StorageTux

Is Western Digital trustful HDD vendor today?

Recommended Posts

For the past couple of years I was maily WD drive user. I still have two WD drives - 640GB caviar black and blue. But I have been watching the rise of WD on the HDD market (now they are on par with Seagate) and I'm starting to be annoyed with their customer treatment. There is no doubt, WD drives are the fastest drives in mainstream consumer market, but the read/write speed is not everything.

I can summarize my observations into few points:

- WD has very sophisticated model labeling system, which completely hide platter density information. If they upgrade the platters in a model of some capacity, and all other parameters remain the same, there is no way how to distinguish the old and new model, except comparing numbers of revisions with benchmarks on the web. I found out Seagate has great model numbering scheme, and it informs customer, how many platters the drive has. On the other hand Samsung does not have such a nice scheme, but they have uniform product line, (F1, F3) where drives share the same platters.

- WD lied about rotational speed of their Caviar Green drives, which supposed to be variable 5400-7200RPM , to better sell these drives, until reviews on the web proved otherwise. Marketing bullshit. Afaik Seagate sell their LP drives with constant 5900RPM information.

- WD loudly advertised their intelipark technology which parks the drive heads to reduce consumption. Their designed this feature so bad, that heads was supposed to park after 8 seconds. On Windows OS this design failure did not lead to any problems because Windows access the drive more often, and the heads did not parked anyway. But on Linux/BSD front the situation was different. These systems access drives less often, thus heads was parked every 8 second witch leads to hundreds of thousands load_cycle_count after few month of using the drive. When users explored this bug, the only way to correct it was WDidle.exe utility, because Caviar Green used some ugly non-standard commands to modify this drive behavior and standard tools like hdparm was useless. WD claimed WDidle unsupported, it it almost disappeared from the web. To get rid of annoying customers questions on support desk, WD cracked their SMART table thus load_cycle_count stopped to count head parking/unparking, although this behavior did not changed in reality. I know that, because I had plenty of these drives with lying SMART, clicking every 8 second! After I found wdidle.exe on some "underground" web and set the timer to 25.5s, the clicking effect completely disappeared.

- WD brought in shiny new models with 4kB sector technology, which is great. But, the drive firmware untruly reports it has 512B sectors. This leads to problems OS is not capable to recognize sector size correctly, use 512B as a default and start first partition on LBA63 - in case of old MS-DOS partition table, or on LBA34 - in case of GPT table which I use on all of my hard drives. Neither LBA63 or LBA34 are not the correct place to start with 4kB sector. It should be LBA64 or LBA36 in this case. As a Linux user, I can use GNU Parted and create my first partition manually on LBA i manually choose. But this is not solution, but workaround.

- WD silently remove TLER from actual revision of WD1001FALS, maybe to force RAID users to buy their RE4 series. They did not informed anybody about this, which is sad, and according to some info it seems this model still claims it is ata8 compliant, which is lie then.

No matter what drives WD produce, it seems I can not trust firmware info in WD drivers anymore. Proprietary incompatible ata commands, lies in drive info output, lies in smart table. Bullshit everywhere. Maybe it is time to buy Samsung and Seagate today. They treat customers noticeably better today. Maybe Seagate learned from their 7200.11 fiasco.

Share this post


Link to post
Share on other sites

Regarding the parking head situation, WD wasnt the only one with this sort of problem. The Seagate had a similar problem with some of their notebooks drives that were overactive with loading cycles that caused premature wear.

Regarding the TLER issue on WD consumer drives... Seagate is doing the same thing to push RAID users to the Constellation series.

I still have a hard time forgiving Seagate over their pain in the butt RMA process. With almost every other manufacturer you can do an advanced RMA with a hold on a credit card. With Seagate it is a flat rate of 20 bucks. Might not seem like much but I was burned on my 7200.2 with the first RMA being bad and I'll be damned if I have to sink any more money into the warranty process.

Share this post


Link to post
Share on other sites

Regarding the TLER issue on WD consumer drives... Seagate is doing the same thing to push RAID users to the Constellation series.

I did not noticed this, have you any additional info? I found out that 7200.12 seem to have ERC enabled.

Share this post


Link to post
Share on other sites

Regarding the numbering: WD hiding the number of platters always annoyed me. For example the old standard 40 GB 5400 rpm drive was available with anything going from 4 x 10 GB platters to a half 80 GB platter. THat's why I like WDs 640 GB drives, as for a long time you could be sure to get the densest platters there. And that's why I won't touch the 500 GB Black, even though it could be really sweet with one 500 GB platter. But you never know, except if they also change the interface and / or cache (like they did with the Black 1 TB).

