I made a video showing the workflow in TortoiseHg Workbench. Here's a description of how it would have avoided the key pain points I had with Plastic SCM.
I've created a dummy project with some text files checked in. For demonstration purposes, the contents of three of these files are identical, to confuse automatic detection of renames.
Move files using Explorer to a different folder
Switch to TortoiseHg Workbench
Perform manual refresh to detect changes in the workspace
TortoiseHg now lists four removed files (in red, prefixed with "!") and four added files (in pink, prefixed with "?")
I can use two different methods for telling the SCM about renamed and moved files. First, I'm going to show the quick manual method, which has no equivalent in Plastic.
Right-click on an added file to bring up the context menu
Select "Was renamed from"
This brings up a submenu that lists all deleted files
I can now quickly pick the correct file from the submenu
No automatic move detection is performed at any point - this process would be fast even with thousands of files. Naturally finding the correct target from the submenu would be slow for the user, but I believe the submenu uses some kind of a quick sorting method, perhaps based on filename similarity, because I often find the correct files near the top of the list
The second method uses automatic move detection, similar to Plastic SCM. However, this is a user-initiated action on the workspace, rather than being a global preference that must be either on or off at all times - again reducing the amount of unnecessary processing compared to Plastic.
Select any number of added files
Right-click to bring up the context menu
Select "Detect Renames..."
This brings up a separate window for processing renames
Just like Plastic, I can now adjust the similarity percentage and then run contents comparison. However, unlike Plastic, I am able to adjust the percentage and run the detection multiple times. Each time, it will only run on the files selected in the left pane, rather than all files (which would be really slow).
Detection results are listed in the right pane
I can now select the matches I want, and manually accept them. If this was a real situation, some matches would have a higher percentage, and these would be prioritised, just like when Plastic chooses the best match. However, unlike Plastic, if the similarity percentage gives the wrong match (for example, a file was both renamed and changed, as happens often with Unity .meta files), I can still quickly fix it by manually choosing a different match. The only way to do this in Plastic is to globally turn off move detection, refresh the workspace, and manually try to find the correct files from among all changes (which are no longer easy to find due to the move detection being turned off).
I can keep making matches within this window until I'm satisfied, then close the window
Once the window is closed, TortoiseHg will refresh the workspace - in Plastic SCM, the workspace would have been unnecessarily refreshed several times, which can take minutes when there are hundreds of moved files.
Hope this helps clarify the workflow that is missing from Plastic.