Hands-On HDV

Remember what DV did for standard-definition video acquisition?
Ostensibly a consumer format, it opened the door for more than just
accelerated home-movie production. Ten years down the line, it’s
enabled everything from hip TV commercials and documentaries to Danish
art films and shark-attack thrillers. Now HDV, which was also
introduced as a consumer format, is catching on even more quickly.
Whether it’s your primary acquisition format or a B-roll solution, if
you shoot smart and edit smart, you can get great value for your money.

But it can be tricky to get the best results. HDV is highly compressed,
which makes it a poor choice in post-production – you’ll probably want
to convert that MPEG-2 stream to something else before you do any
editing or compositing work. And just using HDV means you have to make
a choice between Sony’s HDV recipe and JVC’s- which seems to be
essentially a choice between 1080i and 720p, but with lots of other
considerations in the mix.

To help you make those decisions, we asked HDV early adopters for their
best tips on working in the format. Over the next seven pages, you’ll
get lots of information on shooting and posting HDV in the words of the
pros putting it to the test, out in the field and back in the studio.
These are still early days for HDV, so consider this less a handbook
than a field guide for travelers in uncharted territories- helping to
keep your head above water and the ground beneath your feet.

Nick Tucker
President/director/editor
Golem Films, Greenville, CA
Recent HDV Projects: American Vampire

We really wanted to shoot 25 frames per second since it was the best
alternative to shooting at 24 frames. That’s one of the things that
sold us on the Sony HVR-Z1U, that it shot 50i and 60i. While we did
look at the CineFrame feature on the HVR-Z1U, we weren’t really happy
with the motion quality. But I had shot a couple of projects at 25p
[50i] and slowed down the footage to 24p and had really good results,
so that was what we opted to do. As it happens we’ve had extremely good
results.

When you work with Final Cut Pro and try to put PAL footage into an
NTSC timeline it can do nasty things to your motion quality. It kind of
stutters and it’s not pretty. So we used Cinema Tools to conform the
video to 24p first, and a Nattress plug-in to de-interlace it. After
you slow the audio down four percent, and pitch-shift it to compensate,
you’ve effectively got 24 frames. DVDs support 24p video, and it
displays smoothly on normal NTSC televisions, so it’s the easiest way
to get your PAL footage onto an NTSC monitor.

For the DVD that we’ll use for festival submissions, we’re exporting it
from FCP as a QuickTime 24p movie and using FCP’s Compressor to go to
DVD. At least try some of the major plug-ins before monkeying around
with slowing footage down. I’d recommend that anyone who wants to do it
download all the demo versions of DVFilm Maker and Nattress Film
Effects and try them out. It’s free to try them, after all, and they’ll
give you a good idea of the result.

Our color correction is kind of unique. The film is shot in color but
we’re de-saturating many scenes to black and white, for several
reasons. One is a stylistic choice- we like the look as our movie is
about vampires and has a Gothic feel. We’ve had some amazing results
with pretty dark footage that really looks beautiful in black and
white. We just adjusted in FCP 5 with the three-point color corrector,
first de-saturating and then adjusting the contrast. It was able to
take care of about 90 percent of the color correction that we needed to
do. The color was pretty good out of the camera. We shot a scene in a
comic book store and the colors of the comics are really amazing. Aside
from the built-in Final Cut Pro color corrector, the one thing we have
used is the film gamma effect from Nattress.

Jody Eldred
Director/writer/cameraman
Los Angeles, CA
Recent HDV Projects: Inserts for CBS’s JAG

Get the Right Camera for the Right Job
No one camera is perfect for everything. The lesser the camera, the
more experienced the camera operator required to get good images out of
it. In some respects, it takes a more talented operator to make the
Sony HVR-Z1U look good. Use that camera in places where it’s more
logical to use it than the Sony HDW-F900. We were shooting the big
Sturges motorcycle rally up in South Dakota, and there are a lot of
things the Z1U was perfect for, such as mounting it outside a vehicle
with a wide-angle lens to get up close to a motorcycle. That would have
been impractical, and maybe even dangerous, with an F900. We would use
HDV for shots inside a bar or club that required a lower-profile camera
or greater light sensitivity, and some handheld stuff on motorcycles.
But when it came time to do all the interviews, we didn’t shoot any
with the Z1U. The F900 is more controllable, has shallow depth of
field, and we can make it look how we want pretty easily. But I like
shooting with the Z1U in small places. It’s particularly useful to keep
a wide lens on the camera so you can keep up close to your subject and
stay zoomed out. It gives me a little better effect all the time.