No, they did not lie about the rotational speed of the Green drives. They advertised them as 5400 to 7200. 5400 is within this interval, so it's a correct statement. It's intentionally misleading, though. It's clear that they did it to make the drive appear better. Being the first of this new wave of 5400 (5900) rpm drives they were certainly worried about customer reactions.. as they hadn't seen anything slower than 7200 rpm on the desktop for years. That's not a justification, though. Just an explanation.

As far as I know the new 4kB sector size drives still emulate 512B sectors for compatibility reasons. From all HDD manufacturers.

For me performance counts, so for an OS / programs HDD I'll go with nothing less than a WD Blue. Mass storage is fine with a 'Cuda LP, though.

MrS

Share this post


Link to post
Share on other sites

WD lied about rotational speed of their Caviar Green drives, which supposed to be variable 5400-7200RPM , to better sell these drives, until reviews on the web proved otherwise. Marketing bullshit. Afaik Seagate sell their LP drives with constant 5900RPM information.

As MrSpadge said above, WD didn't lie. They just said the RPM of their Green Power drives was somewhere between 5,400rpm and 7,200rpm. Some people then (wrongly) assumed they were variable RPM drives, despite this being very difficult to do and no manufacturer has made a variable RPM drive for at least 15 years.

Edited by Ace

Share this post


Link to post
Share on other sites

Getting off topic here, but variable speed drives would be pretty sweet, especially for notebooks. Make the speed part of the power profile of the notebook, nice and slow when on battery to eek out a few more minutes and bright and peppy when plugged in.

Share this post


Link to post
Share on other sites

Yes, it sounds like a good idea. I think the main problem is that the head flys over the platters due to aerodynamic buoyancy, i.e. the head design has to match the speed of the air flow, which is related to the rpm. IMO a more practical solution would be hybrid drives, which for some reason haven't taken off yet. Imagine a platter based drive with 30 GB of internal flash at current SSD speeds. The HDD could be powered off most of the time and the flash would buffer most reads and writes, until it's time to wake the HDD up.

MrS

Share this post


Link to post
Share on other sites

Seagate tried hybrid drives if anyone remembers, performance was a disaster. Firmware to implement that has to be extremely complex, plus managing what's on the flash vs. what's on the disk is always going to be hugely hampered by the very slow speed/access time of the mechanical half of the disk.

Variable RPM drives do exist but as pointed out, they've only got one active speed, the reduce speed levels the heads aren't even over the disks inside...

Share this post


Link to post
Share on other sites

Nice read Tux

I too am a linux user, the 4K sector size articles I read had me all excited. Looking at those posts on your link have me worried ...

I had to email wd to find out the platter density on 2 drives I had. The tech support was fast and helpful, but i'd also like to see them clearly print numbers somewhere on the label indicating the platter count/density.

Also it's a shame wd puts so much emphasis on green drives. green-blue-black just seems like they are trying to do too much to me. Why not only green and black, with sizes for each released on the same date. Not green versions coming out a few months earlier, like they are doing now :(

Edited by soulkeeper

Share this post


Link to post
Share on other sites

Well. It is even worse. You can configure behavior of standard HDD using standard ata commands. But that is not case of Western Digital!

You can not configure Advanced Power Management using standard ATA command, you have to use proprietary MS-DOS utility WDIDLE3.EXE:

hdparm -B 255 /dev/sda

/dev/sda:
setting Advanced Power Management level to disabled
HDIO_DRIVE_CMD failed: Input/output error
HDIO_DRIVE_CMD(identify) failed: Input/output error

http://community.wdc.com/t5/Desktop/Green-Caviar-High-Load-Cycle-Cout-after-short-operation-time/m-p/18192

You can not configure Error Recovery Control using standard ATA command, you have to use proprietary MS-DOS utility WDTLER.EXE:

http://www.csc.liv.ac.uk/~greg/projects/erc/

As far as I know, there are not such problems with Seagate or Samsung Drives. They use standard ata commands.

Share this post


Link to post
Share on other sites

One interesting information about WD drives design:

Western Digital's construction makes drives particularly vulnerable to shocks and pressure. Unlike other manufacturers, WD does not secure the hard drive axle with a separate screw to the drive cover. Because of this, pressure exerted on the housing or cover can shift the axle, resulting in it changing its angle, and then damaging the platters. The axle's attachment to the cover is another reason. If the cover is moved, the engine may be blocked. Except for this vulnerability, though, WD hard drives are mechanically and electronically reliable.

source: THW

Share this post


Link to post
Share on other sites

Actually, some of WD's drive designs use what's called a rotating shaft motor while others use one with a fixed shaft. The RE line has a fixed shaft, for example, mainly to help compensate for gyration effects from multiple disks.

