RaptorX

[Bozo]

http://www.wdraptorx.com/


Bozo :mrgrn:

 

[MaxBurn]

I hope they release a cheaper version for normal people. Or hopefully my 15k.4's get here soon!

 

[Tannin]

Bahhh .... They need to stop taking the medication. Seriously, what planet are they on? Seagate had clear, see-through hard drives routinely available to dealers for display purposes years ago. I mean lots of years ago. I forget exactly when they started doing this, possibly before my time, but they were readilily available back around the time that 500 and 850MB drives were the going thing.

 

[CougTek]

There were computers before your time?

500MB hard drives were popular around 1994-1995 IIRC.

 

[Tannin]

There were computers before your time?

Sure there were, only we didn't use silicon to make the chips back then, it was granite, or sometimes flint. Oh, and our montors weren't CRT or LCD, they were SAC (sandstone and chisel). Man, those old SAC monitors were great! 100% flicker-free, and (of course) rock solid.

 

[Will Rickards]

The window would be cool.
But the performance of this drive is also outstanding if it is the same drive that SR reviewed.
If I hadn't already invested in SCSI and a Fujitsu 15K drive I would be considering this drive.

 

[Buck]

Bahhh .... They need to stop taking the medication. Seriously, what planet are they on? Seagate had clear, see-through hard drives routinely available to dealers for display purposes years ago. I mean lots of years ago. I forget exactly when they started doing this, possibly before my time, but they were readilily available back around the time that 500 and 850MB drives were the going thing.

WD had them available back then for dealers too. This new product isn't for dealers.

 

[Gilbo]

I'll just link the most interesting page of SR's review for the lazier among us. I think Will's 'astounding' is not a bad description.

 

[Mercutio]

And I think it's overhypered bullshit for a company that advertises on SR.

But hey, if someone wants to buy a failure-prone drive from a company known to make particularly failure-prone hardware, who am I do anything but point and laugh.

 

[sechs]

It's a drive for the RAID 0 crowd.

 

[Mercutio]

I can't even understand how WD is still in business. I really can't. I know there were people who thought the same thing about IBM during the reign of the deathstar, but that was IBM. WD has one core product and they can't even do that one thing right.

 

[Tannin]

Also, IBM had a long and proud history of flawless execution, and the Deathstar line was soon restored to something approaching its former glory by the perfectly decent replacement models. The Deathstar 75GXP, in other words, was one bad drive out of a very long series of drives which varied from very good to truly outstanding.

WD, on the other hand, .... well, let's put it this way: never mind the latest product, just tell me the last time you remember a Western Digital drive that you really, really trusted. Off the top of my head, the most recent actual trustworthy WD drive I can remember was the Expert - which was, of course, a rebadged IBM Deskstar.

QED

 

[mubs]

I have an 80GB and 120GB WD purchased in 1992 and 1993 that are still working fine. One swallaw does not a summer make, but you said "just tell me the last time you remember a Western Digital drive that you really, really trusted." These have been moved to another PC now, but for ~ 2.5 years these were the only drives in my main PC, held all my mail and data, and I never had a problem.

Actually I've never had a drive die on me. I still have working 80MB Conners somewhere in a box. And working Maxtor 54nn 540MB drives from circa 1994, again in a box. They're slower than half-frozen molasses, but they work. Someday I'll take them apart for them thar magnets!

 

[Mercutio]

You mean MB, right mubs?

Actually, WD's SCSI drives were OK. Too bad they quit making them six years ago.

 

[CityK]

And I think it's overhypered bullshit for a company that advertises on SR.

My direct and indirect thoughts on the SR review are the following:

• I thought it was a much more thorough review then what is usually presented. I like more thorough reviews. Much of what was written I liked and thought was well presented.
• However, here's the flip side of the coin to the previous point: Why was so this review so much more thorough? it smacks me that there is only one reason -- its a WD. Now, if Eugene continues to pay that much attention to all future reviews and articles, I'll be quite pleased and less concerned with meandering thoughts of less then ideal ethics. But sadly, in future, I somehow doubt we're going to see a history of the Barracudas or Cheetahs or anything along those effects. Back to boring is my call.
• Next, I felt like I should have been handed a set of pom-poms before I began reading -- Personally, I could use a little less Rah-Rah-Gooooooo-Raptors! language painted within the review. It stinks, as Merc stated, of overhyped marketing PR bullshit.
• Talking about WD rah-rah, I see that SR did "yet another look at the second generation Raptor" a couple of weeks ago. Consider:

  As the months rolled by, the Raptor continued to hold its own against the latest SATA and SCSI drives to hit the channel.

