Manual / Stories & sharing

Stories & sharing

A story is an ordered set of scenes — saved views with a headline and narrative — played full-bleed in Present mode and shared as a read-only link your audience can actually explore.

Authoring scenes

  1. Save the view.

    Frame the model and press Save scene, on the bottom rail or at the top of the Story tab (Ctrl+B). The rail shows each scene as a picture with its number and name; scenes that came without one (a sample, a drafted story, an older project) show Make previews on the rail, which visits each in turn to capture it and brings you back to your scene. Each tile can update to the current view or be removed (both with an Undo toast). To re-frame the selected scene, set up the view and press Update scene to this view (Ctrl+Shift+B).

    When the view stops matching the selected scene (you orbit, show or hide a layer, or move or turn off the section), its rail tile gets a small gold dot and the Story tab says what changed. Update scene keeps the change in the scene; Back to scene drops it and returns to the scene as saved. A scene that sweeps its section is not flagged for the cut it moves by itself.

    A scene also keeps its map labels as you left them: edited text, labels you dragged or removed, and label size. Each scene shows its own when the story reaches it, however the scenes before it labelled the same layer. A scene saved before you edited a layer’s labels shows the layer’s own labels.

  2. Or make a scene of a layer or a hole.

    Open a layer’s pencil panel in the Layers tab and press Make a scene of this layer: ES3D frames the layer and saves a scene named after it. After you jump to a hole with Find drillhole, Make a scene of DDH013 does the same for the hole. The scene’s name and first headline are the layer name or hole id, your own labels; write the headline you want from there.

  3. Check it as your audience will.

    Play, beside Save scene (Ctrl+Enter), plays the story in Present from the selected scene, skipping the cover. Switch back to Pro and you are on the scene you started from, ready to keep editing.

  4. Write the beat.

    With the Story tab open, the selected scene’s card sits on the canvas in the shape Present gives it, marked As investors see it. Type the headline and narrative straight into it: the lines break where they will for your audience. Its × hides it for the session.

    In the Story tab, give each scene a name, Headline and Narrative, set Duration (0 = fly straight through; or drag the right edge of the scene's tile on the rail, or focus it and press ←/→, a second at a time) and Transition, choose Motion (Static / Orbit / Push in / Push + tilt, a slow dolly while the view leans from overhead toward the horizon) and optionally a Chapter (e.g. Resource · Metallurgy · ESG).

    A scene’s motion follows the clock, not the frame rate, so it keeps the same pace and stays smooth on a slow phone or a busy tablet. While it moves, the scene keeps drawing at full resolution, so drill intervals and traces stay sharp through the narration. A scene’s pan (the slow glide of the view onto a new centre) takes six seconds unless the scene sets its own pan duration, up to a minute, so a section can drift sideways for its whole narration instead of zooming. A flight between scenes draws at one steady resolution from take-off to landing, so the move keeps an even pace. On a screen smaller than a laptop’s, map labels, text blocks, pictures and their leaders are drawn proportionally smaller, in step with the map, so labels you placed to clear each other on your desktop still clear each other on an iPad. Phones and iPads also draw terrain and imagery at a lighter resolution than a desktop, sized to their screens, so a large deck fits in a mobile browser’s memory. Your saved project is unchanged.

  5. Move the section with the story.

    A scene saved with Section on shows a Section block in the Story tab. It says how the cut arrives: it glides in with the camera when the previous scene cuts in the same direction, and jumps when the previous scene has no section, cuts another way, or the transition is a cut. To make consecutive sections glide, aim them the same way (both Along, say). The cut glides over the scene’s whole transition unless the project file gives the scene a shorter slice time, in which case it lands first and the camera finishes after it.

    Tick Sweep the cut while this scene holds and, once the camera lands, the cut slides from where the scene saved it to an end place over the scene’s Duration (six seconds for a 0 s scene), then stays there. The first end is the far side of the scene; to choose your own, move the live cut with Position and press Set end here. Preview flies to the scene again and plays it. Moving the cut yourself, or pressing Stop, ends a sweep where it is. With reduced motion turned on in the operating system, the cut goes straight to the end place. Exported movies play glides and sweeps exactly as Present does.

  6. Mix in other beat types.

    The rail's + Add menu inserts a full-screen text Slide, a 360° scene (an on-site panorama with hotspots and a set landing view), a closing Call to action (booking link + lead capture), or Draft story… — an auto-drafted narrated story from the project.

  7. Narrate it (optional).

    Each scene can carry narration — a hosted audio URL, or a clip you record or upload, stored with the project. A recorded or uploaded clip plays only on your machine until you press Publish for sharing, which uploads it to public storage so everyone with the link hears it. Auto-play waits for each scene’s narration to finish.

