I'm using Backup and Sync in combination with Drive File Stream. There's a work-a-round to see the Backup and Sync folders in Drive File Stream by doing Shift+Z on the folders to make a second location in My Drive. However, it seems Google is throwing this option out in September. So I'm wondering if there's a way to accomplish this though rclone mounting?
I've currently got it working by manually changing the folder id to one of the computers in the "Computers" tab. But I have multiple computers so that won't work.
What is your rclone version (output from rclone version)
go version: go1.14.7
Which OS you are using and how many bits (eg Windows 7, 64 bit)
Windows 10, 64 bit
Which cloud storage system are you using? (eg Google Drive)
The command you were trying to run (eg rclone copy /tmp remote:tmp)
rclone mount --vfs-cache-mode full [name of drive]:/ [drive letter]:
I'm not into command line stuff so I'm not sure what "--vfs-cache-mode full" is doing, or if there's other commands that accomplish the same thing. I don't need it to cache any data on the local drive. I'm just looking to access and edit directly in the cloud. But that's what someone said to use on a YouTube tutorial and it worked.
I tried my original command: rclone mount --vfs-cache-mode full [drive name]:/ [drive letter]:
I then tried it with what you recommended: rclone mount --read-only [drive name]:/ [drive letter]:
The video played in both instances and didn't stutter but it was only a 30MB video. I tried a higher bitrate video just now that was closer to 3GB and it did stutter, even under the read only command. I got a 10ms ping, 28 mbps down and 6 mbps up in the speed test.
I noticed when I use the vfs command and I try to open a folder containing videos it is writing to my local disk. Is this it filling a cache?
Windows Explorer is pure evil once you have more files and folders in one root folder than it wants you to :). That number, at least in my experience, is around 4,000 to 5,000. That's when it slows down to a crawl, unless the contents are indexed. This can be accomplished by using this command, once the mount is up (note that you have to add --rc to your mount command for this to work):
I recommend to set view as "Details" ("Small icons" & "List" can be used too) in your Windows Explorer and then apply that view for all folders of the same type (photos only, music only, mixed etc...). Also enabling "Always show icons, never thumbnails" is good alternative. Probably even better but less flexible.
To apply: File -> Options -> View
It helps because it effectively prevents thumbnail generation which consumes quite a lot of data/requests which you don't want. However metadata (properties) are still being read so Windows Explorer may hang if there are lots of files which specific metadata attributes (apart from standard ones for all files like size, creation/access/modification time, etc.) have to be read (e.g.: you enable dimensions for photos or length for videos in your "Details" view). It is possible to disable thumbnail generation and file specific metadata parsing in effect making Windows Explorer very effective while viewing rclone mounts. Although loosing some convienience as that information won't be available to you through Windows Exporer. You can do that with ShellExView (https://www.nirsoft.net/utils/shexview.html) by disabling required property (MF ASF/AVI/MKV/MPEG/MPEG-4 - for video thumbnails and properties; IPropertyStore Handler for Images - for photos) and thumbnail handlers (Photo Thumbnail Provider). Then restarting Windows Explorer to apply the changes. In my opinion photos thumbnailing (Photo Thumbnail Provider) can be left untouched as they don't affect performance that much and you may not know what photo the file contains if it uses general name. Photos metadata (IPropertyStore Handler for Images) should be disabled or you should use viewing mode which doesn't need file specific metadata information. Else, you risk hanging Windows Explorer with folders which have lots of photo files. All those things apply to music, word and other files from which file specific metadata information can be extracted as every file must be read to get that information compared to general file system metadata (file size, modification date etc.) which don't have to be parsed per file basis.