ide raid does *nothing* to improve access time - which are a function of the HDD. You can stripe a million ide drives together in RAID_0 config, and ur still gonna have sucky access times.
ide_raid is good for things like video_capture, & audio_recording, where ur focusing on *sustained* x-fer rates. ide_raid sux at running an O/S & swap file (cuz it has sucky access times).
ppl who say 'Windows loads x seconds faster with ide-raid' don't get it - it's not about loading windows - it's about running the OS, apps & swap - that's what a boot drive does 99% of the time.
Have read countless horror stories about ide raid at storagereview bbs. The stability/reliability of ide-raid vs LVD SCSI is orders of magnatude lower. If ur value reliability, ide-raid is not the way to go (for running an OS).
Soccerman was on the right track a couple pages back. He got long-winded, but shows that he understoof how access times affect how (efficiently) Windows operates.
Here is a copy-&-paste of a post I ripped from a guy at Ars bbs. Forget his user name - RP I think. Sounds like he knows what he's talking about. I found it interesting. See here:
1) In general, real-world performance of a modern disk drive under a modern OS using a modern file system is far, far more dependent on data access time (RPM and seek) than sustained transfer rate (density), regardless of whether the system is a workstation or a server. You gain very little improvement in performance as density increases. In modern OS such as Windows 2000, multiple threads do I/O independently and virtual memory system accesses paging file in stochastic manner. In addition, modern file system such as NTFS does fragment, and it uses transactional logs, permissions, etc. All these leads to disk access pattern that is random in nature.
2) Because of #1 above, IDE drives do not scale well as load (number of I/O requests) increases. Meaning that - if you put a non-trivial load such as multitasking on an IDE drive, your system will crawl.
3) The improvement in density far outstrip the improvement in access time. In 3 years, density increases by about 800% (double every year), while data access time improves by less than 100% (far less for IDE drives - seek time improves very little for IDE drives). The difference is so large that a modern disk spends most of its time seeking and not actually transferring data.
4) Winbench is largely a lightly-loaded, single-tasking benchmark. And it does not really represent a real-world condition where people do non-trivial tasks. As such, it biases heavily against SCSI drives. Go to storagereview.com, compare the latest Maxtor Diamondmax Plus 40 against the 2-year old Seagate Barracuda 9LP under Winbench and under Intel IOMeter. The Maxtor beats the Barracuda in Winbench by a large percentage, but the 2-year old Barracuda beats the Maxtor decidesively in Intel IOMeter workstation test.
5) It's cheaper for manufacturer to improve density because higher density means less number of platters and lower cost. Improvement in sustained transfer rate is more or less a by-product. Thus, it is in their best interest that mass consumers perceive sustained transfer rate as the most important factor in determining performance, when in fact it is not. They can't do this marketing brainwash with SCSI customers, who are generally more knowledgeable. So SCSI drives generally emphasize data access time and that's also why Cheetah X15 uses very small platters and can only store 18 Gig on a low-profile drive while the new IBM IDE drive, with a much higher density, can store 75 Gig.
