PC Hardware forum

General discussion

Hard drive data transfer rate question...

by bestsealer / December 29, 2006 12:27 PM PST

I have a concern about the speed of a new system I recently bought. It is an Intel Core 2 Duo E6600 processor on an MSI Neo-F P965 motherboard with 2 GB or RAM. The hard drive is a Seagate 320 GB SATA3, (model ST3320620AS). This drive is supposed to have a max throughput of 3 Gbits per second.

I suspected the HD data transfer rate was not as fast as I expected it should be, and today after buying an additional drive, I did some tests. I had a ton of media files on the C: drive, and after installing and formatting the new D: drive (which incidentally, is the same model as the original C: drive) I decided to time the transfer rate by moving the media files from the C: drive to the new D: drive.

It took 124 minutes to copy 243.5 GB of data from the C: drive to the D: drive. By my calculations, that works out to about 32.73 MBytes per second, or slightly less than 300 Mbits per second - I am guestimating this by allowing 9 bits per byte to cover overhead. Still, 300 Mbits is a far cry from 3 Gbits!

When installing the new drive, I noticed there was a 'default' jumper on the drive, that limited the transfer rate to 1.5 Gbits per second. (ref: http://www.seagate.com/support/disc/sata/st3320620as.html ). I also noticed that the original drive also still had this jumper installed. I left both jumpers in place. I spoke with the dealer who built the system about this, and he said they never remove the jumper - that the actual transfer rate is controlled by the motherboard and that my motherboard was very new and would override this jumper and do a 3 Gbit transfer, not 1.5 as the jumper was set.

Now, I have a hard time believing this, especially in light of my recent data transfer test. I wrote to Seagate to inquire about this, but still have not had a reply - it is Christmas time after all, and I don't expect them to get back to me before next year. However, even if the jumper was limiting the transfer rate, and even if removing it doubled the rate to 600 Mbits, that is still a far cry from 3 Gbits.

So, my question is: am I being unrealistic in expecting a faster data transfer rate with this system?

pl

Post a reply
Discussion is locked
You are posting a reply to: Hard drive data transfer rate question...
The posting of advertisements, profanity, or personal attacks is prohibited. Please refer to our CNET Forums policies for details. All submitted content is subject to our Terms of Use.
Track this discussion and email me when there are updates

If you're asking for technical help, please be sure to include all your system info, including operating system, model number, and any other specifics related to the problem. Also please exercise your best judgment when posting in the forums--revealing personal information such as your e-mail address, telephone number, and address is not recommended.

You are reporting the following post: Hard drive data transfer rate question...
This post has been flagged and will be reviewed by our staff. Thank you for helping us maintain CNET's great community.
Sorry, there was a problem flagging this post. Please try again now or at a later time.
If you believe this post is offensive or violates the CNET Forums' Usage policies, you can report it below (this will not automatically remove the post). Once reported, our moderators will be notified and the post will be reviewed.
Collapse -
Sadly, you have an OS, maybe antivirus
by R. Proffitt Forum moderator / December 29, 2006 10:36 PM PST

And what else examining the bits along the way. You also have issues such as "WHO INSTALLED THIS OS?" since many won't fetch current or even the correct drivers.

To do speed tests we may have to examine dark details of the cache size of the drive and test the transfer rate inside that size (transfer rate will be limited when we actually read from the drive.) You also are writing to a drive and even a simple item such as NT/2000/XP timestamping files and making event logs will thwart the tests.

Sorry, I wish there was a simple test for you but I must defer to many fine speed test web pages out there. Compare with those.

Bob

Collapse -
Thanks, but...
by bestsealer / December 29, 2006 11:03 PM PST

Well, Bob, for someone who claims to be a computer 'guru', that is a rather pathetic response. But, thanks anyway for taking the time to reply.

I am not interested in formal tests that produce theoretical maximums.. I am interested in practical results. And if I am only getting approximately 1/10th the rated transfer speed for this drive, then I am questioning that. In other words, is this the expected norm for a 3 Gbit rated drive? If not, what can I do to improve it? Running a formal test will not improve my practical results.

pl

Collapse -
I dumbed it down.
by R. Proffitt Forum moderator / December 29, 2006 11:15 PM PST
In reply to: Thanks, but...

