A player that outputs 4K cannot emit a 5760 x 1080 canvas, because HDMI carries fixed picture formats and none is that wide. Compose the three-screen layout inside one 3840 x 2160 timeline with each panel's picture cropped into its own quadrant, and every 1080p panel shows native pixels instead of the 1280 x 720 a scaled strip delivers.
A player that happily outputs 3840 x 2160 cannot put a 5760 x 1080 picture on the same cable. HDMI carries one picture format at a time from a fixed list, and no entry on that list is 5760 pixels wide. The fix is not a stronger box. Build the three-screen layout inside a 3840 x 2160 canvas, crop each panel's picture into its own quadrant, and let the wall controller slice the 4K output the way it already does. Every panel then shows a native 1920 x 1080 region instead of the 1280 x 720 it gets when the wide strip is scaled to fit.
Why can't a signage player output 5760x1080 across three 1080p screens?
An HDMI output shows one picture format at a time from a fixed list, and through HDMI 2.0 the widest format is 4096 pixels, so a 5760 x 1080 canvas never exists as a signal the cable can carry. To the output side of the player, the wall is one display, and that display has to speak one of the sizes the standard defines: 1920 x 1080, 3840 x 2160 and 4096 x 2160 are entries on the list, and 5760 x 1080 is not. No Android box emits that timing.
A 2x2 wall controller makes the problem concrete. It expects one standard input, usually 3840 x 2160, cuts it into four 1920 x 1080 quadrants, and sends one quadrant down each output. Your three screens want 3 x 1920 = 5760 pixels in a row, and that total exists only inside your content file. The output side never sees it, and no setting changes that. The wiring and splitter side of this setup is covered separately in one Android signage player driving multiple displays.
Why does a box that plays 4K video still refuse a 5760-pixel-wide file?
Because the ceiling has two halves: the output format tops out at 3840 x 2160 on most boxes, and the file format itself has to be one the box is designed to play, so a 5760-pixel-wide master fails at both ends. The pixel budget is not what stops it. The wide strip is 5760 x 1080 = 6,220,800 pixels, while the 4K canvas is 3840 x 2160 = 8,294,400 pixels, a third more, so the canvas has room. The formats are the ceiling.
On the file side, a master wider than the 4K class a single-output box is designed to play has no supported profile to fall into. The file is refused, or it plays as a scaled copy, and that scaled copy is the 720p-per-panel picture on the wall now. The same ceiling shows up in other ways, for example when an Android player renders 1080p and upscales it to 4K. Treat 3840 x 2160 as the largest master a single-output box will take, and design the wall inside it.
What happens to picture quality when the wide strip is scaled to fit?
You lose exactly a third of the resolution: fitting 5760 pixels of content into a 3840-pixel output scales by 0.667, so each 1920 x 1080 panel receives a 1280 x 720 picture, which is the image on the wall now. The arithmetic: 3840 / 5760 = 0.667, a 1920-pixel-wide panel picture becomes 1920 x 0.667 = 1280 pixels, and a 1080-pixel-high one becomes 720. The panels are still 1080p hardware, but a third of the content's resolution is thrown away before the first panel.
For a video loop watched from across the room, the loss is usually invisible. For a price list, a menu, or any small text read up close, it shows: edges soften and fine type breaks up. The quadrant layout below recovers the full 1920 x 1080 per panel, so the choice is about content, not hardware.
Worked example: how do three 1080p screens fit inside one 3840x2160 frame?
By placing each screen's content in its own 1920 x 1080 quadrant of the 4K canvas instead of scaling the whole strip: a 3840 x 2160 canvas is exactly two 1920-pixel columns by two 1080-pixel rows, so three panels fit at native size and one quadrant stays black. The division is exact: 3840 / 2 = 1920 and 2160 / 2 = 1080.
| Canvas region | Position (x, y) | Physical screen |
|---|---|---|
| Top left | 0, 0 to 1920, 1080 | Screen 1, left |
| Top right | 1920, 0 to 3840, 1080 | Screen 2, center |
| Bottom left | 0, 1080 to 1920, 2160 | Screen 3, right |
| Bottom right | 1920, 1080 to 3840, 2160 | Unused, black |
Build it in the editor: crop three 1920 x 1080 regions from the master layout and place region 1 top left, region 2 top right, region 3 bottom left, then export once at 3840 x 2160. The controller's outputs 1, 2 and 3 carry one panel picture each. The screens line up left to right in the order the cables say, so label them by pushing a full-white frame to one quadrant at a time and noting which screen lights.
The arithmetic is the whole trick. Across the wall, screen 1 shows pixels 0 to 1920 of the layout at one-to-one, screen 2 shows 1920 to 3840, screen 3 shows 3840 to 5760. Nothing scales. The scaled path draws 5760 pixels of content into 3840 output pixels and loses 1920 of every 5760, a 33% loss; the quadrant path loses none.
