Hard Disk Speed: Which Is Fast?

[Guest Reimar]

Just had played a bit with Windows 7 build 6956 and have some, maybe for someone interesting results about speed of Hard Disk's!

First HDD I was using was an WD 320 GB 7200 rpm Scorpio SATA 2;

Second a Seagate 320 GB 7200 rpm 7200.10 Barracuda SATA 2;

Third a Hitachi 320 GB 7200 rpm SATA 2!

The peformance Index for the HDD in Windows 7, which has a range from 1.0 - 7.9 shows for :

First WD = 5.6

second Seagate = 3.0

Third Hitachi = 6.0

The time for to install Windows 7 on each HDD was different from 22 min (WD and Hitachi) and 36 min (Seagate).

Even the boot time for the Seagate is more slow than for WD and Hitachi.

But that results I was known before from cloning of HDD's where the Seagate's the most slow HDD's!

Cheers.

[zzdocxx]

What are the determinants of this type of performance?

Just curious.

[Guest Reimar]

What are the determinants of this type of performance?

Just curious.

Read and write access times and speed of transfer from and to HDD.

Cheers.

[dave_boo]

Just had played a bit with Windows 7 build 6956 and have some, maybe for someone interesting results about speed of Hard Disk's!

First HDD I was using was an WD 320 GB 7200 rpm Scorpio SATA 2;

Second a Seagate 320 GB 7200 rpm 7200.10 Barracuda SATA 2;

Third a Hitachi 320 GB 7200 rpm SATA 2!

The peformance Index for the HDD in Windows 7, which has a range from 1.0 - 7.9 shows for :

First WD = 5.6

second Seagate = 3.0

Third Hitachi = 6.0

The time for to install Windows 7 on each HDD was different from 22 min (WD and Hitachi) and 36 min (Seagate).

Even the boot time for the Seagate is more slow than for WD and Hitachi.

But that results I was known before from cloning of HDD's where the Seagate's the most slow HDD's!

Cheers.

I'm confused; it's my understanding that the Scorpio is a laptop drive that ran at only 5400 RPM. It is a single platter though....

Were you talking about the Caviar? There are single platter versions of that too.

The Seagate and Hitachi are dual platter designs.

Obviously the single platter design is going to smoke the dual platter design due to greater areal density. Can't say what generation your drives are (with the exception of the Seagate which is obviously two generations old), are you comparing same generation hard drives? Also, did you remove the SATA II restriction jumper (forces SATA I for compatability) from your Seagate? Doing so should double your burst speed, resulting in a much higher Windows Index score. Won't do anything for large transfers though.

I've always bought Seagates (with the exception of my Raptor array). This has included everything from PATA to wide SCSI to ultra SCSI to SATA. If I could justify it I would get an array of their SAS drives, the Savio 15k.

It's kind of a moot point though. The strides that the SSDs are making are going to obslete the platter based drives shortly. Granted they'll stick around as bulk storage, but reliability will be their defining factor along with capacity. And that's where Seagate wins; reliability. I've never had a Seagate drive fail, in stark contrast to other brands (**cough Maxtor cough**).

[Tywais]

I agree that something seems wrong with those number on the Seagate and the throttle (150) jumper may still be in place. Also the .10 series are a bit dated and the .11 series have significant improvements. If we are talking about desktop (5.25) drives you can see in these reviews the Seagate 7200.11 1TB stands out from the Hitachi and WD.

http://www.barefeats.com/hard94.html

http://www.reghardware.co.uk/2008/10/28/re...cude/page4.html

http://www.anandtech.com/storage/showdoc.aspx?i=3161&p=4

http://www.bjorn3d.com/read.php?cID=1180&pageID=4060

http://hothardware.com/Articles/Seagate-Ba...011-1TB/?page=8

http://www.tweaktown.com/reviews/1261/7/se...rive/index.html

[zzdocxx]

What are the determinants of this type of performance?

Just curious.

Read and write access times and speed of transfer from and to HDD.

Cheers.