  8. Add the descent.

    The Descent panel captures named zoom levels (Regional → Project → Discovery) and plays the map fly-in that opens big.

  9. Reorder freely.

    Drag scenes, use ▲/▼ or Alt+↑/↓, or open the slide-sorter overlay. Per-scene geo animation (build-up / peel / explode / chronology reveal) lives here too.

Terrain imagery is built when a scene reaches it. A terrain carrying an orthophoto, a geological map or other draped imagery has its picture built only while that terrain is on screen, or while the current scene shows it, and the picture is let go once the story has moved past it. So a project can carry sharp, large imagery for just a few scenes without every one of those pictures being held from the moment the story opens — what has to fit is the scenes on either side of a transition, not the whole file. For the smoothest reveal, turn the terrain on at about 1% opacity in the scene before it appears: it is then built during that scene and the next transition simply fades it up. The held terrain is built once the camera has landed on that scene, never during the flight into it, so the move itself stays smooth. A terrain that first appears without that hold is built as its scene begins: on a very large image the transition can pause briefly, and the terrain can begin to fade in plain for a moment before its picture arrives. Terrains the project opens on are ready from the start, and turning a terrain on yourself builds its imagery within a moment. Exports and saved projects are unaffected: a movie or deck export waits for a scene's imagery before recording it, exports still render imagery at full resolution, and every source image stays in the .es3d.

Present mode

The Present-mode cover card with the story headline and Play button.
Fig. 1The cover — headline, brand, Play.
Present mode mid-story with the narrative card and the scene rail.
Fig. 2Mid-story — narrative card, scene rail, disclosures.

Share links

Share (top bar) mints a read-only link — “Recipients can view & explore, but can't edit your project.”

Above the link, Before you send lists what is still worth fixing: a scene with no headline or no narrative, two scenes in a row that show the same view, a scene that shows a map layer you have since removed, a sweep that no longer matches its section, and scenes with no picture. Go takes you to the scene in the Story tab; Make previews captures the missing pictures there and then. When there is nothing to fix it says so.

Options:

Where your audience lands Every link you send — share links, PIN links, live links, published addresses, embeds and the view report — opens on exploration.site, the address set aside for decks. You keep building on app.exploration.rocks. Links made before October 2026 on exploration.rocks keep working exactly as they did.

Publish a fixed address

A share link carries the whole story inside the URL, so changing one beat gives you a different link — and the <iframe> on your website has to be found and re-pasted. Publish (in the Share dialog) fixes that: your story gets one fixed address, exploration.site/e/your-project, and pressing Update re-points it at whatever is on your screen. Nothing on the website changes, ever. The address serves your story while your free trial or plan is active; if the account is paused it shows a “paused” card instead, and your story returns the moment you choose a plan (see Your free trial and plan).

Who can see a published deck Anyone with the address, without signing in — that is what makes an embed work. Addresses are meant to be typed, so treat one as public. Nothing in ES3D lists published decks and our database offers no way to ask for the list, but the address is an ordinary web page: a search engine that finds it linked from a public site can index it, and an indexed address outlives unpublishing. We ask crawlers to skip these addresses in robots.txt and the major engines honour that, which is tidiness rather than a lock. Two more things worth knowing: a published deck carries your project’s internal ID inside it, and one published with project data also carries your account’s internal ID in the download link — opaque IDs, not names or email addresses, and no key to anything; and a story-only share link keeps its narrative in the URL and never reaches a server, while publishing stores that text with us. See the Privacy Policy.

Seeing who watched

Every published deck counts its views — at its own address and in every embed of it, on your website or anyone else’s. The count is anonymous: it records that the tour was opened, which scenes were shown and for how long, whether a scene’s narration played to the end, and whether the viewer moved the section cut. It never records a name, an email address, an IP address or a device, and a viewer whose browser asks not to be tracked is not counted at all.

  1. Open the report.

    In Present, press Share link. Under the published address, the Who watched box has View analytics — it opens the report in a new tab. The report is a web page at exploration.site/r/…; bookmark it and it stays up to date.

  2. Read it top to bottom.

    Four numbers first: visits, total time watched, time per visit, scenes viewed. Then two or three plain sentences — which scene held attention longest, and how often the last scene was viewed. Then visits per day, and a table with one row per scene: how many times it was viewed (the bar compares it with the most-viewed scene), the average time on it, how often its narration was heard to the end, and how often a viewer moved the cut.

  3. Send it to your client.

    Press Copy report link and paste it into an email. Anyone with that link can read the numbers — and only the numbers; it never opens your project. No sign-in is needed to read it.

