Initializing... drag & drop files here
Supports: PNG
PNG (ISO/IEC 15948, third edition June 2025) is a lossless raster format using Deflate compression — perfect for screenshots, scientific captures, and rendered frames where every pixel matters. MJPEG (Motion JPEG) is the opposite kind of beast: a video format where each frame is an independent JPEG with no inter-frame prediction. Wrapping a PNG sequence as MJPEG gives you a video file that surveillance DVRs, machine-vision software, and frame-accurate editors can ingest without re-keying or losing the ability to seek to any single frame instantly.
| Property | PNG | MJPEG |
|---|---|---|
| Type | Still raster image | Video (sequence of JPEG frames) |
| Standard | ISO/IEC 15948 (1996, 3rd ed. 2025) | RFC 2435 (RTP payload, 1998); de facto codec |
| Compression | Deflate (LZ77 + Huffman), lossless | JPEG DCT, lossy per frame |
| Inter-frame compression | N/A | None — each frame is independent |
| Color depth | Up to 48-bit (16 bpc) + alpha | 8-bit per channel YCbCr typically |
| Alpha channel | Yes | No (drops to background color) |
| Typical compression ratio | 2:1 to 4:1 (lossless) | ~20:1 (per-frame JPEG) |
| Frame-accurate seek | N/A | Yes — every frame is a keyframe |
| Storage vs H.264 | N/A | ~5-20x larger (e-con Systems) |
| Common containers | Standalone.png | .avi,.mov,.mkv, RTP stream |
| Browser native playback | All major browsers (image) | Safari, Chrome, Firefox, Edge |
| Setting | Use case | Result |
|---|---|---|
| Image Duration 1/60s | High-frame-rate playback, slow-motion review | 60 fps output |
| Image Duration 1/30s | NTSC video, surveillance recordings, web playback | 30 fps output |
| Image Duration 1/24s | Cinematic look, animation reels | 24 fps output |
| Image Duration 1s | Time-lapse, slideshow, evidence frames | 1 fps output |
| Quality Preset Highest | Archival, scientific accuracy | Largest file, near-lossless JPEG |
| Quality Preset Very High | Recommended default | Visually lossless, ~30-40% smaller than Highest |
| Quality Preset Medium | Web preview, quick share | Visible JPEG artifacts on flat color, half the size |
| Quality Preset Lowest | Thumbnail or reference reel | Blocky in detail areas, smallest file |
It often isn't — but PNG is lossless and MJPEG is lossy per-frame, so it depends on content. Photographic PNGs (microscopy, dashcam captures) usually shrink because JPEG compresses natural-image gradients aggressively. Synthetic PNGs (UI screenshots, line art, charts) often grow because Deflate handles flat color and sharp edges better than JPEG DCT. If output looks blocky around text or solid fills, raise the Quality Preset to Highest, or consider PNG to MP4 with H.264 instead.
It's the inverse of Image Duration. 1/30s per frame produces a 30 fps file; 1/24s produces 24 fps; 1s produces a 1 fps slideshow. Most surveillance recorders expect 15-30 fps, NLEs expect 24/25/30/60, and web players handle anything reasonable. If you set 10s per frame, expect playback to look like a slow slideshow rather than continuous motion.
No. JPEG (and therefore MJPEG) has no alpha channel, so any transparent pixels are flattened against the Background Color you select — Black by default, White and 20+ named colors available. Pre-composite your PNGs against the intended background (or use PNG to WebM with VP9 alpha) if you need transparency in the final video.
XConvert outputs an MJPEG-encoded video in a standard container (typically.mjpeg/.avi/.mov depending on pipeline). Most ONVIF-compliant DVRs, VLC, FFmpeg, and major NLEs ingest MJPEG natively. If your specific device rejects the file, try wrapping it as AVI via PNG to AVI or as QuickTime via PNG to MOV — MJPEG inside MOV is the format Apple historically used for digital-camera video.
MJPEG when you need frame-accurate seeking, low-CPU decoding, surveillance/machine-vision compatibility, or compatibility with pre-2007 hardware. H.264 (use PNG to MP4) when you need 5-20x smaller files, modern streaming, or YouTube/social uploads. Per Wikipedia, MJPEG achieves around 1:20 compression versus 1:50 or better for inter-frame codecs — the gap is real.
XConvert sequences uploads in the order you add them, so the safest approach is zero-padded filenames (frame_0001.png, frame_0002.png … frame_9999.png) and uploading them sorted. Operating systems sort "frame_2.png" after "frame_19.png" lexically, which is why padding matters. Renaming before upload is far simpler than reordering after.
The PNG-to-MJPEG flow is image-only by default — there's no audio track because PNG has no audio. The Advanced settings expose an Audio Codec dropdown for downstream remuxing, but you'd typically add audio after with FFmpeg or an NLE. If audio matters, render to MP4 first via PNG to MP4, then import and dub in your editor.
Up to 4320p (7680×4320, 8K UHD) via the resolution preset, or any custom width × height you enter. Be aware that MJPEG at 8K with 30fps and Quality Preset Highest produces extremely large files — easily several GB per minute — because there's no temporal compression. Drop to 2160p (4K) or lower unless you specifically need 8K archival.
Files upload to XConvert's processing pipeline, convert, and stream back as a download. Nothing is stored long-term. There's no watermark, no sign-up, and no file-count limit. If you need the reverse direction, see MJPEG to PNG for extracting frames back to PNG, or Compress MJPEG if your output file came out larger than expected.