Any properly designed drive will not have a significant "shift in angle" for the motor shaft by simple pressure. That paragraph from Tom's Hardware makes no sense, as it's implying that having a screw leads to problems and that not having one has problems. Every HDD design out there does one or the other.

Share this post


Link to post
Share on other sites

WD is as trustful as anyone else out there.

The higher end WD drives do secure the shaft at both ends, I think they call it StabilTrack or somesuch - I am looking at an RE4 drive right now that has a screw going through the top cover into the shaft.

One complaint I do have is that WD drives don't seem to be as well balanced as other vendors (causing more rotational vibration), but I have not seen any problems due to this.

Share this post


Link to post
Share on other sites

My objections against WD were different ;)

- they lie about rotational speed. "Intelipower" does not mean variable speed as they reportedly claimed some time ago. Today they officially claim Intellipower is: "A fine-tuned balance of spin speed, transfer rate and caching algorithms designed to deliver both significant power savings and solid performance."

source

This is lie. All Caviar Green drives have fixed rotational speed at 54OORPM. It can be easily proved.

- they intentionally mislead customer with their labeling scheme. There can be different drives using different platters under the same model name. Drives can be recognized only by revision numbers, but they constantly refuse to release these information publicly.

- they probably lie about load_cycle_count number in SMART table, instead of solving this "problem".

- They use proprietary commands to change the drive behavior instead of using standard ata commands.

Other HDD manufacturers are more helpful to their customers. ;)

Edited by StorageTux

Share this post


Link to post
Share on other sites

My objections against WD were different ;)

- they lie about rotational speed. "Intelipower" does not mean variable speed as they reportedly claimed some time ago. Today they officially claim Intellipower is: "A fine-tuned balance of spin speed, transfer rate and caching algorithms designed to deliver both significant power savings and solid performance."

source

This is lie. All Caviar Green drives have fixed rotational speed at 54OORPM. It can be easily proved.

I don't get your point. They say that it's a fine tuned balance. Not that the spindle speed is varying. And they have, to my knowledge, never mentioned that it is. Back in the day when the green drives where new and this was a common belief I really doubted it and searched for any claim where WD had said this and didn't find any. What they did was not to specify exactly which spindle speed it used. This was obviously since they would have gotten a bad reputation for being about the only modern 3.5" drive (at the time) with less than 7200 RPM.

This has already been stated in the thread...

- they intentionally mislead customer with their labeling scheme. There can be different drives using different platters under the same model name. Drives can be recognized only by revision numbers, but they constantly refuse to release these information publicly.

Yes thats a shame. I wouldn't say intentionally mislead though.

I'd guess that they do this so that they can phase out the old drives and replace them with new more easily. But anyway, I agree. That kind of information should be easy to obtain.

- they probably lie about load_cycle_count number in SMART table, instead of solving this "problem".

Probably? In your first post you claim it as a fact. Uncool.

Got anything to back it up with?

- They use proprietary commands to change the drive behavior instead of using standard ata commands.

Is there even a standard ATA command that does what WDTLER.EXE does?

- WD brought in shiny new models with 4kB sector technology, which is great. But, the drive firmware untruly reports it has 512B sectors. This leads to problems OS is not capable to recognize sector size correctly, use 512B as a default and start first partition on LBA63 - in case of old MS-DOS partition table, or on LBA34 - in case of GPT table which I use on all of my hard drives. Neither LBA63 or LBA34 are not the correct place to start with 4kB sector. It should be LBA64 or LBA36 in this case. As a Linux user, I can use GNU Parted and create my first partition manually on LBA i manually choose. But this is not solution, but workaround.

As does all other manufacturers of 4KB drives...

Other HDD manufacturers are more helpful to their customers. ;)

You seem to have something against WD. Instead of focusing on one brand you could extend your rant to include basicly all other manufacturers as well, but that wouldn't be as fun or what?

Edited by tjoff

Share this post


Link to post
Share on other sites

Probably? In your first post you claim it as a fact. Uncool.

Got anything to back it up with?