OK but what determines those?

Feel free to ignore this question, I am not an engineer and somewhat of a noob with computers.

It is just that the marketing seems to be that if the RPMs are faster, that is the main thing that makes the HD faster. But that's a big difference in the numbers you got.

Hey thanks for all your work on this forum BTW.

[surface]

The numbers don't match what's in the Storage Review database. Since you gave no model numbers, I couldn't include the Hitachi drive but was able to figure out the WD and Seagate drives.

http://www.storagereview.com/php/benchmark...65&devCnt=2

[dave_boo]

What are the determinants of this type of performance?

Just curious.

Read and write access times and speed of transfer from and to HDD.

Cheers.

OK but what determines those?

Feel free to ignore this question, I am not an engineer and somewhat of a noob with computers.

It is just that the marketing seems to be that if the RPMs are faster, that is the main thing that makes the HD faster. But that's a big difference in the numbers you got.

Hey thanks for all your work on this forum BTW.

Well, actually now-a-days, there's a new way of recording to the hdd. It's called perpendicular recording. What it does is stuff more information into the same area. Since the information is so close together the drive can access it faster. Think of a filing cabinent. If the files are packed in tightly in the various drawers, you could open one drawer and skim the contents much more quickly than you could open one lightly filled drawer, skim the contents and then close that drawer, open another and skim the contents. There's two ways of increasing your access times, increasing the density, the above example, and increasing the speed of the platters. At 3,5", 10k rpm is essentially the maximum speed that the platters can live at due to the stress put on them from their own weight. That's why when SCSI was king (which I still believe it is!), if you wanted capacity you went with 10k rpm. The 15k drives actually used 2,5" platters inside! This was reflected in the much smaller capacity.

Speed of transfer from and to the hdd can be affected many ways. Easiest way for burst speeds or testing programs that sample really small sets is to use small samples. That then fits in the buffer giving artifically inflated examples. There are programs out there that allow you to set parameters so that it actually test the drive rather than your cpu's cache, memory subsystem, and the drive's cache. I highly recommend using IOZone, but if you've got a crazy setup (I have dual Opterons and 12 GB of RAM) it could take quite some time for the test to run since you want to increase the size to be greater than your memory+1 GB(more or less). Following is an example of what I'm talking about; notice that once it gets past your buffer cache the readings drop down to real life performance numbers.

Surface, in his defense you're looking at the 1TB drive. Its areal density is much higher.

[Guest Reimar]

Just had played a bit with Windows 7 build 6956 and have some, maybe for someone interesting results about speed of Hard Disk's!

First HDD I was using was an WD 320 GB 7200 rpm Scorpio SATA 2;

Second a Seagate 320 GB 7200 rpm 7200.10 Barracuda SATA 2;

Third a Hitachi 320 GB 7200 rpm SATA 2!

The peformance Index for the HDD in Windows 7, which has a range from 1.0 - 7.9 shows for :

First WD = 5.6

second Seagate = 3.0

Third Hitachi = 6.0

The time for to install Windows 7 on each HDD was different from 22 min (WD and Hitachi) and 36 min (Seagate).

Even the boot time for the Seagate is more slow than for WD and Hitachi.

But that results I was known before from cloning of HDD's where the Seagate's the most slow HDD's!

Cheers.

I'm confused; it's my understanding that the Scorpio is a laptop drive that ran at only 5400 RPM. It is a single platter though....

Were you talking about the Caviar? There are single platter versions of that too.

The Seagate and Hitachi are dual platter designs.

Obviously the single platter design is going to smoke the dual platter design due to greater areal density. Can't say what generation your drives are (with the exception of the Seagate which is obviously two generations old), are you comparing same generation hard drives? Also, did you remove the SATA II restriction jumper (forces SATA I for compatability) from your Seagate? Doing so should double your burst speed, resulting in a much higher Windows Index score. Won't do anything for large transfers though.

I've always bought Seagates (with the exception of my Raptor array). This has included everything from PATA to wide SCSI to ultra SCSI to SATA. If I could justify it I would get an array of their SAS drives, the Savio 15k.

