Why does my dual camera video look different from the preview?
It depends entirely on how the app is built. Some dual-camera apps keep the two camera streams separate while you shoot and only arrange them into a final layout afterwards, as a rendering step — in that design, the live view is a rough stand-in for a result that gets computed later, and small differences in crop, alignment or timing between the two are normal, because the thing you watched while filming was never the thing that got written to disk.
A dual-camera app that composites both images into one frame in real time and records that exact frame does not have this gap. There is no later rendering pass to introduce a difference, because there is nothing left to render — the file already contains the frame you were looking at, in order, one after another. DualCam works this way: the two camera images are drawn together on the GPU as you shoot, and that composited frame is what gets written to the video.
Two different pipelines, two different guarantees
It helps to be specific about what "the recording" even means in a dual-camera app. One approach saves two source streams and generates the combined picture-in-picture or split-screen layout as a separate step, either right after you stop or when you export — the preview during filming is a live approximation of that eventual layout, not the layout itself. The other approach never separates the two: every frame is composited the moment it arrives from both cameras, and that single composited frame is what gets encoded into the file, frame by frame, for the whole take.
Only the second approach can honestly promise that what you watched is what you get. If an app tells you the file might look slightly different once it is done processing, that is usually a sign it uses the first approach, and the wording is accurate — there genuinely is a processing step between the viewfinder and the finished video.
Why an ordinary single-camera recording can drift from its preview
This is not specific to dual-camera video, and it is worth knowing because it explains a lot of "the recording looks different than I remember" complaints that have nothing to do with a second camera at all. A camera app can apply stabilization that trims and shifts the frame as part of writing the file, which is a slightly different crop than what a smoother, unstabilized preview showed you live. Recording formats that capture a wider dynamic range than a normal screen can show also have to be tone-mapped down for playback on an ordinary display, so the same footage can read a little differently depending on what it is being viewed on and how that viewer processes it.
Neither of these is a bug in the everyday sense — they are deliberate processing steps that trade a live approximation for a better finished result, or that adapt one image for many different kinds of screens. But they are genuine differences between what you watched and what got saved, and they exist independently of whether a second camera is involved at all.
What a real-time compositor removes, and what it cannot
Because DualCam draws the final frame once, at capture time, none of the above applies to the layout itself. The crop between the two cameras, the picture-in-picture position, the filter colour — whatever the viewfinder showed at a given moment is exactly what is in the file at that moment, because they were never two different things to begin with. If a left-right split crops roughly a quarter of the frame from each side to fit a vertical canvas, that crop was already visible while you were filming, not a surprise introduced afterwards.
This also explains why a recording can look different from what you remember, without anything being wrong: layout, the picture-in-picture corner, its size and shape, and the filter can all be changed while a take is still rolling. If you switched from split screen to picture-in-picture halfway through and forgot, the file will faithfully show exactly that change happening partway through — because it is showing you what was actually on screen the whole time, not a single layout decided in advance.
What real-time compositing does not touch is anything that happens to the file after DualCam has finished writing it. Once you send a video through a messaging app or post it to a platform, that app usually re-encodes it on its own terms, which is a separate step outside DualCam entirely — see the guide on why a dual-camera video looks worse after you send it for what that pass actually does.
Common follow-up questions
Can a DualCam recording ever end up different from what I saw while filming?
Not from the compositing itself — the two camera images are drawn into one frame on the GPU as you shoot, and that exact frame is written to the file, so there is no separate rendering step afterwards that could introduce a difference. Anything that changes the file later, like a messaging app re-encoding it before it reaches someone else, happens outside DualCam and after it has already finished.
Why does my split-screen video look more cropped than I expected?
A vertical canvas is narrower than a phone camera's native 16:9 frame, so a left-right split has to trim roughly a quarter of the source image from each side to fit both cameras in. That crop is present in the live preview too — it is worth checking the corners of the frame while filming rather than assuming the saved file cropped something the viewfinder did not.
Why does the layout in my saved video change partway through?
Because you (or whoever was filming) changed it. Layout, the picture-in-picture corner, its size and shape, and the filter can all be adjusted while a recording is still running, and the file simply reflects whatever was actually selected at each moment — it is not re-deciding anything on its own.
Want to just do this?
DualCam records the iPhone front and back cameras at the same time and writes one finished MP4 while you shoot. Free, no account, no ads, nothing leaves the phone.