CityK

WD 74GB Raptor: Desktop in its crosshairs

Recommended Posts

Looking at the preliminary SR benchmark results for the new, bigger Raptor, I find it pretty hard to believe that its intended audience is geared for anything else but the desktop gamer/enthusiast segement. Although this comparison is a little crowded, I think you can get a clear picture of in which ways the new Raptor's firmware has been tuned.

CK

Share this post


Link to post
Share on other sites

That looks GREAT! FDBs with noticably reduced noise paired together with much higher capacity and at least clearly measureable more performance.

That's the point at which I'm willing to bite ... during my upgrade next summer (when no better alternatives emerge).

MrS

Share this post


Link to post
Share on other sites
Guest Eugene

These results have been drawn from a pre-release engineering sample (supplied directly by WD rather than from a third-party reseller as was the case with the first Raptor) running off of our Testbed's standard Promise SATA 150TX4 controller. As a result, command queueing's effects aren't being felt. I'd expect to see much better scaling of performance in the IOMeter suite... the effect on the Desktop DriveMarks, on the other hand, is up in the air ;).

Full commentary will accompany a preview that we'll get up early next week.

Share this post


Link to post
Share on other sites

If WD doesn't pull the same stunt they did with the 360GD (putting out a higher performance version for reviewers than retail), then I am all over this drive. Well, as soon as I can afford it. But I'm going to wait until the Hypermicro sample to get too excited. But those desktop performance numbers look incredible right now. >70MB/s STR is quite a nice little improvement :)

Share this post


Link to post
Share on other sites
These results have been drawn from a pre-release engineering sample (supplied directly by WD rather than from a third-party reseller as was the case with the first Raptor) running off of our Testbed's standard Promise SATA 150TX4 controller. As a result, command queueing's effects aren't being felt.

Doh! There goes my analysis.

Also, the sheer fact that this drive is supplied directly by WD sets off alarm bells for me too. It looks like they tweaked the zones pretty good too (72MB -> 54MB wow!). I really thought what they did the first time around with Raptor production drives was a little sleezy. There is no way of knowing if this will be repeated here with the newer edition.

I'd expect to see much better scaling of performance in the IOMeter suite...
Should be interesting to see the results.
the effect on the Desktop DriveMarks, on the other hand, is up in the air ;).
Judging by the queue depths seen in the TB3 article:

Office - little if any

Highend - again, little if any

Boot - may see some improvement here

Games - err, no individual traces to go by. Combination of the above three? Then little effect here too.

Share this post


Link to post
Share on other sites

Noticed the noise? Now that's impressive for a high-performance desktop drive, moreso considering it's a 10K rpm.

I hope the high-frequency noise pitch that was common to the first gen. Raptor has vanished for the secong generation model.

Share this post


Link to post
Share on other sites

Hurry up, WD! And get the 36 GB version into the Canadian retail channel ASAP! I need a new OS/apps drive as it appears my Adaptec controller is getting flaky and I would like to replace my trusty Atlas 10k3 with a quieter drive.

Share this post


Link to post
Share on other sites

If the 74GB is so much better, then why are the specs practically identical on the WD website? (Noise, transfer rate, etc.)

Or have they changed today?

Did they just copy the info and modify the obvious parts?

:angry:

And regarding the idle "whine" on the 36GB: I'd rather listen to my old 1st gen cheetah! :blink:

Share this post


Link to post
Share on other sites
I really thought what they did the first time around with Raptor production drives was a little sleezy.

eugene got the first drive from hm, not from wd. what they did was 'change horses mid-stream', not cherry-pick a review drive for eugene. if it weren't for sr's quick attention to the drive, it's possible we'd all be none the wiser.

Share this post


Link to post
Share on other sites
If the 74GB is so much better, then why are the specs practically identical on the WD website?  (Noise, transfer rate, etc.)

Or have they changed today?

Did they just copy the info and modify the obvious parts?

It wouldn't be the first time.

Sometimes, the info posted on WD's web site seems so bad and inacurate that one has to wonder if WD hasn't put some under paid folks suffering of mongolism in charge of it. And that's more insulting for people suffering mongolism than for WD's web master.

Share this post


Link to post
Share on other sites

Well, it seems that I will have to hold on before buying my RAID controller... I wonder if having 2 Cheetah X15 36LP 36.7gb in RAID 0 will be faster than a single Raptor 74gb. I currently have 2 Cheetah X15 36LP and all I need is an Adaptec 2100S SCSI RAID card. Heck, the Raptor 74gb is faster on a majority of benchmarks when put against the Cheetah 15K.3. Ouch. The only area where the SCSI drives are superior is the access time. Anyway, I have seen my friend's setup which consist of 2 Raptor in RAID 0 and I did not like the "feel". My system seems to be more responsive so I wonder if going for a 74gb will do better.

