Search the Effects panel for a motion trail and you get a shortlist of names that all sound like the same thing. Trails. Outline Echo. Echo Power. Echo Swag. Ghosting. Picking between them by name is guesswork, and the guess is expensive in time rather than credits, because you only find out what a preset does after you have built the edit around it.
So this piece runs three of them across the same five seconds of dance footage at 16:9, exports each one, and compares them against an export of the same clip with no effect at all.
The clip with the one preset that does what the shortlist promises: 1280 by 720, 30 frames per second, 5.07 seconds. Shown here as frames in sequence; the full video and a GIF version are supplied alongside this document. The dancer and the room were generated for this article.
Five Motion Trail Presets, One Promise
The names come from the panel itself. Searching trail returns Trails; echo returns Outline Echo, Echo Power and Echo Swag; ghost returns Ghosting and Floating Ghosts; motion returns Motion Blur and Neon Motion. Searching afterimage returns nothing.
In the web-editor session used for this test, applying these Effects did not display a separate generation charge. Preset names, availability, controls, and pricing can vary by account, region, and version, so check the current panel and pricing information before planning an edit around a specific preset.
That last one matters: the presets are not organized under a heading that tells you which produce a repeated figure and which produce a light effect. The names are the only signal, and the names overlap.
The Clip They Were Tested On
A trail effect needs something that moves across open space, so the source is one generated still animated into five seconds: a dancer in a dark dress turning in an empty pale room. The dark figure against a pale wall also makes the measurement below straightforward, because the dancer is the dark part of the frame.
The dancer sweeps her raised arm down and across her body in one continuous movement, then turns a quarter turn on the spot and lifts the other arm out to the side. Her long skirt swings out as she turns. She stays near the middle of the frame and fully inside it from head to feet throughout. The room, the wall, the floor and the light do not change, and nothing else in the room moves. Locked-off tripod shot, no zoom, no push in, no dolly, no pan and no crop, and the last frame has the same framing as the first frame.
Three seconds of that clip went into a fresh project five separate times: once with no effect, and once each with Trails, Echo Power and Ghosting. Every export used the same settings, so the no-effect version is a true control — and it measures as one. Compared with itself, it changes 0.0 percent of its pixels.
The same frame through each preset. The bottom-left panel is the only one with a second dancer in it.
Trails Adds Light, Not a Copy
Trails is the name that sounds most like the thing, and it is a lighting effect. Across the clip it changes 43.1 percent of the pixels and lifts the average brightness by 32 levels, drawing a bright streak that sweeps past the dancer.
The giveaway is the dark area. If a preset were duplicating the figure, the number of dark pixels would rise. Under Trails it falls to 0.52 of the control — the streak is bright enough to wash out half of what used to be dark. There is no second dancer; there is a light source that was not there before.
Echo Power Is the One That Duplicates
Echo Power is the quietest of the three by every whole-frame measure. It changes 6.7 percent of the pixels and pulls the average brightness down by 2 levels. What it does is put four more dancers on screen.
Averaged over every frame of the three-second clip. Dark area is the count of dark pixels relative to the control.
Two numbers separate it from the others. Its dark area rises to 1.22 of the control, where both of the other presets cut theirs in half. And taking the dancer's own outline from the control frame and sliding it around the effect frame to find where else it fits, the second-best match scores 0.978 at an offset of 40 pixels straight up, with no sideways shift at all. The control's own second-best match, which is just the outline partly overlapping itself, scores 0.765. Trails scores 0.731, below that floor.
So Echo Power stacks copies at a fixed vertical offset. That is a useful effect, and it is not what the word echo suggests: the copies do not sit where the dancer was a moment ago, they sit directly above where she is now.
Ghosting Is a Flash, Not a Trail
Ghosting changes 87.9 percent of the frame and raises the average brightness by 57 levels. Its dark area drops to 0.50 of the control and, for much of the clip, the dancer is barely visible at all. It is a whiteout with a soft edge, closer to a light leak than to anything ghost-shaped.
