[NEEDS UPDATING SEE MY POST ON PAGE 2 -- David]**Raptors: Game Load-Up Results!**

[Vio1]

[NEEDS UPDATING SEE MY POST ON PAGE 2 -- David]**Raptors: Game Load-Up Results!**

Hello,
I took it upon myself to once and for all figure out if changing the stripe size in raid-0 would change the real-world load-up of games. So for all the people wondering what stripe size to set: LOOK NO FUTHER.

I also decided to see how one drive (1x36gb and 1x74GB) stacks up against raid-0...


Here is the background info:

Games used:
1)UT2004
2)Ground Control 2
3)Lock On
4)Far Cry
5)Unreal 2

Benchmark Utilities used:
1)Sisoftware Sandra
2)PCMark04
3)PCMark 2002
4)HD Tach 2.70

System used:
As discribed in my sig except for the cpu speed (i have it oc to 3.22ghz because its summer here and too hot outside to oc to 3.6ghz)

Procedure used:
I installed a fresh copy of windows Xp pro onto my raid-0 raptors and installed all the critical patches for windows. I installed the newest drivers for all my system components. I then installed all the games and benchmark utilities mentioned above. I then defraged my HDD and used Norton Ghost 2003 to make an image of the drive and copied that to a separate HDD.

I then proceeded to time the load up for the games and programs listed above (restarting the computer between each game/program).

When that was done I restarted the computer and deleted the raid setup and created a new one with the next stripe size up... then copied the image created in Norton Ghost over to the clean HDD and restarted the whole process of timing.

I did this over and over until I tested all the stripe sizes available to me, and then I tried a single HDD.

Update: The first and last tests (16k and single 74GB respectively) were not tested using Norton Ghost, but instead were tested on freshly installed windows XP and freshly installed games/apps.

What did I time exactly?
UT2004 - I timed the load up of the largest map size (DM-1on1-Desolation)
Ground Control 2 - I timed the load up of the tutorial 1 after the video
Lock On - I timed the load up of the F-15C Instant flight
Far Cry- I timed the load up of the first level after the video
Unreal 2- I timed the load up of the first level after the video

Sisoftware Sandra - I timed how long it took the HDD to complete the File System Benchmark
PCMark04 - I timed the HDD test suite
PCMark 2002- I timed the HDD test

Please keep in note: I used a hand-held stop-watch to do these timings... therefore there is a slight margin of error...

AND NOW................. THE RESULTS!

 

[Vio1]

16k Stipe size

UT2004......................12.07s
GC 2..........................11.31s
Lock On.....................30.26s
Far Cry....................1.34.96s
Unreal 2.....................16.92s

Sisoftware Sandra.......49s
PCMark04...............3.30.74s
PCMark 2002..............23s

HD Tach 2.70............8.7ms seek time
max 119.5mb/s
min 63.9mb/s
avg 96.7mb/s

32k Stipe size

UT2004......................13.83s
GC 2..........................11.78s
Lock On.....................26.82s
Far Cry....................1.29.36s
Unreal 2.....................16.46s

Sisoftware Sandra.......45.68s
PCMark04...............3.30.85s
PCMark 2002..............21.27s

HD Tach 2.70............8.8ms seek time
max 119.4mb/s
min 28.5mb/s
avg 59.3mb/s

64k Stipe size

UT2004......................12.12s
GC 2..........................11.81s
Lock On.....................25.96s
Far Cry....................1.23.70s
Unreal 2.....................15.90s

Sisoftware Sandra.......47.17s
PCMark04...............3.42.25s
PCMark 2002..............20.78s

HD Tach 2.70............9.1ms seek time
max 99.9mb/s
min 32mb/s
avg 61.6mb/s

128k Stipe size

UT2004......................12.64s
GC 2..........................12.22s
Lock On.....................25.82s
Far Cry....................1.27.03s
Unreal 2.....................16.29s

Sisoftware Sandra.......46.72s
PCMark04...............3.42.91s
PCMark 2002..............24.92s

HD Tach 2.70............8.9ms seek time
max 91.6mb/s
min 41.4mb/s
avg 59.3mb/s

256k Stipe size

UT2004......................13.65s
GC 2..........................12.08s
Lock On.....................26.57s
Far Cry....................1.25.98s
Unreal 2.....................16.19s

Sisoftware Sandra.......54.88s
PCMark04...............3.33.15s
PCMark 2002..............25.03s

HD Tach 2.70............8.8ms seek time
max 87mb/s
min 34.9mb/s
avg 53.4mb/s

Single 36Gb Raptor

UT2004......................14.32s
GC 2..........................12.37s
Lock On.....................28.16s
Far Cry....................1.30.42s
Unreal 2.....................16.44s