Match Your HDV Settings to Your Workhorse
Cameras
To match the Z1U’s footage with the F900, I prefer to make the picture
a little bit richer so you start out with a little more to work with.
So I turn the chroma up a little bit, and with the phase I warm the
picture up a little bit- just a few points, nothing extreme- and make
sure the sharpness is generally similar to the F900, not too dull or
too crisp. I prefer a look that is richer yet allows me to see into the
blacks, so my personal preference is the Cinematone 2 setting with
black stretch on. That, to me, gets the look that’s closest to what my
F900 does. There are so many things you can do- crank the chroma way
high and do all kinds of stuff to get super high contrast- but
personally, I prefer to capture something really excellent in the field
that’s close to the look I want, and if I want to tweak it in post I
can always do that.

Consider Alternatives to Cineframe Mode
In my opinion, CineFrame 24 is not a good mode. I’ve been shooting at
25 frames per second, or 50i. My F900s were set up for 23.98 but I shot
the Z1Us at 50i, and we’re going to offspeed-play it in post at 23.98.
There will be a tiny speed difference and it should intercut just fine.
That, for us, was the best route to make everything look 24-frame.
Also, there may be a film-out, and this is the best workflow for
getting there.

Steve Gibby
Producer/director/DP
Cut4 Media Group
San Luis Obispo, CA
Recent HDV Projects: Stock footage for Mammoth HD

If you own or plan to rent the current top-performance HDV camcorders,
the JVC GY-HD100U and the Sony HVR-Z1U, how do you decide which one to
use on any particular project? Let’s analyze their key features.

The Z1U records 1080i/60 HDV, and several networks broadcast in 1080i,
including NBC, CBS and HDNET. It also records 480i/60, which looks good
for SD reality TV. The Z1U has the capability to output 480p analog SD
from 480i, but that’s not as good as shooting 480p natively. However, I
use the Z1U for projects where I need a run-and-gun form factor, and
native progressive is not the look I’m after. (The Z1U’s optical image
stabilization is nice for those situations.) The big drawbacks to the
Z1U, especially for indie film production, are the lack of true
progressive 24p and the lack of a removable lens. There are indie
shooters that use the Z1U with a matte box, follow focus, etc. In my
opinion, though, true progressive video is a better choice for indie
production. The 12x fixed lens is also a big drawback for far-away
sports, nature or any long-focal-length needs. I simply don’t use my
Z1U in those situations. I also don’t use my Z1U when I need good depth
of field. I’m used to getting good depth of field easily with
detachable professional lenses on my 1/2-inch and 2/3-inch CCD cameras,
but it’s extra work on the Z1U. I’ve found that the Z1U is excellent
for acquiring mobile 1080i stock footage of scenery, nature and
well-lit indoor subjects. I avoid shooting fast lateral moving subjects
with the Z1U. If I have to shoot them, I increase the shutter speed and
pan with the action.

One decided advantage of the HD100U is its ability to record in
480p/60. For shooting high-motion subjects, 480p/60 beats the heck out
of 480i/60. If the footage is destined for television, slow-motion
sequences will be much better in 480p/60, and frame grabs for a Web
site are crisper. If the HD100U’s motion-smoothing filter for 30p is
effective at eliminating judder, the HD100U may be a good camera for
faster-moving television production (sports, etc.). Another advantage
is the ability to change lenses. With a 2x extender, the HD100U’s 16x
lens could be a good long lens for sports, nature, etc., and the.8
wide-angle converter should make it a decent option for ENG-style
situations (balanced with large batteries on the camera). The 13x lens,
though pricey, should be a great lens for a wide variety of
shorter-focal-length applications. With the 1/2-inch to 1/3-inch
converter, lenses such as the Fujinon S20x6.4 or the Canon YH19x6.7 can
be used. These are $3,500, 1/2-inch SD lenses that work well in SD
modes (480i and 480p), but whether they will resolve 720p adequately
remains to be seen. Both cameras are excellent performers in their
designed functions. For real mobile, short-focal-length, reality-look
production, I’d probably use the Z1, accessorized with a wide
converter, etc. For high-action handheld and tripod work (30p
motion-smoothed), long focal-length tripod work requiring
interchangeable lenses, or work that required 480p/60, I’d choose the
HD100.