Maybe too much. My background is an electronics designer that writes from assembler to higher language code. I've been faulted in the forums for giving too technical responses. Now I try to simplify but some will flog me for that as well.

If you continue to bite the hands that want to help you, I fear you will not put together the answer.

You didn't tell me about WHO INSTALLED THE OS? Which is the key reason I see subpar performance.

Since you flamed rather than engaged in the discussion I hope you can dose the flame and put yourself into gear and see the nuggets I presented and respond so we can see if the situation or test results can be improved or not.

-> Your choice here,

Bob

Collapse -
I'm not flaming..
by bestsealer / December 30, 2006 1:28 AM PST
In reply to: I dumbed it down.

Bob,
As is the case in written language, it is often misunderstood. I was not flaming - sorry you understood it to be that way. If you re-read my reply, I did politely thank you for responding to me.

Once again, I am concerned that I am not getting optimum performance from my HD. The test I ran was rather basic and informal, albeit, a practical data transfer test. And what I saw, indicated I was getting about 1/10th the rated transfer speed... something I find hard to believe is normal. Yes, I fully agree stated rates are maximum theoretical rates usually achieved under specific controlled conditions, and I do not expect to achieve that maximum.. however, 1/10th? That is what I am questioning.. is this normal?

Regarding the OS.. I installed it (XP pro). I have been working with computers for more years than I can remember, and although I would never call myself a guru, I nonetheless do think I know what I am doing, at least far more than the average PC user. Hence my question about the rated speed and drive jumpers - a novice user who never even dream of questioning this.

The question of who installed the OS is moot IMHO.. I have seen so-called professionl dealers and vendors screw up an installation. So, if I said my dealer, what difference would it have made?

A more constructive answer would be to give me some specifics to investigate (or links to such information) that I might use to try and improve performance.. that is, if performance can be improved.. the main question once again being, is 300 Mbits per second normal for these drives in this system?

pl

Collapse -
You installed the OS. Maybe the drives too.
by R. Proffitt Forum moderator / December 30, 2006 3:10 AM PST
In reply to: I'm not flaming..

This is crucial information. You revealed you know about the drive jumpers so we'll skip that issue.

-> Tell me the exact motherboard make and model and the version of motherboard and SATA chipset drivers you installed when you installed XP. To wit, what did you put on that floppy that you feed XP via the F6 key during install.

I'm not going to duplicate web content about how to install XP on SATA systems. I'm only going to note where most gaffe the install.

Bob

Collapse -
no floppy
by bestsealer / December 30, 2006 7:37 AM PST

Bob, thanks for the reply.

I did not have to install anything special prior to XP. I have in the past with other systems, been asked to load up special drivers, but in this case I was not, and just assumed (maybe incorrectly) that XP had all the required drivers. Besides, I would need to find some other method of loading them if needed, because this system has no floppy drive.

If you think the install is incorrect, and I need to pre-load a driver for the hard drive, please advise, or point me to a source on the net. Thanks.

Regarding the jumpers.. I just received a reply from Seagate - they confirmed that the jumpers should be removed. So much for the professional advice of my dealer! I will remove them and re-run some tests over the next few days.

pl

Collapse -
Simple to find. I'll just share the words.
by R. Proffitt Forum moderator / December 30, 2006 7:49 AM PST
In reply to: no floppy

INSTALL XP TO SATA DRIVE

Try that on google.com

I don't duplicate web content.

Bob

Collapse -
thanks again..
by bestsealer / December 30, 2006 10:42 AM PST

Bob, thanks for the reply.

I think we might be going off on a tangent here. I believe the information you are trying to point me to refers to situations where Windows can not detect any Hard drives, so you had to press "F6" (3rd party option) in the Windows Startup/Setup to load up a sepecific driver for the drive. That is not the case for me. I had no problems formatting the drive or installing XP SP2. I believe XP SP2 now has all the required drivers to support SATA and large capacity drives. Although I do remember doing this a few years back, pre SP2, where I had to load drivers before XP would see the drive and install.

I will, however, keep this in mind and continue to investigate.

