A dummy file generator for upload limits, quotas and transfer tests
Testers, developers and IT staff need files of a known size constantly: does the form reject a 26 MB attachment, how fast does a 2 GB copy go over the VPN, what happens when a share hits its quota. A dummy file generator saves you from hunting for a video or a disk image that happens to be roughly the right size.
Set Number of random files, set Each file size in B, KB, MB or GB, pick the Contents, and click the hopper. Sizes use binary units, so 1 MB is 1,048,576 bytes and the file on disk matches to the byte. One run makes as many as 10,000 files.
Before starting, the window shows the arithmetic, for example "12 × 25 MB = 300 MB of 41.2 GB free", and it refuses to begin a set that the chosen drive cannot hold.
Test files that open, not just random bytes
An upload validator that sniffs file types will reject a renamed .bin, so the Contents setting offers eight choices. Five produce genuine files of their format, padded internally with random filler to reach the size you set:
- PNG: a picture of static that opens in Photos. The smallest possible is 84 bytes.
- PDF: a one-page document reading "Test file made by Random File Generator", good for testing a PDF upload of a precise size.
- ZIP: an archive containing a readme that File Explorer can open.
- DOCX: a Word document with the same one-line text.
- MP4: a short clip that plays in Media Player, from about 12.5 KB upward.
- Random bytes (.bin), Zeros (.bin) and Text (.txt, lines of lorem ipsum) start at a single byte.
How to create a large file of a specific size in Windows
The built-in command fsutil file createnew will make a file of a chosen size, but only one at a time and only filled with zeros. Zeros compress to almost nothing, so they give misleading results for anything that compresses or deduplicates in transit, such as cloud sync, backup software or HTTP compression.
With Random File Generator, choose Random bytes for data that cannot be compressed: every byte is drawn from the full 0 to 255 range and every file is seeded separately, so no two in a set are identical and deduplication cannot collapse them. Choose Zeros when you want the opposite: on NTFS those files are written sparse, reporting their full size while occupying almost no disk space.
Naming, stopping and keeping existing files safe
Pick the destination with Browse... under Output. Files are named from the prefix in Settings (File names start with, default test) plus a counter padded to at least three digits and the right extension, so a PDF run produces test_001.pdf, test_002.pdf and onward. If those names already exist in the folder, numbering continues past them; nothing is overwritten.
Each file is shown as a punched card feeding through the hopper, with a live percentage and write speed. Click the hopper again to stop: the half-written file is removed and completed ones stay. Set When it is finished to Open the folder they went to if you want File Explorer to pop up at the end.
