Switch to sequential transfer with big files

[nopria]

Make Rclone transfer small files in parallel while while big files sequentally

Since my upload bandwidth is about 3 MB/s, I would like to transfer big files (say 500M) one by one, to minimize transfer time and error chances. At the same time I would like to keep the parallel transfers for smaller files.

Now I run

rclone sync "%orig%" %dest% --exclude=/__pycache__/** --exclude "/System Volume Information/**" --exclude "/$RECYCLE.BIN/**" --exclude-if-present .rclone_ignore  --max-size 500M --transfers=5 --fast-list --create-empty-src-dirs --checksum --links --progress --track-renames --track-renames-strategy hash

and then the same command with "--min-size 500M."

but this is very inefficient, given the hash checksum. That's why I would like to run one single command. Is there a way to say to Rclone "use simuntaneous transfers with files smaller than X but switch to sequential (one by one) file transfer with files biggert than X"? Or at least to mimic in some way such behaviour?

Version

rclone v1.71.0
- os/version: Microsoft Windows 11 Home 24H2 24H2 (64 bit)
- os/kernel: 10.0.26100.7171 (x86_64)
- os/type: windows
- os/arch: amd64
- go/version: go1.25.0
- go/linking: static
- go/tags: cmountsing? (eg Google Drive)

'rclone config redacted'

[pcloud]
type = pcloud
token = XXX
hostname = eapi.pcloud.com

[kapitainsky]

Only option to control processing order is --order-by

Documentation

But it won’t give you exactly what you want.

You can also split your transfer between two different rclone commands. One with multiple transfers for small files and one with single transfer (--transfers 1) for big files. Use --max-size/--min-size to process only files with given size.

PS:

@nopria

--fast-list

is not supported by Pcloud so it does nothing,