Bernie Mitchell
President, Silver Platter Productions
Knoxville, TN
JVC Pro HD consultant/evangelist
Recent HDV Projects: Test projects only

About a month ago, I received one of the very first JVC GY-HD100U HDV
cameras to shoot with and test out in post. Because it is a pure
24-frame progressive HD camcorder, JVC asked me to introduce it to film
schools.

First, know that this camera is a progressive camera. As a shooter, you
need to understand the difference between interlace and progressive and
understand the type of look you want to get and how to shoot for it.
Remember, high-def sees everything. What I mean by that is, high-def
doesn’t exaggerate things, but it does let you see things you wouldn’t
ordinarily see.

One of the nicest features on this camera, which I use all the time, is
setting the viewfinder to display both safe guides for 16:9 aspect
ratio and 4:3. That way, when you frame for 16:9, you’re also aware of
protecting your 4:3 shot. So, if a project ends up going SD, you’re
fine.

Another feature I use all the time is Focus Assist. This is a great
benefit to a high def shooter. Focus in high definition is critical
and, when you’re looking through a viewfinder it’s difficult to tell if
you’re actually in focus. When you turn Focus Assist on, the viewfinder
goes black and white. As you rack into focus, the outline of your
subject pops into color and there’s a color peaking around the edge of
your object. It makes it very easy, particularly in outdoor situations
where you have extremely bright ambient light, to tell if you’re in
focus. With this camera and its interchangeable lenses you can now get
that cinema look of shallow depth of field. With Focus Assist you can
easily rack focus through from one point of the image to another,
letting you pull off that cinematographic trick of leading the
audience’s eyes through the frame with focus.

One last tip: This camera shoots 24 frame and 30 frame progressive. If
you haven’t shot 24p before, I strongly recommend that you use a tripod
or some sort of camera support. As I mentioned earlier, high def shows
everything. So, you’re really going to see a lot of camera movement in
24p. Support your camera!

Andreas Timmes
Producer, Creative Development
MSS – New York
Recent HDV Projects: Currently shooting a promotion for an upcoming Japanese movie

Shooting with the Sony HDR-FX1, we noticed that the image quality,
although generally very good, becomes a bit noisy (or blurred) when we
panned with the camera. Actually, this is quite typical for MPEG-2
compression, even in HD broadcasts, as we noticed in the Summer
Olympics broadcasts. With the HDR-FX1, the blur/noise is more visible
because it features a less expensive codec. Using an HD-SDI converter,
we then hooked up the camera with a portable HDCAM deck (SONY HDW-250).
This let us bypass the MPEG-2 codec (with a transfer rate of 25 Mbps)
and, instead, record uncompressed HD (with a rate of 140 Mbps). This
resulted in a higher image quality with relatively clean pans.

Based on our experience, we also recommend minimizing horizontal
movement (panning) as much as possible in order to maintain a clean
image. If you need to pan (and want to maintain the image quality),
always pan as slowly as possible!

Tom Camarda
Director of photography
Los Angeles
Recent HDV Project: Busgirl, an independent film with director Steven Schuldt

When I was first approached by director Steven Schuldt to shoot his
film Busgirl, his mind was already made up on what
format we were going to use. Steven is very technically oriented and a
steadfast H.264 proponent. He also had a lot of faith in the HDV format
and, although I was initially hesitant, I gave him the benefit of the
doubt. Steven had seen a film that I photographed called
Grace that was shot on DV using the P&S Technik
Mini 35 adapter, which allowed us to use 35mm cinema lenses. He was
thrilled with that look and wanted the same shallow depth of field for
Busgirl. So off we went to NAB to put together our
HDV camera package.

