Wow... that's pathetic. I run 32 GB RAM in my 2009 Mac Pro Quad and compose music (mostly for TV and film) with 250 tracks (being the most) that have gone as much as 11 minutes... it certainly helps that I also run VSL's VEP5 in a LAN setup with my MacBook Pro (16 GB) as the master (with the Mac Pro acting as the slave) AND I use Logic Pro X (native to Mac)!! All of it running 16 TB (external) direct-from-disk of stored Kontakt and additional samples...
It sounds like you should lighten up your load (if you want to stick with FL) and get a second computer. Use one computer for the DAW and the other to host the sample libraries through a LAN network...

Otherwise, switch to a DAW that's less imposing on your computer... like Logic Pro X, Reaper, Cubase, Studio One, Ableton LIVE, etc.

 

you might also wanna try the smart disable option for unused plugins in tools>macros, but it's only useful if you dont constantly use all the plugs used in your project.

 

140% cpu usage is okay, it's the nature of MacOS metering,
I often get around 580% with Reaper on my 6-core hackintosh

 

I have a shite quad-core i7 MBP 2011 with 16GBs of DDR2 RAM and I can run a metric shit-ton of plugs and VIs along with plenty of 88.2 kHz/24-bit audio tracks all in real time with low buffers for at least the length of an average song with zero dropouts.

 

No, but I've considered taking it for a spin a few times. I'm a fan of the open-source thing for sure. I mainly use Digital Performer 9, which is actually a lot less problematic than some of these DAWs, from the sound of it. I've also gotten into using PT 12 for the audio side, which also runs great (even running lots of AUs through Patchwork or DDMF Metaplugin), and ... (sigh) ... Logic. Once I managed to get it installed properly, the new Logic seems to run okay, although I haven't really tried mixing anything in it.

 

With the newest build of FL20 for mac (57) The cpu issue is gone. My project is now floating around 50% steady through all parts. Indeed it's a K'd version thing... RSA checks running up the usage i guess. Anyways...problem solved for now.