  Nonesense! A pedestrian Samsung drive fell within 5% of the Raptor in two of SR's own desktop benchmarks. But nary a word on this. No banners unfurrowed, and no band struck up the chorus.
  

  Further, like most drives, the WD740GD has quietly received tweaks and revisions here and there. The latest iteration, the WD740GD-00FLC0, boasts improvements of up to 14% in its single-user scores.

  A firmware update, and oh look ! - those amazing Raptors are at it again! And this begs the question: if most drives are recieving tweaks along the way, why not pester all manufacturres for new review samples and revisit their performance too? Did the fact that the new Raptor was to appear in a couple of weeks have anything to do with it? Perhaps? Eh? Whats that? The WD galley's drummer has increased the tempo! Bum-bum-bum-bum-bum-bum-bum.....
• Now then, does my disdain for the overhyping extend to the drive itself? No. I think Eugene's hyping hurts his appearence of journalistic integrity, however, I personally don't believe he has ever fabricated anything. Given WD preferential marketing treatment, yes;, lied about performance, no. Indeed, I actually think it will be a very good performing drive (though its reliability will be another question altogether). I would even actually consider a single platter version of it (despite my full awareness of WD's sketchy reliabilty), if such a product is made available, and depending upon where it is priced. This is the first time I actually think there might be sufficient performance increase (meaning: the former Raptors didn't move/impress me enough to part with some coinage) to make such a risk worthwhile. On the otherhand, this two platter drive will not find a home with me -- there is point on my utility curve at which my intolerence for throwing money away diverges from any number of performace metrics (performance per dollar, GB per dollar, etc etc), and this drive is priced well beyond that threshold. So while (as Sechs so rightly discribed it) the RAID0 club is busy snatching these up (err, if you prefer reality change that last bit to "parting with a good deal of money"), I'll be happily trudging along with some much more pedestrian drives.
• Lastly, I was happy to see little in the way of pontification in the review (or at least I don't recalll being struck by any). You see, I have far too often found Eugene to present strawman arguments to suit his need for forwarding the legitimacy of SR's test suite. For example (while not related to pushing the legitimacy of SR's test suite), consider the strawman presented in the "first" second generation raptor article:

  As users got their hands on retail drives, however, they quickly noticed that their shiny new hotrods topped out with 57 MB/sec transfer rates rather than the 63 MB/sec that SR's WD-supplied sample boasted. Sequential transfer rates (STR), unfortunately, are the easiest HD performance metric to measure and also the least indicative of how the drive will perform under real-world situations. It was obvious from the lower STR figures that the shipping Raptors incorporated the more conservative zone-layout found in our original sample rather than the one used in the following unit.
  
  Conspiracy theories ensued in the community. Eventually, a flawed consensus was reached: If the shipping Raptors featured the lower transfer rates exhibited in SR's first sample, then the write-caching, firmware, and performance of shipping units must also be similar to that of the first drive's. This logic, of course, is fallacious. Zone configuration (and thus transfer rates) is independent of write caching and firmware.

  Newsflash -- I certainly never came to that concensus. And I also know a good number of other posters who never exhibited such fallacy in their logic either. In actuality, this is kind of insulting to the posters who hung around and tried to maintain some sort of order and sainity amongst a population infected with the likes of Raid0 mania and NCQitis.
• As for the test suite itself, I still uphold that its ability to provide inter-drive compariablitiy is in doubt. I raised a number of points of issue in the forums last year, and I've seen no tackling of them yet. I certainly wasn't the only one to spot these incongruities either. On the whole, I like what Eugene is doing (the actual research that is, as opposed to being a part of WD's bolster club....but hey, if you've got to pay the bills, then I guess you've got to pay the bills one way or another), but I'm afraid they aren't hitting their mark scientifically.

Anyways, as I know he views posts here, I hope he will take what I have to say into consideration.

 

[CityK]

Ugh. That's really hard to read with the bullet points squished so close together like that. Sorry, about that. I thought there may have been more white space left in between.

 

[Mercutio]

We forgive you. That was an excellent post regardless.

 

[CougTek]

Anyways, as I know he views posts here, I hope he will take what I have to say into consideration.

Really? I thought he viewed us as a bunch of deserters.

