Just because the network is capable of Gigabit speeds, doesn't mean your HDs are. While I would expect better than 40MB/s w/ a modern HD, I obviously don't know if these are indeed modern HDs. It's also possible anti-malware software can slowdown these transfers as it examines the files in detail, looking for viruses, trojans, etc.
In short, using file transfers as your measuring stick for network efficiency is generally not a good idea. There are just too many other issues involving file access that can mask your true network performance.
Instead, you should be using tools like NetCPS, Iperf, Qcheck, etc., that simply pump raw data from their buffers across the network. You’ll get far more meaningful and accurate results.
Is I get ~950 when i use the -w 64k option, but just ~360 with no option... I googled deeply to find some tweaks but none seemed to work!
So is it necessarily due to the HD? I'm just testing them and the slowest (of the 3 pcs) has average write and read speed of 80MB/s...
Since the Iperf -w options controls the tcp window size, I assume this is the issue. You may need to find a "tweak toy" that allows you to change the default/global tcp window size.
Beware, sometimes changes of this type can have negative side effects. So what might be good for large files transfers, might be bad for smaller file transfers. Or might lead to compatibility issues w/ say, your router. Tcp windows size isn't something I futz with, so I can't say for sure. But usually the default is the default for a good reason.