Now that DSLR-mania has gripped the small format film making community like St. Vitus’ Dance, let us take a moment to step back and reconsider what our gear does for us.

There are two type of film makers: megalomaniacs who want to stroke their egos by having the biggest and best of everything, even if their ideas are small. I nominate James Cameron as the leader of the pack.

Then there are the folks who have an idea and will do whatever they can to realize it.

They need to strip away the excess baggage of traditional film making.

For one thing, they are usually strapped for cash and are paying for the project out of their own pocket. They tend to look for small, low-profile solutions that minimize the overall requirements of gear and crew.

I remember one of my friends who had an actor holding the boom over his own head. The shot was framed so you couldn’t tell. You can’t get many takes before muscle fatigue sets in, but you’ve just saved yourself a sound person for one day. ‘Sorry sound guy- at least you’re booked for tomorrow.’

Never forget that the lapsed guerrilla-style film maker can also come back down from the Hollywood Hills and do something cool, like Rob Patton Spruill, who is revitalizing a media-arts center in Boston that inspired 2 generations of film makers, but closed its doors due to the weak economy.

Historically:

We used to shoot Super-8, later known as ‘Stupid-8.’ Now that is a format that comes with emotion built-in. Try to replicate that in the digital world. It takes a lot of work.

When we were working with small video gear, we opted for Hi8 and then mini-DV. Some of us hopped on the HDV bandwagon and some of us held back, skeptically.

I make this long winded introduction to remind you that your DSLR is a trade-off right now- you’re trading ease of production for pain in post.

Listen to what Sathya Vijayendren has to say about shooting with his DVCProHD camera and the Letus35 adapter:

Depth of Field:

How It’s Making A Mess of the Video Industry

The mild-mannered video person, with a budding desire to film great or not-so-great things, whether they’re people or plants, graffiti or a garbage in the street, will freely admit that images from a video camera never look like what’s in the movies. What it all comes down to, is the camera.

Nowadays with iPhones snapping pictures, the camera industry and the video camera industry are finally in one boat, stranded at sea, with one life preserver. What to do? Release a camera that shoots pictures AND video; problem solved. It was addressing a need many photographers once had: “I want to shoot video along with pictures in the field.” Their prayers were answered with the Canon 5D. The 5D put a notch in the timeline where video cameras and SLR camera’s merged, or consummated in a dirty Vegas hotel room, behind the backs of big video camera companies.
And so was the death of the “Depth-of-field adapter”.

Before I explain that, understand this first :

Depth of field is a photographic term, not a video one. If you pick up a real camera, like Nikon or Canon SLR and snap a photo, the sharpest part of the photo is what you focused on, the blurriest part is the opposite. This simple concept is the science of photography (and migrated into motion picture filmmaking), but never really made it to video. Video cameras were designed so that mostly everything was in focus.  This became part of what was known as the “video look.”

The first shift out of the “video-look” was a few years ago with the Panasonic DVX100, or the “24p” camera. This video camera stirred the industry by using traditional 3:2 pull-down to shoot 24 frames-per-second (the  film standard) onto DV tape.

An industry was born from that camera alone. Most people thought they were staring at something shot on a film camera, but no, the camera was only $5,000 (cheap!). The independent film community transformed overnight with the Panasonic camera. Filmmakers didn’t have to max-out 15 credit cards to make a film, they just bought the DVX and a laptop. But the holy grail was still depth of field. Whereas “24p” imitated the motion of film, the introduction of the depth-of-field adapter flipped it.

The depth-of-field adapter goes on the front of a video camera, replacing the camera’s lens. Depending on the mount installed on the DOF adapter, you can use that corresponding lens. My adapter uses a Nikon mount, but Letus sells other mounts for SLR lenses from Canon, Sony, Contax, Zeiss. There are even mounts for professional cinema lenses!

So why use this adapter instead of buying a Canon 5D or another “DSLR” camera? Simple. These new “DSLR” cameras can’t do the job. A video camera is designed for shooting video and audio, with maximum adjustment. The DSLRs have limited shooting capacity, audio connections, video quality and have a tendency to overheat. The upside is that it’s small, unobtrusive, and promoting more people to films.

