Multimedia
Newsletters
Blogs
Whitepapers
Events
DCP Directory
Mill Directory
Edit Calendar
Advertisers
Reader SurveyHDV@Work
Jan 10, 2005 11:44 AM
An Introduction to 1080i Postproduction
by Steve Mullen
To date I’ve been focusing on the Sony FX1/Z1 camcorder and 1080i HDV technology. However, the real impact of 1080i will be the necessary changes in postproduction. The impact comes about from the consequences of the dramatically greater amount of data incorporated within each frame of 1080 HDV.
That impact may not be obvious given that 1080i is recorded at 25Mbps—the same data rate used for DV25. If your computer can capture DV, it will capture 1080i HDV. Nor will you need bigger or faster disks. The 1080i format will have the same storage requirements as DV.
And as long as a data stream remains MPEG-2—either transport or program stream—you will not encounter editing issues. Of course, eventually MPEG-2 encoded video will need to be decoded into either RGB or YUV digital data. This is where your postproduction procedures become complicated and, potentially, much slower. There are three situations when MPEG-2 must be decoded, pending on the editing system you use.
If you employ a system such as CineForm software with Premiere or Vegas, where MPEG-2 data are converted to another format for editing, the decoding occurs as video is converted to the intermediate codec. This transcoding occurs while the MPEG-2 is being captured. If you have very powerful PC this can be accomplished during capture--you will not even notice the conversion. When this cannot be accomplished in realtime, you will have to wait for the captured MPEG-2 file to be converted to an intermediate format file.
Native HDV editing systems require no conversion because they allow you to import MPEG-2 and place these clips directly in the timeline. However, this only postpones decoding. In order to play MPEG-2, either in a source window or timeline, the MPEG-2 video stream must be decoded on the fly to a video stream so it can be displayed on your computer’s monitor. Moreover, if HD video is to be monitored on an HD display device, the MPEG-2 video stream must be decoded on the fly to a YUV video stream and output via HD-SDI or component analog.
When more than one video stream is required, during a transition, for example, multiple MPEG-2 streams must be decoded on the fly. During a dissolve, each of the two streams must be decoded at greater than 2X faster than realtime to allow surplus CPU cycles to compute the FX.
There is another factor that impacts realtime HDV editing. Engineers have developed ways to compute a realtime “preview” image of a special effect that is less than full quality. Techniques include skipping fields and/or discarding every other pixel within a line. Doing so can reduce the amount of compute power needed to render FX by up to a factor of four.
Significant time savings can also be obtained if techniques are employed that make it possible to not decompress every pixel in a DV frame. For example, with interlace video, DV typically employs field-based compression. During decompression, fields can be skipped, thereby speeding up decompression by a factor of two.
MPEG-2, however, utilizes frame-based encoding so this option is not available. In fact, the smallest skippable chunk is an MPEG-2 “slice” that is eight pixels high. Therefore, with current technology, every pixel in an MPEG-2 frame must be decoded during a preview—no matter the size or quality of the preview window.
There is one other point at which decoding may come to be required. Sony Vegas and Adobe Premiere Pro with CineForm plug-in, plus Ulead’s Media Suite Pro, all generate an entirely new MPEG-2 movie during export. With these NLEs, not only must every MPEG-2 frame be decoded, it must also be re-encoded into MPEG-2. While this need not happen in realtime, you do need to wait for your movie to be ready to record back to your HDV camcorder.
In the coming weeks we’ll look at the four editing technologies that have been developed to cope with the issues of editing 1080i HDV. We’ll also look at how 1080i editing differs from 720p30 editing.
Numbers Part 2, What’s ‘Effective’ Vertical Resolution?
by Steve Mullen
The Sony HDR-FX1 employs “interlace scan dual-line” CCDs. For every upper field, the CCD driving logic selects the top-most row (row #1). The first line scanned out is the sum of each element in CCD row 1 added to the corresponding element in row 2. (The summation is performed within the CCD.) The second line scanned out is the sum of each element in CCD row 3 added to the corresponding element in row 4. And the last line scanned out is the sum of each element in CCD row 1079 added to the corresponding element in row 1080. In this way, the FX1 CCDs output 540 lines (one field) of 960 elements that are digitized (analog-to-digital) within 1/60th of a second.
For every lower field, the CCD driving logic selects the second CCD row (row #2) as the first row. The first line scanned out in a lower field is the sum of each element in CCD row 2 added to the corresponding element in row 3. The second line scanned out is the sum of each element in CCD row 4 added to the corresponding element in row 5. And the last line scanned out is the sum of each element in CCD row 1080 added to the corresponding element in row 1081.
Two benefits arise from summing rows. First, CCDs are made 6dB more sensitive to available light. Second, noise is averaged and thereby canceled, increasing the S/N ratio.