Thanks

pl

Collapse -
Sorry no.
by R. Proffitt Forum moderator / December 30, 2006 11:18 AM PST
In reply to: thanks again..

Without motherboard and SATA drivers... subpar performance. Maybe we expected Microsoft to do this for us?

It's something I see daily. And correct too many times to keep count.

Bob

Collapse -
Can you give me an extimate?
by bestsealer / December 30, 2006 11:34 AM PST
In reply to: Sorry no.

Ok.. can you give me an estimate of what kind of performance I should realistically expect with this SATA drive? (system specs are in my initial post).

I downloaded PassMark Performance Test.. it averages about 47 to 48 MB transfer on both read and write, on both drives. All I am asking is: is this reasonable, or am I being unrealistic in expecting more? What are typical transfer rates for a system such as mine? I mean, if this is typical, then I will drop the issue.

btw.. I removed the 1.5 Gbit limiting jumpers as suggested by Seagate - did not make one bit (pardon the pun) of difference.

pl

Collapse -
more info...
by bestsealer / December 30, 2006 12:01 PM PST

Ok, I found some more information..

the specs for this drive state:
INTERNAL TRANSFER RATE (Mbits/sec) _______up to 1030
SUSTAINED TRANSFER RATE (Mbytes/sec)______up to 72
EXTERNAL TRANSFER RATE (Mbytes/sec) ______up to 300

so, I guess the upper limit is about 72 for practical purposes. Still, about double what I am getting now. Sad

Why the heck do they rate this drive at 3 Gbits?

pl

Collapse -
Sorry I fear you don't see the problem.
by R. Proffitt Forum moderator / December 30, 2006 12:45 PM PST
In reply to: more info...

Maybe you did or didn't install the drivers. There are some that dig in their heels when faced with motherboard drivers and point out that XP installed and runs.

Maybe a car analog? Yup, they run without all the parts too.

Bob

Collapse -
A little fact. . .
by Coryphaeus / December 30, 2006 12:13 AM PST

Transfer rates are in perfect laboratory conditions. Period.

Examples that are fact:

Dial-up 56K modem. Never happen.
USB transfer rate. Never happen.
Broadband transfer rate. Never happen.
HD transfer rate. Never happen.
EPA mileage estimates. Never happen.
The list can go an and on.

All claims are overinflated so the product will sell. As Bob tried to tell you there are variables and all you could do is complain. I suggest you do a little investigating and reading, and when possible, a little live and hands on testing.

There are 10 types of people who understand binary; those that do and those that don't.
Click here to see the CNet faces, learn a little about analog and
digital data, internet connections, spyware removal, and download free software.

Collapse -
I beat "EPA mileage estimates" REALLY!! Constantly.
by R. Proffitt Forum moderator / December 30, 2006 12:24 AM PST
In reply to: A little fact. . .

Call it an ABERRATION.

Our 2003 Civic Hybrid at about 4K miles got a software update. The EPA was from memory 47/45 (city/highway.) On trips to/from Hartford and Boston of 200 miles round trip I hit 50.5 MPG more than once.

Sadly, Honda has gaffed the 2006 and 2007 Hybrid models and when the outdoor temperature is 60F or below... mileage plummets. It's been in the shop for over a week and with a lot of digging I now know why this happens. It's another software glitch from what I've figured out. And they didn't miss it by a little but by miles. Try 35 MPG vs the 49/50 sticker. As it gets colder we saw 31 MPG just recently.

I'm going to attempt a hardware fix before I post the full story.

Bottomline? Honda has lost the formula for hybrids for 2006/2007. Get the Prius.

Bob

Collapse -
RE: Transfer rates
by pcguru4u / February 9, 2007 9:32 AM PST
In reply to: A little fact. . .