No Measurable Past-Position Trail in This Test
The assumption underneath all three names is that the effect leaves something behind along the path of the movement. That is testable. For each frame, take the pixels the effect added, and ask how densely they sit over the area the dancer occupied a few frames earlier — then ask the same question about the area she will occupy a few frames later. A real motion trail is much denser on the earlier side.
Density of added pixels over the dancer's past and future positions. Flat lines mean the added pixels have no relationship to the direction of time.
The ratio of past to future comes out at 1.10 for Trails, 1.03 for Echo Power and 0.99 for Ghosting. All three are within about a tenth of even. Whatever these presets add to the frame, in this run it lands with no regard for where the subject used to be.
That is worth knowing before you plan an edit around one. A repeated figure placed at a fixed offset reads as a stack, not as movement, and it stays a stack however the subject moves.
The Silhouette preset as it appeared in the export preview. Its export did not finish in this session, so it is described here rather than measured.
One more name is worth flagging because it is easy to reach for. In this session's preview, the preset named “Silhouette” appeared as a decorative overlay rather than a verified cutout workflow. Its export did not complete, so do not treat that preview as a conclusive preset test. For subject isolation, use the Remove Background workflow available in your version and review the edge quality before export. If what you want is a clean cut-out over a different backdrop, that is the separate Remove Background tool, not this effect.
What the Two Requests Cost
The settings line and the cost line for the one video request in this piece.
The cost lines as shown under each request, read on 14 September 2026.
Twenty-six credits for the footage. The effects themselves and all five exports cost nothing, because every effect in this piece is applied in the editor rather than generated. The still and the clip were made through the AI image generator and AI video generator flows. These figures are for these configurations on this account and this date and come from the line the composer shows before each request; other models, durations, sizes and modes are charged differently, and prices change.
The One-Minute Test for Any Effect
The comparison above took one generated clip and five exports, and it is worth repeating on your own footage rather than trusting these three results, because a preset that stacks copies 40 pixels apart on a full-body shot will look entirely different on a close-up.
The test is short. Export three seconds of your clip with no effect, export the same three seconds with the preset, and put the two side by side at the same moment. Then ask one question: is there more dark in the effect version, or less? More dark means a second copy of the subject. Less dark means light has been added on top. That single comparison separates the copy effects from the lighting effects faster than any of the names do, and it costs nothing but two exports.
If your goal is a silhouette duet, first isolate the subject, place it over separately licensed footage, and check a short export for edge cleanup, contrast, and timing. Use effect presets as a visual treatment after the cutout has been reviewed.
The planning row for this page asked for three things: a silhouette duet cut against real footage, beat-synced motion trails, and matte and edge settings for the silhouette. Two of those are not in this piece. The presets that carry trail and echo names turned out not to follow the subject's path at all, which removes the basis for syncing a trail to a beat, and the effect named Silhouette is a decoration rather than a matte, so there are no edge settings to report. The title says what the article does deliver, which is what each of those presets actually puts on screen. The address of the page is unchanged.
What is not covered here: background removal and the cut-out route, transitions between two scenes, beat detection and beat-matched cutting, speed ramps, color grading, 4K export, and any dance instruction. The dancer, her clothing and the room were generated for this article; no real person appears, and any footage you run this test on should be your own.
Written 14 September 2026. The still and the clip were generated in CapCut for this article and all effects were applied in its web editor; no real person appears. Interface labels and cost lines reflect a single session on that date and the configurations named above, and may change. Pixels changed, brightness shift and dark area are averages over every frame of a three-second export, each measured against a no-effect export of the same clip made through the same pipeline; the outline match is normalized cross-correlation of the control frame's dark-pixel mask against the effect frame, and the past-and-future comparison is the density of added pixels over the subject's mask at frame offsets of plus and minus eight.