When a CCD adds row-pairs, the image is essentially passed through a low-pass filter. (A filter that removes high frequencies representing fine detail.) This filter helps prevent “interline flicker.” Interline flicker is “aliasing” that occurs when horizontal image detail (signal frequency) approaches that of the CCD row-structure (sampling frequency).
In plain English, this means that when a sharp, horizontal edge (or line) is imaged, you may see a 30Hz flicker on it. The flicker is caused by the horizontal edge being represented by a single interlace scan-line that is displayed 30 times per second.
If a horizontal edge moves vertically, the flicker moves up or down the screen causing “interline twitter.” Both twitter and flicker are reduced, but not eliminated, when horizontal edges (or lines) are softened by the low-pass filter. Essentially an edge is smeared across two lines so that during display, the edge is displayed at 60Hz. Twitter and flicker are two of the most objectionable artifacts of interlace scanning.
Unfortunately, a price must be paid for the benefits of row-pair summing. The low-pass filter decreases an image’s effective vertical resolution by about 25 percent—from 540 lines to about 405 lines per field. Therefore, the FX1/Z1’s effective vertical resolution should be about 810 lines when operating in 1080i60 mode. And, indeed, the measured resolution of the FX1 is almost 800 lines.
Horizontal resolution has been measured at slightly more than 800 lines. You’ll note that both vertical and horizontal resolution are approximately the same. This is as it should be, considering the 1080i format (1920x1080) employs square pixels.
Pinnacle Gear to Feature Dolby Digital 5.1 Creator
Pinnacle Systems is now shipping its digital video editing products with Dolby Digital 5.1 Creator multichannel encoding technology. The technology is currently being included in Pinnacle Liquid Edition and is also available for Studio Plus, the consumer-based video editor with a second video track for A/B editing, chromakey effects and dual-layer DVD support.
With Creator, users can store more audio content on discs and an audio track recording takes up less space. Greg Rodehau, director of PC technology marketing for Dolby Laboratories, said video enthusiasts can now add dramatic sound effects to their video creations without compromising disc space.
Leitch Technology to Acquire Inscriber
Leitch Technology has announced that it has entered into a definitive agreement to acquire Inscriber Technology, based in Waterloo, Ontario.
The companies expect to close the transaction this month, subject to the satisfaction of customary closing conditions and regulatory approvals. The total purchase price is $18 million (Canadian), including the assumption of $1.5 million in debt. The purchase will be funded through Leitch's existing cash balances.
Inscriber extends Leitch's position in the market with a complementary product range focused on broadcast graphics, content branding and logo generation, enabling an Integrated Content Environment that will streamline workflow for broadcast and media companies, the customer base currently served by both companies.
"Our customers will benefit from an extensive range of products that are complementary to our existing Leitch offering," said Tim Thorsteinson, president and CEO of Leitch Technology. "This acquisition is one more step in our commitment to focus on broadcast television customers.”
"Inscriber has a complementary customer base to Leitch and adds technical prowess in the graphics software domain that will only become stronger with this acquisition," said Dan Mance, Inscriber president and founder.
Media 100 HD for Mac Announced
Media 100 has announced that Media 100 HD is now available for Mac OS X. The application brings 10-bit uncompressed HD editing and broadcast-quality format conversion to Mac OS X. Media 100 HD is also a full-featured, 10-bit, uncompressed SD editing system.
With a familiar Media 100 interface, up to 99 video tracks and new HDX technology, Media 100 HD is designed and tuned for the Power Mac G5, giving editors affordable, flexible support for HD and SD applications in a single 10-bit editing system.
Features include:
- Multiple video tracks (up to 99) with opacity controls in the timeline;
- Realtime audio effects with audio dynamics and reverb;
- Advanced keyer with RGB, YUV, and HSL controls— and realtime SD performance;
- Intuitive color correction and full motion alpha channel support with realtime SD performance;
- Broad support for HD and SD standards—with realtime format conversion;
- 10-bit uncompressed SDI I/O: 1080i, 720p, 525/60 (NTSC), 625/50 (PAL);
- 10-bit uncompressed component video I/O 525/60 (NTSC), 625/50 (PAL) ;
- All-On-One Mastering: capture and conform to and from multiple HD and SD resolutions;
- Realtime, broadcast-quality HD to HD, HD to SD, SD to HD format conversion;
- Realtime aspect ratio conversion (when format converting on I/O);
- Simultaneous HD and SD output for program out and monitoring;
- Professional 19in. rackmountable I/O break-out box with Genlock;
- Media 100 HD-native software codec;
- Media 100 i and 844/X compatibility;
- QuickTime support (including Animation, Motion JPEG B, FCP 10-bit, DV codecs);
- Faster-than-realtime rendering of many complex effects;
- Realtime SD editing and multi-stream effects;
- Accelerated HD editing and multi-stream effects;
- Hardware and host-based acceleration support for maximum performance.
