Original film by LeoGuoLabs / Codex Film Studio. Complete final v4 export, 1920 × 1080. The original delivered picture and audio are retained. This is the historical data presentation supplied by the production, including its estimates and methodology.
# Credits and asset provenance - Visual benchmark: [kraton0903's X video](https://x.com/kraton0903/status/2114257652930760928), locally retained for analysis. The delivered film uses newly authored graphics, text, music and sound, and contains no frames or audio copied from the reference. - Code-directed film principles studied from [Lemo-Opuscar](https://github.com/lemomo-ai/lemo-opuscar). No upstream rendering code or composition was used. - Data: Maddison Project Database 2023, Our World in Data processing, World Bank WDI and IMF WEO; definitions, attribution and exact linking policy are in [DATA-SOURCES.md](docs/DATA-SOURCES.md). - Narration: Kyutai Pocket TTS, `alba`, generated through the user's internal nr-test service. The production sent narration text to that internal deployment. No reference-person voice was cloned. [Pocket TTS](https://github.com/kyutai-labs/pocket-tts) is MIT-licensed. - Music and Foley: original procedural composition written for this film. No sampled commercial music or extracted reference soundtrack. - Fonts: Arial, Courier New and Impact from the local macOS installation, used for rendered output. Font files are not redistributed in this project. - Rendering: an independently written Python/Skia renderer; PyAV encoding with explicit BT.709 conversion; FFmpeg audio mastering. Library licences remain those of their respective projects. No publication, release upload or social post was performed.
# Data and historical sources The film compares eight selected major economies, not every country. It measures total real GDP in **trillion international dollars at 2011 prices (PPP)**. It is neither nominal GDP at market exchange rates nor GDP per person. Country trajectories follow the historical source's territorial conventions, including backcasting; the film does not treat a country label as proof of unchanged political borders. ## Historical levels, 1950–2022 [Maddison Project Database 2023](https://www.rug.nl/ggdc/historicaldevelopment/maddison/releases/maddison-project-database-2023?lang=en), by Jutta Bolt and Jan Luiten van Zanden. Citation: Bolt, Jutta and Jan Luiten van Zanden (2024), “Maddison style estimates of the evolution of the world economy: A new 2023 update,” *Journal of Economic Surveys*, DOI [10.1111/joes.12618](https://doi.org/10.1111/joes.12618). CC BY 4.0. The production retains the [original workbook](https://dataverse.nl/api/access/datafile/421302) as `data/mpd2023_web.xlsx` and its original-paper source sheet as `docs/MPD-ALL-ORIGINAL-SOURCES.csv`. These source-specific citations accompany the graphical use of a subset of the database. The plotted total GDP series was downloaded from [Our World in Data](https://ourworldindata.org/grapher/gdp-maddison-project-database), with major processing by OWID: GDP per capita multiplied by population. `data/maddison.metadata.json` retains the definition and processing notes. The historical PPP construction combines 2011 and 1990 PPPs. It should not be interpreted as a new official annual PPP benchmark for each year. ## Linked recent extension, 2023–2025 [World Bank WDI, GDP growth (annual %), NY.GDP.MKTP.KD.ZG](https://data.worldbank.org/indicator/NY.GDP.MKTP.KD.ZG). API snapshot: 2026-10-25; source last updated 2026-10-08. Raw responses are retained in `data/worldbank-growth.json`. For each country, the calculation is `level[y] = level[y−1] × (1 + real_growth[y] / 100)`, starting from its MPD 2022 level. These are **author-calculated linked levels**, not newly published MPD levels. Calendar/fiscal reporting conventions and revisions in the underlying national accounts may differ. The extension preserves the chosen historical level basis; it is not a mix of nominal dollars and PPP values. ## Current-year projection, 2026 [IMF World Economic Outlook, October 2026](https://data.imf.org/datasets/WEO), real GDP growth indicator [NGDP_RPCH](https://www.imf.org/external/datamapper/NGDP_RPCH@WEO). The production retains the fetched response as `data/imf-growth.json`. The 2026 linked level applies the IMF projected growth rate to the linked 2025 level. It is explicitly labelled a projection and drawn with a dashed final segment. It is not a reported 2026 annual outturn, not the IMF's official GDP level, and not a prediction of nominal ranking. ## Historical annotations - 1973 oil shock: [Federal Reserve History, Oil Shock of 1973–74](https://www.federalreservehistory.org/essays/oil-shock-of-1973-74). - 1978 reform and opening: [World Bank, China](https://www.worldbank.org/ext/en/country/china). - 1991 Indian reforms: [Reserve Bank of India, Opening up of India's External Sector](https://www.rbi.org.in/scripts/bs_viewcontent.aspx?Id=2253). - 2001 Chinese WTO accession: [World Trade Organization](https://www.wto.org/english/thewto_e/acc_e/a1_chine_e.htm); membership began 11 December 2001. - 2008–09 financial crisis: [Federal Reserve History, The Great Recession and Its Aftermath](https://www.federalreservehistory.org/essays/great-recession-and-its-aftermath). - 2020 pandemic shock: [World Bank, COVID-19 and the Global Economy](https://www.worldbank.org/en/news/press-release/2020/06/08/covid-19-to-plunge-global-economy-into-worst-recession-since-world-war-ii). The film uses this as event context, not as a source for a contemporaneous forecast number; the plotted outcome comes from MPD. These annotations are context. A line appearing near an event is not a claim that the event alone explains its movement. Interpolation between annual values is purely a display technique.