Jump to content

It’s a question of how you use it.


red930

Recommended Posts

really interesting, I hope you go on with those tests...

I have personnaly never tried raid O : this article in anandtechd had a big influence on me... now I read here on the forums many people (AG, Rgone) speaking about improved loading times in games.

 

would it be just "psychological improvement"? did you test loading times for windows? time to setup a fresh instal of xp, or other softwares?

 

sorry for my bad english, don't find the right words sometimes

Share this post


Link to post
Share on other sites

  • Replies 39
  • Created
  • Last Reply

Top Posters In This Topic

would it be just "psychological improvement"? did you test loading times for windows? time to setup a fresh instal of xp, or other softwares?
You don't need to test that, believe me, you will be able to EASILY tell the difference. And it is a remarkable difference. Window used to take 30 mins or so to install, now w/ raid 0 about 15-20 mins. HL2 loads in almost half the time it used to w/o raid 0.

Share this post


Link to post
Share on other sites

it seems so

but how to explain his results with far cry loading times ? really curious

 

about the install of windows, was it on the same rig? 'cause when I see the difference between nf2 and nf3 to install windows, that's also incredible ( I should say between athlon xp and a64)

 

but you all have convinced me to give it a try , as soon as I get new drives, mines are fed up ; why are those raptors still so expensive? :mad: what are maxtor, hitachi and seagate waiting for?

Share this post


Link to post
Share on other sites

I haven't been able to pry away the two Raptors from my buddy. So the testing is still in limbo.

 

This much is known. The data files for most game loads are compressed. This means the CPU has a direct impact on load times. Once the array is delivering data to the processor as fast as it can, it's up to the processor to get its job done.

 

As processor and memory sub systems have increased their performance, the overall speed of the system has increased. The array spends less time waiting on the delivered data to be processes.

 

I have a single Maxtor 160GB SATA drive 6Y160MO and an array of two Maxtor 160GB 6Y160P0 drives. Simple HD Tach benchmarks show the SATA with a burst of 120MB/sec and a read of 48.8MB/sec. The array has a burst of 105MB/sec and a read of 94.8MB/sec. These drives have the same basic physical properties with the only real difference being the interface.

 

Here are a few real world examples.

 

When encoding a one hour video file my processor runs at 100% utilization. It takes the same amount of time to encode the file whether I use a single SATA drive or an IDE RAID-0 array.

 

Half Life 2 loads almost three seconds faster when installed on the IDE RAID-0 array than the single SATA drive.

 

Windows OS install takes about 28 minutes on the single SATA versus 18 minutes on the RAID-0 array. Of course how often do you install the OS.

 

Battlefield 1942 Desert Combat loads in 22 seconds from the SATA and 14 seconds from the array.

 

Far Cry loads 4 seconds faster from the array than from the SATA drive.

 

Windows loads are a little different. From first drive access to the desktop the SATA drive is 20 seconds. It takes an additional 40 seconds until the last module is loaded. The array loads the desktop in 18 seconds with an additional 20 seconds until the last module is loaded.

 

I believe this is a fair test of the performance of a raid array since the drives and their interface are so evenly matched. The only way to improve the test would be to use a pair of SATA drives for both single and array test but I am constrained by my budget. If anyone would like to donate a pair of Raptors for testing I would be glad to test again.

 

That being said, there is nothing like a pair of Raptors in a RAID-0 array due to their spindle speed increasing access performance.

 

ExRoadie

Share this post


Link to post
Share on other sites

oliwek, you're welcome.

 

Just ordered two Hitachi 80GB SATA II drives from the egg. Looking forward to a mobo upgrade so I figured I'd get the SATA II drives now even though they aren't supported by the nF3 chipset.

 

I'll benchmark as single and RAID-0 to get some current data.

Share this post


Link to post
Share on other sites

from the little testing i could do on the Ultra-D (having to keep pulling it down or kill the OS to reload it for customer etc) those inexpensive SATA II Hitachi drives are not really that impressive...

 

no better than my 36GB Raptors (might possibly be worse than two 36GB raptors on the NF3 since the NF3 doesnt have NCQ/SATA II support). That is probably the sole reason the 2x SATA II Hitachis perform to the Raptor level on my NF4...

 

but i can guarantee if i plugged in either the 74GB Raptors or even better the new super fast 16MB cache Maxtors I could really be rocking.

 

but

 

RAID-0 is good no matter what flavor (unless you using say 40GB drives with 2MB cache and 5400RPM each haha)

Share this post


Link to post
Share on other sites

I'm gonna see if I find these sata2 in Europe already, all my 6 last hdd were ibm-hitachi since the 120gxp. One failed, but it was on a K8 board without pci-lock, not a good thing. They are fast, silent... and cheap here in Belgium ; no better choice (except raptors of course)

Share this post


Link to post
Share on other sites

Cheap is the operative word.

 

I ordered them because I got some CRC errors running spinrite on my main Maxtor drive this weekend. I wasn't able to generate a code using PowerMax since they still don't support nF3 or nF4 chipsets.

 

I got two 80GB 7200RPM 8MB drives for less than the cost of one 36GB Raptor. I know I'm giving up the rotation speed of the Raptor but my needs and wants are a lot less than the Raptor premium.

 

It should be a fun exercise even if the nF3 chipset doesn't support SATA II but I'll be ready for the nF4 chipset.

Share this post


Link to post
Share on other sites

from the little testing i could do on the Ultra-D (having to keep pulling it down or kill the OS to reload it for customer etc) those inexpensive SATA II Hitachi drives are not really that impressive...

 

no better than my 36GB Raptors (might possibly be worse than two 36GB raptors on the NF3 since the NF3 doesnt have NCQ/SATA II support). That is probably the sole reason the 2x SATA II Hitachis perform to the Raptor level on my NF4...

 

but i can guarantee if i plugged in either the 74GB Raptors or even better the new super fast 16MB cache Maxtors I could really be rocking.

 

but

 

RAID-0 is good no matter what flavor (unless you using say 40GB drives with 2MB cache and 5400RPM each haha)

 

hem... I'm gonna check the prices of the small raptors, too (but two drives to get 70gB data :sad: ) ... and try to find a review with those maxtor, thanks for the suggestion

Share this post


Link to post
Share on other sites

100€ for the raptor 36gb

170€ for the raptor 74gb

 

so 110/190$ here in Europe

 

well after all, why not? are they really noisy? (if you use a vantec tornado in your oc'ed rig, or 8 fans, you don't care, I know :D )weren't they about to release a new model BTW?

 

edit :when I said hitachi = cheap and fast, it was comparing to others 7200 8mb, not to raptors :)

Share this post


Link to post
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now

×
×
  • Create New...