Because the producers of Busgirl opted to buy all
the equipment needed for the shoot, the P&S adapter was just not an
option as it is very expensive. Steven found a company called Redrock
Micro, whose M2 35mm lens adapter seemed to be just what we were after.

We had our rig outfitted with a Nikon lens mount so we could use Nikon
SLR zoom lenses the producer owned. We outfitted these with follow
focus rings from Cavsion in Canada, which allowed us to pull focus
either by hand or remotely when using the steadicam. By the time the
whole rig was built, it had the profile of an SRII and an ASA rating of
50! (Not what you’d expect when shooting digital.)

Unfortunately, there is no [true] 24p mode on the Sony HVR-Z1U, so we
shot at 50i and used a 13-inch PAL monitor everywhere we went. We would
watch dailies back at the production office on a 23-inch HD display to
make sure our focus was sharp. Attention to back focus is critical in
any digital production but even more so when shooting with 35mm
adapters. I was really amazed at the latitude of this camera.

Overall, HDV is a great tool for low-budget filmmaking, especially if
coupled with the M2 adapter, which can get you that film-like depth of
field. I’m certainly keeping my eye on the next generation of HDV
cameras because the potential for quality, inexpensive image-making is
only getting better.

Kevin W. C. Wong
Director of photography
Toronto, Canada
Recent HDV Project: Tomorrow is Today

In shooting HDV with Sony’s HVR-Z1U, it is sometimes very tempting to
shoot with low light since the camera does allow for a pretty clean
image without a ton of lights. We tried to shoot in and around T4.0 as
our ideal stop because it allows for a cleaner image as well as
preserving a suitable depth of field. Particularly in HDV, because of
the sharpness of highlights, a slightly soft image is about the most
annoying giveaway imaginable, so shooting with a healthy exposure was
important to the picture quality. Keeping in mind that we were shooting
for a film-out also made us careful in preserving a clean and sharp
image since everything is to be blown up onto a big screen.

On Tomorrow Is Today we made minimal use of filters
(i.e. Polarizer, NDs, ND-Grads and a Sunset Grad.) Our strategy for
filtration was to preserve a clean image for color correction and
timing. We decided not to use softening filters for the same reason-
keeping flexibility for post, especially for the film-out processes.

We used primarily long focal lengths for our shots to avoid barrel
distortions. Since the story called for a naturalistic look, we wanted
the perspective to be as close to the human eye as possible. The
scenery at the Jersey Shore was breathtaking, so we just waited for the
right time of day and let it do most of the work.

Our aim from the start was to have the film end up in 24fps, which is
why we made the decision to shoot in 50i. From consulting with other
filmmakers using the Z1U, we found that shooting in the CineFrame
option simulates the progressive look quite well. However, if we were
to truly de-interlace in post to film, it would be a difficult process.

As for monitoring exposure and composition, we tested the latitude of
the camera in prep and learned to work within our zones. The zebra
lines, along with metering our lights to stay within the levels, kept a
consistency in our image. Compared to my prior experience shooting
another feature last year with the JVC-HD1OU [the first HDV camera on
the market and aimed at the consumer market], the Sony Z1 was miles
ahead in improving its picture quality and most impressively its
latitude. As for monitoring our framelines for the 2.35:1 ratio, we
simply shot a rack leader in prep and stuck tape on our LCD and Monitor
for the shoot.

An interesting feature on the Z1 was the user preset buttons. They
allow for repeatable‘racking’ of focus, focal lengths and exposure over
three selectable intervals of time. At first I was reluctant in using
these features, but as we went along the function allowed for accurate
focus pulls and match framing needs. Towards the end of the shoot, I
used the function almost in every scene (but of course with subtlety).
Frederic [Haubrich, the director] and I had a running joke on set that
we should do a‘slow zoom’ for every scene for kicks. Seriously though,
some of my favorite shots in the film made use of these almost
imperceptible zooms over 18 second intervals. This would normally not
be possible to do unless we had a manual lens system on the camera
along with a micro-force motor to pull it off. This detail in the Z1 is
a definitely plus amongst its peers.