*off topic* I always get a kick out of your tag line coryphaeus!
*on topic*
Burst rate. Theoretical maximum transfer rate - read "laboratory conditions."
Sustained transfer rate will always be much slower. Sectional density, rotational speed, read/write cycles and head speed, OS, cache size, temperature, power fluctuations etc. ad naseum. You get the idea. Makes for nice reading on the package but SATA2 is faster than first gen drives. Add RAID to the mix and there is another set of numbers and conditions.
Kind of like horsepower ratings. Always at the crankshaft, not the rear wheels. There is quite a bit of loss between the motor and the tires.
As for the SATA installation thing, I wasn't aware Windows would load onto a SATA drive w/o third party software (or even recognize the drive) Thank goodness for the good ole' floppy! I recently hosed an XPpro system and during re-install didn't pay attention to the F6 for third party drivers. Kept getting a "no disk recognized" error. After some hair pulling the "duh" alarm went off and one floppy later, new install. How'd you manage to bypass that? On another note, what could you do in that case w/o a floppy drive? I didn't see an option for installation from a CD during the SATA drivers install. (maybe I was crosseyed after pulling hair and missed an option)

Bottom line, 3.0Gb transfer rate is burst mode. Large scale transfers will not reach that, nor sustain that. Still a good drive.

Collapse -
Re the jumpers...if it's set to 1.5 mode...that's the limit.
by VAPCMD / December 30, 2006 12:41 AM PST

They set them that way because often they don't work with older SATA MBs when set to 3.0.

Go to http://www.storagereview.com and see what figures they report for that drive.

Might also try "HDTune V2.51" and see what their benchmark shows.

VAPCMD

Collapse -
They don't list my drive
by bestsealer / December 30, 2006 11:54 AM PST

Thanks for the pointer.

That site does not list my drive, however, there are other Seagate Barracudas there with read transfer rates between 62 and 70 MB. Unfortunately, my test with PassMark shows about 47 for my drive. As I stated earlier, my own basic transfer test (just copying files) was running about 33.. not too impressive!

Collapse -
Hard drive data transfer rate question...
by bob b / December 30, 2006 1:07 AM PST

A number of years ago I played with this sort of thing.

The test setup was.
Ata 100 HD.
Ata 66 channel.

Using a HD speed tester what I saw was.
In cache mode I could get close to 66MB's.
I suppose that told me the hardware did function.

In non-cache mode I was seeing 20-30MB's.
These are just raw speeds without windows and it's overhead getting involved.

Using windows and watching data xfer rate of the HD I'm lucky to see 5MB's.

Perhaps get yourself a HD speed tester and see if the numbers come close.

Collapse -
Finally got it working!
by bestsealer / February 9, 2007 4:18 AM PST

Thought I would stop by again with an update.

With the help of the folks over at Bleedinedge.co,, I finally managed to get the proper SATA drivers installed and this resulted in the drive hitting 73 MBytes per second.

If interested in all the tech details and performace graphs, read the thread at: http://www.bleedinedge.com/forum/showthread.php?t=27223

In a nutshell, the SATA drivers needed to be loaded prior to XP being installed. A couple of setbacks were finding exactly what drivers were needed, as the motherboard had 1 JMircon SATA port, and 4 Intel ICH8 ports (and I had two drives, one on each). I ended up not using the JMicron port. And someone gave me a pointer to where to get the required Intel drivers, since only the JMicron driver was provided with the board.

And, I made the mistake of not getting a floppy drive, since I thought they were obsolete. Mistake! The ONLY way the drivers can be loaded prior to XP, is via the A: drive floppy.. not an external floppy, CD, or USB device... so, I had to buy a floppy drive.

And finally, because I had to reload the OS, I had to reload all my software and data. A total system backup restore was useless, as it would have retored my drive to the prior setting, ie, without the SATA driver.

Been an interesting exercise in frustration.. lots of it!

pl

Popular Forums
icon
Computer Help 47,885 discussions
icon
Computer Newbies 10,322 discussions
icon
iPhones, iPods, & iPads 3,188 discussions
icon
Security 30,333 discussions
icon
TVs & Home Theaters 20,177 discussions
icon
HDTV Picture Setting 1,932 discussions
icon
Phones 15,713 discussions
icon
Windows 7 6,210 discussions
icon
Networking & Wireless 14,510 discussions

Big stars on small screens

Smosh tells CNET what it took to make it big online

Internet sensations Ian Hecox and Anthony Padilla discuss how YouTube has changed and why among all their goals, "real TV" isn't an ambition.