We test the technology on ourselves — and publish what breaks.
Lab is where the studio spends its own money. No client, no brief, no deadline but ours. Every experiment is published with what it proved, what it failed to prove, and the numbers behind both. A studio that shows only its wins is a studio you cannot calibrate.
EXP 01 · VIRTUAL PRODUCTION · 2025
VAQT — a volume test
What we tested
Whether a small team could run a full virtual production cycle — an LED volume, real-time in-camera VFX — and put four different classes of shot into one short film: a digital creature with a chase, an abstract void, a historical exterior and a dialogue interior. The film was made to be shown to the Uzbek film community as a demonstration of what the technology can actually do here.
What it proved
The pipeline holds. Four classes of shot, one volume, one small crew, from drawn boards through previz layout to final master. The backstage is cut into the end of the film itself — the object demonstrates its own method. And the layout survives: you can watch the previz and the finished film side by side, which is the thing every client asks for and almost nobody can show.
What it did not prove
That it is a good short film. Our own breakdown scores it 8/10 as a technology demo and 6/10 as cinema. The story leans on the oldest trick there is — it was all a dream — there is no antagonist, and 47% of the running time is a lecture in a classroom while the sage who carries the whole theme gets seventeen seconds. We publish that verdict because a lab that reports only successes is advertising, not a lab.
What happens next
The findings are already in development. The dream stops being a dream: the artifact stays in the hero's hands after he wakes. The sage becomes a person instead of a function. A sequel is in preparation.
On 14 February 2025 the technology was presented at the O'zbekfilm pavilion. The film was screened — and then the «Bobur» scene was performed and shot in real time in front of the room: an actor on the volume, the sunset built live behind him. A film can be finished in post; a scene played live in front of a hall of professionals cannot. Scene directed by Fuadbey Kopadze, stage and equipment by O'zbekfilm, graphics and budget by United Soft.
EXP 02 · CONTENT PIPELINE · 2026
Süret — a pipeline test
What we tested
Not a film this time — a feed. Agencies rarely need one hero film; they need a channel that publishes every week for months without the cost rising. So we built one and ran it ourselves: Süret, a Turkish-language media about art history — vertical films of paintings coming to life, and photo essays about the painters.
What it proved
The economics hold. A photo essay costs nothing but editorial time; a film runs on tens of credits. Months of continuous output on a budget a single classical shoot would burn in an afternoon. It also proved the editorial part is the hard part: every fact is checked against museum sources, each issue records which anecdotes were dropped and why, and the Ottoman material is handled carefully — Aivazovsky cannot be sold as the sultan's court painter just because it makes a better hook.
What it did not prove
The audience. The channel is young and the numbers are small; the reach we get is the standard test push an algorithm gives anything new. That tells us the problem is not distribution but retention — the first three seconds — and that is what the next round of the experiment works on. We are publishing this while it is still unresolved, because that is the point of a lab.
What happens next
Rebuild the openings, cut the setup, test the hook against the same subject in three different forms. The pipeline is proven; now it has to earn attention.
Every result lands here — including the ones that do not flatter us. If you want the same layout-before-production discipline on your film, that is what a previz is.
EXP 01 · VIRTUAL PRODUCTION · 2025