Frederic Haubrich
Director
Lumià©re Media, Doylestown, PA
Recent HDV Project: Tomorrow is Today

If you get a really good cinematographer to work with, what the HVR-Z1U
camera is able to do with the three CCDs and the lens it comes with is
unbelievable. You still need to take care of your lighting- control
your highlights, contain your environment. In our tool kit we had three
ND filters and a sky filter. But with 1920 x 1080, you essentially have
amazing real estate. We did something that we hadn’t seen done with HDV
before- we shot preserving for 2.35:1 Cinemascope. We felt that it
would showcase the beauty of the Jersey Shore, which is a principal
character in our story. Because Cinemascope is much wider than 16:9, we
did the old trick of putting two pieces of tape on the viewfinder. In
post we clipped the top and bottom.

The field LCD HD monitor [LMD9050] we got from Sony has a 2.35:1 mode
that allows for transparent bands on top and bottom of the screen. This
is important since you always want to give yourself the option to
reframe the shot in post by moving it up or down and therefore keep
booms out of the frame as much as possible.

We’re really looking forward to taking this out to film. People are
going to be floored that this was shot with a $5,000 camera. So far we
haven’t done any color correction on any of our footage and it looks
pretty amazing. If you’re not going to do tons of action and you want
to shoot beautiful landscapes and go for a cinematic look, 1080i is
good. Lots of action can be a problem if you need to turn the footage
to progressive by de-interlacing for a film-out. If the final delivery
is 1080i, then I think interlaced is fine for action. We shot 50i
because it’s a lot easier if you’re going to do a film-out as it’s 25
frames per-second. When you turn progressive you just remove one frame
per-second and you slow down the audio by four percent. Shooting 50i
you’re already getting more of a cinematic feel compared to a 60i. It
has much less of video feel to it.

Marge Janssen
VP, Engineering
Tamberelli Digital Video, New York
Recent HDV Projects: Rents HDV gear to clients

The first question that needs to be addressed in acquisition today is:
what does the post/transfer house require in order to finish the
process? With the greater array of acquisition tools offered in today’s
market, it becomes almost a necessity to answer questions pertaining to
the end of the process first.

Where is the end product going? Whether your project goes directly and
only to film, is meant for television broadcast, video documentary with
possible film-out or webcasting internet use can determine equipment,
format and set-up choices. Talk to your finishing
house. Inquire about frame rates for the easiest transfer.
You may ask if the format is easily edited. Do they require
non-drop-frame or drop-frame time code? Is this format the best for
compositing/green-screening and special effects? Is it better to shoot
straight out and adjust in post or to capture the optimal quality in
the camera and just post out? 720p or 1080i? Is there a benefit or
drawback for the particular project? And, of course, double-system
audio or not?

As an equipment provider, too often we see clients who come in to
evaluate several camera models of different manufacturers side by side
for their upcoming project. Do not make your choice
by what you see in the camera. Any good camera engineer can make a
camera look good on a monitor. Any evaluation requires a look at the
recorded format. If you have the opportunity to do these side-by-side
tests, roll tape and then take it to your finishing house.

There are several manufacturers offering many HDV acquisition choices
and soon to be many more. Your finishing house can give you information
to help you make decisions before you commit to your acquisition tools.

Douglas Spotted Eagle
Managing producer
Sundance Media Group/VASST
Stockton, UT
Recent HDV Projects: Training video for Detroit Diesel, Child Abuse Prevention PSA

Shoot at Higher Shutter Speed if You're Panning A Lot

Shoot higher shutter speeds than 50/60 if you intend on making fast
pans or tilts. The MPEG compression format does not lend itself well to
high speed pans or tilts with lower shutter speeds. Practice shooting
with slower shutter speeds for the beautiful effects this can provide
if you know in advance what you’re looking for.