 

[time]

Excellent post, CityK!

Eugene has frequently used his proprietary benchmarks to claim that WD drives are faster. The JB series topped his charts for ages, despite inferior mechanical specs to some competitors. Hence his mantra that firmware is everything while seeks and STR don't matter (much).

I wonder if the drive manfacturers - particularly WD - have access to his benchmarking tools?

Right now, he's banging the drum about most desktop access being localized. Given the huge increases in areal density, that's hardly surprising. The logical extension of this argument is that there is no point in benchmarking drives, so he may as well shut up shop now!

Instead of benchmarking operations that are already fast (or do not involve waiting for the drive) and are therefore unnoticed by the user, how about measuring the slower operations that users do notice? If a system takes 100mS instead of 300mS to respond to a user event, in most situations the user won't care at all. The difference between two seconds and six seconds, however, may be critical.

 

[Tannin]

Instead of benchmarking operations that are already fast and are therefore unnoticed by the user, how about measuring the slower operations that users do notice?

Can we write this in 72 point flashing type? Engrave it on stone tablets? Get it moulded up and brand it with a hot iron into the thick heads of the jumped-up, self-appointed psudeo-scientists over at SR?

I first raised this point - ths vital, absolutely fundamental point - more years ago than I care to remember. (Not that I'm claiming to have invented it, of course, I was merely one of the more prominent and outspoken participants in the debate.) To my knowledge, no-one has ever responded to it in any meaningful way, other than to echo it with variations, as Time did so eloquently just above. The techno-wankers at SR and other places of that ilk simply ignore it, either because it bursts their FantasyMark bubble, or else because they have not the wit to understand it in the first place. On the whole, I prefer the RAID zero crowd: at least their numerical self-delusion is honest and transparent. Sort of refreshing, in a backwards sort of way.

 

[mubs]

I have an 80GB and 120GB WD purchased in 1992 and 1993...

You mean MB, right mubs?

GB. That should have read 2002 and 2003.

 

[Tannin]

Sorry Mubs, I was ambiguous. I wasn't talking about an individual drive as a single unit, I was saying I can't remember the last time WD made a drive model that I actually trusted, not since their rebadged Deskstars, quite a few years ago now. Matbe their current stuff is good. I wouldn't know. Don't aim to find out either: the last couple of hundred Western Digital drives we sold were crap, and their "service" was appalling.

 

[ddrueding]

In this vein of practical speed, we should put together a list of SCSI cards and how much of a delay they introduce into the POST sequence. Or which motherboards pass you off to the first boot device sooner. Or how to code firefox to be pre-cached during the startup sequence...

 

[CityK]

I first raised this point - ths vital, absolutely fundamental point - more years ago than I care to remember. (Not that I'm claiming to have invented it, of course, I was merely one of the more prominent and outspoken participants in the debate.)

Actually Tannin, I had you in mind when I wrote about the Strawman argument, as I specifically remember you writing something along the lines of "depicting you as a strawman". But there are, no doubt, perhaps several other examples I could have drawn from memory...the fact remains: its pretty easy to blow them down when you line them up yourself.

 

[Gilbo]

In this vein of practical speed, we should put together a list of SCSI cards and how much of a delay they introduce into the POST sequence. Or which motherboards pass you off to the first boot device sooner. Or how to code firefox to be pre-cached during the startup sequence...

Well said.

Which leads me to the fact that I honestly can't think of anything that I do for which hard drive performance is even a concern.
1. Since I moved to Linux, my applications don't swap out just because I alt-tab away from them.
2. I don't care how long file copies and such take. Once I start them I forget about them. They don't need my interaction. Since nearly all of my data is stored on a couple home file servers, the load never affects my work computer.
3. I load all my commonly used applications at boot. They're always in RAM, running when I need them. I never have to wait for OpenOffice to startup, or Firefox, or Evolution. (Linux also seems to load it's equivalent applications much quicker than Windows anyway --I have no idea why I don't even prelink them. Maybe because they're less bloated. Maybe because of the filesystems. Who knows...)

Can anyone guess what the single best performance upgrade I've ever given any system I own has been? Setting up a a caching/ad-filtering proxy server for my home network using Squid and Privoxy. It's likely the largest, most noticeable, most useful performance boost I have ever experienced in all my years of computing. A close second was upgrading my laptop to Linux from Windows. Third was probably moving from 512MB to 1.5GB of RAM in the same laptop.