  4. Take a link back.

    Press Reset link, then confirm. A new report link is issued and every copy of the old one stops working at once. The counts themselves are kept.

What the numbers are — and are not A visit is one opening of the tour, so a person who comes back tomorrow counts twice. A scene counts as viewed when it has been on screen for at least a second — including the last one, when the viewer closes the tab — and a viewer who steps back to a scene counts again. Counting starts from the first visit after 27 September 2026; earlier views were not recorded. Viewers are told on screen that views are counted. Share links (not published) are counted too, but they have no report page — publish the deck to get one.

On phones and tablets

A phone is for watching, not authoring. On a phone, exploration.rocks opens as a viewer: the welcome screen offers the 30-second story, and there is no Pro tab, no sign-in and none of the author’s controls (exports, sharing, Go Live, layers). Build and share your story on a computer; open the link on a phone to see what your investors will see.

Story mode keeps its chrome small. On every screen the scene card sits tight in the top-left corner and is only as large as its text, and the scene rail along the bottom is shorter, so more of the screen is the model. On a phone (and in a small embed or a landscape phone) the colour keys always keep their place: the Mineral Resource Estimate table folds to a single line you tap to open, and the estimate’s method disclosure is in the Disclosures panel — every word is one tap away.

A phone held sideways gives the model the screen. In landscape (and in any short embed) the app header steps aside, the compass card is withheld, and the colour keys rest low, beside the viewing-terms link, instead of climbing under the scene card.

Wide establishing shots open at their authored scale on phones. A phone builds a deck’s map layers scene by scene, and the camera’s zoom-out limit used to be measured before the regional layers existed — so a country- or belt-scale opening beat was held zoomed in on a phone. The limit now uses each layer’s saved extent, so those beats frame as they were saved. Nothing needs re-publishing.

A portrait phone now sees your whole framing. A scene is authored on a landscape canvas; a portrait screen is a third of that width, and it used to keep the framing's height and crop the sides away — callouts and the edges of the deposit simply fell off screen. The camera now pulls back on narrow screens until the full authored width fits with a little breathing room around it, the way a landscape photograph letterboxes on a portrait phone — the frame floats inside the screen rather than pressing against its edges. Nothing is stored or re-authored: the same scene replays exactly as framed on any desktop, and a viewer can still pinch in on a phone for detail. Exports are never letterboxed — they render at their own landscape sizes.

A deck can carry its own phone layout. A portrait phone shows a scene at about a third of a desktop’s width, so a map that names every project in the district has no room for all of them. Each label, picture or callout in a deck can say what a phone-width screen should do with it: leave it out, put it somewhere else, draw it larger (a table or a closing card the viewer must still be able to read), or show a compact version that only a phone sees. A deck without a phone layout plays exactly as it always has. A phone does not show the locator card at all.

Your audience sees the story, not the app. Anyone opening a share link, a published deck or an embed gets no top bar: the story uses the whole height of the screen, and the small Exploration Sites logo sits in the bottom-right corner of the compass and scale-bar box (on a phone, in the bottom-right corner of the frame). The viewing terms and disclosures appear on the cover and the first scene, and fade from the second scene on; going back to the first scene shows them again. When you present your own story you still see the top bar and every control as before.

A scene can carry its own phone framing. Fitting a desktop scene’s whole width onto a portrait phone suits a map, but a cross-section is a long, shallow strip, and fitted whole its intercepts sit small in the middle of a tall, empty screen. A scene in a deck can say how much closer a phone should come in and where the subject should sit on the screen, for example above the caption strip. This applies only on phone-width screens during a story; the desktop view and the saved scene are unchanged, and a viewer can still pinch out to see the whole section.

The story yields the screen to the model on a phone. The scene card collapses to a single line — scene number and title — no wider than the title needs. Tap the strip to read the whole narrative (it scrolls if the beat runs long), tap again to put it away; moving to another scene puts it away for you. The scene rail swaps its thumbnails for numbered chips at phone widths, and the on-screen gesture hints describe touch (drag, pinch, double-tap) rather than a mouse. The gesture hint waits while a story plays, so it never sits over the opening scene's labels; it appears once, the first time the story is stopped or ends. A full-screen button appears in the top-right corner on phones and tablets whose browser supports it — it hides the app header too, giving the model the entire display; press it again (or the device's back gesture) to return. iPhones are the exception: Safari there offers no full-screen control to web pages, so the button does not appear — but an iPhone viewing an embedded story gets an open-the-full-viewer link in the same corner instead, which opens the story in its own browser tab, free of the hosting page’s frame.