Yes I have. A had some Caviar Green Drives in the past. It was 320GB/Platter generation. Load_cycle_count affair started long before those drives, so I thought he problem had been fixed. I checked my drives very often and it was nothing wrong with load_cycle_count. But I realized very soon, all my Caviar Green drives silently click-click periodically every 8 seconds. Load_cycle_count was start_stop_count +2 and never changed individually.

So. I used WDIDLE3 utility, and modified "head park" interval to maximum. Guess what... drives stopped periodically clicking.

I have not any direct evidence, but for me it is obvious, WD digital still park the heads, but not report it in smart table to get rid of wondering customers. That's even worse, because the drives behaves the same way and in addition SMART table lies to you.

Is there even a standard ATA command that does what WDTLER.EXE does?

I suppose it is. It is possible to stop head parking behavior of almost every HDD using hdparm :

"hdparm -B 255"

to disable it, or:

"hdparm -B 254"

to use least aggressive settings.

But Caviar Green drives completely ignore standard APM commands and report Error every time you try it.

WD also abandoned ERC command for sure somewhere between two revisions of WD1001FALS although still claim ATA-8 compliance.

As does all other manufacturers of 4KB drives...

Yes. That is sad. But at the time of my first post, WD was the only manufacturer with 4kB sector drives in the consumer market. I still hope, other manufacturers will bring a better solution (for example jumper to switch reported sector size) or produce drives with correct size 512B or 4kB too.

You seem to have something against WD. Instead of focusing on one brand you could extend your rant to include basicly all other manufacturers as well, but that wouldn't be as fun or what?

As a WD fan I used their drives almost exclusively. A have only few experience with Seagate, and even less experience with Samsung. So I can criticize only WD.

Edited by StorageTux

Share this post


Link to post
Share on other sites

I bought $1,600 worth of 2TB Seagate Constellation ES drives over the WD RE4's due to the constellation having PowerChoice which would allow a low RPM idle which is supported by my Adaptec RAID controllers. When I tried to enable it, it didn't work. I called Seagate, they said that it is not supported on the SATA drives even though it is on the data sheet and suggested I work with my retailer to return the drives.

Seagate also has a proprietary utility to change parameters on their enterprise drives.

I use a lot of hitachi drives, they also have their own proprietary utility to adjust what should be standard parameters.

Edited by cbrworm

Share this post


Link to post
Share on other sites

I bought $1,600 worth of 2TB Seagate Constellation ES drives over the WD RE4's due to the constellation having PowerChoice which would allow a low RPM idle which is supported by my Adaptec RAID controllers. When I tried to enable it, it didn't work. I called Seagate, they said that it is not supported on the SATA drives even though it is on the data sheet and suggested I work with my retailer to return the drives

Strange they claim it support it on SATA. :-(

http://www.seagate.com/docs/pdf/whitepaper/tp608_powerchoice_tech_provides.pdf

How did they explain it does not work, when they claimed it works on SATA drives in tech-spec sheet?

It does not work anyway or it just do not work on your HW RAID controller?

Edited by StorageTux

Share this post


Link to post
Share on other sites

Strange they claim it support it on SATA. :-(

http://www.seagate.com/docs/pdf/whitepaper/tp608_powerchoice_tech_provides.pdf

How did they explain it does not work, when they claimed it works on SATA drives in tech-spec sheet?

It does not work anyway or it just do not work on your HW RAID controller?

My Adaptec SAS controller does support low RPM idle modes, Seagate said it is not supported on the SATA version of the drive despite it being on the spec sheet. They said it was a marketing error and appologized. At the time all the marketing material said both versions of the Constellation ES drives supported it, now some of the marketing material says SAS only.

I would have bought the SAS drives except fot the fear of needing to use them at some point on a SATA controller. Had I known that the low RPM worked only on SAS, I would have bought them instead.

Edited by cbrworm

Share this post


Link to post
Share on other sites

Well. You can let the drives to go into sleep mode, if such a solution is acceptable for you. It should work on all drives. Just beware of Caviar Green line. Some of the low-RPM WD drives do not support advanced power management commands. :(

Edited by StorageTux

Share this post


Link to post
Share on other sites

They do go to sleep, my point was that they are marketed as variable RPM drives (like the WD Green were), and they are not.

There are also special seagate tools required to enable and disable features that should be available through the standard ATA command set.

WD is not unique in misleading marketing and vendor specific utilities.

Share this post


Link to post
Share on other sites

I think they are about equal. I use lots of WD, Seagate and Hitachi drives. Right now WD(black) are my favorite for desktop drives, Seagate and Hitachi for enterprise drives.

All the mfg.'s have screwed up at some point or another. Ultimately they get the job done.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now