Two rules keep it working on a cheap fixed slicer. First, every panel's picture must stay inside its own quadrant: the slicer cuts at x = 1920 and y = 1080, so a wide strip laid flat across the canvas is sliced mid-picture. Second, for a wall of three portrait panels, rotate each asset 90 degrees into its quadrant before export. A portrait column is 1080 pixels wide, so three columns laid flat span 3240 pixels and the cut at x = 1920 lands inside the second column.
What are the options when every panel must run at native resolution?
You have four workable paths, and rebuilding the content for the 4K canvas is the only one that needs no new hardware: keep the existing box and controller, and the other three paths each cost hardware or accept a compromise.
| Path | What you add | Picture per panel | Watch out for |
|---|---|---|---|
| Pre-split 4K canvas | Nothing; one 3840 x 2160 export | Native 1920 x 1080 | Content must stay inside quadrants |
| Player with several 4K outputs | A player built with one output per screen | Native, one canvas across outputs | New hardware and setup |
| One player per screen | Three offline players, three cropped files | Native per panel | Loops drift on long video |
| Accept the scaled strip | Nothing | 1280 x 720 per panel | A third of the resolution gone |
Recommendation: keep the existing box and 2x2 controller and rebuild the content as one 3840 x 2160 timeline with each panel's picture cropped into its own quadrant; move to a player with one output per screen only when the wall grows past four 1080p panels or needs a mapping a fixed 2x2 slicer cannot cut, and accept the scaled 720p picture only for video with no small text.
What changes on the wall when the content is built this way?
The workflow changes, the hardware does not: you compose the wall in a 3840 x 2160 project instead of exporting one 5760-pixel-wide master, and you test the export before the wall opens. Building from one project keeps the three panels consistent, with the same type sizes and colors throughout, and motion that crosses a seam keeps moving across it, because all three slices come from one stream through one controller. That is the quiet advantage over separate players.
Two points on the file itself. A 3840 x 2160 frame has four times the pixels of a 1920 x 1080 frame, so plan for roughly four times the bitrate at the same quality and check how the export actually plays before deployment, as covered in video bitrate for a signage player: test before you deploy. And the box has to start the file on power-up from the USB drive; that behavior is part of choosing the player, which the digital signage player guide walks through.
Can three separate players each play their own cropped file instead?
They can for slide-based walls, but not for one long video: separate players start their loops at slightly different moments and their clocks drift about 0.18 seconds per hour at a typical 50 ppm crystal tolerance, so the seams slide apart within a day. The figure comes from ordinary crystal accuracy: 0.00005 x 3600 seconds = 0.18 seconds per hour, and a day of running adds more than 4 seconds.
For static price lists and slides, drift is invisible, and this path is the simplest of all: a wall of three small offline retail displays, the kind we sell for shelves and counters, each playing its own cropped file from its own card with no subscription. For one video crossing the seams, the drift shows as a rolling tear, and the sliced 4K canvas through one controller is the right answer. What it takes to keep players in step is covered in video wall sync architecture: how players stay in step.
Questions
What goes in the unused fourth quadrant?
Anything, because it never reaches a screen when only three of the four outputs are cabled. Black is the tidy default: if a cable is ever moved, an all-black quadrant makes the mistake obvious in seconds. You can also place a fourth independent message there and cable the fourth output later, which grows the wall to four screens without rebuilding the file.
Do three portrait screens fit the 4K canvas without scaling?
On paper yes: three 1080-pixel-wide portrait columns total 3240 pixels, inside the 3840-pixel width, and a 1920-pixel-tall picture is inside the 2160-pixel height. A fixed 2x2 slicer still cuts at x = 1920, which lands inside the second column, so the strip cannot be laid flat. Rotate each panel's asset 90 degrees into its own quadrant and export the same single 4K file.
Should the 4K export be H.264 or H.265?
Use whatever codec the box already plays at 4K, because the file is an ordinary 4K master and asks nothing new of the playback path. If motion in the rebuilt file shows blockiness the 1080p version did not, raise the bitrate before changing codecs, since a 4K frame carries four times the pixels of a 1080p frame. The exact codec list belongs in the box's specification sheet.
Can a wall controller take a wide file and split it three ways?
A processor that maps arbitrary regions to outputs can take a wide canvas and send three 1920 x 1080 regions to three screens, which is how larger walls are driven. A fixed 2x2 slicer cannot: it cuts one standard input into four equal quadrants and offers no region mapping. Check a controller's specification for region mapping before buying it for a three-screen row.
Planning information, not legal or engineering advice. Rules differ by place: check codes and permits with your local authority and your licensed trades.