I own the Panasonic HVX200 and run a video production business. The camera is fantastic and gives me the “24p” look along with the major HD formats. As an editor, I love that the camera shoots onto cards (no tapes) and preserves the video at an extremely high quality. Most other cameras (including the DSLRs) compress the video into small files (kicking out color and sharpness). The result are files that have to be converted on the computer for further editing (adding even more garbage). The files from the HVX200 don’t have to be converted and can be edited immediately; I love that. Okay, I’m done being preachy.

Q & A

Q: The DSLR allows the use of SLR lenses out-of-the-box. Why invest in a depth-of-field adapter?

A: The DSLR is limited with audio and compresses the video too much. All I want is depth-of-field, not to redesign my whole operation, from shooting to post production. I want an all-terrain vehicle, not a sports car.

Q: Have you used other DOF systems?

A: The main systems out there are Letus35, Brevis, and Red Rock. All of them do the same thing and have mounts for different cameras and lenses. I’ve seen the other systems in action and their results. For me it was price; the Letus35 is the cheapest and most robust. The other systems have brought in computer chips and other innovations that threaten their lifespans. The Letus35 is simple. With the way its made, it seems like it’s 50 years old!

Q: How long to set up?

A: All DOF systems required rails to support the camera, the adapter and the lens. If you know what you’re doing, you can set it up in 5 minutes. Most people keep the rig together throughout the shoot. The camera becomes merely a recording device when used with the DOF adapter; the video iris is kept wide open; aperture and focus are adjusted on the SLR lens. I have a system of rails that mounts on a tripod but is not so good handheld. I want to invest in a better system like a Redrock or Zacuto shoulder mount, which are modular like the cages and rigs for the RED ONE.  Since focus is so critical, I don’t use the camera’s flip out monitor; the SmallHD monitors are great with aluminum construction, 720p resolution, and price ($899 for a 9″ monitor). In the end the camera resembles a serious film camera.

Q: Do you experience vignetting?

A: Letus plugs their product as having “absolutely no vignetting” and I haven’t experienced this problem. I believe that issue depends on the lenses you use, especially wide angle.

Q: What about the spinning ground glass? How does that work?

A: The ground glass inside the adapter is spun by a motor powered by two batteries. Replacing the batteries is easy. The motor is triggered by a red switch on the adapter housing, and is very silent. The motor can be heard on the external camera mike (more like a vibrating noise) but when using a shotgun microphone , it’s non-existent.

Q: How do you like the footage?

A: I think the footage is better than I ever imagined. The HVX200 shoots well already but coupled with the adapter, it’s beautiful. To get great images with the Letus, you need great lenses and appropriate amounts of light. My Letus35 came with the Nikon mount, but I’m switching up for the classic M42 mount so I can use the Carl Zeiss “Jena” lenses made after World War II. On eBay these lenses are around $100 and beautiful. The M42 Jena lenses are meant for medium format cameras, which require larger lenses and sharper glass;  they are closer to cinema lenses, and on the cheap. I’m shooting through the Leica lens built into the HVX camera, so placing a sharp lens on the DOF adapter is key. Sure I could go with Nikon lenses, but the advantage of this setup is using what I couldn’t before; there’s something amazing about that.

Youtube tests:

–
Sathya Vijayendran
www.vijayendran.com

A client requested we use HDV for a shoot. I have never embraced this format, but hey, the editing payload is light.

My colleague Sathya, whose review of the Letus35  is coming up next in this blog, brought along a Sony HVR-Z7U.  It had a big honking piece of Zeiss glass on the front. This always cheers me up.

We lit the scene in a classic 3-point head shot style and taped 2 days of  interviews.

I took the footage home and started capturing through Premiere  CS4 with a Sony HVR-M15 I rented  from the super nice and excellent rental facility in Manhattan, Hello World Communications.

The deck didn’t appear in Premiere’s device support list, but I was not terribly concerned and selected ‘standard’ from the drop-down list.