It's kind of a moot point though. The strides that the SSDs are making are going to obslete the platter based drives shortly. Granted they'll stick around as bulk storage, but reliability will be their defining factor along with capacity. And that's where Seagate wins; reliability. I've never had a Seagate drive fail, in stark contrast to other brands (**cough Maxtor cough**).

Dave you're right, my mistake. The Hitachi and the WD HDD's were in the case. I just was take them out and here the type:

Caviar Blue WD3200AAKS

Deskstar P7K500

Just for to clear, the performance index from Windows 7 was checked in the same computer after installing of Win7 and install all drivers. The whole setup was exactly the same differs only with the used HDD while only one HDD was used and the others disconnected.

The Hitachi and the WD are quite new HDD's, just 1-2 month old while the Seagate is 7 month old.

The Caviar Blue is an double platter same as the Hitachi and the Seagate.

The jumper on the Seagate was removed and it's running with 3 GBps.

Cheers.

[dave_boo]

Dave you're right, my mistake. The Hitachi and the WD HDD's were in the case. I just was take them out and here the type:

Caviar Blue WD3200AAKS

Deskstar P7K500

Ok, thanks for the clarification. I guess I'm not totally nuts.

Just for to clear, the performance index from Windows 7 was checked in the same computer after installing of Win7 and install all drivers. The whole setup was exactly the same differs only with the used HDD while only one HDD was used and the others disconnected.

That's what I understood as far as your testing procedures being.

The Hitachi and the WD are quite new HDD's, just 1-2 month old while the Seagate is 7 month old.

It was bought 7 months ago....the design is quite old (in tech terms).

The Caviar Blue is an double platter same as the Hitachi and the Seagate.

Thanks for clearing that up. Removes some of my argument why my beloved Seagate got its butt kicked.

The jumper on the Seagate was removed and it's running with 3 GBps.

Ouch. And it still did that poorly? Wonder if there's some issue with Seagate's firmware and your controller's firmware. Wouldn't be the first time, after all consider the WD (right?) and Nvidia bug from not that long ago.

Cheers.

[dave_boo]

Oh, and quickly;

How does Windows sever (realising it's still Beta) compare in speed? Is it up to XP or somewhere between it and Vista?

[Guest Reimar]

Oh, and quickly;

How does Windows sever (realising it's still Beta) compare in speed? Is it up to XP or somewhere between it and Vista?

Running on base of Vista SP1/Server 2008 SP1. But looks like the had changed the counter a bit from 1-5.9 to 1-7.9!

But the speed is really different between the Seagate and the other two HDD's. Booting time is nearly double and open an 370 MB Excel worksheet needs also double time!

ill install on all HDD's the same Applications as well and cross check. but that will need a few days because I have not much time in the moment. It's year-end and I have to do the required work for the Gov!

Cheers.

[dave_boo]

Oh, and quickly;

How does Windows sever (realising it's still Beta) compare in speed? Is it up to XP or somewhere between it and Vista?

Running on base of Vista SP1/Server 2008 SP1. But looks like the had changed the counter a bit from 1-5.9 to 1-7.9!

But the speed is really different between the Seagate and the other two HDD's. Booting time is nearly double and open an 370 MB Excel worksheet needs also double time!

ill install on all HDD's the same Applications as well and cross check. but that will need a few days because I have not much time in the moment. It's year-end and I have to do the required work for the Gov!

Cheers.

I'm sorry that was a poorly worded question from my side. I was trying to get a feel for the speed of the Operating System overall compared to XP and Vista. Supposedly it is quick enough to run on Netbooks, something that Vista can only dream about. But is the OS quick enough overall to give XP a run for its money?

[Guest Reimar]

Oh, and quickly;

How does Windows sever (realising it's still Beta) compare in speed? Is it up to XP or somewhere between it and Vista?

Running on base of Vista SP1/Server 2008 SP1. But looks like the had changed the counter a bit from 1-5.9 to 1-7.9!

