly (look at the reflection in the background window.. Shadows are more pleasing. (Also loosing a stop or so of light)

Kicker/Rim Light

Kicker/Rim Light
Bare bulb, helping sculpting/modelling the face against the background

Key + Kicker Light

Key + Kicker Light
Now the background has been flagged off from the Keylight. (Almost, the white part in bottom right of frame is still exposed to key, will be tweaked later).

Hair-/Rim light

Hair-/Rim light
Alone, giving definition to the hair and rims out the shoulders.

Final Portrait

Final Portrait
All lights on, background properly flagged.

So, there you have it – 60 Portraits in 6 hours ;-)

Equipment used:
Elinchrom Ranger RX Speed a/s
Elinchrom Octabox 135cm
Canon Speedlight 580EXII
PocketWizard MiniTT1 / MiniTT5
Canon EOS 5D Mk.II
Canon EF 70-200mm / f2.8
Lee Full CTO Gel
Lenovo W701ds

Lidt indsigt

Jeg har et par blogposts liggende, der nok snart skal se dagens lys ;-) I mellemtiden vil jeg dog gerne lige have lov at linke til denne post jeg faldt over. Den er vel egentligt et direkte respons til det iPhone shoot jeg også har linket til på min blog tidligere. Anyways – Tim Gander tilbyder en lille portion indsigt i hvorfor man overhovedet bør benytte en professionel fotograf hvis man ellers er seriøs omkring sin forretning. Jeg har selv en større blogpost på vej omkring det – den blev trigget af en kommentar på Amino forums (Du ved, der hvor gud og hver mand der vil tjene en skilling starter – og vist nok sjældent bliver så længe igen) – nemlig at en bruger spurgte om hvor meget et produktfoto shoot ville koste, og en fotograf svarede med et tilbud (med en uholdbar lav pris iøvrigt) – hvortil een af kommentarene var: “Hvorfor er det så dyrt? Hvor svært kan det være at tage et skide billede – det er vel bare at købe et kamera og klikke på knappen!” – flot Brian.

(Det er en engelsk blog – sorry ;-)

MicroStock – good or bad?

Microstock… Hvis du spørger mig, er det forkert på så mange niveauer at det kan få blodet til at koge for de fleste professionelle fotografer. Det var, oprindeligt, en fantastisk mulighed for nye up-and-comming fotografer til at få en masse øvelse, måske en smule eksponering (pun intended :p) og hvis de var rigtige dygtige, opleve at få en lille skilling for deres arbejde. Idag, har disse micro stock agencies ødelagt mere end de gør gavn. Med metoder der minder om børnearbejde i u-landende har de en stor del af branchen i et jern-greb der er svært at slippe ud af. Det er beskidt. Enhver professionel fotograf, eller folk med ønsker om at blive det bør læse John Harrington’s “Best business practices for the photographer” for et wake-up call af dimensioner.
Den digitale verden har om noget revolutioneret den fotografiske forretning på mange planer – men der ér en bagside af medaljen; konkurrencen er vokset enormt. Det er normalt en god ting – en rigtig god ting endda – men der nogle forvridende elementer der spiller ind.

Som tingene er idag, er den professionelle fotograf NØD til at differentiere sig selv overfor sit marked for at kæmpe mod horderne af glade amatører med udstyret på rette sted. Kvaliteten som moderne udstyr leverer idag, er nærmest per automatik professionel. Derfor må fotografen der skal leve af det, være bedre end amatøren der gladeligt gør det gratis. Han skal være bedre til at lave billeder, bedre til at køre de projekter som et shoot er og er nød til at differentiere sig selv som værende en forretning som en kunde kan stole på. I sidste ende er det vel i en kundes bedste interesse at kunne skyde en opgave afsted og have en god mavefornemmelse om at der ikke alene kommer et resultat tilbage, men at det også er brugbart – hver gang. Det er her den professionelle fotograf kan differentiere sig. Han kan også skille sig selv ud fra mængden, ved at det arbejde han leverer blæser håret tilbage på enhver kunde – her er problemet så, at der findes masser af kreative ildsjæle der kan levere ligeså god kvalitet; og gladeligt gør det gratis: Det er disse personer der er det største problem for industrien: De gør det i god mening, men i processen skader de de professionelle fotografer der skal leve af det, og endnu værrere, de skader sig selv. De værdsætter ikke deres eget arbejde og får ikke den betaling de fortjener – og nej, en note i et magasin om at xxx har leveret billedet er ikke betaling nok; det er at blive snydt for 5000,- og få bygget sig et renomé som den fotograf der leverer gratis billeder (og når du stopper med det, står den næste klar i kulissen til at gøre det samme – og du tjener stadig ikke en krone). Det er synd for dem, og det er synd for branchen – de eneste der i sidste ende tjener på det, er kunderne – og her tænker jeg bla. på magasiner med op mod flere millioner oplag, der sparer en skilling på deres kreative udgifter. De burde skamme sig og være dem der tager teten – men som med alting, er det cold cash der vinder; og det kan man ikke holde imod dem.

Hvad skal den proffesionelle fotograf gøre? Være professionel selvfølgelig – ikke falde i fælder som at arbejde gratis, levere højeste kvalitet arbejde og gøre det hver gang. Han skal være en forretning kunden kan stole på – hver gang.

Denne post blev lidt længere end det var meningen – i virkeligheden ville jeg bare dele dette link: Time Magazine Cover Photo Ripoff – en artikel på engelsk om Microstock – spændende læsning ;-)

FStoppers.com: Kameraet ér ligegyldigt ;-)

FStoppers.com har lavet en lille video, der viser forskellen på professionalisme og know-how modsat den ivrige amatør, med lidt for mange penge til det dyre kamera. Som jeg har skrevet om tidligere, handler fotografi i langt højere grad om motiv, komposition og lys (at se det, finde det eller selv at lægge det) end det handler om hvilket kamera man bruger.

Selvfølgelig er der basale krav til kameraet, og som med alt andet handler det om finde det rigtige værktøj til det job man udfører. Når det så er sagt, så med på denne video, hvor de gennemfører et ganske fornuftigt fashion shoot med en iPåne – og nej, det er ikke fordi iPåne’n er lækker og fin at det spiller; det er fordi manden ved hvad han gør ;-)

Strobist: Using ND Filters to Kill Depth of Field

Strobist: Using ND Filters to Kill Depth of Field.

David Hobby har skrevet en lille artikel omkring brugen af ND filtre. Der er ikke noget nyt under solen; men han giver en god lille walk-through over den praktiske brug.

Hvis du ikke helt har forstået idéen med hvordan du kan bruge et ND filter til andet end at følge Digital Foto magasinets sædvanlige idéforladte forslag med at få “silkeblødt vand” så læs med her, og se hvordan fænomenet med at begrænse det indkommende lys ud over kameraets tekniske formåen kan bruges med flash til at opnå den kreative effekt du ønsker.