Sisoftware Sandra.......1.21.39s
PCMark04..................3.55.34s
PCMark 2002..............35s

HD Tach 2.70............8.5ms seek time
max 60mb/s
min 36.4mb/s
avg 50.1mb/s



UPDATED RESULTS:



Single 74Gb Raptor

UT2004......................11.56s
GC 2..........................10.82s
Lock On.....................29.10s
Far Cry......................59.12s
Unreal 2.....................15.11s

Sisoftware Sandra.......1.12.91s
PCMark04..................3.56.73s
PCMark 2002..............34s

HD Tach 2.70............7.6ms seek time
max 73.4mb/s
min 38.1mb/s
avg 65.3mb/s

 

[LtBlue14]

omg, i clicked before you made the second post, and was like WHAT ARE THE RESULTS???

wow, sure doesn't make much of a difference does it? 64kb stripe size also seems to be the best/most well-rounded, i think. RAID doesn't make the 50% improvement that people probably hope for with it

 

[Vio1]

Conclusion:

I noticed no change in load up times across the board in the games ( except for Far Cry). Stipe size made no real difference in how long it took to load the levels.
As to how Raid-0 stacks up against a single raptor... the only difference is in the benchmark scores... other then that.... in real world use... there is NO REAL IMPROVEMENT in load up times.

 

[madcow235]

Well great now i only have to buy 1

 

[Randyman...]

Looks like you got an average of 2 seconds slower (about a 15% increase in time) from a RAID0 RAP to a Single RAP on the UT2004. Does NOT look like $150 worth of improvement, does it?

People using RAID0 for DVD transfers to different disks, and like others mentioned - transferring LOTS of data from drive to drive - seems to benefit more than these tests would indicate. I still don't need it. My single RAP is plenty fast for me, and MORE reliable than a RAP RAID0!

That was a LOT of work - We appriciate your difinitive results!

 

[attack]

Awesome, very informative! Thanks for taking the time to do this and posting the results in an organized fashion.

Good to know I only need to buy one 74gb raptor when I upgrade

 

[Busty St. Clair]

i suspected that there wouldn't be much of a difference with raid and i'm right. nothing noticable. nice work.

 

[Shuzzy]

well if your gonna do that you might as well just buy 2 36 raps, and put em in raid 0. take all the little performance benefit you can for your money.

 

[Vio1]

I think 1 74GB raptor would make more sense cause I noticed the improvement in the access time drop (when I switched from raid-0)... windows seems zipper... and the 74GB raptor has an even better access time!

 

[Randyman...]

Maybe this will be different once the "Native SATA" drives are readily avalible and affordable. Correct me if I am wrong (I have been wrong lots lately!), but isn't the Raptor still using ATA133 with an on-board SATA converter?

I just could NOT resist the urge to use the RAID capability of my MoBo. It was short lived for me, too!

I still think the $111 was worth having a DEDICATED 36Gig OS drive that SCREAMS like crazy! Makes B/U OS imaging a peice of cake, too!

 

[Vio1]

The only thing Raid-0 is good for is a higher benchmark result and applications that require higher bandwidth. No real world actual performance was evident for games.

 

[Docta_Z]

there is NO REAL IMPROVEMENT in load up times.

This is also proven on anandtech, but it's nice to see more proof for it.

Good, I never thought raid 0 was that effective anyhow.

Great work.

 

[NoteUser]

RAID controllers have an impact.

 

[djs488]

how bout on a 7200rpm 8mb drive? what are the advantages going to raptor? (retorical)

 

[Mr.Guvernment]

ncie work, spread this info around!

front page article?

 

[Vio1]

How do you make it a front page article?

 

[a c i d.f l y]

I love how everybody disses on RAID 0 like there is no performance gain. Every motherboard RAID controller runs through software drivers -- thus making performance gain merely a drift of the imagination -- limited to CPU.

Stick an Adaptec hardware RAID controller on there -- then test your results. (Granted, not many of us want to spend $299 + cost of a mobo with a 64bit pci slot).

ROMB (RAID on the Motherboard) is stupid unless it is being controlled with hardware -- not the OS -- with the exception of RAID 1, because there's no performance improvement regardless of what is controlling it.

-Frank

 

[Drec]

Vio1...im sure it must have taken atleast a few hours to do all this..and we all thank you, im sure this will effect many peoples purchasing decisions.

*gives Vio1 a pat on the back*
yeah you should get this to be an artical on oc.com

 

[gustav]

How do you make it a front page article?

Email Ed or Joe

Vio1...im sure it must have taken atleast a few hours to do all this..and we all thank you, im sure this will effect many peoples purchasing decisions.

*gives Vio1 a pat on the back*
yeah you should get this to be an artical on oc.com

hmmm, this should be an article on oc.com?