oops
"so you say PCI raid card is way better than onboard raid?"
Well I know it is for me, and I've read of others who experienced the same improvements. However I am sure others would say they have good success with the HP Tech onboard RAID. Well I've read a few peeps say they have had good success with HPT onboard RAID. But overall I've read more complaints on it and have personally experienced the downside of the onboard HP Tech RAID. Anyway I'm sure some people like both, so ymmv.
"...i am using an Abit BE7-Raid mobo.... 3 x 120GB 8mb WDSE, 2 are on raid 0, 1 for back up...."
That was the last mobo I used that had an onboard RAID. It didn't like my Maxtor HDs at all, even though it claimed ata 133 support. I spent a lot of time changing and tweaking the BIOS of that mobo, including a lot of the HP Tech BIOS tweaks and changes trying to get decent performance and compatibility out of it. I never did acheive all my goals with it either. BTW did you ever experience the Cold Boot problem from your BE7 RAID? Part of that problem came from the HP Technologies onboard RAID. Not all but some. I know it has been fixed now though. I even had that part of my BE7 RAID fixed before I changed to my Asus P4PE a couple months ago..
"well as most ppl said, don't really feel much of improvements DUE TO human feeling isn't so sensitive..."
I have heard that very same complaint from other Abit users with HP Tech onboard RAID! And I complained about it...
"i believe anything below double can hardly to be noticed by human feeling(try to use a 1.8GHZ and a 2.4GHZ P4 system without anyone telling you which is which, can you really tell the differences??)......well the benchmark tells me my raid 0 set up is about 30% faster than a single WDSE."
I don't know which benchmark you're using but with Sandra as one example, I get scores of app. 54,000. Now that is almost double single HD performance. But I am using the Maxtor DM +9s with 8 MB cache. Still I can notice RAID differences long before such scores. Even at scores of 40,000 I can notice a little speedier performance. Again I attribute this more to the RAID controller card as much as the HDs. In AATO I get read scores of app. 90,000 - 95,000+ and in read scores I get app. 75,000 scores. This is *considerably* higher than single HD scores.
"I learned that the raid speed is limited by the ATA and FSB. well a single WDSE is ATA100, and my raid ports are ATA133, so it may be fair for just around 30% increase in speed.... now i see some dual mobos have ATA150 raid ports, that may be better than ATA133 raid ports....just my thoughts."
While HP Technologies may claim ata133 compatibility I haven't seen my RAID using it actually benefit with the Maxtor HDs. Matter of fact as I was saying my Maxtor HDs did not run nearly as well with the HP Tech controller. Most people I have read posts from seems to claim WDs gives the best performance and compatibility on the HP Tech controllers. Still not many of them gives extra-oridnary good bench scores. The whole thing is strange to me because if you read and believe the reviewers, then the HP Technologies controllers *supposed* to be a top or the top performer in RAID 0. I and many others have yet to see this as being true in actual use.
Cheers,
Mike
PS There were several stories about this very thing in the Abit forum here last year about this time. Which was also the time I changed over to Promise. After reading all the stories about the HP Tech RAID problems.
One other thing for the UT2003 gamers here. I don't know your load times for the game and mapps, but I can load the game from desktop in about 5 seconds and load a mapp in less than 5 seconds. If I replay the mapp, I load the mapp almost instantly. What are some your load times, app.?