Parabellum

Share this post


Link to post
Share on other sites
I really thought what they did the first time around with Raptor production drives was a little sleezy.

eugene got the first drive from hm, not from wd. what they did was 'change horses mid-stream', not cherry-pick a review drive for eugene. if it weren't for sr's quick attention to the drive, it's possible we'd all be none the wiser.

Which might be true if not for the fact that other sites (Anand comes to mind) had previews of the drive as well at the same time. From Anand's preview:

"We received the drive after Western Digital had stated they had begun shipping the Raptor drives to distributors, so we do believe that performance won't change significantly between what we have here and retail products."

So basically, WD knowingly allowed sites to post benchmarks of a drive they knew was faster than the actual retail drive. Pretty shady in my opinion.

Share this post


Link to post
Share on other sites

No, I take that back. The prerelease drives were from Hypermicro. But those were the poor performing models. The initial retail drive reviewed by SR, was provided by WD (according to the review) and showed inflated performance. Later there was a 3rd article about the Raptor that had the proper benchmarks. So it didn't really have anything to do with SR being on the ball, WD gave them the drive.

Share this post


Link to post
Share on other sites
eugene got the first drive from hm, not from wd
Yes, but it was the second, non beta, drive shenanigans that I found sleezy.
what they did was 'change horses mid-stream', not cherry-pick a review drive for eugene
I beg to differ about WD not cherry picking review drives. Let me illustrate my point:

Previews of Beta Raptors

Storagereviews

The drive tested for this review is a beta unit provided by longtime SR sponsor HyperMicro.com rather than Western Digital itself.

Anandtechs

A special thanks to Hypermicro  for providing us with a number of the hard drives and controller cards for this review.

Production Raptor Reviews

Shortly after we published our preview, we were contacted by a number of other OEMs and system integrators that had been playing around with similar beta samples of the Raptor drive. Everyone that contacted us had similar disappointing performance results
Words echoed by Eugene:
Initial tests run on a pre-production sample supplied by long-time sponsor HyperMicro yielded disappointing results.

But at this point we should take pause to take into consideration what Eugene also wrote:

It is important to note that the market for the Raptor is primarily the entry- and mid-level server segments and not the enthusiast desktop sector.
This has been the stance repeated on every other review site.

But then we see that WD was rather quick to respond to such negative reviews from enthusiast hardware review sites. Anandtech writes:

Western Digital contacted us to arrange for a new sample, with write caching enabled, to be provided. They confirmed what we had heard, as well as mentioned that there were other optimizations present in the shipping firmware that weren't there before
Indeed, it now seems apparent that WD cared very much what the enthusiast market thought about the Raptor. This is no clearly put then by Eugene's own words:
Immediately afterwards, however, Western Digital was eager to get a production-class unit in our hands.

Therefore, it is my contention, that the drives reviewed in the following articles, were very much indeed hand picked (if you will) by WD:

http://www.storagereview.com/articles/2003...WD360GD_sp.html

http://www.anandtech.com/printarticle.html?i=1799

http://www.xbitlabs.com/articles/storage/p.../wd-raptor.html

Take careful note that all show ~63->37 STR. STR sells. The RAID craze is proof in the pudding.

It wasn't for several months (during which time no one else could replicate these same results) until the "official" story broke that WD had changed a few things here and there along the way in their Raptor. Done mid stride or not, WD's silence or failure to annouce the change in the first place was enough in itself to sour my opinion. Why wasn't WD eager to ruch the 3rd iteration to hardware review sites? Simple. It wasn't in their best interests.

----

Oh man, I messed up a quote somewhere and it literally took my forever to get the post straighted out. :blink:

Share this post


Link to post
Share on other sites

Eugene wrote,

These results have been drawn from a pre-release engineering sample (supplied directly by WD rather than from a third-party reseller as was the case with the first Raptor) running off of our Testbed's standard Promise SATA 150TX4 controller. As a result, command queueing's effects aren't being felt.

Hi Eugene. Sorry, but I don't understand this statement. Do you mean that using a PCI SATA solution disables command queueing? Or is just that particular model (Promise 150TX4)? Or am I just totally confused and all you mean is that because it's a pre-release model that the command queuing isn't "active" yet and only will be on the final production version?

Share this post


Link to post
Share on other sites

The drive is nice, what about its price though...? <_<

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