WD1002FAEX 1TB Slow Performing

Hi everyone,

i wanted to ask if the numbers i’m taking are correct for the specific drive.

I have partition the HDD in one C:\ 293GB (primary system) and a second D:\ 640GB.

I have connected this HDD in a sata 3 hole in the mobo (MSI P67a-GD55)and setting IDE to AHCI mode through BIOS, but i’m only taking  113-116 (and 93 in HD Tune)Sequential  read speed, as you can see below !

I have already update my bios to the latest vesrion (7681v4.30).

Does the IDE controller really matter?

Under IDE ATA-ATAPI controllers appears

Intel(R) 6 Series/C200 Series Chipset Family 2 port Serial ATA Storage Controller - 1C08

Intel(R) Desktop/Workstation/Server Express Chipset SATA AHCI Controller.

Not to mention the annoying loud seek-read-write sound that makes when accesing data !

Please tell me if i’m doing something wrong or it’s a hardware malfunction.

Thanks in advance

Lostin out


CrystalDiskMark 3.0.2 x64 (C) 2007-2013 hiyohiyo
Crystal Dew World : http://crystalmark.info/

* MB/s = 1,000,000 byte/s [SATA/300 = 300,000,000 byte/s]

Sequential Read : 113.274 MB/s
Sequential Write : 110.668 MB/s
Random Read 512KB : 40.814 MB/s
Random Write 512KB : 67.583 MB/s
Random Read 4KB (QD=1) : 0.463 MB/s [113.0 IOPS]
Random Write 4KB (QD=1) : 1.001 MB/s [244.5 IOPS]
Random Read 4KB (QD=32) : 1.107 MB/s [270.4 IOPS]
Random Write 4KB (QD=32) : 0.829 MB/s [202.3 IOPS]

Test : 1000 MB [C: 53.6% (157.0/293.0 GB)] (x5)
Date : 2013/10/22 15:33:27
OS : Windows 7 Ultimate Edition SP1 [6.1 Build 7601] (x64)

WD has a habit of fiddling with their platters and heads.

The following benchmarks appear to show drives with two 500GB platters and 4 heads:

http://www.3dnews.ru/assets/external/illustrations/2010/04/29/591174/sata-r.jpg
http://s05.radikal.ru/i178/1009/14/ec15f950fa70.jpg

Your benchmark, OTOH, appears to be consistent with a drive having three 400GB platters and 5 heads. That would result in a reduction in the transfer rate by about 10%.

I’m only guessing, though. If you have access to a faster drive with the same model number, you might like to compare their weights.

First of all thanks for your response Fzabkar.

Unfortunately i’m not in position to compare my drive with another about their weight.

I wonder though how can you assume that my drive has 3 platters and 5 heads (3/5) and not 2/4 as the company describes in here  http://rml527.blogspot.gr/2010/10/hdd-platter-database-western-digital-35_3899.html.

It’s for sure something very wrong here but i cannot find it (yet).

I’m gone proceed to a new format and recheck the unpartition drive of 1 TB to see the results.

I’m also gone try a clean windows installation because this one is by cloning and let you inform about the outcome.

Either way do you think is there a reason to RMA the drive?

I payed for more and get less for some reasons.Isn’t that true?

And what about the software controllers?

Just to let you know,

i contacted with WD support and after i send them the results from WD Diagnostics Tool (quick and extended tests) they’ve answer me that the drive is faulty and has to be returned. (SMART option failed)

I feel more relief now…

Lostin out

Let’s take two imaginary examples. Let’s say we have a drive with a single 500GB platter, and another with four 125GB platters. Since both platters are the same physical size, it is clear that the 500GB platter has 4 times as many bits per square inch (data density) as the 125GB platter.

If we assume that the only difference between the two platters is one of scale, then it is clear that the 500GB platter has twice as many tracks per inch, and twice as many bits per track, as the 125GB platter.

A drive’s transfer rate is dependent on its RPM and on the number of bits per track. Assuming that the two drives spin at the same speed, then once again it is clear that the single-platter drive will have a transfer rate that is double that of the 4-platter drive.

So the rule-of-thumb relationship appears to be …

(transfer rate A) / (transfer rate B) = sqrt(density A / density B)

This rule assumes that all other factors are equal, so it cannot be used to compare AF models with non-AF models.

In your case I’m only offering one plausible explanation. You could probably get the same benchmark result if you took a 2-platter drive and reduced the number of bits per track while increasing the number of tracks per inch. I can’t see WD doing something like that, though.

BTW, Seagate has a 2TB model which has three variants – 4 heads, 5 heads, or 6 heads. The 4-head version is a lot faster than the 6-head. Only the 4-head version was mentioned in Seagate’s Product Manual.

At Tom’s Hardware I saw a WD 500GB drive which benchmarked like a shortstroked 750GB model (two 500GB platters, 3 heads, reduced number of zones).