A shared story opens on a phone or a tablet, and the viewer adapts what it holds in memory to suit the device. The one visible difference is map imagery: satellite and drone orthos, geological maps and rasterised claim boundaries draped over the terrain are composited at a lower resolution on a handheld than on a desktop. At the framings a story actually uses — the whole property, a pit, a corridor of holes — that is at or beyond what the screen can resolve, so it looks the same. Pinch right in and the imagery is softer than the same view on a laptop, and a rasterised boundary line can read up to twice as wide. A very large image — over about 16 megapixels, such as an 8192-pixel orthophoto tile — is read once while the story opens and kept as a screen-sized copy, so the scene that shows it later never has to read the full image in the middle of the story, which is when a phone is most likely to run out of memory. On a deck with several such images that can add a second or two to opening it.

Nothing else changes on screen. Elevations, drill traces, assays, imported solids and geological solids are shown at full detail, the colours and legends are identical, and your imagery is never rewritten — the smaller composite is built for the screen and never written back, so every source pixel stays in the .es3d. Reopen the same story on a desktop and the full-resolution imagery is there.

An export lifts the reduction. A deck slide, PNG still or movie is rendered from the largest imagery the exporting device can build, not from the smaller composite the screen was using. On a desktop that is the source resolution. A phone or tablet has a hard ceiling on how large an image its browser can allocate, so an export taken on a handheld may still be softer than the same export from a desktop — export a client-facing deck from a desktop. See Exporting.

One thing is stored: the still image saved with each scene, captured from what that device is showing. A scene saved or updated on a handheld may keep a softer still, and a deck slide uses that still whenever the beat cannot be re-rendered at export time. Updating the scene from a desktop replaces it.

On a phone, a deck loads as it plays. Phones have a fraction of a laptop's memory, so a story opened in an embed on a handheld builds the map layers each scene needs rather than every layer at once. Layers arrive as you move through the story, usually a scene ahead of where you are, so playing straight through shows nothing waiting. Jump several scenes at once and a layer can arrive a moment after the camera does; while one is still on its way you will see Loading map layers in the corner, and its row in the 3D layers list is marked Loading. Nothing is left out — a layer you turn on yourself is built on the spot, and saving a project always writes every layer whatever is on screen at the time. On a laptop, on a desktop, and in the authoring workspace every layer is built up front exactly as before.

On a phone, the size check now weighs your heaviest scenes — not the whole story. A phone builds each scene's map layers as the story reaches them and releases the ones the story has moved past, so what has to fit is the heaviest short stretch of consecutive scenes, not the file. A story whose total is several times what a phone could ever hold opens and plays there, provided no stretch of it is too heavy — and the check is honest both ways: a link that promises light scenes is re-measured on the device after downloading, before anything is built. Two limits remain, both deliberate. The download itself still has a ceiling (a phone has to hold what it fetched), and a story with even one scene-stretch too heavy for the device declines with a message that says exactly that — loading it anyway would kill the browser tab partway through and leave your audience on a blank page. Spreading heavy layers across scenes, rather than stacking them into one, is now a real authoring lever. Laptops with very little memory keep the whole-story check: per-scene loading runs only on phones and tablets. As always, the new check rides the link — mint a fresh share link (and re-publish a published deck) to pick it up.

Filled map polygons no longer count against a phone. The solid fill of a polygon layer — a country silhouette, a claims block, a tenement package — is rebuilt from the layer's own outlines when a scene first shows it, so a saved project or a share link no longer carries those triangles at all unless the layer is visible the moment the project opens. On map-heavy decks this is most of the weight: fills the size check used to count are often half the story. The saving applies from the next save or share. A link minted earlier keeps the size it was minted with, so if a story is refused on handhelds, open the project, mint a fresh share link and — for a published deck — publish again to the same address; the deck itself does not change. Mint from the project as it opens: recalling a map-heavy scene first turns those layers on, and what is on screen when you mint is what travels. If a story is still refused after that, the remaining route is a lighter deck: fewer or coarser surfaces and geometry layers.

Go Live

● Go Live drives the story to every connected viewer in real time — they follow your scene changes and playback, see a Following presenter notice, and can leave at any time. You get a LIVE pill with relay state and a follow link to send. The share dialog says it plainly: anyone with a live link can also steer which scene connected viewers see.

What your audience sees A shared story opens straight into Present with a quiet “Made with ES3D” mark. An expired link says so; a PIN-protected one asks for the PIN; a link whose data is still downloading shows progress. Anonymous view analytics (which scenes, time on each) go to the story owner, and the strip discloses exactly that to viewers. A published /e/ address and every embed of it are counted the same way — see Seeing who watched.
ES3D Manual · updated 25 August 2026 Open ES3D What’s new Terms