SPCR

Fantastic.

This review came out even faster than I could have hoped. Thanks a lot!

These drives looks fantastic, even with the 1 TB's caveats. But if it's not a big leap in price or logic to the 2 TB and its features, I think that's where I'll be looking next for sure. The efficiency and performance (I include acoustic performance in this) figures are very compelling to say the least.

I can't help but wonder where this leaves the Green and especially the Blue. If we can get this kind of performance out of the "eco" drives, the only reason to buy a Blue would be for the legacy interface support. I guess that's been their role for a while now anyway. The Greens, even with their slight acoustic advantage, will probably have to compete with price, which is so far the case, at least over here: 2 TB Green is 114-ish, 2 TB Red 140-ish. Maybe 5-year warranty Greens?

The Blacks will of course have their limited performance advantage and a 5-year warranty, but the Red is not far behind, and SSDs are so far ahead.

Now only if the 3TB would have pre-flood prices, that is ~120e, i would actually start replacing my greens with these reds. But at ~180e for 3TB model, i don't think so.

Thanks for the review. The 3TB looks to be a winner for a big honking data drive. The 1TB performance seems out of line. I wonder if WD will send you another for re-test?

And if they'd include a 2TB with it!

What are the contraindications to use the Reds as INTERNAL drives in a DESKTOP?

lb_felipe wrote:What are the contraindications to use the Reds as INTERNAL drives in a DESKTOP?

None, afai can see. I'm actually planning to replace a 2TB WD Green in the newer of my 2 desktops with one of these 3TB Red samples. The older Green's seek noise is starting to bother me. I think I will see a little boost in speed -- and less seek noise.

lb_felipe wrote:What are the contraindications to use the Reds as INTERNAL drives in a DESKTOP?

TLER is the only one.

What does TLER means ? Standard desktop drives tries to recover data from hard to read/unreadable sectors until it succeeds. TLER enabled drives tries to recover data only for few seconds, after that it gives up and notifies you that error happened.

Did I miss the explanation as to why the 1TB performs so much worse?

I bought 2x1TB Reds practically on release day because a) my storage needs are extremely modest and b) I wanted redundancy, so although a 3TB drive would have offered significantly more storage at less cost I wanted duplication across both drives. They both went in my new HP N40L using StableBit DrivePool on WHS 2011 for file duplication.

I'm more than happy with my purchase and have no regrets after reading this article as I bought them for their low power consumption and 24/7 purpose, but just wondering why the significant performance drop off?

Jordan wrote:why the significant performance drop off?

It's hard to tell considering Western Digital hides performance-related information and that SPCR's tests are pretty opaque as well.

The relatively small differences detected by HDTune would lead me to believe that there's no major difference between the drives other than firmware and capacity.
I don't know how the controller works but the larger capacity of the 3T drive means that there's potential for larger short-stroking effects which may or may not account for the bulk of the performance differences we see in SPCR's tests.

HFat wrote:I don't know how the controller works but the larger capacity of the 3T drive means that there's potential for larger short-stroking effects which may or may not account for the bulk of the performance differences we see in SPCR's tests.

Except SPCR's test methodology is to make a 50GB partition on a clean drive and all the OS/apps are there. So, they're all tested "short stroked".

Obviously. You'd have a similar effect with any sane methodology anyway (well, not *any*: I guess the methodology could be designed not to show short-stroking effects but it would have to be more complex).
But the effect would be different if the controller used all the platters of the high-capacity drive as opposed to just the one (or are there two?) of the low-capacity drive.

faugusztin wrote:TLER is the only one.

It's ridiculous. That should just be a firmware or SATA option.

Olaf van der Spek wrote:

faugusztin wrote:TLER is the only one.

It's ridiculous. That should just be a firmware or SATA option.

It would make life a lot easier. That's always been a stumbling block for me to use the RE edition drives.

TLER, looks like another way from the marketing guys to keep shareholders happy. As Olaf said, it should be able to be turned on/off via firmware.

I personally don't really mind TLER on or off. With TLER off (Green drives) the drive can lock itself up with many tries to recover data even for minutes. With TLER on (Red drives), it is limited to seconds. But from my standpoint, once drives needs to use special techniques to read the data because the hard drive sector is decaying/damaged, i usually put the drive through a Data Lifeguard Diagnostic test and if it fails, it goes to RMA.

Really nice review, thanks for sharing it.

Once i upgrade my server to 64 bit OS and my HBAs, im probably going to go with WD Reds 3tb.

Regarding performance, I can't see why the 1TB is worse. Same density and same platters, right? Just 1 platter for the 1TB and 3 platters for the 3TB? From what I understand about how the SPCR performance testing was done, I think it was a bit limiting. It shows and concludes the 3TB is pretty decent, keeping up or even surpassing drives like the WD black. OK sure if you only run 1 simple test at a time like boot time, file copy time, etc. Try starting up a few large programs at once for example. The WD black will probably pull ahead.

MikeC wrote:

lb_felipe wrote:What are the contraindications to use the Reds as INTERNAL drives in a DESKTOP?

None, afai can see. I'm actually planning to replace a 2TB WD Green in the newer of my 2 desktops with one of these 3TB Red samples. The older Green's seek noise is starting to bother me. I think I will see a little boost in speed -- and less seek noise.

MikeC, I'm wondering if you've done this swap yet? If so, is it working out well? I'd like to do the same (use a 3TB red) in my desktop, but I've heard discussion that it's not a good idea (due to things like TLER being enabled on these drives... conflicting info on whether or not this can be turned off for desktop use). I think I even saw a note from WD saying these are for NAS use only, and to not use them in desktops.