Hour One captured fine, so I proceeded to the other two tapes. When I was finished, I started reviewing the material and quickly noticed that there was no sound playing back when the clips were popped into a sequence.

Now, a lot of people who gripe about Premiere are going to be laughing at me right now, but I have made this program work for me over the years and the ROI has been very high.

Nevertheless, it became clear that I had to employ a strategy I learned  long ago: bail-out quickly and find another solution.

The solution that I gravitated towards was Sony Vegas Pro 9.0. I figured, ‘I have Sony tape, shot on a Sony camera, capturing with a Sony deck.’ Seemed like a logical choice, and indeed it has been working well. There is a 30-day trial version of Vegas, so you can actually do a real-world project and see how it works for you.

Now, this is not meant to be a ‘shoot-out’ between Premiere and Vegas. I’ll save that for later. This is just a quick tip if you run into this problem. However I have found that another weak point in the Premiere work flow, IMHO, is Adobe Media Encoder. It’s difficult to work with, slow, and unreliable. So far, exporting from Vegas seems simpler.

Oh, btw, I checked into upgrading to CS5. There is only a 64-bit version of the program, so in addition to upgrading the program , I would need to upgrade my computer and OS. That felt a bit like a drastic alternative.

Kodak Zi8 Pocket Video Camera

This is my Kodak Zi8 Pocket Video Camera
rig in what I like to think of as my ‘Steadicam’ configuration. It’s mounted on a Stroboframe, one that I used to use for a medium-format film camera with a Vivitar flash. As you can see, the small Sennheiser MKE400 Shotgun Microphone attaches neatly to the shoe on the top.

Mounting the camera and mic in this fashion gave me a great deal of flexibility in capturing shots that involved movement and distance from the subject. I was able to do jib-style shots and to walk backwards with ease. The grip on the frame, combined with the electronic image stabilization on the camera, gave really nice results.

I am including two videos I made with SneakGeekz so you can see what I am talking about. We shot them at 720 30P. I am finding that this is about as high of a data rate that I am willing to take-on with my Dell Precision M90 Core2 Duo laptop.

As a matter of fact, what I have been doing is cutting in Adobe Premiere Pro CS4 at widescreen DV resolution and then conforming to 720P for export through Adobe Media Encoder.

The way to do this is to set-up your project as AVCHD 720, then create a new sequence with widescreen DV settings. Rough cut the clips by adding the shots in order- you can even do it without trimming. Select all of the clips, right-click and check ‘scale to fit.’

Then, render the whole timeline. After it’s finished, you’ll be able to do your fine cut easily and smoothly. When it’s done, select everything in the timeline and copy and paste it into a new sequence with AVCHD 720P settings and it’ll be ready for export.

Let me address a couple things I have noticed about the Zi8.

You’re sort of stuck when you want to do a shot that spans the range of both focus zones. It’s not really possible to flip the switch on top of the camera in the middle of the shot when you want to move in close. You’ll see in the second video below that some of the close-ups are a little soft.  Because the viewfinder doesn’t tilt, if you move off-axis, you’ll also be hard-pressed to notice whether you’ve lost focus, too. These aren’t deal-breakers, just things to keep in mind.

If you are shooting light skin and dark skin in the same shot, be careful when using face detection to set the exposure. There may not be enough latitude under natural lighting conditions to achieve proper exposure for both. Think about a small kit for augmenting your set-up. Even a collapsible reflector on a stand could prove to be immensely helpful.

While the shotgun performs nicely when the subject is at a distance from the camera, for close-range shots, the on-board mic is actually quite nice. I would have hi-pass filtered out the wind noise from my outdoor shots, but I ran out of time- there was a deadline.

Overall, there is a lot to like about the Zi8, as I have said before. I think the thing I appreciate the most about it is the astoundingly accurate colorimetry. It really renders colors believably, even under varied lighting conditions. Seems like Kodak has capitalized on their pre-digital technological achievements.

Hope you like the vids.

  A basic design element of the DSLR, inherited of course from the SLR, is a mirror that diverts light from the imager to the optical viewfinder. When shooting video, this mirror remains in the ‘up’ position so that light can continuously reach the chip. Naturally, this means you no longer get the benefit of the optical viewfinder and you need to use the LCD.

