mn98

For me, as an end user, there's an argument for two flavours of `broadcastThrough` to exist, e.g. a `raceThrough` for cases where performance is key and a `coherentlyThrough` where control...

Sorry, yes, that's what I meant, no constraints on faster pipes running ahead by any number of elements. If that's already achievable via `prefetch` then great. Does that mean `broadcastThrough`...

Great, makes sense, thanks very much both for sharing your thoughts on this! In conclusion I suppose it just remains for this issue to be addressed at some point.

> Opened #2621 to, if not correct the behaviour and add the new variant, at least make the documentation a bit less inaccurate. @mn98 would that be more informative? Changing...