Pay Close Attention to Your Audio in HDV
HDV has an MPEG-1 layer II audio format that records at 384 Kbps. This
is very close to CD quality audio, except that it’s 48 KHz rather than
44.1. This compressed format is one that many HDV naysayers decry, but
it’s perfectly useable for most shooting scenarios. However, acquiring
audio at too low a level is like shooting images at too low a
resolution. This becomes a problem when the audio is processed.
Therefore, acquire audio in the HDV realm, as loud as possible without
crossing the 0 dB brick wall. Level processing is a challenge in all
formats, but HDV audio is slightly less forgiving than PCM audio (the
DV standard). Therefore, you’ll want to pay closer attention to the
audio if you’re shooting HDV.

Shoot On Set with a Black-and-White Monitor

Use the uncompressed outs to feed a flat screen with the color pulled
out, or better yet, rent a production monitor. This will help when
focusing HD cameras. Even with large on-camera displays, or enlarged
viewfinders, anything out of focus in the HD world is immediately
noticeable, particularly when viewed on a reasonably large display.
Pulling focus properly is critical at higher HD resolutions.

Tim Kolb
Producer/director/co-founder
Kolb Syverson Communications
Appleton, WI
Recent HDV Projects: CopperBox "Finally Found You" music video

Before you shoot in simulated progressive or progressive scan HDV, you
need to consider your post pathway. Right now progressive scan support
for HDV source material does not exist in every NLE that can handle HDV.

For instance, if you’re going use the Sony HVR-Z1U camera, and you’re
going to use it in one of the CineFrame modes, then it’s important to
look at your post pipeline, and see if it’s going to support
progressive on the timeline or not.

The Sony camera is not actually recording progressive; it records
interlaced. There’s a flag on the data to specify a playback mode on
the footage. You shoot in CineFrame 30 mode, but many editing systems
will see it as 60i on ingest. These modes in the camera create a look.
It’s not changing how it writes the file. If you shoot at CineFrame 30
you most likely want a 30p look, therefore you’d probably like to keep
that during post.

We use CineForm’s Aspect HD running in Adobe Premiere to post. Aspect
HD can be set to capture HDV footage as 60 or 50 fps interlaced, but
can also be set to capture at 24, 25 and 30 fps progressive. You can
bring in the HDV based on how you shot it. One particularly nice
feature in Aspect is that you can bring in CineFrame 25 footage as 25
fps progressive and slow it to 24 fps and adjust for the audio pitch
change, bypassing the 60i-to-24p pulldown from CineFrame 24 mode if you
prefer.

Frederic Haubrich
Director, Lumià©re Media
Doylestown, PA
Recent HDV Projects: Tomorrow is Today

With Final Cut Pro 5, ingesting and editing HDV is basically plug and
play – you can cut HDV in real time. Here are the choices: from the
timeline when you decide to generate a final output you could go out
uncompressed HD or back to HDV. You can also generate full-resolution
H.264.

But going back to HDV involves some MPEG artifacting when you start
layering and doing effects and transitions. I’ve done a little bit of
color correction in HDV and it looks terrible. I don’t think they’re
insurmountable issues if you turn that timeline into a 4:2:2 color
space codec with Media Manager before your output. Then, when you
render and do color correction, you’re doing it off 4:2:2 where each
frame is now a frame, you’re not working with MPEG any more. Now you
can do color correction and get some pretty decent results. It’s still
4:2:0 to 4:2:2.

We experimented with Nattress’s latest FCP plug-ins for HDV and I must
say that the Film Gamma plug-in is a really easy and impressive
solution to making your footage even more film-like. From there we’ll
do a final render to QuickTime 24p or whatever we need.

Douglas Spotted Eagle
Managing producer
Sundance Media Group/VASST Stockton, UT
Recent HDV Projects: Training video for Detroit Diesel, Child Abuse Prevention PSA

Don't Go Native, Go DI
Regardless of the NLE being used, consider editing in an intermediary
format. In the Windows environment, consider using the CineForm DI
(Digital Intermediary) as it converts the 4:2:0 m2t stream (native HDV)
to a 4:2:2 short-GOP stream. Several software companies tout their
ability to edit native HDV, but this isn’t a good idea due to
color-shift issues of the GOP format. Converting the stream to an
intermediary eliminates these concerns. Further, the transport-stream
format at any bit rate loses quality after a single recompression,
therefore making the m2t stream less than ideal for compositing. At
Sundance Media Group/VASST, we convert immediately to 4:2:2 YUV for our
compositing and editing.