TLER is something which is relevant only when the hard drive is going bad.

With no TLER, hard drive tries to recover the data in the bad sector and tries to move it to one of it's backup sectors. This can take seconds (if you are lucky), minutes or even hours. And during that time, your hard drive is pretty much stuck rescuing data, so you can't use that hard drive in meantime. In other words, bad situation.

All TLER does is that it limits this recovery time to 7 seconds. If data can be recovered in that time, the data is lost, sector marked as bad.

So, do you really care if your sector is marked bad after minutes or 7 seconds ? I don't.

flyingsherpa wrote:

MikeC wrote:

lb_felipe wrote:What are the contraindications to use the Reds as INTERNAL drives in a DESKTOP?

None, afai can see. I'm actually planning to replace a 2TB WD Green in the newer of my 2 desktops with one of these 3TB Red samples. The older Green's seek noise is starting to bother me. I think I will see a little boost in speed -- and less seek noise.

MikeC, I'm wondering if you've done this swap yet? If so, is it working out well? I'd like to do the same (use a 3TB red) in my desktop, but I've heard discussion that it's not a good idea (due to things like TLER being enabled on these drives... conflicting info on whether or not this can be turned off for desktop use). I think I even saw a note from WD saying these are for NAS use only, and to not use them in desktops.

No I have not done the swap yet... but I really can't see any serious reason for not using it on a desktop drive.

I had some spare time and scoured the WD site up and down. There is no notice or information to suggest that the Red drives were somehow unsuitable for desktop drives. The only notice there is is concerning the use of consumer drives in RAID installations versus RE enterprise drives.

Das_Saunamies wrote:I had some spare time and scoured the WD site up and down. There is no notice or information to suggest that the Red drives were somehow unsuitable for desktop drives. The only notice there is is concerning the use of consumer drives in RAID installations versus RE enterprise drives.

Hmm, ok, good to know. I swear I saw something somewhere though... perhaps it was a customer service email someone posted on another forum where they had asked WD about this. But if it's not on their site anywhere, it's probably not a big deal.

on page 3, it says:

In our application testing, the Red 3TB delivered superb performance edging out the VelociRaptor 600GB, the second faster hard drive we've ever tested. The 1TB model performed as one would expect, faster than the Scorpio Blue, but slower than the 7200RPM Barracuda XT.

i think it should be fastest. a comma after 'performance' might help too. better yet, a semicolon. nobody uses them anymore. maybe:

"...superb performance; it even edged out the VelociRaptor 600GB, the second fastest..."

VeganThug wrote:on page 3, it says:

In our application testing, the Red 3TB delivered superb performance edging out the VelociRaptor 600GB, the second faster hard drive we've ever tested. The 1TB model performed as one would expect, faster than the Scorpio Blue, but slower than the 7200RPM Barracuda XT.

i think it should be fastest. a comma after 'performance' might help too. better yet, a semicolon. nobody uses them anymore. maybe:

"...superb performance; it even edged out the VelociRaptor 600GB, the second fastest..."

You're right, corrected; but semicolons? We use them all the time.

concerning tler, wd's official response is that there will be no performance loss in a non-raid environment. here's the link:
http://wdc.custhelp.com/app/answers/detail/a_id/1397

they don't say anything about an increased chance for data loss due the recovery operation getting timed out (if there is, i bet it's a pretty low probability). they also say that tler cannot be turned off (i read elsewhere that in older drives it can be):
http://wdc.custhelp.com/app/answers/det ... RhbA%3D%3D

amazon has the 3tb red's for $165 as of this writing. i saw the 2tb ones for as low as $119 at newegg and amazon, but they have since raised prices. the funny thing about newegg is that i set a price alert for $140 for the 2tb, but i was never notified of the temporary price drop; i just happened to have one in my cart with the lower price. the 3tb one is tempting, but i think i'll see if they go even lower in the coming weeks.

Just bought the WD RED 2TB for a new build, following the SPCR recommendation. What a great hard drive. I have purchased several drives recommended by SPCR in the past 7 years or so. The WD RED is, hands down, the quietest mechanical hard drive I have ever had the pleasure of not hearing!

It's barely audible over the fans in my system. For reference, the loudest fan in my case is a 12 cm Noctua @ 700 rpm. The seek noise is inaudible in a closed case. I had my previous hard drive (a WD Green) in an enclosure, but for this one, there's no need. I had no idea mechanical drives could be so quiet.

VeganThug wrote:they also say that tler cannot be turned off (i read elsewhere that in older drives it can be):
http://wdc.custhelp.com/app/answers/det ... RhbA%3D%3D

I was surprised to come across this:
http://www.anandtech.com/show/6157/west ... -premium/2

It shows that it can be adjusted or disabled with smartctl.

Also described here:
http://hardforum.com/showthread.php?t=1590200

Ah, but TLER doesn't survive power, as I've just found myself with WD20EFRX. I could have sworn I saw accounts stating that it did. I suppose it could be put in Task Scheduler as a startup task easily enough.

One other thing: for some reason, I don't see the same throughput as measured by HD Tune that the review does and every other screenshot that I've seen from this drive does. Mine starts at about 125 and heads down, while the other start at 150. On a clean install of Win8 with nothing else happening.

I'm certainly on a 6G port (it even says so in the BIOS--it's a Sandy Bridge-era system). I also have an SSD in the system that far exceeds these numbers.

Any ideas?