But the speed is really different between the Seagate and the other two HDD's. Booting time is nearly double and open an 370 MB Excel worksheet needs also double time!

ill install on all HDD's the same Applications as well and cross check. but that will need a few days because I have not much time in the moment. It's year-end and I have to do the required work for the Gov!

Cheers.

I'm sorry that was a poorly worded question from my side. I was trying to get a feel for the speed of the Operating System overall compared to XP and Vista. Supposedly it is quick enough to run on Netbooks, something that Vista can only dream about. But is the OS quick enough overall to give XP a run for its money?

I would say that the speed of Win 7 and Server 2008 is similar but Vista is more slow. But that all didn't tell anything because until now the OS Kernel is the one from Server 2008 and until that is changed to it's own, it would be difficult to say anything!

Win 7 even isn't in Beta stage right now, just M3 but for that the OS works very well.

Cheers.

[zzdocxx]

This is an interesting discussion, though alot of it is over my head. Like the part where R. took the HDs out of their cases and posted the names and numbers.

OK I'm going to keep my mouth shut now.

One more thing though, Win 7 is supposed to be the next version? As far as Vista is concerned, it seems like half the people swear by it, and the other half swear at it.

I'm using XP but of course am curious about what's next.

Alright that was a little off topic.

[Tywais]

Supposedly it is quick enough to run on Netbooks, something that Vista can only dream about.

I'm running Vista Ultimate on my MSI Wind with no problems. Initially installed XP but thought, after reading a couple of reviews of experiences with Vista on netbooks, to try Vista. I find that Vista is at least as fast if not more so then my experience with XP on the netbook. Startup and shutdown included. Of course I made sure to have the latest Vista drivers for it to be sure it is optimum plus 2GB of memory.

I may have to look at the 2.5" 7200 drives that Reimer experimented with to boost its performance more. I'm a tweaker.

[Guest Reimar]

Supposedly it is quick enough to run on Netbooks, something that Vista can only dream about.

I'm running Vista Ultimate on my MSI Wind with no problems. Initially installed XP but thought, after reading a couple of reviews of experiences with Vista on netbooks, to try Vista. I find that Vista is at least as fast if not more so then my experience with XP on the netbook. Startup and shutdown included. Of course I made sure to have the latest Vista drivers for it to be sure it is optimum plus 2GB of memory.

I may have to look at the 2.5" 7200 drives that Reimer experimented with to boost its performance more. I'm a tweaker.

You may should try the Win 7 Netbook version?! That version should run faster than Vista and XP anyway. I use the full version and it is quite stable, which means I havn't found some errors yet. And I really like it and wait for the next "release", the running one is build 6956 now.

Cheers.

[Tywais]

You may should try the Win 7 Netbook version?! That version should run faster than Vista and XP anyway. I use the full version and it is quite stable, which means I havn't found some errors yet. And I really like it and wait for the next "release", the running one is build 6956 now.

That does have my interest and will look further into that. That's one of the things I use the Wind for, trying different things on it to "tune" it.

[dave_boo]

Supposedly it is quick enough to run on Netbooks, something that Vista can only dream about.

I'm running Vista Ultimate on my MSI Wind with no problems. Initially installed XP but thought, after reading a couple of reviews of experiences with Vista on netbooks, to try Vista. I find that Vista is at least as fast if not more so then my experience with XP on the netbook. Startup and shutdown included. Of course I made sure to have the latest Vista drivers for it to be sure it is optimum plus 2GB of memory.

I may have to look at the 2.5" 7200 drives that Reimer experimented with to boost its performance more. I'm a tweaker.

You may should try the Win 7 Netbook version?! That version should run faster than Vista and XP anyway. I use the full version and it is quite stable, which means I havn't found some errors yet. And I really like it and wait for the next "release", the running one is build 6956 now.

Cheers.

They actually plan a Netbook version? That must be the version I was reading about before. Also, earlier I had written Windows Seve® but meant to put Seven.