Expand the format label before counting.
A specification such as “1080, 59.94” leaves an important question unanswered: does the rate refer to fields or complete progressive frames? Write the raster, scanning method and exact picture rate together. Then record the timecode convention separately.
| Label | Picture / frame rate | Other temporal units |
|---|---|---|
| 1080i/59.94 | 30000/1001 interlaced frames/s | 60000/1001 fields/s |
| 1080p/59.94 | 60000/1001 progressive frames/s | One complete progressive picture per frame |
| 1080PsF/23.98 | 24000/1001 progressive pictures/s | Two transported segments per picture |
| 720p/50 | 50 progressive frames/s | No interlaced field pair |
| 2160p/59.94 | 60000/1001 progressive frames/s | UHD raster; transport specified separately |
These are common labels, not a universal notation grammar. Some specifications write an interlaced frame rate as “1080i/29.97”. Resolve the convention with the sender rather than inferring it from punctuation. In an interlaced signal, two successive fields form a frame and represent different temporal samples. Dividing an interlaced-frame count by 59.94 instead of 29.97 halves its calculated duration.
The familiar decimals are shorthand. Use the rational rates in calculations: 30000/1001 is approximately 29.97003, rather than exactly 29.97. Integer 30 and fractional 29.97 are different rates. The same distinction applies to 60 and 59.94.
Separate capture, scanning and transport.
ITU-R BT.709-6 distinguishes progressive capture, interlaced capture and progressive segmented-frame transport. PsF carries one progressively captured picture as two segments. Those segments do not represent two successive capture moments. They must be recombined in the correct relationship to recover the original progressive picture.
This distinction matters when inspecting motion. A PsF signal may pass through equipment using an interlace-compatible interface, yet applying motion-adaptive deinterlacing indiscriminately is not equivalent to correctly reassembling the segments. Conversely, genuine interlaced motion does not become progressive merely by changing a metadata flag.
The image system also does not specify the entire connection. SDI describes a signal interface with particular mappings; IP describes a different transport environment. A connector’s presence or a nominal link speed does not prove that a particular raster, rate, sampling structure and mapping are supported end to end. Write those requirements into the equipment and delivery plan.
UHD adds another reason to be precise. “4K” alone leaves raster and rate ambiguous. BT.2020 describes UHDTV image parameters; it does not supply every receiving facility’s file, transport or delivery requirement. For a practical handoff, name the actual pixel dimensions and the intended rate before discussing cable or file choices.
Distinguish the address from its carrier.
Timecode identifies positions on a defined counting system. Its carriage describes how that information reaches another device or file. SMPTE ST 12-1:2014 covers time and control code, including LTC and VITC. ST 12-2:2014 specifies ancillary timecode carriage.
| Path | How the information travels | Practical check |
|---|---|---|
| LTC | Serial electrical signal, often carried or recorded through an audio path | Correct input, level, rate and decoding |
| Legacy VITC | Code in the vertical interval of an analogue television signal | Expected lines and preservation through processing |
| ATC | Timecode packets in digital video ancillary space, including SDI applications | Packet presence, type and preservation across devices |
| File metadata | A file or track’s stored timecode representation | Actual metadata, interpretation and continuity in the receiving software |
Do not infer that all four are present because a monitor displays a label. A converter, capture card or transcode may preserve one path and omit another. Test the complete route using a known source, then inspect the recording made at its end.
Timecode also does not by itself establish picture phase or a common audio sample clock. Read timecode, genlock and word clock for that separate synchronization problem.
Use each standard for its stated layer.
This limited map cites specific published editions. It is a reading route, not a claim that these are the newest editions or that one table covers a complete broadcast facility. Check the edition required by the contract and the standards publisher before implementation.
| Document | Scope relevant here |
|---|---|
| SMPTE ST 12-1:2014 | Time addresses, counting conventions, LTC and VITC |
| SMPTE ST 12-2:2014 | Transmission of timecode in ancillary data space |
| SMPTE ST 12-3:2016 | High-frame-rate timecode and ancillary formatting |
| SMPTE ST 296:2012 | 1280 × 720 progressive image family and its representations |
| ITU-R BT.709-6 (06/2015) | HDTV image parameters, including capture / transport distinctions |
| ITU-R BT.2020-2 (10/2015) | UHDTV image parameters for production and programme exchange |
High-rate displays need particular care. ST 12-1 includes frame-pair methods for rates above 30 fps, while ST 12-3 addresses further high-frame-rate cases. A software display counting every 59.94 fps picture with a full-rate frame number does not describe how that address is packed into LTC or ancillary data. Do not impose a universal “FF must be below 30” rule, or assume that a two-digit display documents the entire wire representation.
Check elapsed time with whole-frame examples.
Take a continuous 1080i/59.94 sequence containing 30,000 complete interlaced frames. It contains 60,000 fields. A continuous 1080p/59.94 sequence needs 60,000 progressive frames for the same elapsed duration:
Progressive: 60,000 ÷ (60,000/1001) = 1001 seconds
1001 seconds = 16 minutes 41 seconds
These examples use whole counted units. Ten physical seconds at 30000/1001 fps spans approximately 299.7003 frame periods; it is not a valid instruction to deliver a fractional image file.
Now count 1800 frame advances from 00:00:00:00 at 30000/1001 fps. The elapsed interval is exactly 1800 × 1001 / 30000 = 60.06 seconds. NDF labels the resulting boundary 00:01:00:00; DF labels it 00:01:00;02. The same pictures and physical duration exist in both cases. DF skips designated labels, not recorded pictures.
Neither label says whether the pictures are 1080i, 1080p or another image format. Keep that information in the specification. See drop-frame timecode for counting rules and timecode duration for endpoint conventions.
Calculate the interval, then check the delivery.
Use the free timecode calculator for label arithmetic and the duration calculator for elapsed-time checks. Select the actual frame rate and DF/NDF mode. The tools’ full-rate 59.94 counting is an arithmetic convention; it does not validate high-rate signal packing.
Timecode Pro helps with production timing and frame-count calculations. It does not inspect SDI packets, decode an incoming LTC signal, measure genlock, convert interlace / PsF media or certify standards compliance. A broadcaster’s delivery specification, equipment documentation and measured acceptance checks remain necessary. Confirm file format, image parameters, audio, captions and required metadata with the recipient before delivery.
Technical references
The linked editions in the standards table are the primary references: SMPTE’s published documents and ITU’s official recommendation records. BT.709-6 Attachment 2 explains segmented-frame transport; ST 12-1 sections 5 and 12 distinguish counting from higher-rate representation; ST 12-2 covers ancillary carriage. The numerical examples are original calculations using exact rational rates.