Convert to 24p in Vegas
If your final output is to film, shoot 50i with your HDV camera. 50i
converts to 24p for output much more nicely than does 60i. Merely
converting, and in some software applications, slowing the audio and
video by 4 percent, provides a wonderful 24p output with no hassles.
Sony Vegas 6 does an outstanding job of the 50i to 24p conversion
without the extra renders and slow-downs, so if you plan on working
with film-out, consider Vegas as an option. Otherwise, simply render
the 50i to 25p, and then slow the footage and audio.

If Your Computer's Slow, Use a Proxy
Be sure your computer is up to the task of editing HD. If it’s not up
to speed, consider proxy-based editing. Editors have done this for
years in various NLEs, and proxy-based editing allows slower NLEs to
edit HD. The way a proxy works is that the HD file resides on the
drive, waiting to be used at final render. The proxy stream, which
could be plain-vanilla DV in widescreen mode, or a more complex
low-resolution MJPEG file in a full 1440 x 1080 (anamorphic) format, is
used for editing, color correction, titling, etc. The proxy files are
swapped out for the real HD files at the time of final render. Be sure
your proxy format can properly manage the ITU-709 color space that HD
uses.

Michael Cioni
Post-production supervisor
PlasterCITY Digital Post
Los Angeles
Recent HDV Projects: The Biker Chef, All In

One of the most common questions from HDV filmmakers is how to conform
the HDV format from 29.97 to 23.98 frames per second for large format
23.98 Psf 1080 x 1920 exhibition. After extensive testing and examining
the results on a 1.3K DLP projector, I recommend three options for
conforming HDV to the world of uncompressed HD.

Option 1: Shoot in PAL
Though PAL acquisition quickly disappeared when 24p DV was made
available, HDV does not yet have a satisfactory solution for 24p
acquisition. For director Nick Vallelonga’s poker film All
In, HDV cameras were brought in to supplement three Sony
HDW-F900’s required for a large car crash sequence. The Sony HVR-Z1U
was chosen for its high resolution and captured the crash at
1080/25/50i. The footage was ingested natively into Final Cut Pro at
HDV PAL, 25/50i. Afterward, it was brought in to Adobe After Effects
where it went through a 4 percent slow-pass conversion by re-mapping
the time base from 25 to 23.98 progressive. This was then rendered out
uncompressed 10-bit and laid to HDCAM SR, creating a seamless
1080/23.98 Psf camera master for editorial.

Option 2: Use Reverse Telecine
On Christopher Coppola’s traveling cooking show, The Biker
Chef, HDV became a handy solution. Using the Panasonic
VariCam, Biker Chef is shot documentary-style on the
road, and critical insert shots are often missed in the field. Dealing
with budget constraints, the Sony Z1U in CineFrame mode was used for
small stage set-up cooking insert shots instead of the VariCam. Though
the CineFrame mode is not true progressive capturing, the Z1U does lay
3:2 pulldown to the HDV tape. Once ingested into Final Cut Pro, the
footage is taken to Cinema Tools and a reverse telecine is performed.
The footage is then re-laid back to HDCAM at 23.98 Psf for editorial
and later color corrected to match the VariCam.

Option 3: Use the Teranex Xantus
For optimum HDV image acquisition for a feature in development, the RGB
output of the Z1U was fed through an AJA HD10A RGB to HD-SDI converter.
The RGB feed, essentially an uncompressed HD stream, was recorded to
HDCAM SR at 4:4:4 10-bit. Afterwards, the SR tape was fed through the
Teranex Xantus frame-rate converter and de-interlaced and converted to
1080/23.98Psf. Working right off the Z1U chip allowed us to bypass the
heavy HDV compression and still achieve a 24p master with stunning
color, resolution and clarity.