While these cameras allow for magnified views on the display that enhance focusing ability, when shooting under conditions where light hits the LCD, it can be difficult to view the screen.

A whole host of products are available to address this problem, at various price points and with various features. In the video below, Carlos Sanchez, aka SneakGeekZ, an icon of the world of sneaker collecting, reviews the LCDVF affixed to the Canon EOS 7D.

There is a lot to like about this product, as Carlos explains. My only complaint was the lack of a diopter, that would allow you to adjust the focus if you need to wear glasses. Contact lens wearers needn’t worry about this, as they will easily be able to put their eye up to the glass.

This is our third avidemux demo. Who knows how many more we’ll need to get through the large number of useful features of this piece of freeware?

Why just yesterday, this program saved my butt. In a story that will be all too familiar to many of you, I was up against a deadline and had little time for things to go wrong. This is when software smells your fear- and breaks. We exported our final cut from Adobe Premiere CS4, a program that I like, and one that is feature-rich and smartly designed.

For the encoding, we used Adobe Media Encoder. This program is a little new to me. I don’t know how long it’s been around, and certainly it’s on track to be a very useful tool, but for some unknown reason our final mpeg4 file was missing the audio. Oh, the file claimed it had audio, but no player programs agreed.

With our hard-out in sight we didn’t have time to export another pass of the mpeg4. It was a 20-minute compression, and we had only ten minutes in our budget. What I did instead was to export the audio portion of the cut to a windows waveform file, something that was quite speedy and would be easy to check when it was finished. That worked just fine: we had our program audio.

Media encoder had produced a great looking H.264 file for us, so we had our video too. The only problem was that the audio and video were in two separate files. The solution? Use Avidemux to marry the two together. We’d be able to stream-copy the video track without re-encoding, and add and compress the audio track, combining them into the final deliverable file. I picked Avidemux for the job because I knew it would be fast, and it was.

Now some of you will recognize that ‘demux’ in ‘avidemux’ means to de-multiplex, or to take apart two streams of information, such as a video track and an audio track, or even a stereo audio tracks into a pair of mono tracks. In this example we are actually using the program to ‘mux,’ or multiplex the two tracks, and this versatile piece of freeware is definitely fit for the job. Check out the video demo to see how it’s done!

We’ve started shooting with the Canon 7D and the results are pretty impressive. The camera records video compressed with the AVCHD H.264 codec, and 48kHz linear PCM audio, all wrapped in a quicktime MOV container. We used Quicktime Pro,  version 7.6.6, which supports H.264, to extract the clip below.

The data rate is 47.10mbits/sec. To put this in perspective, standard definition video on DVD clocks in at a peak of 8mbits/second, while the Panasonic format DVCPro HD, which does not utilize temporal compression, records to tape or P2 cards at 100mbits per second. Of course the Canon 7D, body and lens, costs only a fraction of any of the Panasonic camcorders.

Given the large amount of data and the complexity of the AVC codec, you’ll need some real horsepower to edit this format, unless you just need to trim and append clips. I love this back-to-basics approach to editing. It reminds me of cutting Super 8.

You can see the complete file size specifications in the video description on YouTube. This 1 minute 42 second clip came in at 564 MB, but now that a terabyte of storage can be had for a hundred bucks, this isn’t as frightening as it perhaps once was:

Iomega Prestige 1 TB USB 2.0 Desktop External Hard Drive 34275

A couple of things to note: the camera only shoots 60P, even at 640x480. The only mode that offers 24P is 1920x1080. Technically, you are shooting at 23.976fps so the camera can accomplish fitting the frames into the 60P timebase using the pull-down method.

When shooting video, the optical view finder is not available because light has to reach the APS-C size CMOS sensor continually. While the LCD panel does offer a toggle through 5X and 10x magnification to aid in focusing, under bright light conditions you will probably want a hood with a lens. We’ll review one in our next post.

I won’t say a lot about the audio right now, but we are off to shoot with it today, so I’ll add something in the comments. Please feel free to share your experiences or questions.

