Ok, I put the drives in another machine, found the idle time was set to 8 seconds, and changed it to 300 as advised. Honestly, it looks like performance has gotten worse.
Here's another round of tests:
100MB
500
1000
2000
4000
Another thing I noticed that's weird is the way...
Well, looks like two of the disks may be having some S.M.A.R.T. issues:
Read Error Rate rawValue="0"
Spin-Up Time in millisecs rawValue="4391"
Start/Stop Count rawValue="366"
Reallocated Sectors Count rawValue="0"
Seek Error Rate rawValue="0"
Power-On Hours rawValue="3720"
Spin Retry Count...
The reason it is recommended to switch the individual disk cache off and have the controller cache on is for protection if there's a power outage. If you're transferring data the info in the disk cache will get lost (if you don't have a UPS) whereas the controller has the ZMM attached. With a...
I've previously had nothing but great performance from them in software RAIDs and even a hardware RAID 0 that's currently in a Server of mine (though I know 0 and 5 are VERY different beasts) so I figured why not? Despite the months of digging I've NEVER stumbled across the post you linked from...
Before I start I just want to say that I've done a TON of research on this and have tried everything that I can find, short of getting a different controller, to resolve the write speed issues that I am seeing.
Current Rig:
OS: Windows 7 Professional 64-Bit
Case: Corsair Obsidian Series...