Post your hard drive benchmarks!
- Thread starter Scoobie Dave
- Start date
Soldato
does that look ok for a Raptor?
http://img404.imageshack.us/my.php?image=hdtachit0.jpg
http://img404.imageshack.us/my.php?image=hdtachit0.jpg
Been a bit of a learning curve setting up raid, but its getting there nicely.
It seems making sure 'write back caching - ON' makes a difference.
Also disabling my usb flash drive using readyboost helped get a better curve and slightly more speed.
162MB/s average
It seems making sure 'write back caching - ON' makes a difference.
Also disabling my usb flash drive using readyboost helped get a better curve and slightly more speed.
162MB/s average
Caporegime
- Joined
- 18 Oct 2002
- Posts
- 33,188
scoobie dave, what drives are those? on my 975x system with 2x, errm, 320gb seagate 10's 128kb stripe and i get 135mb sustained which i thought was pretty sweet. according to intel matrix its running 3.aac firmware.
now the questions i guess, i need to replace drives in main rig which are much slower about 95mb/s raid 0 80gb sata 2 hitachi's i got cheap a long while back.
so at first i was thinking the hitachi's 1tb, or more like the 750gb version of the new drives, but are anandtech results fairly standard, is 140-145mb sustained in raid 0 about all you'd get out of them. so not noticably better than the barra 10's with the right firmware.
which are the bad firmwares and why the heck do seagate constantly have drives with awful firmwares in them for raid 0? i remember back in the day ibm's won because their raid firmware was second to non, WD weren't great, maxtor were an easy second and seagate were so horribly bad they needed to be sent off to have a new firmware flashed for raid setups.
has anyone tested 7200.11's in single/raid and gotten anywhere near say 100mb single, 180-200mb raid 0? currently my 7200.10's are giving me very close to double the performance, and i've generally gotten around double performance by buying the best raid drives around at any given time.
so why are people hitting only 140-160 sustained with the 7200.11's. or is it simply a case that they can't get anywhere near the claimed 105mb sustained single drive in the first place?
basically trying to decide, higher price on the 7200.10 250gb with the .11 platter, wait for the 500gb's .11's(any timeframe?), WD never prefered and not seeing noticably better results on their raid drives over my old 7200.10's so can't see the point there.
now the questions i guess, i need to replace drives in main rig which are much slower about 95mb/s raid 0 80gb sata 2 hitachi's i got cheap a long while back.
so at first i was thinking the hitachi's 1tb, or more like the 750gb version of the new drives, but are anandtech results fairly standard, is 140-145mb sustained in raid 0 about all you'd get out of them. so not noticably better than the barra 10's with the right firmware.
which are the bad firmwares and why the heck do seagate constantly have drives with awful firmwares in them for raid 0? i remember back in the day ibm's won because their raid firmware was second to non, WD weren't great, maxtor were an easy second and seagate were so horribly bad they needed to be sent off to have a new firmware flashed for raid setups.
has anyone tested 7200.11's in single/raid and gotten anywhere near say 100mb single, 180-200mb raid 0? currently my 7200.10's are giving me very close to double the performance, and i've generally gotten around double performance by buying the best raid drives around at any given time.
so why are people hitting only 140-160 sustained with the 7200.11's. or is it simply a case that they can't get anywhere near the claimed 105mb sustained single drive in the first place?
basically trying to decide, higher price on the 7200.10 250gb with the .11 platter, wait for the 500gb's .11's(any timeframe?), WD never prefered and not seeing noticably better results on their raid drives over my old 7200.10's so can't see the point there.
Soldato
Got my machine set up today. Running X64 Vista and 4 x Western Digital Caviar RE2 500gb.
Not sure if Sata II is enabled, how would I check?
Any suggestions for setting this raid up better?
Not sure if Sata II is enabled, how would I check?
Any suggestions for setting this raid up better?
Last edited:
Permabanned
You hit 240 so its is over 150 so SATA 300 seems ok (SATA II is not SATA 300 BTW), and use HD-TACH anyhow, more accurate.
Soldato
HD-Tach in windows XP SP2 compatibility mode
Last edited:
Caporegime
- Joined
- 18 Oct 2002
- Posts
- 33,188
hdtach runs in compatibility mode for xp just fine. as for the over 150mb's, its 4 drives, over 4 ports, so over 150mb's wouldn't make a difference would it, as each drive contributing fairly there is doing less than 50mb each.
are we hitting controller maximum's here then? or are the RE2 last gen ones and not as good or something, i only tried a 4x raid 0 a long time ago and still saw a fairly scaled performance. if one of those drives could do say 60-70mb average then that seems really low for 4 drives.
just noticed it was 190 average, and 240 max, still seems kinda low, with 2 7200.10's giving me 135mb average.
intel raid matrix storage dodah array jobby hard drive access terminal program (meh, its got some of those words in) should be able to tell you if its running in sata 1 or 2.
are we hitting controller maximum's here then? or are the RE2 last gen ones and not as good or something, i only tried a 4x raid 0 a long time ago and still saw a fairly scaled performance. if one of those drives could do say 60-70mb average then that seems really low for 4 drives.
just noticed it was 190 average, and 240 max, still seems kinda low, with 2 7200.10's giving me 135mb average.
intel raid matrix storage dodah array jobby hard drive access terminal program (meh, its got some of those words in) should be able to tell you if its running in sata 1 or 2.
Last edited:
Permabanned
Yes why not simply use XP-SP2 mode 
What have you got running in background, Jesus thats Mount Everest lol.
I get a smooth degrading curve.
What have you got running in background, Jesus thats Mount Everest lol.
I get a smooth degrading curve.
Soldato
ok, just redid my last post changed to XP SP2 and the longer test. Seems to be a little better, is this about right then. I'm a RAID n00b.
Permabanned
Do the Short one so peaks are not as close, Im curious and its also harder on modern PC as so many auto Processes running in background, AV can decide to scan etc lol.
Soldato
Last edited:
Permabanned
Your speeds are up and down aint they ?, 190 then 230 then 199 
Soldato
I know, what's all that about then??
Does that help anything?
Does that help anything?
Last edited:
Man of Honour
Try the Long Benchmark in HDTach, the short bench shown above can give those big swings sometimes. I guess it doesn't take as many data points along the graph or something.
Soldato
Man of Honour
Sorry, didn't realise you'd already posted the long benchmark. It's a wee bit peaky but as previously noted that could be down to other processes accessing the disks at the same time as the benchmark was running.
I wouldn't complain too loudly though, your minimum speed is still above the maximum achieved by some people's arrays - mine included
I wouldn't complain too loudly though, your minimum speed is still above the maximum achieved by some people's arrays - mine included
Soldato
okey, as long as there's nothing wrong with it, that was my main concern in case I'd done a wrong setting or not used some switch somewhere. I have to say, it is pretty damn nippy in Vista. Especially after using a single 5400 laptop drive.
Associate
Raid 0 of 2x 7200.10 160gb.
Thats after I dropped one of them 1m onto a hardwood floor also
Thats after I dropped one of them 1m onto a hardwood floor also
This is the one with the long benchmark
That is outrageously quick - I know the new intel controllers are good, but 230mb/s average from 4x500Gb WDs? Awesome - I guess my scsi days really are nearing their end. Any idea why these are so much quicker than dedicated Pci-e controller based solutions - I know I have an advantage for raids other than 1 and 0, but still - hardly seems worth it if the onboards are this good