Overall, this is a rather large beast compared to the AVCHD cameras we’ve been reviewing, and it takes a bit more commitment to work with this camera. With a zoom lens attached it is also, naturally, kind of front heavy, which is atypical for the video shooter.

The price is a bit steep, but the sensor is 18 megapixels- kind of crazy, really. The CMOS revolution has really made high-res imagery available to everyone. Now we need media that can record higher bitrates so we don’t have to edit long-GOP video. Otherwise, post-production is headed toward a bottleneck.

Check out this video, shot for http://SneakGeeKZ.tv

Here is a cool little device I saw a while back at NAB. It was invented by a DP and he was demonstrating it at the Lowel booth. Apparently Lowel decided they liked it, because they are now selling it as part of their product family.

Basically it’s a matrix of cool LEDs and warm LEDs. You control the white balance by dialing-up the intensity of one set or the other. Simple, yet effective. The output was bright enough for a head shot from several feet away.

clipped from lowel.com

Lowel Blender^TM Brings Intuitive Lighting to Your Location.

2 sets of LED’s in Tungsten & Daylight color are quickly & easily blended. Perfectly match the mixed light sources of your real world location, or create a contrasting color blend to make your subject stand out. Let your creative instincts be your guide.

Powered by AC or battery for maximum flexibility.

Twin sets of 5000k Daylight and 3000k Tungsten LED’s, with rotary dimmer controls for each on the back of the fixture.

Blender packs more punch than you’ve come to expect from compact LED lights.

The lamphead’s compact size (4 x 3 x 3″) makes it small enough to fit on a pistol grip or compact light stand.

Powered by AC, Camcorder Battery Sleds, or conventional 12v connections.

Supplied with a set of front diffusers for variable softening & diffusing of the blended output.

Is it worth it to process poorly-shot H.264 encoded video? I mean, AVCHD is so highly-compressed and adding filters requires a decompression-tranformation-recompression step. Would the result be so bad as to be unusable? That was my question.

I shot a demo of how to pull pin-bones out of a salmon fillet with a pair of needle-nose pliers for my food blog, Obsessed Chef. I did a poor job shooting it because I was just too lazy to light it. The result was blurry and the color rendition suffered.

I had been playing with AVIDemux, a freeware cuts-only editor, transcoder, and image processor. I like it a lot. If it had more compression options for Flash and allowed you to make Shockwave files it would be perfect. You can (and should) download it at http://avidemux.org.

It includes some filters that are borrowed from Avisynth, a reliable old pal that allows you to script all kinds of video transformations. But because of the recompression step, I just wasn’t sure if the end result would be better or worse.

You’ll find that many of the filters work in YUV color-space, which is unintuitive to use and impractical without a scrubbing preview window. I checked all of the filters and discovered that mplayer eq2 had just the controls I was looking for. You can perform operations in the more familiar RGB space and, by fiddling with the overall gamma and then the individual channel gamma settings, you’ll be able to do a nice job of color-correction without clipping the levels. The controls are shown below:

As for the blurriness, well my hope was that shrinking the size would minimize the problem. I did perform a light sharpening using the sharpen filter. Be careful with this one, and definitely scrub through the clip. It’s easy to overdo it, and you won’t really get anything better.

So below are the results and I leave it to you to decide whether it was worth it. You can provide a guess about what I think in the comment section!

Here’s a speedy and feature-rich piece of freeware that you should have in your video toolkit. It’s arguably the best thing to come along since Avery Lee’s VirtualDub. What do these programs have in common? They’re both very streamlined and perform their operations more quickly than even some of the best commercial software.

In our first video, we show how Avidemux can be used to trim mpeg4 files shot with H.264 codec. This is a long-GOP codec, so the efficiency of being able to trim without reencoding turns out to be a real time saver.

Besides cutting, appending and transcoding, Avidemux can also be used to replace the audio in an mpeg4 clip and to make conversions to iPhone/iPod formats with one-button simplicity.

While there are a couple quirks to be alert to, it certainly makes sense to add this fine piece of software to your video editing software portfolio today.